Torture as Tort: Comparative Perspectives on the Development of Transnational Human Rights Litigation 9781472562340, 9781841130606

Citation preview

Acknowledgements I am grateful to the Social Sciences and Humanities Research Council of Canada for generous funding of two human rights projects, each of which has fed in some way into this project. The Faculty of Law at the University of Toronto provided editorial assistance needed in the final stretches. I was also fortunate to have been able to spend part of the 1999–2000 academic year on leave at the European University Institute, which proved to be both a relaxing and a stimulating venue in which to complete the editing of the contributions to the book. Boris Nevelev provided excellent research assistance. The perceptive and demanding copy-editing of Janesse Leung has helped whip many a chapter into shape. Thanks also to Michelle Everson for her translation work, Ted Tjaden for helping track down materials, Robert Howse for lending his students to this project, Taisier Ali for making Sudan real, and to all the contributors for their openness to editorial feedback and their general good humour. Special thanks are owed to two people. To Marie Chen for being there even when I wasn’t. And to Richard Hart for the enthusiasm and alacrity with which he responded to the proposal to do this book, and for both his patience and efficiency in making sure it came to pass. Finally, as we go to press, allow me to make an odd use of an Acknowledgements section, namely by drawing the reader’s attention to a case that reached the International Court of Justice in December 2000, a case which sees the Democratic Republic of the Congo impleading Belgium as a result of a Belgian investigating magistrate having issued an international warrant for the arrest of the DRC’s (then) Minister of Foreign Affairs for alleged war crimes. Outside the global linguistic spotlight of the English-language proceedings against Pinochet, a former head of state, much of interest has begun to occur in Belgian courts, including this attempt to lay the groundwork for the extradition of a current member of a foreign state’s Cabinet. The reader may accordingly wish to follow developments before the ICJ in Arrest Warrant of 11 April 2000 (Democratic Republic of Congo v. Belgium) on the ICJ’s web site at http:// www.icj-cij.org/icjwww/idocket/iCOBE/iCOBEframe.htm. A pronouncement by the ICJ in the context of an assertion of immunity to one state’s criminal law jurisdiction over a high official of another state will have an immense impact on the course of transnational justice more generally, including that pursued by way of transnational tort litigation. Craig Scott Osgoode Hall Law School, York University, Toronto

Contributors ADAMS, WENDY—Wendy Adams is an assistant professor at the Faculty of Law, University of Western Ontario. She received her LL.B. from the University of Toronto and LL.M. from the University of Michigan. Her principal areas of interest are property theory and information technology, particularly within the context of comparative and international law. Her writing is currently focussed on the commercial exploitation of technology, including the associated ethical and social concerns engaged by the transition to a knowledge economy. BAXI, UPENDRA—Upendra Baxi, LL.M. (Bombay), LL.M. (Berkeley), J.S.D. (Berkeley, California), is Professor of Law and current Director of the Law in Development programme of the Faculty of Law, University of Warwick. He served as Professor of Law, University of Delhi (1973–1996), Vice Chancellor University of Delhi (1990–1994), University of South Gujarat (1982–1985). He was Visiting Professor at Duke Law School (1987–1988) as well as American University, Washington D.C. (1994–1995), and he taught at Sydney Law School (1968–1973). He has also been a Global Law Fellow at New York University. His areas of specialist interest include comparative constitutionalism, social theory of human rights, law in globalisation, science, technology and human futures. He recently delivered a course at the Hague Academy of International Law on transnational corporate accountability which has just been published. Oxford University Press will soon be publishing a new book on a theory of human rights. Together with Professors A. R. Blackshield and Oliver Mendelsohn, Professor Baxi is also currently engaged in the preparation of a treatise on Indian constitutional theory and practice (forthcoming, New Delhi, Oxford University Press, 2000). BÜHLER, MARTIN—Martin Bühler was originally from Vancouver, British Columbia. He attended Simon Fraser University and received a Bachelor of Arts (Criminology) in 1995. In 1996 Martin moved to Toronto, Ontario, to attend the Faculty of Law at the University of Toronto, and received his LL.B in 1999. He articled at Borden Ladner Gervais LLP and will be called to the Ontario Bar in 2001. Martin is also an officer in the Canadian Naval Reserve, holding the rank of Lieutenant (Navy). BURNETT, MICHAEL—Michael Burnett (LL.B. (Hons) (Adel), LL.M. (Alberta)) is a former Adelaide partner of the law firm Minter Ellison. He was the Australian lawyer representing Deborah Parry and Lucille McLauchlan, the British nurses charged with murder in Saudi Arabia. He is currently the

xii Contributors Director of Enforcement (Queensland) for the Australian Securities and Investments Commission BURNS, PETER—Peter Burns, Q.C., LL.B. (Otago) 1963, LL.M. (Hons.) (Otago) 1963, called to the Bar of B.C., 1975, appointed Queen’s Counsel, 1984, is Professor of Law at the Faculty of Law, University of British Columbia. From 1982 to 1991, he served as Dean of the Faculty of Law. Since 1987, he has been a member of the United Nations Committee against Torture, and is currently Chairperson of that treaty body. He has also served on the B.C. Law Reform Commission (1985–92) and has been a board member of the B.C. International Commercial Arbitration Centre since 1986. Professor Burns was appointed Chair of the International Centre for Criminal Law Reform and Criminal Justice Policy in 1993 and is the current President of the Society for the Reform of Criminal Law. Professor Burns’ scholarship centres on the fields of Torts, Criminal Law and Criminal Procedure and International Criminal Law. BYRNES, ANDREW—Andrew Byrnes, B.A. (Hons) 1979, LL.B. (Hons) 1982, ANU; LL.M. Harvard 1984; J.S.D. Columbia 1988, is Director of the Centre for Comparative and Public Law at the Faculty of Law, University of Hong Kong. Professor Byrnes’ interests include human rights law, international law, constitutional law, administrative law, feminism and the law, conflict of laws. On the professional front, he has been a barrister and solicitor, Australian Capital Territory, since 1982 and a member of the New York Bar since 1987. Prior to joining the Faculty of Law of UHK he held positions in the Australian Attorney-General’s department, Canberra, in 1982–83 and as lecturer at the University of Sydney in 1985–86. He has also served as consultant, Attorney-General’s Chambers, Hong Kong Government 1990–91. His publications include Hong Kong Public Law and Human Rights Documents: A Sourcebook (edited, with Johannes Chan) (Singapore: Butterworths, 1993). CLAPHAM, ANDREW—Andrew Clapham is Associate Professor of Public International Law at the Graduate Institute of International Studies in Geneva. He teaches international human rights law, humanitarian law and public international law. He is Special Adviser on Corporate Responsibility to the UN High Commissioner for Human Rights. He is an academic member of the barristers’ chambers Matrix in London. He was formerly the Representative of Amnesty International to the United Nations in New York from 1991 to 1997. EVANS, MALCOLM—Malcolm D Evans is Professor of Public International Law at the University of Bristol. He has a particular interest in the work of the European Committee for the Prevention of Torture and is co-author and editor with Professor Rod Morgan of Preventing Torture (OUP, 1998) and Protecting Prisoners (OUP, 1999). His other interests include the freedom of religion and aspects of the international law of the sea.

Contributors xiii FLAH, ALEJANDRA—Alejandra C Flah (Hon. B.A. Int’l Stud., LL.B.) is an Associate in Toronto with the Canadian law firm of McCarthy Tétrault, focussing on International Trade Law. GERSTENBERG, OLIVER—Born 1964, Dr. iur (JWGoethe-University Frankfurt/M.), research fellow at Zentrum fuer Europaeische Rechtspolitik, Bremen University, Germany, Jean-Monnet-Fellow at EUI (1999–2000), visiting fellowships at Harvard University (Philosophy) and at Centre de Philosophie du Droit, Louvain-la-Neuve. Currently finishing a book on Limits of Private Ordering (habilitation-thesis) and co-authoring a bookproject (together with Charles Sabel) called “Directly-Deliberative Polyarchy as an Institutional Ideal for Europe.” HYLAND, EDWARD—M.B.A. (St. Francis Xavier, Antigonish, N.S.), M.A. (Toronto), S.T.B. (Centre Sèvres, Paris), S.T.M. (Regis College, Toronto), LL.B. (Toronto). A former executive director of a human rights and international development agency, he is now an associate with the law firm Iler Campbell (Toronto), specializing in legal services to charitable and nonprofit organisations. KAHANA , TSVI—Tsvi Kahana is a Visiting Scholar at Yale Law School. As of August 2000, he is the Executive Director of the Centre for Constitutional Studies, at the University of Alberta in Edmonton, Canada. He received his LL.B. and LL.M. (Magna Cum Laude) from Tel-Aviv University and his S.J.D from the University of Toronto. He has published in the areas of international law and human rights. KLABBERS, JAN—Dr Jan Klabbers is professor of international law at the University of Helsinki, and deputy director of the Erik Castrén Institute of International Law and Human Rights. Previously, he taught international law and EU law at the University of Amsterdam and the Amsterdam School of International Relations. LLEWELLYN, JENNIFER—Jennifer Llewellyn, M.A.(Queen’s) and LL.B. (Toronto), was a law clerk at the Federal Court of Appeal in Canada during 1999–2000. In 1997 she worked in the Research Department at the Truth and Reconciliation Commission National Offices in Cape Town, South Africa. She is currently completing her LL.M. degree at Harvard Law School. MCBURNEY, SEAN—Sean McBurney is a graduate of the Faculty of Law, University of British Columbia. MCONVILLE, ANNE—Anne C. McConville graduated from the University of Toronto Law School in 1998. After articling with the firm, she joined Fasken Martineau DuMoulin as an associate in 2000. Currently, she practices in the area of civil litigation, with an emphasis on disputes involving breaches of contract, confidence and fiduciary duties. Her practice also includes constitutional litigation and matters before administrative and regulatory tribunals.

xiv Contributors MORAN, MAYO—Mayo Moran is an Associate Professor and Associate Dean at the Faculty of Law, University of Toronto. She holds an LL.B. from McGill, LL.M from Michigan and S.J.D. from Toronto. She has written in tort law and theory as well as comparative constitutional law and legal theory. Rethinking the Reasonable Person will be published by Oxford University Press in 2001. Her current work is on the changing relationships between public and private, domestic and international law in an era of legal globalisation. MORGAN, ROD—Rod Morgan is Professor of Criminal Justice and Director of the Centre of Criminal Justice in the Faculty of Law, University of Bristol, UK. He is an expert advisor on custodial conditions and processes to Amnesty International and the Council of Europe. He is co-author (with Evans) of Preventing Torture (OUP, 1998) and Protecting Prisoners (OUP 1999) and coeditor of the Oxford Handbook of Criminology (2nd Edition, OUP 1997). He is a full-time researcher on aspects of criminal justice ranging from policing to sentencing. OOSTERVELD, VALERIE—Valerie Oosterveld (B.Soc.Sc., LL.B., LL.M.) is a Legal Officer with the Canadian Department of Foreign Affairs and International Trade, focussing on international humanitarian law issues. She is a member of the Canadian delegation to the ICC Preparatory Commission and also served on the Canadian delegation to the ICC Diplomatic Conference. Prior to joining DFAIT, she served as Director of the International Human Rights Programme at the University of Toronto Faculty of Law and practiced labour and employment law at the Toronto firm of Fasken Campbell Godfrey. She is currently pursuing her J.S.D. (Columbia Law School) on the topic of command responsibility under international humanitarian law. ORANGE, JENNIFER—Jennifer A Orange is an associate in the litigation department of Torys, Toronto. She has a B.A. in Asian Studies from the University of Pennsylvania and an LL.B. from the University of Toronto. She practices in the areas of commercial, administrative, constitutional and labour litigation and has a keen interest in private and public international law. She has spent several years in Japan, including one year on University of Toronto’s Work in Japan Programme at the firm of Hamada & Matsumoto, and speaks Japanese. RAPONI, SANDRA—Sandra Raponi graduated from the University of Toronto, Faculty of Law, in 1999. As a student in the Combined LL.B. and Ph.D. (Philosophy) Programme, Sandra is completing her graduate studies in philosophy. Her research areas include legal and political philosophy, with an emphasis on international human rights. In the summer of 1999, Sandra was an intern at the United Nations High Commission for Refugees in Geneva.

Contributors xv REICHMAN, AMNON—Amnon Reichman holds an LL.B. from the Hebrew University in Jerusalem, an LL.M. from the University of California at Berkeley, and an S.J.D. from the University of Toronto. He is a 2000–2001 Faculty Fellow at the Center for Ethics and the Professions at Harvard University. Prior to pursuing graduate work, he served as a clerk to Justice Aharon Barak of the Israeli Supreme Court. His main areas of interest are constitutional theory and comparative human rights, communication between the media and the courts, theories of adjudication, and aspects of globalization in constitutional law. SCOTT, CRAIG—Craig Scott is Associate Professor, Osgoode Hall Law School, York University, Toronto. Before joining Osgoode, he was a member of the Faculty of Law, University of Toronto, from 1989 to 2000 and a Jean Monnet Fellow at the European University Institute in Florence in 1999–2000. He served as law clerk to the former Chief Justice of the Supreme Court of Canada, Brian Dickson, in 1988–1989. SORNARAJAH, MUTHUCUMARASWAMY—Muthucumaraswamy Sornarajah (LL.B., Ceylon; LL.M., Yale; LL.M., London; Ph.D., London; LL.D., London) is Professor of Law at the National University of Singapore (NUS) where he has been on faculty since 1986. Amongst his appointments prior to coming to NUS, he served as Head of the Department of Law of the University of Tasmania. Apart from general international law, he specializes in various aspects of international economic law, and, in particular, is widely considered one of the top few scholars working in the field of the international law on investment. He is the author of some six books in the field including International Law on Foreign Investment (Cambridge University Press, 1994). He is currently editor in chief of both the Singapore Journal of Legal Studies and the Singapore Journal of International and Comparative Legal Studies. SWAN, MICHAEL—Michael Swan is a graduate of the Faculty of Law, University of Toronto. TERRY, JOHN—John Terry is a litigation and international trade lawyer at Torys in Toronto. He teaches Public International Law at the University of Toronto. Prior to joining Torys, he taught international human rights and international trade law in New Zealand and worked on constitutional and aboriginal issues for the government of Ontario, Canada. VERDIRAME, GUGLIELMO—Guglielmo Verdirame is Junior Research Fellow, Merton College, Oxford. He has previously worked as a research officer at the Refugee Studies Programme (University of Oxford), and is in the process of submitting his Ph.D. thesis at the London School of Economics. He has participated in various fact-finding missions on human rights for the most part in Africa on behalf of Article 19, Human Rights Watch and the Lawyers Committee for Human Rights.

xvi Contributors VIRGO, GRAHAM—Graham Virgo, M.A. (Cantab.), B.C.L. (Oxon), is a Fellow of Downing College, Cambridge, and a University Lecturer at the Faculty of Law, University of Cambridge. He researches in the fields of Restitution, Criminal law and Private International Law. He is author of Principles of the Law of Restitution (Oxford University Press, 1999). WAI, ROBERT—Robert Wai, B.Com.(McGill), LL.B.(UBC), M.Phil.(Oxon) S.J.D.(Harvard), is Assistant Professor at Osgoode Hall Law School, York University, Toronto. His teaching and research interests include international trade regulation, private international law, and contract law. He previously served as law clerk to Justice Gerard La Forest of the Supreme Court of Canada and practised law in Vancouver and New York. WELLS , BELINDA—Belinda Wells (LL.B. (Adel), LL.M. (Lond)) is a lecturer in Constitutional Law and Human Rights at the Flinders University of South Australia. She has previously worked as a senior law reform officer at the Australian Law Reform Commission in Sydney and with the South Australian Solicitor-General and Crown Solicitor’s Office. She is currently on leave from Flinders University and working in the field of refugee law with the South Brisbane Immigration and Community Legal Centre. WICKREMASINGHE, CHANAKA—Chanaka Wickremasinghe is a lecturer at the University of Bristol, and was previously Senior Research Officer at the British Institute of International and Comparative Law. He is currently Rapporteur for the study of transnational human rights torts being conducted by the Human Rights Committee of the International Law Association (British Branch).

Table of Cases 1. INTERNATIONAL Aloeboebe v Suriname (1992) 13 HRLJ 140.................................................418 Ambatielos Arbitration (Greece v UK) (1956) 12 RIAA 83 ..................................................................................177 Barcelona Traction, Light and Power Co. Ltd, —see Belgium v Spain. Belgium v Spain [1970] ICJ Rep 3 ..................................173, 379, 397, 434, 435 Bosnia Genocide [1998] ICJ Rep. 595..........................................................501 Celiberti di Casariega (Lilian) v Uruguay, 1981–2 Yearbook HR Cttee II 1989, 327 ...................................36, 505, 510 Consuelo v Argentina, Inter-American Ct HR 28/92 ...................................568 Corfu Channel [1949] ICJ Rep. 4................................................................503 Cumaraswamy, (ICJ), 29 April 1999...................................................477, 482 Elmi v Australia (UN), 14 May 1999 ...........................................................548 ELSI [1987] ICJ Rep. 15..............................................................................508 France v Turkey [1927] PCIJ Reps., Series A No. 1 ..............................167, 169 Garay Hermosilla v Chile, Inter-American Ct HR 36/96..............................568 Home Missionary Society (1920) 6 RIAA 20 ...............................................509 Legal Consequences for States of the Continued Presence of Africa in Namibia [1971] ICJ Rep. 16 .......................................................36 Manderlier v UN and Belgium, 45 ILR 446 .................................477, 478, 479 Mavromattis Palestine Concessions [1954] PCIJ Reps. 12............................497 Mendoza v Uruguay, Inter-American Ct HR 29/92 .....................................568 Mexico v US (Quintanilla Claim), US-Mexico General Claims Commission, 4 RIAA 101 .......................................................................112 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v US) [1986] ICJ Rep. 134 .............................................392, 394 Nicaragua [1984] ICJ Rep. 4 ......................................................501, 505, 507 Paraguay v USA [1998] ICJ Rep 248 ...........................................................563 Prosecutor v Anto Furundzija, International Criminal Tribunal for the Former Yugoslavia, 10 December 1998 ................119, 459, 473, 646 Prosecutor v Delalic Mucic Delic and Landzo, ICTY Case IT–96–21-T, 16 November 1998 .....................442, 445, 446, 447, 448 Prosecutor v Kayishema and Rozindana, Case ICTR–95–1-T ........................................................................446, 460 Rainbow Warrior (1987) 26 ILM 1346................................501, 505, 509, 510 Sadiq Shek Elmi v Australia, UN Doc.A/54/44 ............................................276 Sambaggio (1903) 100 RIAA 499.................................................................508 Schufeldt Claim (1930) RIAA 1079 .............................................................498 Short v Iran (1987) 16 Iran-US CTR 76 ...............................................500, 510

xviii Table of Cases Stanvays v UN [1971] Revue Belge de Droit International ...................451, 483 Steamship Lotus—see France v Turkey Tadic, Case IT–94–1, 7 March 1997 ...................................................473, 501 US v Field Marshall Wilhelm List, UN War Crimes Commission, (1948) 11 Law Reps. of War Crimes 1271........................444, 445, 447, 461 US v Iran [1980] ICJ Rep. 3 ........................................................................392 US v Mexico (Roberts Claim), US-Mexico General Claims Commission, 4 RIAA 77 .............................................................112 US v Wilhelm von Leeb, UN War Crimes Commission, (1948) 12 Law Reps. of Trials of War Crime 76.......................443, 444, 461 US Diplomatic and Consular Staff in Tehran—see US v Iran Velasquez-Rodriguez v Honduras, Inter-American Ct. HR (1988) 9 HRLJ 212, (1989) 11 HRLJ 127...........112, 381, 418, 512, 518 Vienna Convention on Consular Relations—see Paraguay v USA Yamashita (1945–6) US Military Commission at Manila ...............................................................................443, 444, 461 Yeager v Iran (1987) 17 Iran-US CTR 92.....................................................500 Youmans Claim (1926) 4 RIIA 110 .............................................................509 Zafiro Claim (1923) 6 RIAA 160.........................................493, 495, 509, 510

2. EUROPEAN COURT OF HUMAN RIGHTS Airey v Ireland, 1979 Series A No. 32..........................................................521 Aydin v Turkey, App. No. 23178/94 ...................................................144, 503 Beer and Regan v Germany, 18 February 1999 ....................................479, 486 Delazarus v UK, App. No. 17525/90 ...........................................................145 Golder v UK, 1973 Series A No. 18 .............................................................521 Helmers v Sweden, 1991 Series A No. 212...................................................528 HLR v France, 29 April 1997 ..............................................................529, 534 Ireland v UK, Series A No. 25 .............................................................404, 619 Loizidou v Turkey, Series A No. 310 ............................................23, 530, 531 Matthews (Denise) v UK, App. No. 24833/94..............................................486 Moreira de Azavedo v Portugal, Series A No. 189 .......................................527 Osman v UK, 28 October 1998............................................22, 486, 515, 516, 521–523, 525, 527, 528, 534 Peers v Greece, App. No. 28524/95 .....................................................145, 146 Raphaie v UK, App. No. 20035/92 ..............................................................145 Selmouni v France, 28 July 1999 .................................................................146 Smith and Grady v UK, 27 September 1999.........................................532, 533 Soering v UK, Series A No. 161 ............................23, 529, 530, 531, 534, 616 Tosunoglu v Greece, App. No. 21892/93.....................................................145 Waite and Kennedy v Germany, 18 February 1999 ..............................479, 486

Table of Cases xix

3. EUROPEAN COURT OF JUSTICE 189/87, Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst und Co [1988] ECR 3365 ........................................................................336 C–415/93, Bosman [1995] ECR I–4921.................................31, 687, 688, 690 4. NATIONAL Australia Attorney-General v T S Gill & Son Pty Ltd [1926] VLR 414 ........................621 Attorney-General (on the Relation of Daniels and Others) v Huber [1971] 2 SASR 142 .......................................................................621 Australian Capital TV P/L v Commonwealth (No. 2) (1992) 66 ALR 695 ............................................................................................664 Australian Conservation Foundation Inc. v Commonwealth (1980) 146 CLR 493 ..........................................................................................621 CSR Limited v Cigna Insurance Australia (1997) 189 CLR 345 ....................622 Dietrich v The Queen (1992) 67 ALR 1 .......................................................664 Ditford (Re), ex parte Deputy Commissioner of Taxation (NSW) (1988) 83 ALR 265..................................................................................619 Gibson v Parkes District Hospital (1991) 26 NSWLR 9 ...............................412 Gimson v Victorian Workcover Authority [1995] 1 VR 209 (Sup.Ct.) .......................................................................412 Koowarta v Bjelke-Peterson (1982) 153 CLR 168 ........................................619 Mabo v Queensland (No. 2) (1992) 175 CLR 1....................................682, 683 McLauchlan v Gilford (1997) 69 SASR 269 .........................605, 610, 611, 623 Minister of State for Immigration and Ethnic Affairs v Teoli (1995) 183 CLR 273 .........................................................620 Tasmanian Wilderness Society Inc. v Fraser (1982) 153 CLR 270.................619 Taylor Holdings Ltd v Bond (1993) 59 SASR 432 ........................................611 Voth v Manildra Mills Pty Ltd (1990) 171 CLR 538 ............................623, 624 Wentworth v Woolahra Municipal Council (1982) 149 CLR 672 .................621 Canada Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v Canada (AttorneyGeneral), 6 July 1999, unreported (Ont. Sup. Ct.)................ 3, 6, 8, 33, 34, 35, 36, 38, 39, 104, 130–133, 169, 474, 487, 657 Acronym (Cayman) Inc v Ontario Lottery, Corp (1997) 12 CPC 4th 331 (Ont. Gen. Div.)...............................................................................186 Ahani v Canada (Ministry of Citizenship anmd Immigration), 2000 Doc. A–413–99 (FCA) ....................................................................180, 427

xx Table of Cases Alberta Union of Provincial Employees et al. and the Crown in Right of Alberta (Re) (1981) 120 CLR (3rd) 590 (Alta. QB)......................420 Amchem Products Inc. v. British Columbia (Workers’ Compensation Board) [1993] 1 SCR 897 ................................122, 189, 190, 191, 192, 226 American Farm Bureau Federation v The Canadian Import Tribunal (1990) 74 DLR (4th) 449 .........................................................................421 Antares Shipping Corp v The “Capricorn” [1977] 2 SCR 422 ......................189 Arrow River and Tributaries Slide & Boom Co Ltd (Re) [1932] SCR, 495 ...............................................................................................420 B (J) v D (E) (1993) 16 CPC 242 (Ont.Gen.Div.) ..........121, 184, 190, 192, 195 BG Chico International Ltd v BC Hydro and Power Authority (1993) 99 DLR (4th) 577 (SCC) ...............................................664 Baker v Canada [1999] 2 SCR 817 .......................................37, 129, 249, 256, 257, 258, 259, 260, 263, 271, 376, 383, 387, 388, 421, 425, 620, 676, 682 Bazley v Curry [1999] 2 SCR 554 ................................454, 455, 456, 462, 463 Bhadauria v Board of Governors of Seneca Ciollege of Applied Arts and Technology (1979) 105 DLR (3d) 707 (Ont.CA) ...................390, 391, 392, 393, 414, 416, 417 Bittel v Yim (1978) 20 OR (2d) 617 .............................................................315 Board of Governors of Seneca College of Applied Arts and Technology v Bhadauria 1981) 124 DLR (3d) 193 (SCC)........................................310, 376, 390, 415 Bonaventure Systems Inc v Royal Bank of Canada (1986) 57 OR (3d) 270 (Div.Ct.) ...................................................................................191 Canada (Attorney-General) v Ward [1993] 2 SCR 689 ..................................43 CAIMAW v Paccar of Canada [1989] 2 SCR 983 ........................................413 Canada Labour Code (Re) [1992] 2 SCR 50 ................................217, 219, 243 Canada Trust Co and Ontario Human Rights Commission (1990) 69 DLR (4th) 321 (Ont.CA) .........................................................390 Canadian International Marketing Distributing Ltd v Nitsuko Ltd (1990) 68 DLR (4th) 318 (BCCA) ............................................................164 Capital Cities Communications Inc v Canadian Radio-TV Commission [1978] SCR 141 ..........................................256, 376, 387, 421 Carrato v US, (1992) 40 OR (2d) 459 (Ont.HC)..........................................420 Central Trust v Rafuse (1986) 31 DLR (4th) 481 (SCC) ...............................664 Chitat Ng v Canada (CCPR/C/49/D/469/1991)............................................616 Cook v Ip (1995) 52 OR (2d) 289 (CA)........................................................182 Dagernais v CBC [1994] 3 SCR 833.............................................................664 Daniels v The Queen [1968] SCR 517 .........................................................387 Davidson Partners v Citibank NA (1990) 72 OR (2d), 450 (Master) ............185 Davidson Tisdale Ltd v Pendrick, (1997) 10 CPC (4th) 336 (Ont.Gen.Div.), 1998 OJ No. 5308..................................................190, 322

Table of Cases xxi De Savoye v Morguard Investments Ltd [1990] 3 SCR 1077..................................................121, 161, 162,163, 164, 170, 182, 185, 186, 194 de Vlas v Bruce (1994) 18 OR (3d) 493 (Gen.Div.) ...............................184, 185 Dino v Albertson’s Inc (1994) 28 CPC (3d) 15 .....................................186, 192 Dobson (Litigation Guardian of) v Dobson [1999] 2 SCR 753......................423 Drummond Wren (Re) (1945) 4 DLR 674 (Ont.CA) ....................................390 Dunlop v Connecticut College (1996) 50 CPC (3d) 109 (Gen.Div.).......186, 190 Dy 4 Systems Inc v Diamond Point International Inc (1997) 14 CPC (4th) 99 (Ont.Gen.Div.) ..................................................................190 Ecolab Ltd v Greenspace Service Ltd (1998) 38 OR (3d) 145 (Div.Ct.) ................................................................................................186 Ell v Con-Pro Industries Ltd (1992) 11 BCAC 174 .......................................164 Estonian State Cargo and Passenger Steamship Line v SS Elise (The Elise) (1948) 4 DLR 247 (Exch. Ct.) ................................................347 First City Trust Co v Inuvik Automotive Wholesale [1993] NWTR 273 (SC) .................................................................................................164 Francis v The Queen [1956] SCR 618..........................................................421 Frymer v Brettschneider (1994) 19 OR (3d) 60............................122, 164, 182, 189, 190, 195 Gilmour v Barry-Wehmiller Co (1996) 34 OR (3d) 548 (Ont.Gen.Div.).......................................................................................122 Grimes v Cloutier (1989) 69 OR (2d) 641 ....................................................292 Hanlan v Serenesky, 1997 OJ No. 3658 (Ont.Ct. (Gen.Div.)).......................................................................300, 301 Harrison v Carswell [1978] 2 SCR 200........................................410, 411, 412 Hercules, Managements Ltd v Ernst & Young [1997] 2 SCR 165 ...................................................................................414 Hill v Church of Scientology [1995] 2 SCR 1130 ...........423, 424, 425, 426, 664 Holodeck Adventurer Ltd v Orbotron Inc (1996) 8 CPC (4th) 376 (Gen.Div.) ..................................................186, 191 Hunt v T & N plc [1993] 4 SCR 289 ..................................120, 121, 161, 163, 164, 166, 182, 185, 349, 350, 382, 432 Jacobi v Griffiths [1999] 2 SCR 370.............................................................454 Jaffe v Dearing (1988), 63 OR (2d) 133 (HCJ)....................170, 184, 192, 193 Jaffe v Miller (1993) 13 OR, (3d) 745, 103 DLR (4th) 315..................................................368, 369, 370, 651, 652 Juelle v Trudeau (1968) 7 DLR (3d) 82 (Que.SC) ........................125, 348, 360 Kindler v Canada (Minister of Justice) [1991] 2 SCR 779 ............426, 427, 616 Laane and Baltzer v Estonian State Cargo and Passenger Steamship Line (1949) 2 DLR 641 (SCC) ........................125, 347, 351, 354, 356 Libman v The Queen [1985] 2 SCR 178 ......................................170, 312, 313

xxii Table of Cases Lucas v Gagnon (1992) 11 OR (3d) 422........................................296, 297–298 M (A) v Ryan [1997], 1 SCR 157 ................................................................423 M (K) v M (H) [1992] 3 SCR 3....................................................................314 MacDonald v Lasnier (1994) 21 OR (3d) 177 (Ont.Gen.Div.) .....121, 164, 184, 185, 186, 192 MacDonald v Vapor Canada Ltd [1977] 2 SCR 134 ....................................421 McLean v Pettigrew [1945] SCR 62 ...................................291, 292, 293, 294, 296, 298, 299 Moran v Pyle National (Canada) Ltd [1975] 1 SCR 393 ......161, 170, 313, 314 Moreno v Norwich Union Fire Insurance, Society Ltd (1971) 1 OR 625 (HCJ) ...........................................................................182 Morguard Investments Ltd v De Savoye [1990] 3 SCR 1077 ..........................................................377, 382, 432, 433 National Bank of Canada v Clifford Chance (1996) 30 OR (3d) 746 (Ont.Gen.Div.) ....................186, 187, 190, 191, 192 Noble and Wolf (Re) (1949) 4 DLR 375 ......................................................390 Ontario New Home Warranty Program v General Electric, Co (1998) 36 OR (3d) 787 (Ont.Gen.Div.) .......................................186, 190 Orelien v Canada [1992] 1 FC 592 (CA) ......................................................420 Philippines (Republic) v Pacificador (1999) 60 CRR (2d) 126, (Ont.Gen.Div.).......................................................................................427 Power v Probert (1987) 19 CPC (2d) 142 (Dist.Ct.)......................................184 Public Service Employee Relations Act (Alta) (Re) [1987], 1 SCR 313 .........387 Pushpanathan v Canada [19987] 1 SCR 982..........................................43, 478 Queen (The) v Brocklebank, Court Martial Appeal, Court of Canada, 2 April 1996................................................................473 Queen in Right of Canada (The) v Saskatchewan Wheat Pool [1983] SCR 205....................................390, 391, 414, 415, 416 R. v. Cook [1998] 2 SCR 597..............................167, 168, 170, 180, 427, 428 R. v. Crown-Zellerbach [1986] 1 SCR 401 ....................................................43 R. v. Ewanchuk [1998] 1 SCR 330 ........................................................43, 260 R. v. Finta [1994] 1 SCR 701.......................................................172, 179, 180 R. v. Harr, [1993] 3 SCR 562 .....................................................................299 R. v. Libman [1985] 2 SCR, 178 .................................................................239 R. v. Oakes [1986] 1 SCR, 103 ...................................................................633 R. v. Salituro [1991] 3 SCR 654 ..................................................424, 425, 426 R. v. Swain [1991] 1 SCR 933 .....................................................................424 R. v. Terry [1996] 2 SCR 207 ......................................................................299 Recherches Internationales de Québec v Cambion [1998] QJ No. 2334 (Que.SC) ...................................................123, 189, 194, 195 Reference re s.94(2) of the Motor Vehicle Act (BC) [1995] 2 SCR 486 ..........260 Reference re Power of Municipalities to Levy Rates on Foreign Legions and High Commissioners’ Residences [1943] SCR, 208.....................................................................................420

Table of Cases xxiii Reference re Whether, Members of the Military or Naval Forces of the USA are Exempt from Criminal Prosecutions in Canadian Criminal Courts [1943] SCR, 483...........................................................420 Regina and Palacios (Re) (1984) 45 OR (2d) 269 (CA) .................................420 Retail, Wholesale and Department Store Union Local 580 v Dolphin Delivery Ltd [1986] 2 SCR 573 .................................180, 391, 392, 423, 424, 664 Royal Bank and Corriveau (Re) (1983) 117 DLR (3d) 199 (Ont.HC) ...............................................................................................219 Said v Canada (1999) 48 Imm. LR (2d) 9 (Fed.Ct.Trial Div.) .......................423 Schavernoch v Foreign Claims Commission [1982] 1 SCR, 1092 .................387 Ship “North” (The) v The King (1906) 37 SCR 385 .....................................420 Slaight Communications Inc v Davidson [1989] 1 SCR 1038 ..........................................................259, 387, 423, 424 Suresh v Canada (Ministry of Citizenship and Immigration), 2000 Doc. A–415–99 (FCA) ............................................................180, 636 Suresh v R. (1999) 38 OR (3d) 264 (Ont.Gen.Div.) ..............................423, 427 Tolofson v Jensen [1994] 3 SCR 1022 .......................14, 15, 38, 40, 123, 132, 161, 162, 165, 166, 191, 226, Ch.11, 328, 329, 377, 378, 432, 433, 434, 436, 549, 598, 599 Trepanier v Kloster Cruise Ltd (1995) 23 OR (3d) 398 (Ont.Gen.Div.) ........186 Upper Lakes Shipping Ltd v Foster Yeoman Ltd (1992) 12 CPC 31 (Ont.Gen.Div.), (1993) 14 OR (3d) 548 ...........................................122, 186 Vile v Von Wendt; Zurich Insurance Co (1979) 103 DLR (3d) 356 (Ont.HCJ) .............................................................................................184 Vriend v Alberta [1998], 1 SCR 493 ...........................................................260 Webb v Hooper (1994) 19 Alta LR (3d) 269 ................................................164 Wilson v Moyes (1993) 13 OR (3d) 302 (Ont.Gen.Div.).......................164, 184 France Burgat v Ministry of Foreign Affairs (1977) 104 JDI (clunet) 630 .................485 Germany Lüth, BVerfGE 7, 198 .................................................................................689 Universal Jurisdiction over Drug Offences (1987) 74 ILR 166 (BGH)....................................................................................................406 Israel ‘Abd al-Halim Bilbeisi v General Security Service, HCJ 7964/95, 11 January 1996......................................................................................124 City of Kiryat Gatt v State of Israel, 37 (3), PD 832 ....................................639 Khader Mubarak et al. v General Security Service, HCJ 3124/96, 17 November 1996..................................................................................124

xxiv Table of Cases “Kol Ha’ami” Co Ltd v Ministry of the Interior, 7 PD 871 ..........................639 Muhammad ‘Abd al-Aziz Hamadan v General Security Service, HCJ 8049/96, 14 November 1996 ....................................................................124 Public Committee against Torture in Israel v State of Israel and General Security Service, 1999 HC 5100/94 (Sup.Ct.) ............................................124, 631, 632, 633, 634, 635, 636, 638, 639, 640, 641, 642, 643, 644, 645, 646, 647, 648, 649, 650, 654, 656, 657, 658 United Mizrahi Bank Ltd v Migdal Cooperative Village, 49 (4) PD 221 (1995)...................................................................633 New Zealand Bradley v Wignut Films Ltd [1993] 1 NZLR 415 (HC).................................412 South Africa Azanian People’s Organisation v President of the Republic of South Africa, 1998 (8) BCLR 1015 (CC)............................................................579 Switzerland Marcos and Marcos v Federal Department of Police (1989) 102 ILR 198 (Fed.Ct.) ...................................................................................652 United Kingdom Abidin Daver (The) [1984] AC 398 .............................................................195 Adams v Cape Industries plc [1990] Ch. 433 (CA) .......................................161 Al-Adsani v Government of Kuwait (1996) 107 ILR 536 (CA) ...................................3, 188, 269, 270, 336, 337, 341, 505, 506, 534, 538, 652 Amin Rasheed Shipping Corp. v Kuwait Insurance Co (The Al-Wahab) [1983] 2 All ER 884 (HL) ..............................................161 Attorney-General v Nissan [1969] 1 All ER 629 (HL) ..................................487 Attorney-General of Canada v Attorney-General of Ontario (Labour Conventions) [1937] AC 326 (PC) ......................................180, 421 Attorney-General of New Zealand v Ortiz [1984] AC 1...............................655 Bank voor Handel en Scheepvaart NV v Slatford [1953] 1 QB 248 ...............355 Banque Keyser Ullmann SA v Skandia (UK) Insurance Co Ltd [1991] 2 AC 349......................................................................................412 Bonancina (Re) [1912] 2 Ch. 394.........................................................327, 331 Boys v Chaplin [1969] 2 All ER 1083, [1971] AC 356 (HL)...........329, 597, 598 Buck v Attorney-General [1965] Ch. 745 (CA) ...................348, 349, 353, 367, 652, 653 Buttes Gas v Hammer, [1981] 3 WLR 787 .................................345, 346, 354, 356–358, 367, 370

Table of Cases xxv Candler v Crane Christmas & Co [1951] 1 All ER 426 (CA) ........................401 Castree v Squibb, Ltd [1980] 1 WLR 1248 ..................................................329 Chaplin v Boys [1971] AC 356.....................123, 292, 295, 296, 319, 320, 338 Chung Chi Cheung v R. [1939] AC 160.......................................................264 Code v Turner (1704) 6 Mod. 149...............................................................331 Collins v Wilcock [1984] 1 WLR 1172.........................................................331 Cook v Sprigg [1899] AC 572......................................................................357 Congresso del Partido (I) [1983] 1 AC 244..........................226, 365, 366, 367 Connelly v RTZ Corp P/L [1998] AC 854, [1997] 3 WLR 373 (HL) ..........................................122, 192, 194, 204, 207 Cranstoun v Bord (1896) 4 BCR 596 (Div.Ct.).............................................354 Derbyshire County Council v Times Newspapers Ltd [1992] 3 WLR 28 ......664 Distillers Co (Bio-Chemicals) Ltd v Thompson [1971] AC 458 ....................329 Donoghue v Stevenson [1932] AC 562 (HL) ................................................411 Dubai Bank Ltd v Galadari (No. 5) [1990] TLR 490 (HC) ...........................349 Duke of Brunswick v King of Hanover (1844) 6 Beav. 1...............................357 Emperor of Austria v Day and Kossuth (1861) 3 De GF & J 217 .................420 Empresa Exportadora Azucar v Industria Azucarera Nacional SA (The Playa Larga and Marble Islands) [1983] 2 Lloyd’s Rep. 171 (CA) .......................................................366, 370 Gouriet v Union of Post Office Workers [1978] AC 435...............................621 Heathfield v Chilton (1767) 4 Burr. 2015 .....................................................420 Hedley Byrne & Co v Heller and Partners Ltd [1963] 2 All ER 575 (HL) ........................................................................401 Helbert Wagg & Co Ltd [1956] 1 Ch. 323...................348, 351, 352, 353, 355 Hesperides v Muftizade [1978] 2 All ER (HL) .............................................125 Khorasandjian v Bush [1993] QB 727 (CA) .................................................412 Kidd (Trial of Captain William), 14 Howell State Trials 147 (1701).............128 Kingdom of Spain v Agusto Pinochet Ugarte, 8 October 1999 (Bow Street Magistrate) ...................37, 525, 529, 572, 599 Kuwait Airways Corp v Iraqi Airways Co and Iraq [1993] 2 Lloyd’s Rep. 317 ..................................................................................652 Lorentzen v Lydden & Co [1942] 2 KB 202 .........................................351, 355 Lubbe et al. v Cape plc, QBENI98/0192/1, 30 July 1998...............................207 Luthor v Sagor [1921] 3 KB 532 ..........................................................348, 351 McGovern et al. v Attorney-General [1981] 3 All ER, 493 (Ch.D)...............413 Machado v Fontes [1897] 2 QB 231 (CA) ....................................295, 296, 596 Macmillan Inc v Bishopsgate Investment Trust plc (No. 3) [1996] 1WLR 387, ..............................................................326, 327 Maharanee of Baroda v Wildenstein [1972] 2 QB 283 (CA) .................161, 182 Metall und Rohstoff AG v Donaldson, Luffkin and Jenrette Inc [1990] 1 QB 391......................................................................................329 Mohammed v Bank of Kuwait and the Middle East [1996] 1 WLR 1483 (CA) ........................................................122, 193, 194

xxvi Table of Cases Nagle v Feilden [1966] 2 QB 633 (CA).........................................................416 Nissan v Attorney-General [1970] AC 179 ..........................................359, 366 Oppenheimer v Cattermole [1976] AC 249 ................................338, 351, 353, 354, 355, 655 Overseas Tankship (UK) Ltd v Morts Dock and Engineering Co Ltd (The Wagon Mound) [1961] AC 388 .................312–313 Phillips v Eyre, (1869) LR 6 QB 1......................................110, 123, 294, 295, 296, 332, 338, 595, 596 Pinochet (No. 2) ..................................................283, 343, 344, 366, 525, 678 R. v Bartle and Commissioner of Police for the Metropolis, ex parte Pinochet (No. 3) [1999] 2 All ER 97 .........................................5, 13, 16, 17, 20, 28, 37, 39, 84, 110, 114, 115, 119, 122, 126, 252, 277, 280, 282–286, 328, 330, 341, 343, 344, 366, 367, 368, 478, 484, 488, 492, 525, 533, 549, 572, 599, 626, 627, 643, 644, 646, 648, 649, 651, 652, 662, 678 R. v Bow Street Metropolitan Stipendiary Magistrate, ex parte Pinochet Ugarte (No. 1) [1998] 3 WLR 1456, [1998] 4 All ER 897 .......................................283, 284, 285, 343, 344, 366, 525, 571, 572, 573, 643, 648, 678 Red Sea Insurance Co Ltd v Bouygues SA [1995] 1 AC 190..............................................................123, 292, 437, 549 Regazzoni v KC Sethia (1944) Ltd [1957] 2 Lloyd’s Rep. 289 ...............353, 355 Schaik Willem Burger Lubbe v Cape plc, HL 20 July 2000, unreported ..............................................................39, 40 Seaconsar Far East Ltd v Bank Markazi Jomhouri Islami Iran [1994] 1 AC 438......................................................................................336 Secretary of State in Council of India v Kamanchee Boye Sahaba (1859) 13 Moo. PC 22 .............................................................................357 SK & F Laboratories Ltd v Bloch [1983] 2 All ER 72 ...........................110, 204 Spiliada Maritime Corp v Cansulex Ltd [1987] 1 AC 460 (HL) ....................................................161, 189, 206, 226, 336, 550, 623 Tallina Laevauhisus A/S v Tallina Shipping Co (The Vapper) (1946) 79 Ll.LR 245 (KB), (1947) 80 Ll.LR 99 (CA)..................346, 347, 348 Thornton v Kirklees Metropolitan Borough Council [1979] QB 626 (CA) ................................................................................414 Trendtex Trading Corp Ltd v Central Bank of Nigeria [1977] QB 529 (CA)................................................................227, 243, 356 Triquet v Bath (1764) 3 Burr. 1478..............................................................420 Watson v Prager [1991] 3 All ER 487 ..........................................................416 Westland Helicopters v AOI [1995] 2 All ER 387 ........................................478 Wilkinson v Downton [1897] 2 QB 57 ........................................................331

Table of Cases xxvii Williams & Humbert Ltd v W and H Trade Marks (Jersey) Ltd [1986] AC 368 ........................................................................................338 United States Abebe-Jira v Negewo 72 F 3d 844 (CA, 11 Cir.1996) .................76, 87, 91, 98, 448, 562, 661 Adra v Clift 195 F Supp. 857 (D Md.1961) ..................................................673 Agency Holding Corp v Malley-Duff & Associates 483 US 143 (1987) ...................................................................................103 Agent Orange Product Liability Litigation (In re) 580 F Supp. 690 ..............................................................204, 206, 210, 211 Agunda v Texaco 175 FRD 50 (1997)..........................................................593 Alabama Claims Arbitration (1872) 1 Moore 495 .........................21, 494, 499 Alejandre v Cuba 996 F Supp. 1239 (1997) ..................................................493 Alomang v Freeport-McMoran Inc 718 So. 2d 971 (La App., 4 Cir. 1998) .............................................................................331 Alvarez-Machain v US 107 F 3d 696 (9 Cir.1996).........................................410 Amerada Hess Shipping Corp v Argentine Republic 830 F 2d 421 (2 Cir.1987) ............................................................................................409 Argentine Republic v Amerada Hess Shipping Corp 488 US 428, 109 S Ct. 683 (1989)................................................67, 70, 267, 268, 269, 270 Askir v Boutros Ghali and Others 933 F Supp. 368 ......................................484 Babcock v Jackson 12 NY 2d 473 (1963) .....................................................300 Baker v Carr 369 US 217 ........................................................................81, 82 Banco Nacional de Cuba v Sabbatino 376 US 398 (1964)...............................................78, 79, 80, 86, 125, 244, 357, 360, 361, 362, 385, 591 Bank of the United States v Planter’s Bank 22 US (9 Wheat.) 904 (1824) ...............................................................................72 Beanal v Freeport-McMoran Inc 969 F Supp. 362 (ED La. 1997) ..........................................................34, 67, 74, 76, 91, 92, 93, 94, 101, 506 Bell v City of Milwaukee 746 F 2d 1205 ........................................................85 Bernstein v NV Nederlandsche-Amerikanische Stoomvaart-Maatschappij 210 F 2d 375 (2 Cir. 1954)...............................161 Bernstein v Van Heughen Frères SA 163 F 2d 246 (2 Cir.1947) .....................361 Bivens v Six Unknown Named Agents 403 US 388 (1971) ............................111 Bradvica v INS 128 F 3d 1009 (7 Cir.1997) ..................................................410 Burger King v Radzinowicz 471 US 462, 103 S Ct. 2174 (1983) ......................78 Burnham v Superior Court 493 US 604, 109 L Ed. 2d 631, 110 S Ct. 2103 (1990) ..............................................................................................78 Byung Wha An v Doo-Hwan Chun 1998 US App. LEXIS 1303 (9 Cir.) .............................................................................................77

xxviii Table of Cases Cabiri v Assasie-Gyimah 921 F Supp. 1189 (SDNY 1996) ...........83, 84, 91, 92 Chuidian v Philippine National Bank 912 F 2d 1095 (9 Cir. 1990) ...........................................................................................626 Demjanjuk v Petrovsky 776 F 2d 571 (6 Cir. 1985).......................................128 De Sanchez v Banco Central de Nicaragua 770 F 2d 1385 (5 Cir. 1985) .............................................................................................96 Doe v UNOCAL, 963 F Supp. 880 (1977), 27 F Supp. 2d 1174 (1998) ....................................................................................364, 492, 506 Dow Chemical Co, and Shell Oil Co v Domingo Castro Afarao 786 SW 2d 000 (Tex.1990)...................................................207, 210 Erie RR Co v Tompkins 304 US 64...............................................................85 Fernandez v Wilkinson 505 F Supp., 787 (D Kan.1980)...............................409 Filártiga v Peña-Irala 630 F 2d 876 (CA, 2 Cir.1980) .............3, 14, 68, 69, 74, 75, 76, 79, 80, 81, 82, 84, 87, 89, 90, 91, 92, 95, 96, 97, 106, 107, Chap.4, 151, 176, 247, 291, 292, 293, 305, 318, 363, 374, 382, 386, 396, 397, 399, 407, 409, 410, 436, 538, 560, 562, 587, 661, 665, 667, 669, 670, 671, 672, 673, 674, 678, 679, 681 Flatlow v Islamic Republic of Iran 999 F Supp. 1 (DCCir.1998) ......................................................................71, 77, 78, 363 Forti v Suarez Mason 672 F Supp. 1531 (Forti I) (N Cal. 1987) ............................................73, 93, 94, 95, 96, 97, 98, 103, 104, 106, 126, 408, 409, 448, 627, 661, 673 Forti II ....................................................................................96, 97, 98, 675 Gallagher v Neil Young Freedom Concert 49 F 2d 1442 (10 Cir,, 1995).........93 Garcia-Mir v Meese 788 F 2d 1446 (11 Cir.1986) .........................................409 General Building Contractors Association v Pennsylvania 458 US 375 (1982) ..............................................................................................451 Gulf Oil Corp v Gilbert 330 US 501 (1947)............................................83, 226 Guinto v Marcos 654 F Supp. 280 ................................................................96 Hatch v Baez 7 Hun. 596 (1876)..................................................................277 Heikkila v Barber 308 F 2d 558 (9 Cir.1962) ................................................105 Helicopteros Nacionales de Colombia SA v Hall 466 US 408 (1984) ..............77 Hilao v Estate of Ferdinand Marcos 25 F 3d 1467 (Ca, 6 Cir.1994)............................................................66, 73, 82, 91, 104, 105, 107, 363, 364, 562 Hilton v Guyot 159 US 113 (1895) ......................................162, 165, 385, 622 International Association of Machinists & Aerospace Workers v OPEC 649 F 2d 1354 (9 Cir.1981) .............................................79 International Shoe Co v Washington 326 US 310 (1943) ..............................164 Iwanowa v Ford Motor Co 67 F Supp. 22 (1999).................................424, 665 Jota v Texaco 157 F 3d 153 (1998) ..............................................................593

Table of Cases xxix Kadic v Karadzic 70 F 3d 232 (CA, 2 Cir.1995) .............67, 74, 75, 78, 80, 81, 82, 84, 87, 88, 91, 92, 93, 94, 99, 100, 101, 102, 120, 363, 386, 448, 661 Kirkpatrick (WS) & Co v Environmental Tectronics Corp 493 US 400, 110 S Ct. 701 (1990) ....................79, 360, 361, 362, 592 Klaxon v Stentor Electric Manufacturing Co 313 US 487 (1941) ...........................................................................................85 Kline v Kaneko 685 F Supp., 386 (SDNY 1988) ............................................70 Kor v Bayer AG (US DC, SD Ind.) ..............................................................665 Lafrontant v Aristide 944 F Supp. 128 (EDNY 1994) .............................84, 661 Letelier v Chile 499 F Supp. 665 (1980)........................................................505 Linder v Portocarrero 963 F 2d 332 (CA, 9 Cir.1992) ........................73, 81, 82 Liu v Republic of China 642 F Supp. 297 (ND Cal., 1986), 892 F 2d 1419 (CA, 9 Cir.1989)...........................79, 80, 452, 453, 454, 455 Lochner v New York 198 US 45 (1905) .......................................................241 Lugar v Edmondson Oil Co 457 US 922, 102 S Ct. 2744 (1982) ......................................................................................................94 M (Mary) v City of Los Angeles 814 P 2d 1341 (1991) .........452, 453, 454, 463 Maganlal (R) & Co v, MC Chemical Co Inc 942 F 2d 164 (2 Cir.1991)..............................................................................................83 Marcos (In the Estate of Ferdinand), Human, Rights Litigation 978 F 2d 493 (9 Cir. 1992), 25 F 3d 1467 (9 Cir.1994) ........................................127, 410, 626, 661, 673 Marcos Estate II 25 F 3d 1475 (9 Cir.1994) ..........................................661, 672 Marino v INS 537 F 2d 686 (2 Cir.1976) ......................................................569 Martinez-Baca v Suarez-Mason 87–2057 SC (ND Cal.1988) ........................106 Mendaro v World Bank 717 F 2d 610 (DCCir.1983) ....................................466 Meritor Savings Bank v Vinson 477 US 57 (1986) ........................................451 Monroe v Pape 365 US 167 (1960)...............................................................111 Mushikwabo v Barayaguiza, 1996 WL 164496 (SDNY 1996) ...............448, 562 National Coalition Government of the Union of Burma v UNOCAL 176 FRD 329 (DCC Cal., 9Cir.1997) .....................................82, 101, 102, 106, 107, 220 New York Times v Sullivan 376 US 254 (1963)............................................664 Nicaragua v Reagan 859 F 2d 929 (DCCir.1988) ...........................................92 Oetjen v Central Leather Co 246 US 297, 38 S Ct. 309 (1918) ........................81 Osborn v Bank of the United States 22 US, (9 Wheat.) 738 (1824) .................72 Palestine Infrormation Office v Shultz 853 F 2d 932 (DCCir.1988) ...........................................................................................78 Paul v Avril 812 F Supp. 207 (SD Fla.1993) ..........................107, 112, 364, 673 Paull F Newton & Co v Texas Commerce Bank 630 D 2d 111 (5 Cir.1980) .................................................................................451 Pellatier Claim (1887).................................................................................499

xxx Table of Cases Persinger v Islamic Republic of Iran 729 F 2d 835 (DCCir.1984) ..................652 Piper Aircraft Co v Reyno 454 US 235 (1981)................................83, 226, 623 Princz v Federal Republic of Germany 307 US App. DC 102 (1994) ..............................................................................................76, 652 Quiros de Rapaport et al. v Suarez-Mason C87–2266 (ND Cal.1989)........................................................................................106 Rasoulzadeh v Associated Press 574 F Supp. 854 (SDNY 1983) ...........................................................................................83 Republic of Argentina v Weltover Inc 504 US 607 (1992) ............................................................................217, 222, 242, 243 Republic of Philippines v Marcos 806 F 2d 344 (2 Cir.1996) ............72, 73, 363 Republic of Philippines v Marcos 862 F 2d 1355 (9 Cir.1988).......................363 Rios v Marshall 530 F Supp. 351 (SDNY 1981) .............................................70 Saltany v Reagan 702 F Supp. 319 (1988) ....................................................652 Sampson v Germany 975 F Supp. 1108 (ND Ill.1997)...................................652 Saudi Arabia v Nelson, 307 US 349 (1993)..................................221, 234, 237 Schooner-Exchange v McFadden 11 US 116 (1812)......................................249 SEC v Management Dynamics Inc 515 F 2d 1277 (2 Cir.1973) .....................451 Siderman de Blake v Republic of Argentina 965 F 2d, 699 (Ca, 9 Cir.1992) ......................................................71, 72, 79, 91, 92, 102, 106, 267, 268, 410, 652 Skirotes v Florida 313 US 69 (1941).............................................................493 Smith v Libya 886 F Supp. 406 (EDNY 1995) ..............................................652 Sola Electric Co v Jefferson 317 US 173 (1942) ..............................................86 Tel-Oren v Libyan Arab Republic 233 US App. DC 384, 726 F 2d 774 (1984) .........................68, 75, 87, 96, 101, 375, 378, 382, 383, 386, 389, 399, 408, 539, 562, 669, 673, 674 Textile Workers of America v Lincoln Mills 353 US 448, 77 S Ct. 912 (1957) ......................................................................................................87 Todd v Panjaitan CV–92–12255-PBS (D Mass.1994)........................................................106, 112, 113, 114, 506 Trajano v Marcos 878 F 2d 1439 (9 Cir.1989)......................................127, 562 Trajano v Marcos 978 F 2d 493 (1992).........................66, 70, 76, 82, 96, 106 Underhill v Hernandez, 168 US 250 (1897)...................78, 590, 592, 593, 652 Union Carbide Corp Gas Plant Disaster at Bhopal (In re) 809 F 2d 195 (2 Cir.NY 1987) .......................192, 198, 205, 208, 210 US v Matta-Ballesteros 71 F 3d 754 (9 Cir.1996)..........................................102 US v Soemu Toyoda...........................................................................444, 461 Verlinden BV v Central Bank of Nigeria 461 US 480, 103 S Ct. 1962 (1983) ................................................................................67 Von Dardel v USSR 623 F Supp. 246 (1985) ..................................................72 Wilson v Garcia 471 US 261, 85 L Ed. 2d 254, 105 S Ct. 1938 (1985) ..............................................................................103

Table of Cases xxxi Xuncax v Gramajo 886 F Supp. 162 (DC Mass.1 Cir.1995)........................................76, 84, 85, 89, 91, 92, 94, 97, 98, 99, 101, 103, 105, 106, 107, 364, 409, 448, 449, 450, 562, 661, 669, 673, 674, 677, 678 Yosofa Alomang v Freeport McMoran 1996 US Dist. LEXIS 15908...............67 Zapata v Quinn 707 F 2d 691 (2 Cir.1983)...................................................553

Table of Legislation 1. INTERNATIONAL ACP-EEC Lomé Convention 1989 502 African Charter on Human and Peoples’ Rights 667 Art. 5 .....................................................................................................403 21(2) .......................................................................................112, 304 American Convention on Human Rights 116, 403, 518–519 Art. 5 ...............................................................................................96, 403 10....................................................................................................111 45....................................................................................................116 63(1) ...............................................................................................418 Basle Convention on Control of Transboundary Movement of Hazardous Wastes and their Disposal 1989 502 Basle Convention on the Transfer of Hazardous Waste 501 Convention on the Elimination of All Forms of Racial Discrimination ...............................................................................116, 547 Art. 11 ...................................................................................................116 12....................................................................................................116 Convention on the Elimination of Discrimination against Women ...........................................................................................519–520 Art. 3 .....................................................................................................519 17 .....................................................................................................43 Convention on the High Seas 1958 Art. 15 ...................................................................................................128 Convention on the Law of the Sea 1982 44 Art. 99 ...................................................................................................173 110..................................................................................................128 Convention on the Prevention and Punishment of the Crime of Genocide 1948 .............................................................................279 Art. IV ...................................................................................................100 11....................................................................................................100 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment 1984.............6, 8, 12, 13, 28, 63, 97, 98, 109, 116, 137, 139, 159, 175, 176, 231, 251, 252, 253, 255, 256, 258, 261, 262, 263, 264, 265, 267, 269, 270, 271, 272, 273, 275, 278, 280, 283, 286, 287 288, 304, 305, 306, 330, 332, 367, 373, 394, 395, 403, 404, 405, 410, 419, 458–459, 461, 525, 528, 534, 537, 538, 539, 541, 543, 550, 603, 614, 615, 617, 618, 619, 626, 650, 667, 671, App.1

xxxiv Table of Legislation Art. 1 .......................................................33, 88, 251, 261, 262, 276, 285, 288, 459, 541, 644, 646 (1)..................................................................................................404 (2)..........................................................................................404, 463 2.............................................137, 173, 178, 276, 285, 342, 646, 647 (3)..........................................................................................332, 459 Arts. 4–7 ................................................................................................525 4–8 ................................................................................................276 4 .....................................................................138, 147, 167, 277, 619 (1)..................................................................................................110 Arts. 5–7 ................................................................................................138 5–8 ................................................................................................277 Art. 5 .............................................................................167, 281, 282, 648 (2)..........................................................................................110, 261 6 .....................................................................................................110 7 .....................................................................................110, 282, 287 (1)..........................................................................................261, 533 (2)..................................................................................................118 8 .....................................................................................................281 11....................................................................................................542 12...................................................................................................286, 13 ...................................................................................286, 406, 542 14.............................................18, 23, 24, 36, 37, 40, 111, 124, 172, 174, 252, 262, 263, 264, 265, 266, 286, 287, 305, 310, 398, 406, 417, 418, 419, 429, 431, 540, 542, 543, 544, 545–549, 615 15....................................................................................................541 16 .............................................................................28, 459, 542, 647 17....................................................................................................275 20....................................................................................................140 21 ...........................................................................................116, 175 22....................................................................................................175 39(2) ...............................................................................................284 Declaration on the Protection of All Persons from Enforced Disappearance .........................................................................584 Declaration on the Protection of Women and Children in Emergency and Armed Conflict ..............................................................304 Art. 2 .....................................................................................................477 V(18)......................................................................................................488 VI(22) ....................................................................................................488 Geneva Conventions 1949 ..........................................................................473 Prot. 1, Art. 86 .......................................................................................445 87........................................................................................445

Table of Legislation xxxv Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (Geneva I)..............................................................100, 279, 537, 538 Art. 3 .....................................................................................................538 (1)..........................................................................................101, 132 12....................................................................................................538 49....................................................................................................537 Geneva Convention for the Amelioration of the Condition of the Wounded, Sick, and Shipwrecked Members of Armed Forces at Sea (Geneva II).................................100, 279, 537, 538 Art. 3 .....................................................................................................538 (1)..........................................................................................101, 132 12....................................................................................................538 50....................................................................................................537 Geneva Conventions for the Protection of War Victims 1949 .......................................................................................................172 Geneva Convention on Refugees Art. 1F(c) ...............................................................................................478 Geneva Convention Relating to the Protection of Victims of Non-international Armed Conflicts 1949 Additional Prot. .............................................................................101, 132 Geneva Convention Relative to the Protection of Civilian Persons in Time of War (Geneva IV) ......................100, 279, 304, 537, 538 Art. 3 .....................................................................................................538 (1)..........................................................................................101, 132 146..................................................................................................537 147..................................................................................................538 Geneva Convention Relative to the Treatment of Prisoners of War (Geneva III).................................................100, 279, 304, 537, 538 Art. 3 .....................................................................................................538 (1)..........................................................................................101, 132 17....................................................................................................538 129..................................................................................................537 Hague Convention on Private International Law ..........................................52 Hague Convention on Traffic Accidents .....................................................303 ILO Convention (No. 105) Concerning the Abolition of Forced Labour 1957................................................................................159 ILO Convention (No. 29) on Forced or Compulsory Labour 1930 .......................................................................................................159 Inter-American Convention to Prevent and Punish Torture.........304, 404, 667 Inter-American Convention on the Prevention, Punishment and Eradication of Violence against Women ..................................................................................................519 Art. 1 .....................................................................................................519

xxxvi Table of Legislation International Covenant on Civil and Political Rights (ICCPR)............................................................6, 36, 58, 88, 90, 92, 116, 124, 159, 175, 176, 250, 251, 304, 378, 389, 392, 394, 395, 396, 402, 516, 517, 527, 588, 615, 618, 625, 667, 682 Art. 2 .....................................................................................................504 (3)..................................................................................120, 178, 588 3 .............................................................................................480, 504 4 .............................................................................................304, 403 6 .......................................................................................96, 517, 615 7 ....................................................................137, 296, 304, 373, 402, 403, 404, 616, 617 9(3) .................................................................................................111 14 ...................................................................................589, 616, 618 (1)..................................................................................................526 41....................................................................................................116 Optional Prot.................................................................................175, 177 Optional Prot. 2 .....................................................................................617 North American Agreement on Environmental Co-operation (NAAEC)...............................................................................................433 North American Agreement on Labour Co-operation (NAALC)...............................................................................................433 North American Free Trade Agreement 243 Chap., 11...............................................................................................243 Rome Treaty on the World Court 1998 ......................................................575 Slavery Convention, 1926 ..........................................................................159 Space Objects Convention..........................................................................202 Supplementary Convention on the Abolition of Slavery and the Slave Trade and Institutions and Practices Similar to Slavery 1957 ...........................................................................159 Treaty of Guarantee 1960...........................................................................487 Treaty of Lausanne 1923 ............................................................................278 Treaty of Sèvres 1920 .................................................................................278 Treaty of Versailles 1919 Arts. 228–230 .................................................................................278, 467 UN Charter .......................................................................................230, 479 Chap. VII...............................................................................469, 470, 481 Art. 1 .....................................................................................................474 (3)..................................................................................................478 103..................................................................................................477 104..................................................................................................477 105..........................................................................................477, 481 UN Convention on the Rights of the Child...................................37, 256, 425

Table of Legislation xxxvii Art. 3(1) .................................................................................................257 Universal Declaration on Human Rights ......................92, 250, 303, 402, 667 Art.3 ........................................................................................................96 5 ......................................................................................137, 373, 402 10 ....................................................................................................479 Universal Islamic Declaration of Human Rights .........................................304 Art. VII ..................................................................................................607 Vienna Convention on Diplomatic Relations Art. 39(2) .......................................................................................283, 284 Vienna Declaration on Human Rights 1993 ................................................250 Vienna Convention on the Law of Treaties 1969.........................................272 Art. 27 ...................................................................................................525 30....................................................................................................477 31 ...........................................................259, 262, 263, 264, 265, 282 32 ...........................................................................................262, 285 53 .............................................................91, 253, 272, 302, 378, 405 64 ...........................................................................................173, 598 2. EUROPEAN European Convention on Human Rights and Fundamental Freedoms ..............................................................145, 304, 340, 341, 403, 513, 522, 675, 681 Preamble ................................................................................................531 Art. 1 .....................................................................................................529 2 .............................................................................522, 523, 524, 526 3...............................................99, 143, 144, 146, 339, 373, 403, 524 5(5) .................................................................................................111 6 ...............................................................................................23, 479 (1) ....................................22, 523, 524, 526, 527, 528, 529–531, 534 8(2) .................................................................................................533 13 .............................................................................23, 480, 526, 532 (2)..................................................................................................403 19 ...........................................................................................531, 534 34....................................................................................................340 50 .....................................................................................................58 53 .....................................................................................................58 Prot. 1, Art. 3 .........................................................................................486 3 .....................................................................................................304 5 ......................................................................................................304 11 ....................................................................................................532 European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment ............................................................139, 140, 149, 404, 667

xxxviii Table of Legislation Art. 2 .............................................................................................139, 140 3 .....................................................................................................140 7(1) .................................................................................................141 8 .....................................................................................................140 10(2) ...............................................................................................142 Prot. 1 ....................................................................................................141 European Convention on State Immunity and Additional Protocol 1972 .........................................................................................215 3. EUROPEAN COMMUNITIES EC Treaty Art. 5 .....................................................................................................553 48....................................................................................................688 EEC Convention on the Enforcement of Judgments in Civil and Commercial Matters 1968 (Brussels Convention)......................336, 340 Art. 2 .....................................................................................................120 3 .....................................................................................................120 E.C. Convention on the Enforcement of Judgments in Civil and Commercial Matters (Lugano Convention) .....................................................................336, 340 4. NATIONAL Australia Crimes Act 1914.........................................................................613, 621, 623 s.5 ..........................................................................................................614 74 ........................................................................................................614 Crimes (Torture) Act 1988 ..................................603, 614–620, 623, 626, 628 s.6..................................................................................................614, 626 7 ..........................................................................................................626 Foreign State Immunities Act 1985 .............................................215, 625, 626 s.9 ..........................................................................................................625 13 ........................................................................................................625 Statutes Amendment (Capital Punishment Abolition) Act 1976 (S. Australia) ...............................................................602, 613, 621 s.4 ..........................................................................................................613 Supreme Court Act 1935 (S. Australia) s.29 ........................................................................................................620 30 ........................................................................................................620 31 ........................................................................................................611

Table of Legislation xxxix Canada Automobile Insurance Act 1977 ss. 3–4 ....................................................................................................297 Charter of Rights and Freedoms .........299, 387, 391, 411 421, 423, 633, 664 s.1............................................................................................427, 636 637 ss. 7–14 ..................................................................................................428 s.7..........................................................................180, 424, 427, 636, 637 10(b)....................................................................................................168 12 ................................................................................................424, 426 14 ........................................................................................................426 Compensation for Victims of Crime Act 1990 305 s.5 ..........................................................................................................305 16(1) ....................................................................................................305 Constitution Act 1982 Part 1 .............................................................................................299, 421 Courts of Justice Act 1990 s.11(2) ....................................................................................................182 103 ......................................................................................................188 106 ..............................................................................................182, 189 Criminal Code .....................................................3, 168 , 169, 170, 305, 312 s.7 ...............................................................................................117, 180 737 ......................................................................................................305 Crown Liability and Proceedings Act 1985 .............................38, 39, 130, 132 s.32............................................................................................38, 40, 132 Family Law Act 1990 s.61(4) ....................................................................................................132 Geneva Conventions Act 1965 ....................................................................132 Human Rights Code ..................................................................................415 Immigration Act ........................................................................................388 s.114(2) ..................................................................................................256 Income Tax Act 1985 .................................................................................413 National Defence Act 1985 s.7............................................................................................................39 269(1) ............................................................................................39, 132 Ontario Human Rights Code .............................................................390, 415 Public Authorities Protection Act 1990 .........................................................38 s.7 .............................................................................................38, 39, 132 State Immunity Act 1982 ....................................................129, 215, 255, 258 s.2 ..........................................................................................................217 3 ..........................................................................................................254 4..................................................................................................219, 267 5..................................................................................................215, 254 6..........................................................................................216, 254, 453

xl Table of Legislation Trustee Act 1990 s.38(3) ....................................................................................................132 Yukon Torture Prohibition Act..................................................417, 418, 431 s.3 ..........................................................................................................418 Chile Law 18342 Art. 5(3) .................................................................................................570 Legislative Decree 2191 ..............................................................................570 Art. 1 .....................................................................................................571 2 .....................................................................................................571 France Code of Criminal Procedure Art. 1 .....................................................................................................528 689..................................................................................................528 Germany Bundesgesetzbuch para. 133 ................................................................................................696 138 ................................................................................................696 157 ................................................................................................696 Grundgesetz ......................................................................698, 699, 701, 702 Art. 1.I ...................................................................................................695 1.III.................................................................................................692 India Bhopal Act.................................................................................................208 Israel Basic Law:, Human Dignity and Liberty ...........................632, 633, 634, 635, 637, 642, 646, 655 s.2..................................................................................................634, 635 3..................................................................................................634, 635 4 ..........................................................................................................634 5 ..........................................................................................................635 8..................................................................................................637, 642 Civil Wrongs (Liability of the State) Law....................................................645 s.2 ..........................................................................................................645 3 ..........................................................................................................645 5 ..........................................................................................................645 7 ..........................................................................................................645

Table of Legislation xli Civil Wrongs Ordinance 1972.....................................................................645 Criminal Procedure Statute (Testimony).....................................................638 Art. 2(1) .................................................................................................638 Foreign Judgments Enforcement Law 1958 .................................................654 Penal Code ................................................................................641, 645, 657 Saudi Arabia Basic System ..............................................................................................607 Art. 7 .....................................................................................................606 8 .....................................................................................................606 23 ...................................................................................................606 26 ...................................................................................................607 48 ...................................................................................................606 South Africa Constitution s.8 ..........................................................................................................514 36 ........................................................................................................514 Indemnity Amendment Act 1992 ................................................................582 Indemnity Act 1990 ....................................................................................582 Switzerland Constitution Art. 33 ...................................................................................................514 United Kingdom Bill of Rights 1689......................................................................................613 Canada Act 1982........................................................................................633 Sched. B .................................................................................................299 Common Law Procedure Act 1852 .............................................................161 Constitution Act 1867 s.92 ........................................................................................................180 Criminal Justice Act 1988...........................................................................644 s.134 ..............................................................................335, 644, 647, 648 (1) ....................................................................................................284 135 ......................................................................................................524 Foreign Limitation Period Act 1984 s.1(1)......................................................................................................334 Human Rights Act 1998...............................................15, 339, 340, 341, 342 s.3 ..........................................................................................................341 6..................................................................................................339, 340 7..................................................................................................339, 340

xlii Table of Legislation Private International Law Act ......................................................................15 Private International Law (Miscellaneous Provisions) Act 1995 .........................................................123 226, 294, 319, 321, 323, 329, 335, 337, 597, App.3 s.9..................................................................................................320, 327 10 ........................................................................................................597 11 ................................................................................................319, 338 12 ........................................................................................................320 14 ................................................................................................320, 338 Protection from Harassment Act 1997 s.3 ..........................................................................................................332 Sex Disqualification (Removal) Act 1919 ....................................................416 State Immunity Act 1978 ...................................................188, 215, 267, 270, 272, 293, 337, 341 s.1(1)......................................................................................................269 2..................................................................................................267, 368 (2)......................................................................................................219 3(1)..............................................................................215, 217, 218, 219 (2)......................................................................................................218 (3) .............................................................................217, 219, 242, 243 5 ..........................................................................................................216 Statute of Edward 1354 ..............................................................................613 United States Alien Tort Claims Act 1789 ...............3, 4, 7, 14, 16, 19, 25, 27, 28, 30, 31, 32, 63, 66, 67, 68, 69, 70, 73, 74–76, 86, 90, 91, 97, 99, 100, 102, 103, 107, 110, 112, 113, 115, 118, 126–127, 130, 133, 220, 247, 291, 318, 374, 379, 380, 383, 389, 407, 408, 409, 410, 418, 448, 506, 538, 539, 560, 575, 590, 615, 661, 670, 675, 679, App.2 Alien Torts Act ..........................................................................................203 Civil Rights Act .........................................................................................451 Constitution ..............................................................................................671 Art. III ............................................................................................67, 69, 73 Foreign Corrupt Practices Act ....................................................................508 Foreign Sovereign Immunity Act ............67, 70–72, 79, 84, 86, 96, 215, 265, 267, 268, 269, 453 s.1603.....................................................................................................217 1605 ............................................................215, 216, 219, 265, 267, 452 Judiciary Act 1789 .......................................................................................74 Immigration and Nationality Act .................................................................75 Racketeer Influenced and Corrupt Organizations Act .................................362 Securities Exchange Act 1934 s.10(6) ....................................................................................................450

Table of Legislation xliii Torture Victim Protection Act ...............3, 7, 41, 63, 68, 74, 76, 84, 86, 90, 92, 97, 103, 104, 105, 110, 112, 113, 115, 118, 126–127, 130, 133, 247, 273, 291, 318, 383, 410, 418, 506, 538, 539, 543, 546, 560, 575, 590, 615, 661, 675, 679, App.2 s.2 ............................................................................................74, 84, 104, 8............................................................................................................96 US Code Annotated Title 28 (Judiciary and Judicial Procedure) §1331 ................................................................................72–73 USS §1350 ................................................................................86, 87, 95–102 USS §1983..............................................................................................93, 94 28 USC §1332...............................................................................................73

1

Introduction to Torture as Tort: From Sudan to Canada to Somalia CRAIG SCOTT 1

1 BEGINNINGS H E P R E S E N T V O L U M E began in a somewhat unusual way in the winter of 1998. Members of the Sudanese community living in exile outside Sudan approached me to ask about possible avenues within Canada for seeking justice against two members of the Cabinet of the Government of Sudan. Both men were also said to be Canadian citizens, their citizenship having been acquired following time spent as graduate students in Canada. One avenue was to alert the Federal Government of Canada about the possibility that criminal proceedings might need to be considered on the basis of crimes in Canada’s Criminal Code which attract universal jurisdiction. The other avenue that needed to be explored was that of bringing a civil action in Canadian courts against either man, and possibly against the state of Sudan itself, for human-rights-violating conduct in Sudan. As with most international lawyers who work in the human rights field, I was familiar with the unique phenomenon of transnational human rights litigation that has been developed in United States federal courts since the time of the precedent-setting Filártiga case brought under the Alien Tort Claims Act (ATCA).2 Bracketing for the moment the desirability of this paradigm of human rights protection, it seemed clear that a thorough study needed to be done of the myriad issues that arise in such actions in order to assess the potential for such an action to be brought successfully in Canada.3 The question of employing

T

1

Osgoode Hall Law School of York University, Toronto Filártiga v. Peña-Irala, 630 F 2d 876 (CA, 2d Cir. 1980). The case involved a tort action brought in US courts against a former Paraguayan police officer by the family of a Paraguayan who had died under torture in police custody in Paraguay. Alien Tort Claims Act (of 1789), 28 USC §1350, [“ATCA”] since supplemented by the Torture Victim Protection Act, also consolidated at 28 USC §1350 [“TVPA”]. 3 Only two cases seem to have been brought in Commonwealth courts where the underlying conduct of the alleged tort was torture in another country, and neither case proceeded on the basis of the human right not to be tortured as being the direct cause of action: Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. Canada (Attorney General) (Unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J.); Al-Adsani v. Government of Kuwait and Others (1996) 107 ILR 536 (Eng., CA). 2

4 C Scott “private law” to vindicate “public law” norms is an approach that has met no small degree of judicial resistance in Canada with respect to Canada’s own public law human rights norms.4 The doctrinal and theoretical issues involved with bringing the public international law of human rights into Canadian law through tort law would present at least as many hurdles—and likely more. At the time I was teaching at the University of Toronto Faculty of Law. It so happened that I was approached with the Sudan query just as winter term was beginning. While I was not teaching a course from which I could draw students, a colleague was. Robert Howse (now of the University of Michigan) was teaching a course called International Law II (Private International Law) and readily agreed to permit a team of LL.B. students to work with me, for credit, on the various “interface” issues that linked conflict of laws to the question of transnational human rights torts. The papers produced by the end of that term were of such a quality and undoubted value that I felt it was important that this work see the light of day in some academic venue. We had the beginnings of this book. But, it seemed myopic to limit the catchment area to Canada and, so, the book project evolved into an effort to bring together a group of non-American scholars to reflect on a variety of dimensions of the American phenomenon. At the most general level, the animating question for any given author was whether such human-rightsbased actions were either already possible in other legal systems, primarily Commonwealth common law jurisdictions, or, whether or not possible, desirable. In this collaboration, the papers of the student research team became “foundation papers” for the scholars who agreed to contribute to the volume.5 The original hope was that the foundation papers would help pave the way for several scholars from fields other than public or private international law, most notably tort theorists (both common law and civilian), to feel sufficiently comfortable with the issues to agree to contribute. In the end, this ideal proved premature as a significant proportion of the private law theorists approached to contribute were not actively working on the nexus of tort law with public law values. While generally intrigued by the subject, most were reluctant to dive into what seemed to them a normative minefield given that comparative law, private international law, public international law, and constitutional law—and not tort law—were the prevailing discourses in the literature generated in the United States around the ATCA phenomenon. To a few others, the inquiry of the planned volume seemed like much ado about nothing; either because tort law exists by definition in some near-pure non-national space or because moral rights were seen as so clearly the foundation of tort law no less than “human rights law”, these tort scholars tended to see any barriers to being able to sue for personal injury caused by torture wherever conducted in the world as merely 4

See the discussions on this point in Hyland, infra note 33, and Raponi, infra note 32. These foundation papers have evolved into chapters in this volume: see Swan, infra note 17; McConville, infra note 21; Orange, infra note 28; Bühler, infra note 31; Raponi, infra note 32; Hyland, infra note 33; and Flah (going on to co-author with Oosterveld), infra note 34. 5

Introduction 5 “jurisdictional” or “procedural”—and thus conceptually uninteresting from a substantive tort law perspective.6 As such, the initial ambition that this book would include not just the comparative perspectives of private and public international lawyers from different national and legal backgrounds but also the analysis of a wide range of “domestic” tort scholars has had to be postponed. That said, the central objective of the book remains unchanged: to generate analysis, debate and dialogue in policy and academic communities, both national and transnational, over the question of the feasibility and legitimacy of transnational human rights litigation.7 It is very much hoped that tort theorists, both civilian and common law, will find in this book something to sink their teeth into by way of both specific response and their own future work.8 Undoubtedly such consideration will either dovetail with, or parallel, much of the debate generated by the recent Pinochet case, discussed by a number of authors in this volume.9 A case primarily about universal jurisdiction in the criminal law sphere (and immunity therefrom), Pinochet resulted in general objections to universal justice through the vehicle of foreign courts, reflecting concerns that will almost certainly resurface should the US model begin to be taken up by courts and legislatures in other states.10 6 A third group were keen to contribute, but could not free up the time to be able to make a commitment. 7 And including in the United States itself where a volume devoted to “external” scrutiny of an American phenomenon may well help clarify or redirect analysis. 8 This is not to say that tort theory is absent from the volume; indeed, the book closes with chapters by two leading younger-generation tort theorists, one from both sides of the Atlantic with one from a common law system and the other from a civil law system. See Moran and Gerstenberg, infra notes 49 and 51, respectively, and further mention, infra, this chapter. It is hoped—even expected— that the discussions of judicial and scholarly discourse found in the Moran and Gerstenberg chapters will provide the bridge to tort theory. Note also the engagement with the foundations of tort law in the Raponi and Hyland chapters, infra notes 32 and 33, respectively, as well as less extensive and more indirect discussions in a good number of the other chapters. And note that Gerstenberg’s chapter, the final in the book, will additionally provide an entry-point for civilian law scholars whether they be public or private lawyers. 9 See especially the final of three House of Lords decisions: Regina v. Bartle and the Commissioner of Police for the Metropolis and others, ex parte Pinochet (No. 3), [1999] 2 All ER 97; also (1999) 39 ILM 581. 10 In the criminal law context, the clear legal duty to exercise universal jurisdiction, and not just the power to do so, meant that many of the objections made in the Pinochet affair were objections on grounds of legitimacy and not, or less so, objections on grounds of legality: for a discussion of the undesirability of British and Spanish courts trying Pinochet, see the nuanced article co-authored by Ricardo Lagos, the opponent of Pinochet who was elected as President of Chile while Pinochet was still under house detention in the UK: see R Lagos and H Muñoz, “The Pinochet Dilemma,” (1999) Foreign Policy 26. On the interaction of international criminal law and transnational civil law, it is interesting to note that one of the subjects of the original Sudan query has declined to come to Canada because of leaks that let it be known that the War Crimes Unit of the Canadian Department of Justice was actively investigating allegations of his involvement in a range of practices in the Sudan, including torture and slavery. While arrest in and extradition from a third country remain a possibility, especially should the current Sudanese regime fall, the ability to avoid criminal law prosecution in common law countries (where jurisdiction over the person requires the physical presence of the accused before criminal proceedings can move forward) points to the possible role of civil law remedies as filling the breach when criminal law is stymied. This is because courts can sometimes assume jurisdiction over civil suits by permitting service of process on a defendant who

6 C Scott Several final observations should be made. First of all, the present volume does not hold itself out as exhaustively covering every issue relevant even from standard public international or private international disciplinary perspectives. While the volume’s coverage of issues does border on the comprehensive, there are some matters that do not receive sustained scrutiny, for example diplomatic immunity. Secondly, the main title of the book—Torture as Tort—warrants comment. In order to provide a focus and also in order to make a meaningful contribution to the literature on the protection of one of the key human rights, the prohibition of torture was selected as both a focus and as an exemplar for other core human rights. By virtue of that choice, many of the concrete discussions in the book necessarily relate to the specificities of the emerging law on torture as set out primarily in the Convention against Torture (CAT) and the International Covenant on Civil and Political Rights (ICCPR).11 That being said, it is expected that the bulk of the discussion will be relevant to all norms that are plausibly characterizable as jus cogens (peremptory) norms of international human rights law and not simply relevant to rules and principles applicable to torture.12 However, readers should be aware that, even with respect to a “single” norm such as the prohibition on torture or the illegality of slavery, there are a range of factors that can affect determinations on both jurisdiction and applicable law: (a) different actors in terms of kind and degree of connection to the territorial jurisdiction of the court; (b) different kinds of conduct, including acts and omissions; and (c) different degrees of severity of infringement of the relevant norms. These and other variables may have considerable relevance to the persuasiveness of the justifications, both legal and philosophical, for a given transnational human rights action. The primary purpose of the remainder of this Introduction is to provide an overview sense of the topics covered and arguments developed in the 25 chapters that follow the present one. The title of each sub-section corresponds to that of each Part in the book. Once this narrative is completed, an additional subsection seeks to place the normative vistas opened up by the volume’s chapters into a concrete context. Sub-section 7 takes the opportunity to comment on some of the problems with the reasoning of the judge of first instance in the first tort case to come before a Canadian court that involves allegations of torture in another country, Arone v. Canada.13 is outside the jurisdiction and proceed to hear the case even if that person refuses to take part in the trial: see, inter alia, Terry, infra note 18, on the relative virtues of criminal law and civil law as avenues to vindicate human rights and McConville, infra note 21, on the conditions for ex juris service. 11 International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, GA Res. 2200 (XXI), UN GAOR, 21st Sess., Supp. No. 16, UN Doc. A/6316 (1966), 999 UNTS 171 (entered into force 23 Mar. 1976)[“ICCPR”]; Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, GA Res 39/46, UN GAOR, 39th Sess., Supp. No 51, UN Doc. A/39/51 (1985), 1465 UNTS 85 (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 (“CAT”). 12 Including the more severe end of the adjunct right, that of the prohibition of “cruel, inhuman or degrading treatment or punishment” that does not amount to “torture.” 13 Arone, supra note 3.

Introduction 7

2 FRAMES AND FOUNDATIONS

The team of contributors who prepared this volume’s foundation papers were asked to approach their slice of the questions surrounding transnational human rights litigation by having regard to the conceptual barriers existing theories and legal doctrine may create to the reception of international human rights norms into domestic legal processes as the foundation of civil law causes of action. What does it take for (internationally-prohibited) torture to be transformed into (domestically-sanctioned) tort? A series of dichotomies that structure juridical thinking to varying degrees—such as those between public law and private law, between international law and domestic law, between “horizontal” application of human rights norms and “vertical” application, between public international law and private international law, between the power to regulate and the duty to regulate, and so on—were treated as “problems of translation” between legal categories and fields. In order to provide the reader of the present volume with a frame of reference for understanding the degrees of normative resistance that may exist to the translation of torture norms into tort norms, the second chapter by the present editor provides a kind of narrative of the translation issues, ending by borrowing from the literary context to suggest a notion of “mutual translation” as the best way to understand the negotiation of the space between torture and tort.14 Chapter 2 situates its otherwise-abstract discussion in a concrete focus on the accountability of a particular kind of non-state actor, the corporation and especially the multinational corporation.15 As such, this chapter seeks to make an independent contribution to the evolving analysis of human rights obligations of transnational corporations as well as to provide a lead-in to three further chapters in the volume which focus on civil law remedies against corporations, those by Upendra Baxi, Robert Wai, and Muthu Sornarajah.16 Having thus set up some of the conceptual debates that surround treating torture as tort, Part I of the volume moves on to an overview study by Michael Swan who surveys the state of play in the United States with respect to how US courts have handled claims grounded in international human rights norms.17 For those readers unfamiliar with the experience to date under the 18th-century ATCA and the modern Torture Victim Protection Act (TVPA), the Swan chapter is an essential primer. Its intended audience is non-American, with its 14 C Scott, “Translating Torture into Tort: Conceptual Divides in the Debate on Corporate Accountability for Human Rights Harms,” chapter 2. 15 Use of the corporate accountability context has the additional benefit of helping make clear that the field of transnational human rights litigation need not be an all-or-nothing question of “pure” universal civil jurisdiction—that is, foreign nationals suing foreign nationals for human rights violations that occurred in a country other than that of the court asked to hear the case—but can also be one way of conceptualising a special form of civil law accountability of nationals for their conduct abroad. 16 Baxi, infra note 23; Wai, infra note 24; and Sornarajah, infra note 37. 17 M Swan, “International Human Rights Tort Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines,” chapter 3.

8 C Scott selection of topics, introductory nature and over-all organisation being designed to help set the scene for references to the US phenomenon by various authors in their chapters. With these two framework chapters in place, Part I ends with a pair of pieces containing contrasting, although not firmly opposed, perspectives on the role of tort law in protecting international human rights. John Terry discusses tort claims as “third country legal actions” that can play an integral role in providing remedies for conduct that violates international human rights law.18 In arguing for the special virtues of the tort remedy (and the availability of this remedy in most common law jurisdictions without further need for legislative intervention), he makes a case for its superiority to universal criminal law enforcement in important respects. In the course of making this case, he points out, inter alia, that civil actions may provide important symbolic victories even if actual compensation is unlikely and that civil law gives a special voice to actual victims of human rights harms in a world where states’ interests in pursuing criminal law prosecutions is hedged and sporadic. In terms of the translation question, Terry also makes an argument for doctrinal continuity and moderation when he says that existing domestic tort law and private international law are adequate for vindicating international human rights values without necessarily needing to go the next step and make human rights and associated universal jurisdiction the direct bases for the civil actions in question. A special contribution of the Terry chapter is its discussion of the action brought against Canada by the family of a young Somali man, Shidane Arone, for Arone’s torture-death in Somalia at the hands of Canadian soldiers operating under the auspices of a UN military mission.19 Malcolm Evans and Rod Morgan are less convinced about the ability of transnational tort actions to effectively contribute to the eradication of the practice of torture.20 While accepting the compensatory goals of tort law, the authors see tort actions as involving de facto punishment objectives as much as restitutionary ones. As such, a universal civil law remedy is seen as part and parcel of an after-the-fact legal approach that it shares with criminal law prosecution, a reactive approach they contrast to a focus on prevention. Of course, each area of individual legal accountability, civil liability and criminal responsibility, have as part of its rationale the deterrence of future—both specific and general deterrence. However, it is here that Evans and Morgan make the case for a pride of place to be accorded to preventive mechanisms and then explain why the positive aspects of civil redress need to be carefully considered alongside the potential for civil liability before foreign courts to actually undermine preventive mechanisms. By drawing on the extensive experience of the European Committee on the Prevention of Torture, Evans and Morgan suggest how it is 18 J Terry, “Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad,” chapter 4. 19 Arone, supra note 3. 20 M Evans and R Morgan, “Torture: Prevention versus Punishment?”, chapter 5.

Introduction 9 that an increased risk of legal action may result in increased secrecy and unwillingness for state officials to cooperate with international prevention-oriented processes. While acknowledging that a nuanced approach that harmonizes transnational tort actions and international preventive mechanisms may be possible, Evans and Morgan conclude that it would be a mistake to proceed as if there is not, at present, a tension between the two approaches to protection against torture.

3 JURISDICTION AND IMMUNITY

In standard private international analysis, the first question one must ask is whether a court before which a claim has been brought may, or even must, hear the claim, and then whether a court having jurisdiction may decline to exercise it. The chapters in Part II deal with that question in several dimensions. Anne McConville comprehensively examines how principles of private international law interact with principles of public international law within Canada’s current combined statutory and common law framework for the assumption of jurisdiction by courts in tort cases which have a foreign element. Her specific focus is Ontario which is Canada’s largest jurisdiction.21 Much of the Ontario jurisdictional framework is identical or similar to that prevailing in other non-American common law jurisdictions, such that her survey and analysis of the Ontario state of affairs will have a broader relevance. She concludes that these rules provide sufficient scope for plaintiffs resident in Ontario to bring transnational human rights actions, but takes the view that a territory-centred and comity-oriented theory of private international law structures the Supreme Court of Canada’s “new” conflicts jurisprudence to such an extent that there may be considerable judicial resistance to reading the existing rules in a facilitative way.22 Especially important is the case McConville makes on three fronts for how public international law should inform the “domestic” interpretation of jurisdictional principles in a conflict of laws situation. First of all, she notes that the residual power of Ontario judges to grant leave for ex juris service (service outside of a jurisdiction, here Ontario) can dovetail with the universal prescriptive jurisdiction that adheres to the most fundamental human rights norms protected by international law. Secondly, she outlines the way in which the doctrine of forum non conveniens (according to which judges may decline to exercise a jurisdiction they otherwise have) shares rationales with the doctrine of exhaustion of local remedies, a procedural condition for individuals to access 21 A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality,” chapter 6. 22 The focus on an Ontario resident is less in terms of this being a formal condition for assumption of jurisdiction and more in terms of it being an important factor when judges have to decide whether to stay proceedings on grounds of forum non conveniens in those cases where this doctrine is applicable (i.e. where justice can be done in a foreign court system).

10 C Scott international human rights mechanisms through individual petitions claiming rights violations by their states. By drawing on the extensive international jurisprudence as to when local remedies are both existing and effective, courts in transnational court cases have a solid doctrinal basis, as well as greater legitimacy, in finding that a foreign court system does not get over the threshold in forum non conveniens analysis of showing there is an alternative forum capable of dispensing justice. And thirdly, she notes the especially powerful claim supporting Canadian courts taking jurisdiction based on the Canadian nationality of the perpetrator of the alleged human rights violations. Quite apart from universal jurisdiction—whereby any person can come to Canada and sue any other person for violation of certain human rights norms wherever committed—there arises the question of the special onus on home states to have their courts hear human-rights-based torts brought either against the state itself or the state’s nationals. Upendra Baxi’s discussion picks up on this question of the special place of nationality jurisdiction by focussing on those nationals whose activities abroad present the most pressing need for access to justice for victims of human rights abuses, namely corporations.23 The Baxi chapter offers a stinging critique of the stasis and myopia that dominates the academy, and the judiciary, in the private international law realm. Beyond his general account of the non-receptivity of much of conflict of laws to human rights values, he picks up on McConville’s detailed discussion of the availability of the forum non conveniens doctrine to defendants seeking to have cases removed from the jurisdiction of Canadian courts. Drawing on extensive previous work on the invocation of forum non conveniens by multinational corporations in transnational litigation, from the Bhopal mass deaths in the mid-1980s through various cases in the 1990s, Baxi discusses forum non conveniens doctrine, especially American doctrine, in terms of the role it can, and often does, play in upholding a dehumanising form of capitalism according to which the risk of harms caused by corporate activities is borne by Southern societies least equipped, from a regulatory and legalsystem perspective, to prevent and seek compensation for those harms. One important reason forum non conveniens doctrine is so susceptible to being employed to remove human-rights-related suits against MNCs from Northern courts is the ease with which judges in home states of MNCs persist in thinking in highly parochial terms about whether the home state has an interest in being the venue for global justice against its national MNCs. In contrast, Baxi makes the case for aligning forum non conveniens doctrine with an ethical discourse of home state (and home society) responsibility to provide judicial forums for 23 U Baxi, “Geographies of Injustice: Human Rights at the Altar of Convenience,” chapter 7. Note that, whatever the nature of disparate holdings and global activities, Baxi implicitly works from the premise that multinational corporations still have, and will continue to have, home states or, indeed, sometimes several home states if we consider not only the state of formal nationality of the parent company but also those states whose communities benefit from the externalisation of costs of a corporation’s global activities.

Introduction 11 claims brought against MNCs. Yet, Baxi’s analysis hovers between the view that human rights discourse can produce some kind of “redemptive transformation” of private international law and the concern that it is a field of law that has proven to be remarkably impervious to making ethical discourse part of its own being. Baxi’s analysis of how the political economy of private international law resists transnational corporate accountability for human rights violations dovetails with the chapter by Robert Wai who examines immunity from jurisdiction, a doctrine which lies at the intersection of public and private international law.24 The starting-point for his discussion is the fact that domestic legislation invariably treats foreign states as generally immune from suit in courts of other states. This includes, on the face of most statutes, an immunity from suit for personal injury or death torts if such torts occur outside the state where the court in question is located. In contrast to this situation regarding torts, states do not benefit from immunity where a suit relates to “commercial activity” of the state or its agents. Wai’s point of departure then is the sharp distinction drawn in the law of state immunity between commercial and non-commercial activity, with his purpose being to lay bare and query both the policy justifications for such a distinction and the ideologies of (global) political economy that underpin immunity doctrine. Two questions may be asked in this respect. First of all, why should commercial claims be treated more favourably than claims relating to personal harm, especially when the harm is also characterizable as a human rights violation? Secondly, may not the commercial activity exception provide at least one normative opening for the civil liability of foreign states to the extent that there is increasing recognition (and perhaps also incidence) of human rights violations which occurring in the course of, or in support of, commercial ventures in which states are either the sole actor, a joint partner, or a closelyinvolved regulator?25 In approaching these questions, Wai demonstrates that the preferential treatment of plaintiffs suing states with respect to “commercial activity” is grounded in understandings about state consent stemming from contract law’s imbrication in commercial matters as well as theories of the international co-operative benefit achieved in transnational economic relations. He reveals those premises to be ideological constructs which do not support the commercial/non-commercial distinction. However, in so doing, he is careful to be sceptical about the view that the appropriate response (whether through legislation or judicial interpretation) is to equalize the situation by removing immunity for human rights torts as well as commercial claims. Instead, the incoherence of the distinction points to the need to revisit the purpose of immunity more generally and, in the process, take a hard look at whether removal of 24 R Wai, “The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legalism,” chapter 8. 25 Here, the record of the failure to employ this exception in a human-rights-oriented way offers further evidence of Baxi’s thesis of the pervasive impact of political economy and judicial ideology on the actual workings of private international law.

12 C Scott immunity contributes to a problematic form of legalism centred on court-based adjudication—a form of legalism which, when internationalised, tends to harness itself to a broader substantive agenda of international neoliberalism pushing for more and more delegitimation of an active state. Could the removal of state immunity for the most egregious kinds of human rights violations become the opening for a broader human-rights-validated neoimperialism in which not just economic but also social and political regulation become the concern of foreign courts when they should, rather, be left to the domestic processes of state-communities around the world? Wendy Adams’ chapter represents an interesting spin on this question of national self-determination as it relates to the operation of state immunity doctrine in transnational litigation.26 Adams’ task is to analyse how the assumed jus cogens (peremptory law) status of the international prohibition of torture should interact with the law of state immunity as set out in states’ immunity statutes. In the result, the key doctrinal question is: should domestic court judges read an implied exception into immunity statutes such that foreign states are not immune from suits grounded in violations of jus cogens norms, in statutory contexts where it is clear the legislature specifically provided for some exceptions (such as the commercial exception) but not this one? In answering this question, Adams argues that two theories of legitimacy are needed in order to provide a baseline for understanding how activist domestic judges in Westminster systems should be, a theory justifying one state’s courts judging other states for the latter states’ conduct within their own borders and a theory of the proper role of judges versus legislatures in taking initiatives in the name of international law. Precisely because of the interventionist nature of an immunity-disregarding interpretation based on domestic law reception of jus cogens norms, Adams reasons that the lack of clear legislative authorization speaks against judges forging such an exception to immunity. When this reason is combined with her view of the limits, in Canadian jurisprudence, of the principle that statutes should be interpreted in conformity with international law (here, in relation to what normative use can be made of the Convention against Torture), Adams concludes that it is up to the legislatures of Canada, and those of other Commonwealth jurisdictions, to reconcile the various tensions in play by amending immunity statutes to allow for foreign states to be sued for violations of jus cogens norms. Peter Burns and Sean McBurney tackle the issue of jurisdictional bars to civil suits for torture in terms of whether a state violates the Convention against Torture by refusing to either extradite or prosecute a person accused of torture (for whom there is adequate evidence of guilt) on grounds of either immunity under international law or national amnesty having been granted by the state where the torture occurred. They present their analysis as applying, in all sub26 W Adams, “In Search of a Defence of the Transnational Human Rights Paradigm (May Jus Cogens Norms Be Invoked to Create Implied Exceptions in Domestic State Immunity Statutes?),” chapter 9.

Introduction 13 stantial respects, to the issue of civil liability as well as criminal responsibility but choose to the focus on the latter due to the more developed body of doctrine and jurisprudence generated to date with respect to torture protection through criminal law mechanisms.27 In general terms, their answer is that failure to bring an accused torturer to justice for either reason is a breach of CAT and, as such, amounts to granting (illicit) impunity to torturers; the fact that the CAT does not expressly rule out immunity is argued not to preclude such a prohibition as a necessary implication of the treaty, read in light of parallel international law. Working as they do at the level of state responsibility under international law, Burns and McBurney do not address, as Adams does, the allocation of institutional power or duties within states with respect to giving effect to the CAT’s try-or-extradite obligations. That said, their analysis has clear implications for both domestic courts (making clear what international illegality needs to be avoided if interpretively possible to do so given a statute’s language and scheme) and domestic legislatures (which have full constitutional power to rectify any clash between domestic law and the CAT, and, as such, would have, on the Burns/McBurney analysis, an international duty to make sure state and diplomatic immunity statutes do not allow individuals to invoke immunity). Burns and McBurney’s narrative relates the Pinochet judgment of the House of Lords to a concluding observation adopted by the Committee against Torture vis-à-vis the UK a very short time before the House of Lords ruled in Pinochet; the Committee let it be known that, in its interpretive view, the UK had a duty to consider the feasibility of prosecuting Pinochet should it decide not to extradite him (assuming the requisite evidence). With respect to amnesties granted through tailored and genuine national reconciliation commission processes, Burns and McBurney are more equivocal about whether such amnesties always amount to unlawful impunity from a CAT perspective. Unequivocally contrary to the CAT are the kinds of “self-amnesties” granted by outgoing repressive regimes or incoming regimes who decide to mollify those who ran the previous regime. However, while harbouring some doubts as to whether any amnesty can meet the CAT obligations, they do leave open the possibility that failure to prosecute in the context of a truth and reconciliation process may be characterisable as an acceptable decision to the extent admission of guilt and compensation for any torture victims are built into the process. In this respect, Burns and McBurney implicitly see civil redress as a sine qua non of international torture protection obligations even where criminal law prosecution may (possibly) be foregone; such a civil redress bottom-line would seem to lend interpretive support to the argument that article 14(1), reproduced in Appendix 1, of the CAT involves universal civil jurisdiction. 27 P Burns and S McBurney, “Impurity and the United Nations Convention against Torture,” chapter 10. Sorting out the complicated question of when individuals can invoke state immunity is beyond the scope of the Burns/McBurney chapter, especially when some states treat head-of-state immunity (at issue in the Pinochet case) as an extension of state immunity and some as a sui generis immunity that blends classical state immunity with analogies drawn from the law of diplomatic immunity.

14 C Scott

4 CHARACTERISATION , CHOICE OF LAW AND CAUSES OF ACTION

Assume that the jurisdictional hurdle is overcome, or that the question of whether to retain jurisdiction in a given case will be affected by an assessment of what law will be applied to resolve the tort claim should the case go to trial. Part III of the volume deals with this question of the governing law in transnational human rights tort claims. The question of applicable law can be thought of as being answerable either through a private international law analysis (according to which rules of characterisation and choice of law point to one or more domestic legal systems as providing the tort law that will govern either the case as a whole or specific elements of it) or through a substantive analysis of whether “international” or “transnational” law grounds in “domestic” law a cause of action for foreign human rights torts. As will be seen in several of this Part’s chapters, these two avenues may interlock in interesting ways. It should be noted that all contributions are grounded in an analysis of private international law and constitutional law principles relevant to most Commonwealth common law jurisdictions, with a heavy emphasis on the situation in Canada and England. Jennifer Orange’s chapter picks up from John Terry’s brief argument that existing tort law categories and conflicts law-selection methods are adequate for vindicating international human rights violations through civil law, making unnecessary either new legislation or a common-law forging of a specific new kind of international or transnational human rights tort.28 Orange starts with the basis facts in the landmark US case of Filártiga and asks whether, if those facts were transposed to a Canadian context, Canadian judges would find a law to apply to achieve the same result as the US court was able to do by virtue of the presence of ATCA. She reviews the evolution of Canadian tort choice-oflaw doctrine. This evolution has culminated, to date, in the Supreme Court of Canada’s Tolofson case which applied the lex loci deliciti in an interprovincial car-accident context and stated in dicta that this rule should be presumed to govern international tort claims as well.29 Drawing on the reasons in Tolofson about the rationales of tort choice of law and the comments on the differences that might distinguish interprovincial cases from “pure” international conflicts claims, Orange notes that the strict application of the lex loci delicti rule to international fact situations can lead to injustice, especially when one can characterise the wrong done as a human rights violation and not merely a civil wrong. She compares the approach suggested in the American Law Institute’s Second Restatement on the Conflict of Laws with that of the recently-adopted Private International Law Act in the UK, and concludes that the best approach is to treat lex loci delicti as a presumption sufficiently flexible to accommodate 28 29

J Orange, “Torture, Tort Choice of Law, and Tolofson,” chapter 11. Tolofson v. Jensen [1994] 3 SCR 1022.

Introduction 15 non-application of the lex loci delicti where injustice would result from its application. She recommends that Canada follow the UK lead and clarify the tort choice of law principles through legislation, while making clear that dicta in La Forest J’s judgment in Tolofson are themselves sufficient in any pre-legislative period to sustain a justice-based approach where “overriding” norms (specifically mentioned by La Forest J) of international law—in other words, jus cogens norms—would be compromised by Canadian courts’ application of a foreign lex loci delicti. While Orange’s chapter assumes that the existing tort law of one or more domestic legal orders is to be applied to torture-related claims made in a transnational context, that of Graham Virgo queries whether “tort” is necessarily the correct category to be working with.30 He confronts this issue by tackling one of the most vexing conceptual questions in traditional conflict of laws, that of the “characterisation,” or classification, of a legal claim for purposes of then deciding what choice of law rule, principles or method to adopt. After discussing the tendency to say the lex fori is the law which governs the characterization process, Virgo suggests that this may be inadequate to the realities of human-rights-violating conduct such as torture. He argues that it is not the lex fori that governs as much as it is the discretion of the judge of the forum to characterise the claim, and, as such, insights from both public international law and comparative law should play a persuasive role in that exercise. When approached in this way, it becomes possible to arrive at a more nuanced characterisation of torture as a special kind of tort claim akin to existing torts but not identical to any, with this characterisation then feeding into the need for judges to fashion a choice of law process that fits this sui generis claim. In this way, characterisation has the potential to reorient the entire private international law analysis, and allow creative (while principled) recourse to multiple sources in deciding what norms to apply to claims based on a tort of torture even if one must integrate that analysis into application of the generic choice of law principles in England’s Private International Law Act that prima facie apply to a tort of torture as to other torts. Virgo ends his analysis with a detailed set of arguments as to how the justice-based flexibility of the Private International Law Act as well as public international law principles with respect to redress for torture provide the basis for avoiding application of tort rules that thwart that ideal. He then goes on to develop argument as to how the application of English law is particularly appropriate due to a broad interpretation of the new UK Human Rights Act. On Virgo’s interpretation, the Human Rights Act recognises not only torture as tortious when committed in the UK but also torture as a transnational tort if committed within the jurisdiction of any state party to the European Convention on Human Rights. He ends by noting that, even if the courts do not agree that the Human Rights Act itself creates a transnational (intraEurope) tort of torture, the interaction of the Act and traditional private 30

G Virgo, “Characterisation, Choice of Law, and Human Rights,” chapter 12.

16 C Scott international law rules is conducive to recognition of a kind of universal tort of torture. By using principles of public international law and forum-based public policy to reject the application of foreign law and to choose English law, courts can apply the new (statutory) tort of torture to foreign conduct just as they would apply domestic common law rules to many other foreign torts when English law is chosen as the governing law. Virgo’s approach was to create a more cosmopolitan approach to English tort law by starting with a clarification of the notoriously underdeveloped characterisation concept. The next contributor, Martin Bühler, argues for a more rigorous understanding of another doctrinal concept, the act of state doctrine, which has begun to lose its jurisprudential way and, in the process, is in danger of becoming complicit with a very human-rights-unfriendly approach to choice of law.31 Bühler uses as his jump-off point comments made by several judges in the Pinochet cases which suggest that they understand there to be an AngloAmerican doctrine according to which courts are precluded from inquiring into the legal validity of conduct by foreign states or their agents. If the act of state doctrine as understood in these dicta were to be adopted in future cases, argues Bühler, the path-breaking decision of the House of Lords with respect to nonimmunity of a former head of state for conduct he claimed was “official” could easily be circumvented at the merits stage of a case if courts refuse to look behind the claimed “acts of state” which resulted in the human rights violations at issue in a suit. Bühler reviews the development of the notion of “act of state” in Anglo-Canadian law as an adjunct to choice of law analysis. He shows how this doctrine is distinct from a doctrine of non-justiciability and also how English and Canadian courts have consistently been willing, first of all, to evaluate the formal validity of foreign officials acts, laws and decrees according to a foreign state’s own constitutional and statutory framework and, secondly, to refuse to apply formally-valid foreign law if substantively repugnant to English or Canadian “public policy.” With respect to the content of public policy, Bühler argues that public international law informs the content of public policy rather than only being more narrowly relevant by way of invocation as a background justification for a territorialist theory of comity from which would flow near-complete deference to foreign law. He goes on to show how the US has a very distinct “act of state” doctrine which is less willing to inquire into the formal validity of foreign laws and which is heavily tied to American constitutional considerations and only secondarily to international law. While noting that US case law has been hostile to invocation of the act of state doctrine in ATCA human rights tort cases and to that extent is helpful from a comparative law perspective, Bühler cautions against full-scale borrowing of the doctrine from the US without an appreciation of the distinctive trajectory of English and Canadian common law to date. Most significantly, a proper understanding of 31 M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of ‘Act of State’ in AngloCanadian Law,” chapter 13.

Introduction 17 the Anglo-Canadian version of the act of state doctrine will alert courts to the problematic confusion that seems to have developed in some recent decisions in Canada and the UK. In these cases, officials of foreign states have been found to be able to claim in civil cases a state immunity for acts carried out by them in ostensible exercise of their functions as agents of the foreign state. If this line of reasoning were to continue, Bühler notes, not only would it be inconsistent with the approach to immunity taken in the criminal law context in Pinochet but it would also amount to a covert dismantling of the non-deferential AngloCanadian approach to foreign “acts of state” through the doctrinal back door of jurisdictional immunity. In the end, Bühler contends, a more principled understanding of Anglo-Canadian conflict of laws, including its legal history, would be promoted if we were to avoid altogether the terminology of “act of state doctrine” with all the American baggage and evolving confusion attendant upon it. The question becomes one of deciding what law should be applied in light of relevant principles of choice of law and of the way in which public law values should inform private international law analysis. Part III moves on to Sandra Raponi’s account of how we may wish to consider a way of transcending choice of law (as we have come to know it) by having courts treat “transnational law” as opposed to domestic private law as the normative source of the cause of action in tort claims for torture.32 Raponi argues that a notion of a transnational human rights action conceptually harmonises domestic and international law as well as public and private law: a transnational human rights tort is one which seeks to harness the private law processes of the forum by grounding a claim that in substance is primarily derived from rules and principles of public international law. After setting out the virtues of such a conceptualisation, Raponi then engages in a detailed analysis of the doctrinal resistance that proponents of such a cause of action might face in Canada due to the received judicial understanding of the relationship between private law and public law human rights norms as well due to the Westminster tradition of treating treaty norms as being directly received into Canadian law only through the vehicle of legislative action. One ground-breaking aspect of her analysis is that which looks at the question of the direct reception of customary international law into Canadian law as bound up in the tensions in Canadian case law on how the common law causes of action interact with statutory human rights law. She pays particular attention to the extent to which the existence of statutory human rights remedies covers the field with respect to recognition of common law human-rights actions. Her critical analysis demonstrates why these barriers are false ones, but, recognising that some Canadian judges may take a different view, she argues that the increasingly active use of international human rights law in interpreting Canadian law may serve as a starting point from which “Canadian” law is developed under the influence of “international” law, in the process forging “transnational” law without necessarily calling it such. 32

S Raponi, “Grounding a Cause of Action for Torture in Transnational Law,” chapter 14.

18 C Scott The analysis in Ted Hyland’s chapter in many ways overlaps with, and otherwise complements, that of Raponi.33 He asks whether international human rights law can be said to be received in some fashion by Canadian law such that it provides a right of action in Canadian courts against torturers whose conduct occurred abroad. In order to do this, he assesses the textual evidence with specific respect to torture, notably article 14 of the Convention against Torture which speaks of the duty of states to ensure “redress and . . . an enforceable right to fair and adequate compensation” without expressly stating this right is to apply to torture committed outside the territory of the state party whose obligation is at issue. Hyland then discusses the state of the debate in US courts and commentary on whether international law can itself be understood to create a right of action. He then moves on to the question of whether international law can directly found a cause of action before Canadian courts (assuming such an action can be said to arise on some account of international law), a question he addresses through an extensive survey and discussion of all the various ways in which Canadian courts might be willing to recognise a new tort of torture. Such recognition is clearly open to Canadian courts, Hyland argues, but he feels that a further argument about the converging trajectories of contemporary public and private international law is needed in order to convince the average judge to take the step of recognizing a universal tort of torture. He argues that the first step is for Canadian courts to at least accept that public international law on the duty to redress torture impacts enough on Canadian law to create a common law tort of torture with respect to conduct that takes place in Canada. The next step is to tap into the principles of private international law on choice of law in tort which allow extraterritorial application of Canadian tort law in those circumstances where the foreign lex loci delicti does not conform to the jus cogens imperative that remedies must be provided to torture victims who are tortured in a state’s jurisdiction. In much the same way as Virgo suggests an internationalised choice of law exception allows a UK statutory tort of torture to be applied to foreign conduct, Hyland argues for a similar syllogism with respect to the application of “Canadian” tort law on torture to foreign conduct. But the most innovative aspect of Hyland’s analysis is his concluding argument that, should neither the lex fori nor the foreign lex loci delicti adequately promote the torture-reparation goal of public international law, then a judge should apply international human rights law principles, mutatis mutandi, as the rule of decision. By adopting a “principled instrumentalism” in choice of law whereby law selection is substantively conditioned—choose that system of law which best integrates compensation for torture into its rules on civil liability—Hyland has in effect arrived by another route at a special hybrid of the public and the private as well as the domestic and the international that fuses public international law and private international 33 E M Hyland, “International Human Rights Law and the Tort of Torture: What Possibility for Canada?” chapter 15.

Introduction 19 law in a “transnational” space in which general principles of law and justice intermingle.

5 EVOLVING INTERNATIONAL LAW ON CIVIL RECOURSE AGAINST NON - STATE ACTORS

Part IV consists of five chapters treating various dimensions of the question whether public international law permits, encourages or requires civil recourse against non-state actors for serious human rights abuses—actors ranging from regular individuals to high state officials to military officers to intergovernmental organisations to corporations. This question is inextricably bound up in the parallel issue of the existence and scope of an obligation of states (as well as interstate entities) to ensure that the procedural avenues to vindicate such “horizontal” claims are in place in the relevant legal order, or, possibly, the duty to provide compensation to victims where there is no avenue for victims to directly claim such compensation against their torturers or abusers. The chapter of Valerie Oosterveld and Alejandra Flah involves argument by extrapolation from, and analogy to, two fields of law—international criminal law and shared US/Canadian principles of civil liability—on the question of the derivative liability of officials, both civilian and military, for human rights violations committed by their subordinates.34 Oosterveld and Flah first explore the doctrine of command responsibility as it has evolved mainly from the time of the Nuremberg and Tokyo trials to current case law of the International Criminal Tribunals for the Former Yugoslavia and for Rwanda, noting the nascent migration of this doctrine into the field of human-rights civil liability in US ATCA cases. By the doctrine of command responsibility, officials may be held liable for crimes against humanity, war crimes and genocide—and, posit Oosterveld and Flah, also torture—not for having been the torturers nor for even having ordered the torture but for the failure to stop ongoing abuses or prevent future ones where the officials had the requisite knowledge of the abuses and the necessary degree of control that could have been brought to bear on their subordinates. Command responsibility thus involves a fault standard. The authors then go on to consider the doctrine of respondeat superior—and, more particularly, the closely related doctrines of “enterprise risk liability” and “enterprise liability” as they have been developed in leading cases in Canada and California, respectively. By these doctrines, a form of strict liability is involved given that specific knowledge of subordinates’ or agents’ conduct is not required. Rather, the trigger for liability is the creation of a zone of risk for third parties by virtue of the operation of the “enterprise” whether it be corporate, governmental or non-governmental. Oosterveld and Flah acknowledge the need 34 V Oosterveld and A Flah, “Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability”, chapter 16.

20 C Scott for additional principles that would allow these enterprise-related doctrines to apply not only to the liability of collective actors such as corporations and governments but also to key decision-makers within such enterprises, leaving identification and development of such principles for future work.35 The chapter ends with a careful and quite detailed application of command responsibility and respondeat superior to the facts with regard to torture practices in the Sudan. We move to the evolving framework which governs the accountability of the UN for human rights violations committed under its auspices, specifically in the course of UN field operations. Chanaka Wickremasinghe and Guglielmo Verdirame venture into poorly charted terrain with their twinned analysis of the international (public law) responsibility and the (civil) liability of the UN for serious human rights and humanitarian law violations.36 Notwithstanding the specificity of the legal regimes applicable to the UN, much of the authors’ analysis is directly relevant to other intergovernmental organisations (IGOs). At the level of prediction based on existing, rather ad hoc case law in several legal systems, Wickremasinghe and Verdirame conclude that the prospects for successful claims against the UN through the vehicle of transnational tort litigation are limited due primarily to jurisdictional immunities before national courts that are systematically in place throughout the world. In this regard, the authors are wary of the applicability by analogy of the finding by the House of Lords in the Pinochet judgment that acts such as torture do not constitute “official” acts for purposes of immunity analysis, in part because they see courts—at least UK courts—as not unlikely to see limitation of UN immunity as a non-justiciable determination of a matter of UN law by domestic courts. They further note the very recent European Court of Human Rights cases in which, through the vehicle of an action against a member state of an IGO, IGO immunity was questioned as an unjustifiable barrier to access to justice. The Court ruled that immunity from domestic courts was justified at least where the IGO’s internal law provided alternative procedures for resolving the disputes in question (here, employment disputes). At the same time, easy invocation of the UN’s responsibility on the international level is complicated by limitations recently placed on UN tort liability within claims procedures recently established by the UN General Assembly. On the other hand, Wickremasinghe and Verdirame make clear that it is unlikely that serious human rights conduct such as torture would fall within liability exclusions and that limitations on heads of recovery (such as 35 Here it may be that a fusion of command responsibility and enterprise risk principles will lead to a doctrine whereby something less than specific knowledge of torture or a torture practice is required while still requiring evidence of why a particular official is, alone or collectively, responsible for having set up or maintained a situation of risk which they could alter and which they know could well lead to torture, but which they permit (indeed, often intend) to function sufficiently “autonomously” to allow them to deny the existence of the kinds of command links needed in command responsibility doctrine. 36 C Wickremasinghe and G Verdirame, “Responsibility and Liability for Violations of Human Rights During the Course of UN Field Operations,” chapter 17.

Introduction 21 exclusion of claims for pain and suffering) may in many cases not be relevant to human-rights-based claims because the General Assembly rules lift the remedial limits in cases of gross negligence and wilful misconduct. In relation to both claims before national courts (assuming jurisdictional immunity is circumvented) and international claims (whether within UN-established claim procedures or at the level of claims of diplomatic protection brought by torture victims’ states of nationality), there remains the persistent problem of formulating the principles which should govern the allocation of responsibility as between the UN and states acting for, or under authorization by, the UN. Like Wickremasinghe and Verdirame, Muthucumaraswamy Sornarajah analyses the question of accountability for serious human rights violations within a framework in which international responsibility and civil liability are closely connected.37 Sornarajah argues that home states of corporations committing serious human rights abuses abroad themselves incur responsibility for those abuses by virtue of the way in which both global power realities and the link between corporation and national state-society interact with classic international law doctrine and modern general principles. His argument as to the nature of the home state’s responsibility is that such responsibility can arise in one or both of two ways. Firstly, there can be responsibility for breach of the positive duty to provide for civil recourse by victims of jus cogens human rights violations against the corporation in its home state courts. Secondly, there is a form of indirect responsibility for failure to use mechanisms of control available, or potentially available, to the home state to prevent or stop abuses of which relevant state officials were forewarned or came to have knowledge. Here, especially in relation to the latter form of responsibility, Sornarajah takes great care to debunk the common view that the notion of liability of states for extraterritorial harm caused by nationals is a new normative phenomenon. He reminds us of older arbitral case law that was premised not simply on the extraterritorial application of a positive duty to prevent harm as between nonstate actors but more radically on a form of direct responsibility of home states for harm caused by corporations operating abroad. On this approach, including as invoked by the US in its successful claim against Britain in the Alabama Claims, the conduct of the national is assimilated to that of the state itself by virtue of a notion of the state’s tacit approval of the national’s conduct by not seeking to prevent that conduct where the state had knowledge of it nor punish the national once the harm was caused.38 Borrowing from older law, an approach of structural state complicity in corporate activity thus becomes a central focus of Sornarajah’s analysis. He juxtaposes the evolution of late 19thcentury and 20th-century positive international law in the image of capitalexporting and trade-based states—and thus away from the notion of indirect responsibility of states for corporate nationals abroad, let alone a form of 37 M Sornarajah, “Linking State Responsibility for Certain Harms Caused By Corporate Nationals Abroad to Civil Recourse in the Legal Systems of Home States,” chapter 18. 38 Alabama Claims Arbitration, (1872) 1 Moore 495.

22 C Scott vicarious responsibility—to a counter-movement that is sending new normative signals, including in areas as diverse as transfer of hazardous waste and sex tourism. Proceeding from this base, he builds a strong argument for viewing the failure to provide some system of home-state remedies for the victims of corporate harm abroad as sufficient to actually attribute the corporate conduct to the remedially-recalcitrant state. Such imputation occurs by virtue of viewing the state’s failure as amounting to the kind of implicit, tacit or constructive approval of non-state conduct spoken of in the pre-20th-century era before state positivism and global capitalism fused into mainstream international law. By thus linking home-state responsibility to civil recourse, Sornarajah takes the view that the deterrent effect with respect to future conduct is enhanced beyond that which is attributable solely to modification of corporations’ own conduct due to the risk of transnational legal accountability. It is also enhanced because the home state’s potential state responsibility should lead it to become a more active regulator of problematic conduct of its corporations abroad. Whereas Sornarajah examines the question of civil recourse for torture and similarly serious human rights violations in relation to general principles of international law and the example of corporate actors, Andrew Clapham’s discussion of access to civil remedies is largely centred on evolving principles within the sub-field of international human rights law and is, as well, largely focussed on civil liability of non-state actors who unquestionably act in concert with the state—that is, state officials such as political leaders and police.39 Clapham’s starting-point is his earlier study, Human Rights in the Private Sphere, which remains the leading account of the conceptual and doctrinal dimensions of international law’s duties with respect to “private” violence and other harms. In the present chapter, Clapham uses that work as well as the Pinochet affair as the starting points for asking how “horizontal” application of human rights (claims by individuals against individuals) interacts with a “vertical” axis according to which states have duties both to prevent violence and also to ensure “horizontal” access to human rights justice. A centrepiece of Clapham’s analysis is the decision in the European Court of Human Rights case of Osman v. United Kingdom that article 6(1) of the European Convention on Human Rights, which protects the “right to a court” in civil disputes, includes a prohibition on granting police authorities immunity from negligence suits brought on the basis of the police failure to protect one person’s right to physical security, most notably the right to life, from harm at the hands of another person. This case is discussed against a backdrop of similar provisions in a number of other international human rights instruments. From there, Clapham narrates a recent effort, in which he was involved, to persuade the European Court to rule that failure of the UK to either extradite or prosecute Pinochet would amount to an interference with the rights of victims to have their article 6 “civil 39 A Clapham, “Revisiting Human Rights in the Private Sphere: Using the European Convention on Human Rights to Protect the Right of Access to the Civil Courts,” chapter 19.

Introduction 23 rights and obligations” adjudicated before a court. Clapham’s argument as to how failure to pursue criminal proceedings undermines civil rights is too intricate to convey in an overview of the present sort, other than to say it involves the contentions that article 6 civil rights can include “human rights” not protected as such as positive law rights in a given state’s legal system and that criminal law remedies can, in some circumstances, be viewed as a protected legal avenue for vindicating “civil rights” either where this avenue is the only one specifically provided by the legal system in question or where it acts in concert with a civil remedy process. He outlines the argument that the release of Pinochet, as opposed to his extradition to various requesting countries (not just Spain but also France and Switzerland), would violate not only the rights to access to justice of UK-based victims but also victims residing in other countries (the extradition-requesting countries) who lose the opportunity to have Pinochet brought to justice—again, especially where the legal system of the country has a hybrid approach that uses criminal law as the vehicle or threshold for civil remedies. Recognising that some might see this argument for protection of a transnational right of access to justice under article 6 of the European Convention as tenuous, Clapham constructs a careful argument that pulls together normative strands from the Soering case (states may incur indirect responsibility for human rights violations committed, or to be committed, by other states) and from the Loizidou case (stating the European Court’s view that the European Convention is a “constitutional instrument” for a normatively unified Europe).40 On this foundation, Clapham argues that one state party can have duties towards persons located in other states parties beyond the limited Soering situation of the duty not to deport or extradite someone to a humanrights-violative situation. He ends by bolstering the normative force of his arguments by reference to the deepening content being given to the European Convention’s article 13 right to have effective remedies for Convention breaches in place within the national legal systems of states parties. The Clapham chapter focussed on the normative specifics of the European Convention on Human Rights. In a similar vein, Andrew Byrnes takes on the issue of (transnational) access to civil remedies within a single treaty, the Convention against Torture, and primarily by reference to the proper interpretation of article 14 of the CAT which provides that each state party to the CAT “shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation.”41 The central question animating his inquiry is whether a state party to the CAT is obliged not only to exercise a form of universal criminal law jurisdiction (that is, to try or extradite an alleged torturer found on its territory) but also to exercise a form of universal civil law jurisdiction by ensuring civil remedies are 40 Soering v. United Kingdom, ECHR (1989) Series A, No. 161; Loizidou v. Turkey (Preliminary Objections), ECHR (1995) Series A, No. 310. 41 A Byrnes, “Civil Remedies for Torture Committed Abroad: An Obligation under the Convention against Torture?”, chapter 20.

24 C Scott available within the state’s national legal system for acts of torture abroad. Except in situations where those acts of torture were committed by agents of the forum state (for example, the torture-death of Shidane Arone in Somalia at the hands of Canadian soldiers), Byrnes concludes that article 14 of the CAT does not go so far as to make mandatory civil remedies for extraterritorial acts of torture or, at least, for extraterritorial acts of torture that were not committed by the forum state or its agents. His conclusion is based on an analysis of the drafting history of the treaty, the other provisions of the treaty, the CAT’s overall goals and interpretive practice to date. However, he is careful to point out that treaties, especially human rights treaties, are living instruments such that their interpretations can evolve over time especially where an independent body established under the treaty to monitor its implementation puts forward a more expansive interpretation than had previously been generally understood and states parties gradually accommodate themselves to that new interpretation. Here, it might well be added that the willingness of domestic courts to also push the normative envelope, whether before or after the Committee against Torture, will likely be as significant in adjusting normative expectations as to the territorial scope of article 14 as the reactions of the executive branches of states parties. As Byrnes notes, the first step in this direction might be for the Committee against Torture to read article 14 as not necessarily setting out an obligation to establish universal civil jurisdiction for torture but as establishing the permissive basis for the exercise of such jurisdiction by those states willing to do so.42 Here again, some domestic courts may well see it as their appropriate role to act on behalf of their state by recognising that article 14 provides interpretive encouragement for them to develop their common law or their interpretation of the scope of civil codes. And, finally, it bears reminding that Byrnes does not view the scope of article 14 as being entirely bound up in a traditional spatial tension between the territorial and the extraterritorial; foreign acts attributable to the forum state would trigger the article 14 duty to provide for redress and compensation.43 Byrnes ends by noting that the standard private international law methods for selecting applicable law will still be available even absent a freestanding international law duty to provide for a universal tort of torture in domestic law, and, in such selection, the CAT may exercise an interpretive influence.

42 Here the wording of article 14(2) assists in some respects: “Nothing in this article shall affect any right of the victim or other person to compensation which may exist under national law.” 43 In this respect, the next step after recognising article 14 as permitting universal civil jurisdiction may be to read article 14 in light of evolving principles of state responsibility for acts of nationals abroad, as discussed by Sornarajah by reference to corporations, such that article 14, purposively read, requires states to provide for claims for redress and compensation in their legal systems for all acts of torture committed by their nationals anywhere in the world—whether or not those acts are imputable, as in the Somalia torture example, to the state itself.

Introduction 25

6 LEGITIMACY , INTERVENTION AND THE FORGING OF NATIONAL HISTORIES

From the quest for the universal we move to four chapters that are premised, to varying degrees and in different ways, on the view that universal jurisdiction over human rights torts raises special concerns in relation to national processes of governance, ownership of the key decisions that forge a society’s future, and normative disagreement over the content and scope of even core human rights such as the torture prohibition. Without embracing statism, each chapter argues that sovereignty of state-societies, self-determination of peoples, and culturally distinctive understandings cannot be simply ignored in a rush to justice. Jan Klabbers opens Part V with the volume’s strongest critique of transnational tort adjudication in the name of universal human rights.44 As will Mayo Moran in her chapter,45 Klabbers starts by noting how undertheorised transnational human rights litigation has been. He attributes this in part to the gospel-like stature of human rights discourse in our times. His purpose, then, is to draw attention to legitimacy problems with this phenomenon. His primary contention is that American-style tort adjudication over human rights abuses occurring abroad amounts to a doubly unjustified intervention—of foreign states into the affairs of other polities and of judges into politics. In developing his case against the propriety of ATCA actions and thus of any mirror processes that might arise in other states, Klabbers sketches out, and then relies on, a conception of human rights as, in the main, guarantors and facilitators of politics. He ties a theory of politics into the long-standing concern of international law with the self-determination of peoples (it bears emphasising that there is no defence of state sovereignty per se in Klabbers’ account) which leads to “a certain primacy” that must be given to freedom from foreign decision-making about choices that are more legitimately made from within a given historicallysituated political community. This concern slides into a parallel view that, whatever universal sociological consensus may exist for abstract formulations of a human rights norm, it is only in the context of a particular community’s historically-situated understandings that more concrete content can legitimately be given to human rights norms. Klabbers is not naïve about the consequences of the political spaces he seeks to preserve for self-determining and largely nonjudicial decisions, noting as he does that human rights violations through and in the name of politics is a price that has to be paid for keeping a separate niche for politics. A form of ethical realism appears to emerge from Klabbers’ account, one that is not hesitant to point out hypocrisies that undermine claims to moral high ground, as when he notes that a state like the United States can generally refuse to allow even the International Court of Justice to function as an appellate court while failing to see that the same line of thinking should apply 44 45

J Klabbers, “Doing the Right Thing? Foreign Tort Law and Human Rights,” chapter 21. Moran, infra n.49.

26 C Scott a fortiori to (US and other states’) domestic courts acting in effect in a review capacity of foreign states-societies. He ends by affirming that human rights have a crucial role to play in political discourse, just not as legalised claims let alone as claims generated from afar. Jennifer Llewellyn tackles one slice of the question of the deference that foreign societies, and international society as a whole, should give to selfdetermining choices made by societies. How should the granting of amnesties from civil liability be treated in foreign torts claims?46 Proceeding from the unavoidable political reality that countries previously ruled by repressive regimes often compromise and co-operate with members of the outgoing regimes, Llewellyn seeks to show how it is that amnesties are not necessarily simply pragmatic ways to buy a stable context on which a new democratic order can be built. Rather, she draws on her own and others’ work in the field of restorative justice to argue that certain amnesties, notably those emerging from well-constructed truth and reconciliation processes, are morally worthy of respect. That being the case, she addresses the need for private international law to accommodate itself to justified amnesties when it comes to foreign courts deciding whether or not to recognise the liability-shielding effects of amnesties accorded by the country where a human-rights tort such as torture has occurred. To do this, she also canvasses the underdeveloped approach to amnesties taken by public international law, partly in order to see whether that body of law affords enough leeway for just amnesties to be considered lawful and thus for domestic courts deciding private international law tort cases to defer to the foreign state’s laws (or associated foreign court judgements) by reference to this public international law acceptability. Public international law has not firmly crystallised a view on amnesties, although certain developments like the Statute of the International Criminal Court do not appear welcoming of amnesty for the most serious of human rights violations amounting to crimes under international law. This somewhat open normative state of play allows Llewellyn to argue for an interpretation of international human rights law that recognises the justice-promoting dimensions of some kinds of amnesties. Those amnesties which are (restoratively) just, according to principles argued for by Llewellyn, deserve not to be circumvented by foreign tort proceedings. Amnesties such as those granted in South Africa would pass muster while blanket amnesties such as that in Chile would not. The foregoing is an ideal normative argument, but Llewellyn recognizes that there is a follow-on institutional question with respect to whether foreign courts, removed as they are from the national social context of the country of amnesty, should be empowered to make nuanced interpretations as to which side of the just/unjust line a given amnesty falls. Put differently, it might be argued that institutional reasons transform a middle-ground approach to amnesty into a kind of total deference by foreign courts—a transnational non-justiciability. However, Llewellyn contends that, on balance, the 46

J Llewellyn, “Just Amnesty and Private International Law,” chapter 22.

Introduction 27 risks of abuse are outweighed by the virtues of holding national feet to foreign fires; in this respect, she enumerates several reasons why the potential for foreign courts to conclude that an amnesty is unjust serves an important function in ensuring that amnesties are structured and operated in a just way. We move to Australia with the discussion by Belinda Wells and Michael Burnett of what they call the Saudi nurses case.47 The issue with which they grapple is the extent to which Australian law should defer to Saudi Arabian law on the implementation of the death penalty (of beheading) when such an act would be characterised in Australia as murder, torture, inhuman punishment, or all three. The context is not a typical ATCA-like situation in which a person accused of torture in one country (Saudi Arabia) is sued in tort in another country (Australia). Rather, the facts involve a complex interweaving of territoriality and nationality, and a remedy which has not yet become part of the ATCA scene: the injunction to prevent an imminent human rights violation in another country. Wells and Burnett narrate a fascinating chain of events according to which, in summary, an Australian citizen was, under Saudi shari’a law, being asked to decide whether the death penalty should be imposed on a British nurse convicted in a Saudi court of having murdered an Australian nurse. One of the authors, Burnett, found himself acting as counsel for the convicted British nurse who was seeking to find a way for an Australian court to exercise territorial jurisdiction over the Australia-located brother of the murdered nurse and order him not to exercise his prerogative under shari’a law, thereby having an extraterritorial impact: saving her life. Thus it is that the chapter offers a marvellous window not only into the legal possibilities that may exist for civil suits in Australia to deal with foreign human rights situations. It also offers a glimpse of a kind of existential limbo in which a lawyer representing a client in transnational human rights litigation might find herself or himself if they hold a nondogmatic view that “there is clearly a tension between the principle of the universality of international human rights and the need to appreciate the cultural context of rights,” as the authors put it in the first paragraph of their chapter. In the end, the authors embrace an ideal of cross-cultural dialogue around radically-different conceptions of human rights, but point out how very difficult it is for such an ideal notion to be worked into substantive interpretations in a pressing litigation context. That said, there was an interesting outcome to the case—the brother, on the eve of the Australian injunction trial, agreed to exercise his right under Saudi law to accept compensation from the convicted person in lieu of the death penalty, and then donated most of the money to a public hospital. The authors present this outcome as having emerged from an extracourt dialogue between the victim’s brother and the convicted person (through her lawyers), a dialogue that only occurred by virtue of the legal proceedings having been initiated. Legal coercion interacted with a quest for compromise. 47 B Wells and M Burnett, “Cultural Challenges: Injunctions in Australian Courts and the Right to Demand the Death Penalty under Saudi Arabian Law,” chapter 23.

28 C Scott The compromise solution that ultimately emerged can be interpreted as substantively dialogical in nature; as the authors conclude, “the victim’s brother c[a]me to a decision that combined Islam’s preference for forgiveness . . . with a principle of charity which may well have owed much to the brother’s own moral convictions.” We turn to another concrete context, by returning to the Middle East with a chapter by Amnon Reichman and Tsvi Kahana on the implications for foreign tort actions of a recent judgment by the Israeli Supreme Court which held that the use of physical force by Israel’s General Security Service (known as GSS or Shin Bet) in the course of interrogations was illegal under Israeli law.48 That judgement clearly places the methods used by Israel within the notion of “inhuman” treatment but is ambiguous on whether the Court views some or all of the methods to pass a gravity threshold so as also to be characterisable as “torture.” The chapter begins with a form of interpretive inter-Israeli (inter-author) dialogue over whether the judgment can be read as contemplating that the Court left open that new legislation could well be constitutional if it expressly permitted the currently-illegal methods to be used in “ticking bomb” situations in which it is judged that information is urgently needed in order to prevent deaths or serious injuries. This question is relevant for two reasons: first of all, because it speaks to the question of whether individual members of Shin Bet would be treated in Israeli law not just as having a “necessity” defence if charged with a crime, but as acting within the law and thus as having a defence to civil claims as well; and, secondly, because any ambiguity in Israeli law on this point may have implications with respect to how interventionist foreign courts would view themselves as being should they choose to find a member of Shin Bet liable in an ATCA or ATCA-style human rights tort claim. Before moving on to discuss transnational civil liability against this unclear normative backdrop, the authors discuss the reasoning in the House of Lords Pinochet case with respect to whether “official” torture can be a defence in criminal proceedings based on universal jurisdiction under the Convention against Torture. They note as well interpretations handed down by the UN Committee against Torture with specific respect to Israeli’s state responsibility for the use of the interrogation techniques that were at issue in the GSS case. Here, the authors note that characterisation of the conduct as “torture” versus “cruel, inhuman or degrading treatment” is key because article 16 of the CAT appears to contemplate the availability of some defences that are excluded when the conduct amounts to torture. Any foreign court would have to make a choice as to where to seek normative guidance. From the Committee against Torture which characterised the interrogation techniques as torture and not only as cruel, inhuman or degrading treatment, and thus as absolutely prohibited even in ticking-bomb situations? From the foreign court’s own best interpretation of what the Israeli Supreme 48 A Reichman and T Kahana, “Israel and the Recognition of Torture: Domestic and International Aspects,” chapter 24.

Introduction 29 Court found on the characterisation of the conduct? From its own interpretation of how the facts as described in the Court’s judgment mesh with the test for torture? Quite apart from sources of interpretive authority, there would be, the authors argue, a question of deciding whether it is timely for foreign courts to adjudicate tort claims against Israeli officials when the GSS case represents evidence that the Israeli Supreme Court is no longer prepared to accept that GSS interrogation techniques are currently lawful absent specific legislative authorization for them nor to accept, it seems, they can ever be lawful outside tickingbomb situations (which the court avoids pronouncing clearly on), and thus that this is not (or no longer) a situation of legal-system-failure such as might justify foreign courts stepping into the breach. The authors come down on the side of foreign tort proceedings when the state under whose regime torture has been committed fails to hold its officials, or itself, civilly “responsible”—although the authors do not clearly assert that monetary compensation to victims must be a sine qua non. They find that foreign tort liability for the use of the GSS interrogation methods in non-ticking-bomb (“normal”) situations presents a comparatively easy case, should Israel not act soon to create civil liability for harm caused in such situations (including retrospectively, it would seem). Indeed, these are comparatively easy cases because a foreign court can simply rely on the fact that the conduct was illegal not only under international criminal law but also under Israeli law itself as authoritatively found by Israel’s highest court. With respect to ticking-bomb situations, Reichman and Kahana consider there may be good reasons founded in comity for foreign courts to find themselves forum non conveniens to the extent it remains an open possibility that Israeli courts will find the state or the GSS officials themselves to be civilly liable for harm through interrogation even if the “necessity” defence remains a shield from criminal responsibility. Some deference—at least in terms of a wait-andsee delay (which would, one assumes, be possible to accommodate through conditional stays by foreign courts)—would have the practical benefit of increasing Israeli courts’ sense being part of a transnational human rights dialogue with foreign courts and thus of making it more likely Israeli courts will enforce foreign damage awards once foreign courts decide the wait-and-see period has lasted long enough. But, in a vein similar to Llewellyn, Reichman and Kahana emphasise the importance of foreign courts being prepared to assess the justice of any non-liability within Israel’s legal system; they argue that foreign courts should refuse to accept any subsequent Israeli Knesset legislation which in effect makes lawful the currently-unlawful interrogation methods in “normal” situations. Reichman and Kahana end by noting the inevitability (and, if I read them correctly, legitimacy) of foreign courts also passing judgment on whether ticking-bomb situations warrant different treatment. Whether eventual foreign courts’ determinations are “right” or “wrong”, they will necessarily become part of the normative context which both Israeli courts and legislators will have to take into consideration as Israeli society grapples with the issue of the limits of interrogation.

30 C Scott

7 ON THE BORDERS OF TORT THEORY

In Part VI, which concludes the volume, we enter the domain of private law theory on the relationship between tort law and public law human rights values. The volume thus ends by leaving the reader in a border zone, as it were, between a medley of (public and private) international law perspectives and various private law traditions around the world which may have much to say about the enlistment of tort law into the service of international human rights protection. That being said, this border region is much less of a minefield than it might otherwise be due to the fact that the contributors of the two final chapters are tort theorists who see a symbiosis between tort law and public law norms rather than a strong tension, let alone a radical disjunction of function. Mayo Moran takes the view that the ATCA human rights tort cases, despite often being both “overwritten” and “undertheorised”, do not deserve the complete lack of discussion by tort theorists or complete omission from leading texts on the law of torts.49 Instead, there is a case to be made that the torture-as-tort cases in the US provide “a glimpse of the challenges that will face the private law of the future” in which the “geography” of the law is being reworked in such a way as to present a challenge to the very foundation of modern legal traditions. Linked to complicated transnationalisation of context, she perceives a richness in the strongly rhetorical nature of judgment, and associated reasoning, that animates the ATCA tort cases—a field that is reminiscent of the “multiple, diverse and often conflicting” sources upon which judges had to draw in fashioning persuasive reasons in old common law decisions when prior judicial precedents, and interpretive exegesis thereof, were much less of a central plank in providing parameters for judgment. From this starting point, Moran points to broader trends into which transnational human rights litigation dovetails, notably increasing comparativism that is creating an interpenetration of legal norms from various legal systems (and an associated sense of the relevance of “external” sources to “internal” norms) and a discernible breaking-down of a “separate spheres” understanding of private law and public law. The emerging legal world is “increasingly characterized by interactive and even hybrid norms” with the US international-human-rights-oriented private law actions epitomising this development. And it is not just purist understandings of private law that are becoming anachronistic. The phenomenon discussed by Moran also strikes at the very heart of traditional conflict of laws analysis which sees norms as being capable of selection from autonomous national systems of law and then straightforwardly applied to binding effect. It also challenges rather static understandings of many international lawyers that domestic systems are in essence mere enforcement vehicles for international norms.50 Moran engages in 49

M Moran, “An Uncivil Action: The Tort of Torture and Cosmopolitan Private Law,” chapter 25. Even as, it might be added, it equally challenges the sense of constitutional lawyers of international law as somewhere “out there”, disembodied and of questionable relevance. 50

Introduction 31 a close discourse analysis of the reasoning in several of the ATCA cases in order to show judges grappling—never successfully—with all these boundarycrossing implications and with the promiscuity of potentially-meaningful normative touchstones. Ultimately, Moran concludes, the torture-as-tort and other human rights cases are pulling us toward a more cosmopolitan legal world in which judges (and other decision-makers) no longer can seek comfort in the notion of discrete normative spheres of private law and public law, of international law and domestic law, of national law and of foreign law. She sees evolving a “more multifaceted integrative understanding of sources and a broader persuasive approach to authority” in which the fundamentally rhetorical nature of the older common law may have far more useful things to say than many of the current self-understandings of the nature of legal judgment dominant in both the judiciary and legal academy. A certain sense of the common law tradition as particularly well-suited to grasping, adapting to and working with the “new private law” (if it might be called that) flows from Moran’s analysis, which necessarily leads to the question of how various civilian legal traditions and cultures might react to the normative messiness of the ATCA phenomenon. One rumoured truth for those given to references to tendencies in other legal systems without having had the benefit of studying or working in systems is that the “horizontal” effect of constitutional norms (that is, drittwirkung—their application in “private” relationships governed by areas of law such as tort, contract, and property) is taken for granted in Germany. However, in reading Oliver Gerstenberg’s contribution with which we conclude the volume, one would be forgiven for concluding that the blurring of the public/private boundaries will face considerable conceptual resistance from tort law scholars in Germany (how representative this would be of reactions in other civilian systems cannot of course be known).51 Gerstenberg eases into his critique of various German legal scholars’ view of the sources and function of private law by starting with a case (already somewhat uncivilian of him, some might be thinking). The Bosman case, decided not by a German court but by the European Court of Justice, decided that the freedom of movement guarantee in the European Community’s constitutive treaty applied to non-state actors and not only to EU member states. Bosman dealt with a challenge to transfer rules in the European football world which required payments between clubs before a player could move to another team. Holding that this fell afoul of the freedom of movement protection, the Court, at the remedial level, declined to specify what rules a transfer regime would have to contain in order to strike the requisite proportionality between the interests of the players and the football association. The ECJ “regarded its role as one of promoting self-regulation experiments in private legal 51 O Gerstenberg, “Private Law, Constitutionalism and the Limits of the Judicials Role,” chapter 26. For both a very useful account of German drittwirkung cases and for an expression of concerns about over-constitutionalisation of German Private Law, see S Oeter, “Fundamental Rights and their Impact on Private Law-Doctrine and Practice under the German Constitution,” (1994)12 Tel Aviv University Studies in Law 7.

32 C Scott relations that will foster fundamental rights.” Notwithstanding the softness and democratic-process-prodding nature of the remedy, the ECJ’s injection of “public” law rights into “private” relations “contrasts starkly”, notes Gerstenberg, with the mainstream scholarly view in Germany on the effect of fundamental rights on private law. Gerstenberg then seeks to demonstrate this disjuncture through a detailed criticism of the positions taken by leading scholars on the German Constitutional Court’s view that constitutional rights are “objective legal principles” governing “private legal order” no less (albeit perhaps differently) than state/individual relations. Such scholars appear to be challenging the wisdom of this drittwirkung (third-party effect) step, some making the case that “the direct binding of private actors by fundamental rights would, in effect, make private law superfluous.” Gerstenberg counters with an account of a rights-oriented “constitutionalisation and democratisation of private legal relationships” in which rights adjudication serves as the path by which courts make certain that private actors forge the principles and rules that bind them to each other in a process of dialogue that amounts to democracy in the private sphere. As such, Gerstenberg’s theory of the direct application of human rights to private law is profoundly more radical than what has so far occurred under the auspices of ATCA-based adjudication. This is because the vast majority of individuals sued in the name of international human rights law in the ATCA cases have been, at some point, actors operating within a state apparatus, whereas Gerstenberg envisages rights as pervading law and decision in the relations amongst “private” actors with no formal connection to the state and within the legal orders of private organisations. As such, the German legal situation and Gerstenberg’s defence of it (or, perhaps more accurately, his argument for its principled deepening) have much to say to emergent attempts to use ATCA as a basis for suing multinational corporations for human rights violations (as referenced in the chapters by Robert Wai and Muthu Sornarajah), providing some theoretical foundations for moving beyond the sterile formalism whereby US courts have found that the vast majority of human rights torts must have been generated in some way characterisable as “state action.”52 Taken in tandem with Moran’s call for normative openness and creativity, the German constitutional approach to private law (as distinguished from its challenge by the private law scholars engaged by Gerstenberg) may have much to say about the trajectory that a transnationalised process of tort claims grounded in international human rights norms and values could take in the years to come.

52 For an overview of US litigation up to 1999, see S Zia-Zarifi, “Suing Multinational Corporations in the US for Violating International Law”, (1999) 4 UCLA J. Int’l L. & For. Aff. 81. See, more generally, S Joseph, “Taming the Leviathan: Multinational Enterprises and Human Rights”, (1999) NILR 171. For one of numerous human rights NGO reports beginning to emerge on corporate conduct, see, by way of example, Human Rights Watch, The Enron Corporation: Corporate Complicity in Human Rights Violations (1999).

Introduction 33

8 STUMBLING ON THE FIRST STEP : OBSERVATIONS ON THE ARONE CASE

I began this introduction with mention of the Sudanese context as having been the proximate motivation for this volume. To bring this beginning to an end, I now invite the reader to take a short flight from the Sudan over Ethiopian territory to Somalia. Allusion has already been made in this Introduction to proceedings brought before a Canadian court on behalf of the parents of Shidane Arone who was tortured to death, in the most brutal fashion, by Canadian soldiers who caught him inside the perimeter of their camp.53 Sadism aside, it appears that one purpose was to send a message to others about trying to engage in theft from the Canadian UN peacekeeping contingent. Let it be assumed, then, that there is no question but that the torture of Arone is covered by the definition of torture found in article 1 of the Convention against Torture and that the acts, however unlawful under Canadian military law, were carried out by agents of the Canadian state. As such, it appears reasonable that Arone’s parents would bring a tort action against the government of Canada in Canada’s own courts. Intuitively reasonable perhaps, but the result of the case—the thinly-justified dismissal by the trial judge on the basis of no reasonable cause of action—is a case in point as to the distance that must be traveled before many judges will be prepared to treat torture as a transnational tort absent an express legislative directive to do so, even where Canadian nationals are the respondents or, as here, Canada itself. The statement of claim in Arone starts by stating that Arone’s parents were seeking “general damages for torture, wrongful death, and/or murder, negligence, and breach of fiduciary duty” and then, rather cryptically, ends with a list of statutes on which the plaintiffs “plead and rely” including the Canadian statute which legislates into Canadian law the 1949 Geneva Conventions on the laws of war, together with separate mention of the treaties themselves.54 Cunningham J. granted the government’s motion to dismiss the action on the primary ground that it had been brought on behalf of Arone’s Somalia-based parents (by one Abdullahi Godah Barre) and not by them directly, there being no provision under the relevant Canadian law for such a “Litigation Guardian” for persons who were not disabled or unavoidably absent from the jurisdiction (with the condition precedent that the person was normally resident in the jurisdiction). The judge went on to hold that there were two further grounds on 53 The facts are set out by Terry, supra note 18, including reference to criminal proceedings in Canada against the torturers which saw one soldier convicted of the crimes of torture as well as manslaughter, and three others convicted of negligence in Arone’s torture and death. 54 Statement of Claim, in Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. Canada (Attorney General), Notice of Action issued on March 15, 1999, Ontario Court (General Division), Court File No. 99-CV-9387 (“Statement of Claim”). Many thanks to Ted Tjaden of University of Toronto Law Library for having securing this document. The Geneva Conventions are incorporated into Canadian law in the Geneva Conventions Act, RSC 1965, c.G-3 as amended.

34 C Scott which he would also have dismissed the case even had the parents directly brought the claim: the failure to state facts sufficient to reveal Canada owed a duty of care in tort to Arone; and the fact that the action was barred by relevant Canadian statutes of limitations applying to the Crown (Canadian government) and soldiers. While it is not my role as editor of this volume to take a position on all the issues discussed herein, I cannot let pass the opportunity at least to point out what very relevant arguments were totally missed in this case—to the point that the Arone case represents an extreme example of limited lawyering and noncosmopolitan judicial formalism. If the present volume comes to have any impact at all on the direction the law takes in jurisdictions outside the United States, especially in Commonwealth countries, it is hoped that it will at minimum mean that cases such as Arone will in future be decided in much fuller awareness of what is legally at stake (public and private international law, and their transnational interface, and not simply domestic statutes with limited direct relevance to the context) as well as of the rich range of normative considerations that should prompt courts to consider viewing claims such as that of Arone’s family as crying out for the common-law-like development spoken of by Raponi, Moran and others in this volume. Since John Terry has more than adequately set out the basic mistakes made by the judge in Arone, it is not my intent to replicate that analysis, but, rather, merely to supplement it by noting some of the arguments and doctrinal areas canvassed in this volume which are relevant to deciding this case on appeal—or relevant for a future trial judge should the case be recommenced.55 The judge ruled that “[n]o facts are set out in the statement of claim that would remotely give rise to a duty of care that would support the allegations of negligence.” Keeping in mind both that the soldiers were actually convicted of criminal negligence and that statements of claim need only set out the basic facts, it is nothing short of stunning that Cunningham J. would find the following passages (none of which were even alluded to, let alone quoted, by the judge) in the statement of claim as lying beyond even remoteness in terms of establishing a duty:

55 Whatever the justifiability of the first ground of dismissal—the lack of standing for a Litigation Guardian—it is assumed that it is within the discretion of both the Ontario Court of Appeal and a future trial judge to allow the case to be recommenced by the Arone parents acting as direct plaintiffs. Arguably, given the realities of litigating a case such as this when one is located in Somalia in the circumstances of Arone’s parents, it was unaccommodating of the judge not to have conditionally dismissed the case in order to have allowed a refiling. This comment applies equally to the second ground of dismissal—the lack of sufficiently pleaded facts to show the existence of a duty of care. Here, Terry is surely correct in his observation that the statement of claim is a textbook example of a sufficient pleading of facts for purposes of commencing an action, but, even were the judge correct to say clearer facts had to be pleaded, he should at least have acted as did the judge in the case of Beanal v. Freeport-McMoran, Inc., 969 F.Supp. 362 (E.D. La. 1997), by giving Beanal (a plaintiff living in Indonesia) a chance to revise the statement of claim before considering final dismissal.

Introduction 35 “The Plaintiffs state and it is a fact that, during the period in which he was detained and prior to his death, Shidane Abukar Arone was in the custody of various [Canadian] military personnel [listing the four soldiers convicted by military tribunal back in Canada]. . . [B]etween the time of his capture and death, a period of several hours, Shidane was blindfolded, bound at his wrists and ankles and systematically, brutally beaten and tortured by members of the Canadian Armed Forces . . . [A]t periodic intervals . . . , Shidane’s captors took turns posing for photographs with their blindfolded, bloodied, beaten and helpless victim . . . [T]he aforesaid prolonged torture and beating of Shidane was witnessed by other members of the Canadian Armed Forces stationed at the base, including [nine others names] . . .[M]any [other] Canadian soldiers including [a further six are listed] were informed of the ongoing torture and beating of Shidane . . .[N]one of the soldiers who observed or were informed of the beating and torture of Shidane took steps to protect him or put an end to this action . . . [A]t some point during the detention, Shidane Abukar Arone died as a result of the beating and tortured which he suffered at the hands of the Canadian soldiers . . .”56

Recall that it is Canada being sued in the Arone case. If one were to be charitable, one might attribute the judge’s total failure to connect these facts to a duty of care owed to Arone to his profound confusion over who owed a duty to whom. The judge seemed to think a separate duty of care in negligence had to be shown as running between the state of Canada and Arone other than the one clearly running between Canadian state agents (the soldiers) and Arone. Not only is Canadian law clear on the vicarious liability of the Crown for torts of its servants (and keep in mind that the judge seemed to think the only relevant law was Canadian), but the discussion in the Oosterveld and Flah chapter on respondeat superior (applying, at minimum, to the Canadian state itself) normatively bolsters the case for liability of Canada for its agents’ torts, as does the general responsibility in international law of a state for harm by soldiers acting ultra vires as well as the positive duty of a state to prevent harm even by private actors where relevant state decision-makers have requisite knowledge of pending or ongoing harm. One reason Cunningham J. is able to be so cavalier, it would seem, is his curious insistence on discussing the action only in terms of a duty of care in negligence, completely leaving behind his initial observation that “assault [and] battery” (his distillation of the statement of claim’s reference to “torture, wrongful death, an/or murder”) were pleaded along with negligence. The closest he comes to revisiting these intentional torts is the cursory statement: “Nor are any facts pleaded which would support the plaintiff owed ‘legal and moral obligations’ to Shidane Abukar Arone.” Had the judge openly addressed the core claims relating to the torture which do not in the least involve an issue of negligence but rather direct and brutal intent to maim, he would perhaps have been less likely to have discussed the case as if it were about whether some abstraction called “Canada” had to be shown to be negligent in controlling its own soldiers. 56

Extracts from paras 9–16, Statement of Claim, supra note 54..

36 C Scott There is a hint in Cunningham J.’s reasons that he felt the duty of care did not arise because of the transboundary context, Canadian troops somehow not having a sufficient nexus to persons outside Canada, at least to family members of persons harmed by those troops: “If it can be said that members of the Canadian Forces on a peace-keeping mission in another country owe a duty of care to all citizens of that country beyond their ordinary duties as soldiers, it cannot be said that they owe any duty of care to . . . the parents.”57 On the face of this passage, Cunningham J did not feel he needed to decide the question of the extraterritoriality of duties of care, but at the very least he does not affirm that such duties exist and it may well be that doubts on this score had a sub-textual influence that bolstered his sense that no duty ran to the parents. Before addressing the dutyto-parents question, the duties-beyond-borders point warrants further mention. It bears emphasising that international human rights law is as clear as it can be that a state incurs state responsibility for the human-rights-violating conduct of the state’s own agents abroad, the duties in, for example, the International Covenant on Civil and Political Rights flowing from a mix of jurisdiction and opportunity to control and not from the blunt fact of territoriality.58 The just-quoted passage indicates that Cunningham J. had problems with the duty of care not simply because he seemed to fixate on some sort of duty of care on Canada independently of vicarious liability for harm caused by its agents’ breaches of their own duties of care, but also because he seemed to see the case as about whether there was a legal duty of care running directly between Canada and Arone’s parents. Terry notes how this misses the point that this is an action primarily about harm to Arone, with the issue of the parents suing being more one of standing to sue for that harm than an assertion of substantive duties owed directly to them.59 The judge has clearly said no substantive duty runs to the parents, a point to which I shall return, so it is clear he has erred in thinking this is the central question. But, then, he complicates matters by implic57

Arone, supra note 3 (emphasis added). On this, see Byrnes, supra note 41, who views article 14(1) of the Convention against Torture, on civil redress, as clearly applicable to torture committed by a state’s agents outside the state’s own territory. See also the reference to the ICCPR Casariega case, cited in Sornarajah, supra note 37, dealing with abduction and abuse carried out in Brazil by agents of Uruguay. Communication No. 56/1979, Lilian Celiberti de Casariego v. Uruguay, Yearbook of the Human Rights Committee 1981–1982, Vol. II, 1989, pp. 327–329, at paras 10.1–10.3. Another standard example of state responsibility for extraterritorial harm by its agents is the bombing of the Rainbow Warrior (and killing of a Greenpeace photographer) by French security agents. See also, as matter of general international law, the finding by the International Court of Justice in the Namibia case that South Africa was legally responsible for violation of a host of rights of persons in, and the people of, Namibia even though South Africa’s presence in that territory was an illegal occupation: “By occupying the Territory without title, South Africa incurs international responsibility arising from a continuing violation of an international obligation. It also remains accountable for any violations of the rights of the people of Namibia, or of its obligations under international law towards other States in respect of the exercise of its powers in relation to the Territory.” Legal Consequences for States of the Continued Presence of Africa in Namibia (South West Africa) notwithstanding Council Resolution 276 (1970), [1971] ICJ Reports p 16 (Adv Op). 59 Terry, supra note 18. 58

Introduction 37 itly acknowledging that there is the issue of Arone’s rights in tort surviving his death. Here, Cunningham J. resorts to an application of “trite . . . common law” that the death of an individual extinguishes existing tort claims unless such claims survive by virtue of legislation. One statute he deems relevant transmits some rights in tort to the estate of a deceased, but, since the parents did not purport to be suing in that capacity, their claim could not be recognised.60 Had the Arone family’s lawyer drawn the judge’s attention to the Convention against Torture (nowhere mentioned in the statement of claim), he would have been duty-bound to have regard to it in accordance with the principle that statutes must be interpreted, where possible, in such a way as to allow Canada to respect its international law obligations.61 Article 14(1), apart from speaking to the duty of Canada (and thus Canadian courts) to facilitate an “enforceable right to fair and adequate compensation” at least where the torture is committed by Canada (its human agents), also speaks to the issue of the right of the torture victim’s survivors. Its second sentence reads: “In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation.” This provision is creatively ambiguous in that it does not state the exact basis of the compensation, whether vicarious compensation for the family member’s death (and pre-death suffering) or whether compensation for their independent harm (pain and suffering, loss of an income earner, and so on)—or both.62 Without having to decide whether the family has freestanding rights for their own harm, article 14(1) would provide a clear reason for a Canadian judge to look long and hard at a statute in order to see whether “estate” should be interpreted liberally on facts such as these or even whether the context in tandem with article 14(1) leads to the interpretation that the issue of survival of rights and the capacity in which family members can sue falls outside the scope of the statute so as to be subject to common law (development).63 60 Keeping in mind that this is an alternative ground and the judge has already found the parents were not properly before the court because this is not the kind of case properly brought by a Litigation Guardian. The judge also dismisses actions related to the death of a family member found in the (deemed) relevant family law statute, but this shall be left aside for the moment because he does not decide the parents have no claim but, rather, that (a) the statute was not even pleaded and (b), even if it had been, its two-year limitation would have barred the claim. 61 Baker v. Canada, [1999] 2 SCR 817 at paras 70–71 (majority opinion of L’Heureux-Dubé J.), is the leading case decided by the Supreme Court of Canada, and discussed or noted in a number of chapters in this volume. In that case, the Court held that the Convention on the Rights of the Child, although not expressly incorporated into Canadian law by legislation, should be used to interpret the scope of discretion of an immigration official deciding whether a deportation of a mother of four Canadian children should go ahead. 62 Here, note the final judgment in the Pinochet saga—that of the extradition judge deciding that sufficient facts were made out by Spain to justify Pinochet’s extradition for, inter alia, torture: The Kingdom of Spain v. Augusto Pinochet Ugarte, Unreported Judgement of 8 October 1999, Bow Street Magistrates’ Court. The magistrate noted that the suffering of a family during the period when a family member has been disappeared (and is not known to be dead or alive and is feared to be undergoing torture) itself may be an independent human rights violation sufficient to amount to a form of “mental torture.” 63 This latter line of reasoning—i.e. is this even a case in which either existing Canadian domestic law or existing Somalian domestic law are fully applicable?—would arguably also be relevant to

38 C Scott Finally, time limitations. Cunningham J. found an array of Canadian statutory limitations periods which had not been met. In addition, he did not even seem to feel the need to address whether he had any authority to waive them. Right from the outset, there is a serious error of law in assuming Canadian law on limitations is the applicable law. As noted by Terry, in Canadian conflict of laws, limitation periods are matters of substantive law and the presumptive choice of law rule for an ordinary tort is the lex loci delicti (Somalian law), both these rules having been laid down by the Supreme Court of Canada in Tolofson.64 We will return to what law or rules should, or could, have applied, but note for the moment that the judge may, unknowingly, have been justified in turning to Canadian law because Somali law was not pleaded; traditional conflicts analysis would suggest that the lex fori is the default law when the applicable law is either not pleaded or not proved. So, assume that Canadian law should apply to the time bar question. The Arone family lawyer managed to file the claim on March 15, 1999, which is one day shy of six years from the date of Arone’s death. This is significant because the Crown Liability and Proceedings Act establishes a time limitation for tort actions of six years.65 I will return to the question of whether even this statute should apply, but must first note that the judge avoids finding the claim was brought within time by turning to what he considered lex specialis cutting back on the “generosity” of the six-year rule. He invokes the very short sixmonth rule of the Public Authorities Protection Act, which states in section 7: “No action, prosecution or other proceeding lies or shall be instituted against any person for an act done in pursuance of execution or intended execution of any statutory or other public duty or authority, or in respect of any alleged neglect or default in the execution of any such duty or authority, unless it is commenced within six months

the Litigation Guardian issue. Should Canadian law on this point be interpreted as exhaustive or merely permissive? Thus, even if the Arone family did not fit under existing Canadian law provisos for a Litigation Guardian (disability, absence), the statutes need not be interpreted as barring suits through a Litigation Guardian in such a novel transnational context as this. The force of analogy, combined with fairness and perhaps necessity, might suggest that the difficulty of instructing a Canadian lawyer from one’s home in a quintessentially “failed state” like Somalia (especially for a family of limited means) comes very close to satisfying the underlying rationales related to the existing bases for using a Litigation Guardian, disability and absence of a resident. Indeed, treating the absence condition as fused to the parallel condition that the absentee must first have been a resident of the jurisdiction is arguably irrational, at some general normative level, if the persons who are absent are within their rights to turn to Canadian courts but cannot reasonably be on site to instruct counsel. (All this being said, cryptic comments by the judge suggest this particular Litigation Guardian either was not truly authorized or would not be acceptable as a Litigation Guardian even if he were: “In fact, on the material available to me, no court would have appointed Abdullah Godah Barre as a Litigation Guardian.” This suggests far more going on in the case than meets the eye. In light of this throwaway comment, one might be forgiven for thinking Cunningham J. should have elaborated on this concern rather than going on to decide on the duty of care and limitations points in the truncated way in which he did.) 64 Tolofson, supra note 29. 65 Crown Liability and Proceedings Act, RSC 1985, c. C-50, s.32.

Introduction 39 after the cause of action arose, or, in the case of continuance of injury or damage, within six months after the ceasing thereof.”66

Section 269(1) of the National Defence Act, also found applicable by the judge, is materially identical to the just-quoted passage.67 There are at least three questions that the judge should have answered before assuming these two statutes covered these facts. Firstly, and most significantly, this being an action against the government and not against “any person”, do sections 7 and 269(1) even apply as lex specialis or should not the Crown Liability and Proceedings Act apply, as it states on its face, to “any proceedings by or against the Crown”? It is of some interest to note that the first and only time Cunningham J. clearly states that it is the vicarious liability of the Crown that is at issue is when mention of this fact seems, to the judge, to assist the Crown. In reference to the applicability of section 7 to the Arone family action, the judge simply asserts that “their claim . . . against members of the Canadian Armed Forces and vicariously the crown [sic]” is barred by section 7, when nowhere does section 7 indicate it covers actions against the Crown.68 Secondly, even if Crown liability is barred, the words “execution of . . . any statutory or other public duty or authority, or in respect of any alleged neglect or default in the execution of any such duty or authority” begs a major normative question which fairly shouts out for decision in a post-Pinochet world: should conduct constituting torture be interpreted as falling within this phrase or, just as the House of Lords found torture to be normatively incapable of being treated as “official” conduct, should these words be interpreted both to exclude torture as “duty or authority” and to mean that the words “neglect or default” should be read down so as to not include intentional violations of jus cogens human rights norms? In deciding whether to adopt a redress-facilitative interpretation of these six-month limitations provisions, it is worth bearing in mind that the issue is interpretation of these statutes for the non-dispositive purpose of allowing a case to go ahead and not in order to dispose of the case on the merits. Here, of course, a court must fully face up to the consequences of following Cunningham J.’s line, namely that dismissal of the case in Canada will almost certainly mean, in effect, the end of the case given the almost complete lack of a functioning state, including judicial, structure in Somalia. That is, the dismissal is not simply denial of access to Canadian courts, but denial of access to justice full stop. As he did not appreciate the case has a conflict of laws dimension, Cunningham J. likely did not turn his mind to this result, but future courts may do well to bear in mind the words of the House of Lords in the just-decided Lubbe case on the question of dismissing cases on grounds of forum non conveniens: “[A] stay will not be granted where it is established by cogent evidence

66 67 68

Public Authorities Protection Act, RSO 1990, c.P.38, s.7 (emphasis added). National Defence Act, RSC 1985, c.N-5, s. 269(1). Arone, supra note 3 (emphasis added).

40 C Scott that the plaintiff will not obtain justice in the foreign forum.”69 This succinct statement would seem suited to stand as a general principle of access to justice in a transnational context not necessarily limited to its core application in the forum non conveniens context. If so, it would seem reasonable to request courts to consider not only article 14(1) of the CAT but also this general common law standard as interpretive baselines in reading statutes, or applying common law rules, which have the effect of dismissing a transnational case once and for all. Thirdly, should supplementary principles of (transnational) common law be either read into limitations statutes or applied in lieu of the statutes, so as to permit the reasonableness of when an action is brought in all the circumstances of the case to be taken into account—a principle not of discoverability but of reasonable feasibility? Factors to take into account in the Arone/Somalia context could be: whether out-of-court payment discussions delayed the realisation that legal action should be taken; whether it seemed hopeless to persons such as the plaintiff, given local realities and general questions of power imbalance, to consider legal proceedings until such time as they became aware that it could be possible to sue in the courts of the state whose soldiers were the torturers; whether there were inordinate difficulties in communicating with lawyers in Canada; and so on. Admittedly, if the applicable statute is still taken to be one with a six-month rule, then a feasibility principle will only take one so far. Which returns us to the error in assuming, without more, that Canadian limitations law applies at all. It would be possible at the outset, on reasoning such as set out in the chapters of Moran and Raponi, to have immediate recourse to a (regulative ideal of a) transnational normative sphere in order to determine what limitations period, if any, should govern the case. A less radical position would be to start with Terry’s point that, by Canada’s own conflict principles, Canadian limitations law would not normally apply to foreign tort actions. If this is so, this suggests an extra level of freedom on the part of judges in a case such as Arone to apply, forge or otherwise draw on the principles most appropriate for the case, if it should turn out that Somali law is either not discernible or underdeveloped on the point. This includes a freedom not to apply existing Canadian law in contrast to mechanistically defaulting to it. Even then, if a judge insisted on Canadian law as the most principled default law, the strongest case is for the Crown Liability and Proceedings Act to be the most applicable to actions against the Crown.70 That being assumed, a careful reading of section 32 suggests that section 32 itself appears implicitly to embody the very principle of Tolofson (that limitations periods in torts are, presumptively, governed by the law under which the cause of action arises: the lex loci delicti) when it refers 69 Speech of Lord Bingham of Cornhill (for four of five Lords) in Schalk Willem Burger Lubbe (suing as administrator of the estate of Rachel Jacoba Lubbe) and 4 others v. Cape plc and related appeals, Unreported Judgment of the House of Lords, 20 July 2000. 70 This is especially the case when it is noted that none of the potentially-applicable statutes specifically speaks of the situation of Crown liability incurred outside Canada. That being so, it seems reasonable to resolve doubts as to which statute shall be deemed to govern in favour of the most human-rights-respectful statute—in other words, that with the longer period of limitation.

Introduction 41 to the six-year limitations rule as applying to “any cause of action arising in [a] province” of Canada. We come full circle. Even if direct recourse to the transnational sphere is not embraced, section 32’s circumscription seems to send us in that very direction. Where would one turn if one were to seek to forge a transnational limitations rule for this case? With John Terry, it seems reasonable as a matter of fairness to first affirm that a foreign tort plaintiff should not be worse off than would be the case had the cause of action arisen in a Canadian province; so, six years should be the floor and, as such, the Arone family claim was actually brought in time.71 Rather than stop with such a case-specific pragmatic result, however, it is worth briefly drawing attention to the kinds of multiple normative footholds identified by Moran as paradigmatic to the “new private law” she sees emerging. We might first note that, as a matter of both first principles and doctrinal generalization, limitations periods tend to be longer the more serious the underlying legal cause, especially in the criminal law. In terms of crimes which overlap with international human rights law, we could then note the capacity to bring criminal proceedings for crimes against humanity and war crimes that occurred as far back as the Second World War. As a matter of close analogy, this reasoning would seem to apply a fortiori to legal proceedings relating to similar underlying conduct where the consequences (financial liability) are less grave than for criminal convictions (imprisonment). At minimum, we might take a comparative-law look south of the border and note that the US Congress has set 10 years as the limitations period for civil actions under the Torture Victim Protection Act.72 Finally, we could look to see if public international law offers any established or emergent norms which Canadian and other courts would find relevant. On this front, most worthy of note is the rapidly-emerging consensus on principles governing the right to reparation for victims of serious human rights abuses. In particular, a set of principles formulated and then revised by Theo Van Boven, Special Rapporteur of the (then-named) UN Sub-Commission on Prevention of Discrimination and Protection of Minorities, is making its way through the UN Commission on Human Rights process. It appears to have attracted wide support, sufficient for Van Boven’s successor as rapporteur, Cherif Bassiouni, to have presented a final report in 2000 containing his draft set of principles which the Commission on Human Rights is invited to adopt at its 2001 session.73 Van Boven’s 1997 Second Revised Basic Principles and 71

Terry, supra note 18. TVPA, supra note 2. 73 The final of three drafts prepared by Van Boven can be found as “Note prepared by the former Special Rapporteur of the Sub-Commission, Mr. Theo van Boven, in accordance with paragraph 2 of Sub-Commission resolution 1996/28 [13 January 1997]”, Annexed to Question Of The Human Rights Of All Persons Subjected To Any Form Of Detention Or Imprisonment, Note by the Secretary-General, UN Doc. E/CN.4/1997/104 (16 January 1997), Appendix: [Second revised set of] Basic Principles and Guidelines on the Right to Reparation for Victims of [Gross] Violations of Human Rights and International Humanitarian Law (“Van Boven Second Revision, 1997”). The 72

42 C Scott Guidelines on the Right to Reparation for Victims of Gross Violations of Human Rights and Humanitarian Law included the following provision: “9. Statutes of limitations shall not apply in respect of periods during which no effective remedies exist for violations of human rights or international humanitarian law. Civil claims relating to reparations for gross violations of human rights and international humanitarian law shall not be subject to statutes of limitations.”74

This appears to have been narrowed somewhat in Bassiouni’s proposed principles: “6. Statutes of limitations shall not apply for prosecuting international human rights and humanitarian law norms that constitute crimes under international law. 7. Statutes of limitations for prosecuting other violations or pursuing civil claims should not unduly restrict the ability of a victim to pursue a claim against the perpetrator, and should not apply with respect to periods during which no effective remedies exist for violations of human rights and international humanitarian law norms.”75

It can be seen that the Bassiouni has, while maintaining a hiatus for so long as there are no effective remedies, confined the no-limitations rule to prosecutions for crimes under international law while opting for a no-undue-restriction test for all human-rights-related civil claims, whether or not the violation in question also constitutes a crime under international law. However, the principle he states is clearly inconsistent with excessively short periods such as six months and otherwise-arbitrary limitations rules. There is, furthermore, nothing in his formulation that precludes drawing analogies between criminal and civil remedies. What is an “undue restriction” for a torture claim must surely take into account the fact that there are no limitations at all for criminal prosecutions related to the same conduct. In a context such as Somalia, the further emphasis in the Bassiouni principles on availability of effective remedies much certainly play a significant role for a judge in Canada deciding whether it would be an “undue restriction” to cut off the claim of a torture victim’s family after six months, and indeed even after six years. As noted at the outset of one of his reports, Van Boven’s mandate was to draft these guidelines “in the light of existing relevant international instruments.”76 Thus, the principles he has put forward are mainly distillations of, and to some final report of the subsequent rapporteur, Cherif Bassiouni, was delivered for the March 2000 session of the UN Commission on Human Rights: Final report of the Special Rapporteur on the right to restitution, compensation and rehabilitation for victims of grave violations of human rights and fundamental freedoms, Mr. M. Cherif Bassiouni, submitted in accordance with Commission resolution 1999/33, UN Doc. E/CN.4/2000/62 (18 January 2000), Annex: Basic Principles and Guidelines on the Right to a Remedy and Reparation for Victims of Violations of International Human Rights and Humanitarian Law (“Bassiouni Final Principles, 2000”). 74 Van Boven Second Revision, 1997, ibid. (emphasis added). 75 Bassiouni Final Principles, 2000”, ibid. (emphasis added) 76 UN Doc. E/CN.4/Sub.2/1996/17 (24 May 1996), [First revised set of] Basic principles and guidelines on the right to reparation for victims of gross violations of human rights and humanitarian law prepared by Mr. Theo van Boven pursuant to Sub-Commission decision 1995/117, para. 1.

Introduction 43 extent principled extrapolations from, the current legal state of affairs. The more conservative formulations of Bassiouni suggest that, after consultation with states, the final principles he recommends come closer to tracking (albeit by clarifying and elaborating) existing law than engaging in much of the kind of law-forging that often occurs in a codification process, and such as Van Boven seemed more inclined to engage in. This suggests that, while the Bassiouni principles need not be taken as a ceiling for Canadian judges (especially when it is the government of Canada being sued and not a human defendant with her own rights to consider), they should be viewed as uncontroversially providing a normative floor. Of course, when the set of principles becomes an instrument adopted by the Commission on Human Rights and perhaps also by the General Assembly, it will have even greater normative relevance, albeit as a “soft law” instrument. “Soft law” refers to instruments whose normative force derives not from formal bindingness but from the substantive worth of their contents in tandem with relevant understandings that grow up around it in that regard. But, norms that are neither “hard” treaty norms nor clear customary law norms have often been invoked by Canadian government lawyers before Canadian courts or referenced by courts themselves. As such, drawing on these just-quoted draft principles and, later, on any finally-adopted principles would be a legitimate way to seek normative guidance in the course of deciding on what transnational legal principles to embrace.77 77 To make sure this point about the flexibility of international normativity is clear as is the legitimacy of invoking “soft law” in the course of judicial reasoning, a few examples from the practice of the Supreme Court of Canada may assist. Numerous other examples could be given from this Court other than those that follow. Canada (Attorney General) v. Ward, [1993] 2 SCR 689 (“While the drafting history of the [Geneva] Convention [on Refugees] may not go far in justifying the exclusion of state complicity from the interpretation of “Convention refugee”, other sources provide more convincing support. A much-cited guide on this question is paragraph 65 of the UNHCR Handbook on Procedures and Criteria for Determining Refugee Status (“UNHCR Handbook”). While not formally binding on signatory states, the Handbook has been endorsed by the states which are members of the Executive Committee of the UNHCR, including Canada, and has been relied upon by the courts of signatory states.” Per La Forest J for the Court); Pushpanathan v. Canada [1998] 1 SCR 982 (See the invocation of numerous—a total of 20—soft law documents by Cory and Iacobucci JJ in their dissenting opinion; these included General Assembly resolutions (no less than nine), draft treaties, International Law Commission draft principles, UN reports and so on, all of which were pleaded and put before the Court by the Government of Canada lawyers.); R v. Ewanchuk, [1998] 1 SCR 330 (“The Committee on the Elimination of Discrimination against Women, G.A. Res. 34/180, U.N. Doc. A/47/48 (1979), established under Article 17 of the Convention [on the Elimination of Discrimination Against Women], adopted General Recommendation No. 19 (Eleventh session, 1992) on the interpretation of discrimination as it relates to violence against women [going to quote passages] . . .”, per L’Heureux-Dubé and Gonthier JJ.at para 71; note, as suggested by the word “Recommendation” that the Committee’s authority does not include the power to directly bind states parties to its interpretations of the meaning of the Convention, although states conforming in good faith with their treaty obligations would need to offer persuasive reasons why they would not adhere to the Committee’s view. Also, in the same opinion: “On February 23, 1994, the U.N. General Assembly adopted the Declaration on the Elimination of Violence against Women, G.A. Res. 48/104, U.N. Doc. A/48/49 (1993). Although not a treaty binding states, it sets out a common international standard that U.N. members states are invited to follow . . .”: at para 72.) R v. Crown Zellerbach, [1986] 1 SCR 401 (“The internal marine waters of a

44 C Scott state are those which lie landward of the baseline of the territorial sea, which is determined in accordance with the rules laid down in the United Nations Convention on the Law of the Sea (1982)” per LeDain J at para 38. Note that this convention was not in force in 1986, that even Canada had yet to ratify it in 1986, and that its relationship to customary law was highly controversial four short years after adoption. As such its textual provisions referenced by the Supreme Court of Canada were classic examples of “soft law.” In the same paragraph, the Court makes use of another soft law source: “This impression is reinforced by the United Nations Report of the Joint Group of Experts on the Scientific Aspects of Marine Pollution, Reports and Studies No. 15, The Review of the Health of the Oceans (UNESCO 1982) (hereinafter referred to as the “U.N. Report”), which forms part of the materials placed before the Court in the argument.”).

2

Translating Torture into Transnational Tort: Conceptual Divides in the Debate on Corporate Accountability for Human Rights Harms CRAIG SCOTT 1

1 INTRODUCTION R O M V A R I O U S P E R S P E C T I V E S and on various specific doctrinal issues, each contribution in this collection seeks to grapple with the relationship between two core categories of “public law”, on the one hand, and “private law”, on the other—torture and tort, respectively. The foundational premise of the present chapter is that, in understanding a relationship such as this, categorical distinctions tend to exert a constant pressure on jurists to organize the normative world of human rights in terms of (unduly) dichotomous ways of thinking. And, even when a jurist is open to seeing distinctions amongst categories less in terms of dichotomies and more in terms of differences along various continuua, the tendency to organise judicial minds around a strong contrast between opposite ends of a continuum—that is, in terms of polarities—may result in instinctive either/or dichotomising. Bracketing for the moment some of the analytical virtues of thinking in terms of oppositional categories, the extent to which one adheres to dichotomies or polarities as a way of ordering one’s juridical universe will affect the ease and justifiability of moving from one legal realm to another. One way to think of the movement across boundaries, between categories, or from one pole to another is in terms of “problems of translation”. Can a human rights norm which, according to (one reading of) public international law, applies only to states be translated into a civil cause of action within a domestic legal system? If so, can this translation go so far as to include civil suits for foreign delicts and not only for conduct occurring (or, at least, causing harm) in the other state whose courts are seized with an issue? And so on. The more one

F

1

Osgoode Hall Law School of York University, Toronto.

46 C Scott thinks in terms of categorical oppositions as somehow real (as opposed to simply useful starting points), the more likely one is not only to perceive problems of translation, but also to perceive those problems as serious. Conversely, in the context of the movement from (or, more accurately, between) torture and tort, the more that one approaches poles with a healthy degree of suspicion and willingness to inquire into the normative spaces that lie between them, the more likely it is that one will see value in forging (at minimum) overlapping categories of accountability or (more maximally) hybrid categories of accountability. The purpose of the present chapter, then, is to present a series of common contrasts (whether approached in terms of dichotomies or polarities) and to do so with as much neutrality as can be mustered. An important qualification is that this chapter also seeks to make a contribution to the doctrinal debates on transnational tort liability of corporate actors. The chapter’s premise is that an important measure of analytical clarity can be achieved by situating this book in terms of possible stances that one could take on any given doctrinal or theoretical “translation” problem as a function of each contributor’s approach— whether implicit or explicit—to each in this series of relevant pairings. The poles provide stylised reference points which help construct the views of two kinds of juridical characters on the issue of liability for human rights breaches. We might call these two characters the “restrained conservative” and the “activist radical”. The (stylised) position of the restrained conservative insists that international human rights standards are a matter of public law (vertical) applicability wherein corporate conduct is regulated through indirect state responsibility which attaches only to corporate harm caused within a state’s own territorial space and which is assessed at the level of international treaty law by quasi-judicial institutions whose role is to apply, with no binding authority, the existing law (including any new additions to it by interstate agreement). By contrast, the (stylised) activist radical insists that international human rights standards are (not only, but) also a matter of private law (horizontal) applicability wherein corporate conduct may be regulated through direct civil liability which is capable of attaching to harm caused by corporate conduct outside a state’s own territorial space and which may be assessed at the level of domestic law (receiving international law values) by national courts whose role is to interpret the law so as to creatively advance human rights values and to render decisions directly binding on the impleaded corporate actors. Thus described, these characters are, of course, as much caricatures as anything else, but they serve an important analytical function: highlighting the complex range of positions that any given person might adopt, in the real world of juridical decision and analysis, on the issue of transnational corporate accountability for human rights violations. To take just one example, a less-than-radical conservative position would be to accept a measure of involvement of national courts in extraterritorial corporate conduct (beyond existing private international law involvement revolving around already-standard private law causes of action) but only pursuant to specific statutory authorization which demonstrates the state’s legisla-

Translating Torture into Transnational Tort 47 ture has taken responsibility for making its domestic legal system the agent of international legal values in this way.

2 STATE RESPONSIBILITY : DIRECT VERSUS INDIRECT RESPONSIBILITY

“State responsibility”, simply put, is the name public international law gives to the normative state of affairs which occurs following a breach by a state of one of its international legal obligations (whether that obligation derives from treaty law, customary law or other recognised sources such as “general principles of law”). The obligations in question can be negative or positive; that is to say, a state incurs “state responsibility” when it fails to do something international law requires it to do (put differently: when it breaches a positive obligation) no less than when it does something which international law prohibits it from doing (put differently: when it breaches a negative obligation). Thus, a state may breach international law by act or omission, depending on what obligations an international norm places on it. In this sense, a state may breach a positive obligation (e.g., the duty to put into place, within national law, a particular kind of system for regulating the export of hazardous wastes abroad) no less “directly” than it may breach a negative obligation (e.g., the duty not to use police force against a foreign embassy). That being said, international lawyers do commonly use a form of shorthand which distinguishes between so-called direct responsibility and so-called indirect responsibility. This distinction does not in fact refer to any difference in the status of the responsibility a state may incur but, rather, simply signals the existence of one species of duty that international law can place on states, namely positive duties to protect some nonstate actors from being harmed by other non-state actors. For instance, we say that a state is indirectly responsible if an unruly mob or a group of criminals causes physical harm to a person, where that state was in a position, if it had exercised due diligence (reasonable care), to have prevented the harm from taking place. One way to understand why this term is used is to think of state actors not as the direct or immediate agents of harm, but as indirect agents or secondary authors of harm. To reiterate, the state has an obligation which it “directly” breaches, but the substance of the obligation is such that it is the harm (or harm-risking conduct) of non-state actors which triggers the state’s obligations. To return to the example used above, a corporation from Canada may dump a load of toxic waste in a West African country. By current international law (the Basel Convention and perhaps customary law as well), it would be Canada, not the corporation, which would incur responsibility under international law if it failed to regulate the company in the required ways (e.g., if it had handed over responsibility for issuing waste-exporting permits to a selfregulating industry body). Finally, and perhaps most importantly, the existence of this shorthand category called “indirect (state) responsibility” attests to a wholesale substantive gap in mainstream conceptions of the actors to whom

48 C Scott international law normally applies. Because, as a general rule, corporations are not thought in mainstream circles to be capable of directly violating public international law, it is by way of regulating states (requiring them to regulate corporations) that public international law creates a degree of regulation of corporations—indirect regulation.

3 AXES OF LEGAL ACCOUNTABILITY FOR HUMAN RIGHTS VIOLATIONS : THE VERTICAL

( PUBLIC

LAW ) VERSUS THE HORIZONTAL

( PRIVATE

LAW )

The final sentence in the preceding section leads us to a more specific version of the generally perceived juridical incapacity of privately-owned corporations, as non-state actors, to breach international law. International human rights law (at least, international human rights treaty law) has adopted a tendency from the constitutional systems of a large number of states to see human rights as generating obligations for states only. In this sense, human rights protection is seen as part of the domain of public law, with the legal relations in question running between individuals or groups and the state. A private actor seeking to use a constitutional bill of rights in national law or a human rights treaty in international law as the source of rights on which to found some cause of action (e.g., a human rights tort) against another private actor has, by and large, fallen outside the terms of mainstream legal discourse except in a few countries such as Germany. In recent years, helped along by constitutional debate in the new South Africa, it has become common to distinguish between the “vertical” as opposed to the “horizontal” axes of human rights obligations. Human rights run vertically when it is only the state owing the obligation(s) to respect, protect and fulfil them. Here, it is also worth noting that the standard form of legal vindication of the rights is the public law process of judicial review; comparatively rare are the situations in which the affected individual or group can use the human rights norm as a basis for a civil suit in which monetary compensation is sought from the state for the violation. Partly as a result of the insights of comparative legal theory and partly due to growing sensitivity to the issue of private power, new or revised constitutions, such as that of South Africa, are now more likely to stipulate that the rights in the constitution apply to private no less than public actors, at least mutatis mutandi (making necessary modifications that reflect the differences in extent or nature of obligations that may be generated for different actors). This application of human rights in states’ basic laws to private actors is spoken of as a situation of “horizontality”. By and large, international human rights law has neither led nor tracked this movement towards horizontality and has been content to leave the regulation of corporate harms to the indirect responsibility of states. The duty “to ensure” human rights (a duty that appears in a number of human rights treaties) or, amounting to the same thing, the duty to “protect and fulfil” human rights (which scholarly doctrine has added to legal discourse) provides the conceptual basis for interpreting

Translating Torture into Transnational Tort 49 states to have positive duties to prevent and respond to harms caused by corporate actors no less than the traditional unruly mob besieging a foreign factory or residential enclave. The slow percolation of thinking (and some corresponding international human rights case law) around the content of states’ duties with respect to corporate activity is occurring in a process of cross-pollination with the development of positive state obligations in the “private” family sphere. In this latter sphere, analysis of the duties attendant on protection of women’s and children’s rights is considerably more advanced (or, at least, it has been more successful in influencing official human rights discourse).

4 SOURCES OF LAW : INTERNATIONAL LAW VERSUS NATIONAL LAW

Were international human rights law to develop an understanding of at least some human rights obligations owed directly by corporations as a matter of international law (whether general or treaty), would this translate into enforcement of these human rights within national legal systems? The answer to this question is relatively simple: it depends on whether international and national law are treated as separate sources of legal obligation and, further, on whether international obligations are treated as automatic sources of obligation in domestic law. More specifically, it depends (mostly) on the governing (constitutional) conception within a given national legal system of the system’s relationship with international law and also (somewhat) on whether international law’s development of the corporate obligations is accompanied by a reciprocal duty on states to give direct effect to these obligations within the domestic legal order. As such, the extent to which a dichotomy operates as between international and national law is largely determined by both the forms and the ethos of reception in each jurisdiction. The discussion in the following paragraphs assumes that the issue is whether a state’s domestic law will recognize international law’s development of human rights obligations owed directly by corporations as applying to harms caused by corporations and occurring within that state’s own territory. (The gloss of extraterritorial application of the same norms will be dealt with in subsequent sections.) In common law countries which have inherited the UK’s Westminster conception of government, such as Canada or India or Singapore, there exists by and large a radical formal separation between international treaty law and its implementation by domestic courts, such that a treaty norm does not have the direct force of law but rather must be given effect by legislation; even if a statute in such a country specifically and in verbatim terms provides for an entire treaty to be applied by its courts or designated tribunals, it is still the case, in formal terms, that those institutions are applying national (statute) law and not international (treaty) law. Although not free of conceptual difficulties, it is more common for it to be said in such countries that customary law is automatically part of the law on the same level as, and indeed embedded in, the common law;

50 C Scott but, even here, it is probably just as true to say that custom is part of the local law only when the judiciary makes it so by adjudicating in a concrete case upon the existence and content of a claimed customary obligation. In legal systems outside the Westminster tradition, for example some European continental systems, and also in the United States, treaty obligations can be directly adjudicated upon, although the conditions that must be fulfilled (in terms of the nature and “justiciable” quality of the obligation in question) can vary considerably from system to system. The status of the treaties that are received in this direct fashion will also vary, from being treated as superior even to the constitution, to being treated on a par with constitutional norms, to being treated as akin to statutory norms. Leaving aside the US, customary international law is less easily treated as a direct part of the law of the land in those civil law jurisdictions in which codification is the essence of law and which accordingly hold resolutely to the notion that all law stems from either the (written) constitution or from statutes. That being said, it is not unknown for some of these systems to accord custom a superior legal status, either as equivalent to constitutional norms or as hierarchically superior to statutory norms; while such treatment is invariably something provided by the constitution itself, it remains, for these systems, a potentially significant port of entry for the enforcement of non-treaty corporate obligations. It should finally be noted that there is a significant element of commonality in how most, if not all, domestic legal systems handle the “national versus international” tension. Specifically, one principle of statutory interpretation seems so widely shared across legal systems that it probably is itself part of international law qua general principle of law: the presumption that the legislature does not intend to place the state in a position of breaching international law and the associated judicial duty to strive to interpret legislation, without distorting it so as to avoid an interpretation that would have this effect. Much can be done in all legal systems to create a harmonisation between international legal obligations and domestic law if the courts actively and rigorously rely on this principle of interpretation. It could be claimed that this same presumption, grounded as it is in the ideal of the rule of law, should be equally available as a directive to courts in common law countries when they are faced with the choice of whether or not to develop the common law in situations where the legislature does not appear to have implemented an international legal obligation and where that obligation is capable of enforcement in a way analogous to existing common law causes of action. However, to the extent that the international legal obligation in question is a treaty norm and cannot plausibly be said to have a parallel existence in custom, this claim would be strongly resisted by many as something which it should be presumed a court may not do; on this contrary view, the act of using a treaty to influence the judicial creation of new positive law is simultaneously an act of circumventing the formal rule of reception that requires the content of treaties to enter the domestic law through legislative enactment.

Translating Torture into Transnational Tort 51

5 FIELDS OF INTERNATIONAL LAW : PUBLIC AND PRIVATE

A whole book would need to be written to adequately convey the multifarious and complex relations which do, or could, exist between the two fields of law called “public international law” and “private international law”. This relationship is at the very core of any development of corporate liability within one state’s legal system for human rights violations that take place in another state’s territory. Both fields of law deal with legal relationships which transcend the single legal system(s) of any given state. Public international law is that collection of norms the primary function of which is the regulation of states’ relations inter se—and, to that extent, regulation of states. The closest analogy is to constitutional norms that govern the way federal units within a given state function so as to respect each other’s “sovereign” jurisdiction while permitting the units to act in concert for a common interest. In terms of obligations placed by public international law on non-state actors, where the conduct of other actors is directly regulated (for example, ships or international organisations or natural persons), it is generally thought that the regulation takes the form of regulation by states. Primarily since the advent of the UN Charter era, public international law has increasingly begun to assign rights to individuals qua human beings, thereby reaching inside the governance structure of states to an extent unknown when the primary focus of public international law was on providing rules of coordination to make interstate coexistence and mutually beneficial cooperation possible. Yet, crucially, this attribution of rights under international law (primarily through multilateral treaties with a core of general international law norms also being generally regarded as having crystallised as well) is still very state-centred, in that the holders of the duties that correlate with human rights are, as noted above, thought to be primarily states. Here the public law analogy within state systems is, of course, constitutional bills of rights which bind government in its relations with those it governs. It is a marginal, and some would say non-existent, function of public international law to regulate the substantive rights and obligations in private law relationships between private persons. Private international law is a field of law which seeks to regulate private law relationships across borders—but structurally and not in terms of providing substantive rules that govern relationships. Although an oversimplification, private international law (often also called “conflict of laws” or “conflicts”) deals with and seeks to answer three main questions. First of all, when a plaintiff initiates proceedings (e.g., a tort claim) before a given court, does the court have the requisite connections to the dispute for it to have jurisdiction to hear the case (adjudicative jurisdiction) and, if it does, should it nonetheless exercise some form of discretion to decline to keep jurisdiction (for instance, on the ground of it being an inappropriate forum—a forum non conveniens)? Second, if the court does assume jurisdiction, which legal system’s rules and principles will it apply

52 C Scott to resolve each of the private law issues submitted for resolution? (If the most significant negligent conduct leading to harm occurs in a state different from the state in which the harm occurs, which state’s law should govern?) Third, in situations where a foreign court has decided a private law dispute with geographically complex facts and issued a judgment, and one of the parties to the foreign case seeks to have a court in another state give legal effect to that judgment (by recognition or enforcement), must that latter court do so or may it decline to do so? Thus, it can be seen that private international law, as commonly conceived, deals with framework questions that allow decisions to be made as to which courts may lawfully determine a private law dispute and as to what system of norms must be applied. But, like public international law, it does not (again, on the common conception) lay down the substantive private law rules. It only tells you where to find them, namely, in which domestic legal order or other. Beyond a mainstream understanding that national law creates the substantive rules and private international law simply determines which national law to apply, it is also commonly said that public international law does not regulate the framework rules of private international law. Thus, it is said, each national legal system has its own body of private international law rules which may happen to be informed by comparative law doctrine and even by background norms of public international law (as to the limits of one state seeking to regulate another state’s internal legal order), but general (non-treaty) public international law does not constrain what rules on adjudicative jurisdiction, on choice of law and on recognition and enforcement of foreign judgments each state’s legal order may adopt. States may consent in treaty form, and increasingly do (through the Hague Conference on Private International Law and the European Union) to certain framework rules, but the exercise of express consent to be bound still keeps this phenomenon of public international legal regulation of private international law within a paradigm of each state choosing what rules it will adopt. There was a time in the history of international law, prior to the height of state-centred domestic and international legal thought, when that body of law was thought to regulate the rights and obligations of all relevant “transnational” actors and relationships. Where international law did not directly provide rules that regulated the substance of a private law relationship, at the very least it provided framework (choice of law and jurisdictional) principles as to which law governed actors of different nationalities in different territorial contexts when one actor alleged a wrong against another. More recently, a school of international legal thought in the United States began to speak of international law, public and private, in conjoined terms often using the term “transnational law” to describe principles that are neither purely national nor international and which apply to private law and public law relationships. What results, in effect, is law that is neither national nor international nor public nor private at the same time as being both national and international, as well as public and private. In the realm of international commercial transactions, it is increasingly said that we are seeing the resurgence of the old “law merchant”

Translating Torture into Transnational Tort 53 (lex mercatoria) according to which a fusion of the practice of business actors and various degrees of informal acquiescence and formal endorsement by states has produced a hybrid body of transnational law in this substantive field, a body of law that increasingly is articulated through international commercial arbitration that outflanks both domestic and international dispute settlement through courts. The notion that such a transnational law exists in areas of private law such as personal injury torts remains marginal, as there is not a community of practice that has forged consensus on desirable norms; in the absence of such self-regulatory practice, victims of personal injuries have begun to rely on international human rights law values as the foundation for a kind of transnational law of delictual civil liability.

6 GEOGRAPHICAL SCOPE OF NORMS : TERRITORIAL VERSUS EXTRATERRITORIAL

The previous section noted that it is generally assumed that international law does not lay down substantive private law rules. Nor does it specify what legal system must be applied to a given private law dispute where there are geographically complex facts. There is, however, a growing consensus that the radical separation of public international and private international law is no longer sustainable, if it ever was, at least in respect of the public law principles of the limits on jurisdiction. That is to say, public international law provides at least abstract principles derived from the foundational norms of sovereignty and the (presumptive) territoriality of jurisdiction—although not necessarily many precise rules—that structure the obligation of one state and its judicial system not to regulate subject-matter and activity that is more properly regulated only by some other state’s (or states’) legal order(s). Territoriality is not the sole basis for regulation, although it provides the benchmark against which the weight and justifiability of other potential “extraterritorial” grounds for jurisdiction can be assessed. Prima facie grounds for one state having sufficient connection to an extraterritorial matter to seek to regulate it in some manner range from the nationality of the harm-causing actors to the universality of concern to regulate certain categories of activity. It is common in some legal systems, for instance, to make that system’s criminal law applicable to conduct by a state’s nationals outside its own borders. However, assessing competing claims to jurisdiction is not neutral as to the subject matter being regulated, and some subject matter attracts louder cries of “extraterritoriality!” than others. The more that one moves away from areas of common criminality, the more that states become sensitive when other states seek to regulate the conduct of their nationals, notably their corporate nationals, in the economic realm. One reason for the sensitivity is a dominant view that economic policy belongs in some more intrinsic way to the very idea of sovereignty (for instance, as signalled by the notion of “permanent sovereignty over natural resources”). Economic

54 C Scott self-determination is thought to include the sovereign prerogative of states to make decisions on what economic development path to pursue, including the right to take active advantage of an ability to attract investment capital through minimalist labour rights protection. It is also the case that economic policy can vary significantly from one jurisdiction to another, such that extraterritorial regulation of corporate conduct abroad is much more likely to interfere with, or pre-empt, policy choices the host state has made or has chosen not to make. In the context of transnational corporate accountability, a further crucial spin on the limits of reasonable assertion of extraterritorial jurisdiction is the issue of jurisdiction over which corporate entity for which conduct. This is the problem posed by the principle of separate legal personality of corporations and their shareholders (including shareholders that are parent companies). It may well be that it is acceptable for Canada to assert jurisdiction over a Canada-based parent company (Company P) where conduct in Canada has a sufficient causal relationship to harm in, say, Malaysia. And it is without question reasonable for Malaysia to assert jurisdiction over that company’s Malaysia-based and Malaysia-incorporated subsidiary (Company S) for Company S’s conduct in Malaysia that leads to harm. But may Canada reasonably regulate the conduct of Malaysian Company S, either directly or by piercing the corporate veil and deeming Company P directly or vicariously responsible for Company S’s conduct? And may Malaysia assume both prescriptive and adjudicative jurisdiction over Company P if Company P has taken care to have no legal presence in Malaysia qua Company P? By asking these questions in this way, it can be seen how separate corporate personality can produce gaps in liability if one or both interested states are unwilling or legally unable to exercise a measure of extraterritorial jurisdiction. Three general kinds of jurisdiction may be spoken of, each intruding in different ways into foreign states’ affairs when they operate extraterritorially. “Enforcement jurisdiction” generally refers to non-judicial forms of state enforcement of the law through police or similar power. For Canada to send its governmental inspectors to factories of Canadian companies in Malaysia and order the companies to obey Canadian law (say, on workplace safety) would generally be viewed as unreasonable, indeed intolerable. Enforcement jurisdiction is, then, essentially territorial; only another state’s consent can operate as a defence to the unlawfulness of extraterritorial enforcement. The next most intrusive form of extraterritorial jurisdiction is “prescriptive jurisdiction”, that is to say the power of a state to have its constitutional, statutory or common law rules apply to persons or activities outside its own territory. Here the assumption is that the rule applies abroad (say, Canadian law on workplace safety) but is enforced in Canada, for instance against a corporate officer in charge of the foreign plant. The intrusiveness is less if the officer is Canadian rather than, say, Malaysian. But, even though enforcement may take place on Canadian territory, by that fact it is still seeking to alter behaviour taking place in another country. There is finally “adjudicative jurisdiction”. This is placed as the third and least intrusive

Translating Torture into Transnational Tort 55 form of jurisdiction in the private law realm because, under the basic structure of private international law, the simple fact of a judge taking jurisdiction over a case does not mean that the lex fori (law of the forum) is applied. Rather, there is always the possibility that the choice of law process will lead to Malaysian law (e.g. on workplace safety) being applied. By applying the foreign law, foreign sovereignty would, in a sense, be affirmed and the only element of intrusion would lie in having a foreign court be the one to interpret the foreign law (no insignificant matter) and apply it, once interpreted, to the facts of the case. To the extent that assets of the person against whom judgment is then issued are located in the foreign country (e.g. Malaysia), the fact that Malaysian courts will have the final decision as to enforcement of the (Canadian) judgment against those assets also mitigates the sovereignty-intruding aspects of adjudicative jurisdiction per se. If, however, the court not only takes jurisdiction but also applies its own law, then the intrusiveness of extraterritorial prescriptive jurisdiction kicks in, exacerbated by the actualization of that jurisdiction in a concrete case.

7 STATE REGULATION OF NATIONALS ’ CONDUCT ABROAD : LIBERTY VERSUS OBLIGATION

Following on from section 6, mention should be made of the necessity to distinguish between powers and duties. The foregoing discussion of the limits of jurisdiction concerned the jurisdictional power of states to regulate matters with an extraterritorial dimension. The more consensus there is of a common international interest in a specific form of legal sanction with respect to specific subject matter the more that this will count in favour of the acceptability of extraterritorial regulation. The point at which that consensus becomes so widespread and clear that states are no longer simply permitted to regulate a matter but required to do so is the point at which we move from the realm of state jurisdiction to state obligation—i.e. to state responsibility which is incurred when a state fails to provide for jurisdiction in its domestic law and to exercise it where the triggering facts are present. For example, normative discourse has progressed to the point with respect to the problem of child sex tourism that some states, such as Canada and Australia, have made it a criminal offence for their nationals to have “sex” with children anywhere in the world. Little if any protest from states afflicted by the sex-tourism trade, such as Thailand and Sri Lanka, has occurred, and the debate has rapidly gone to another level. The real question now is not whether states are permitted to regulate their nationals’ conduct but whether they have a duty to do so as an extension of their duty to ensure human rights. The more debate focuses on this question, the more it is reasonable to assume that states at least have (prescriptive and adjudicative) jurisdiction over their nationals’ behaviour. However, the truly interesting question from the perspective of the theme of this chapter is whether two variants on the just-described sex-tourism

56 C Scott regulation would meet with the same general acquiescence. The first variation would be to take the regulation out of the context of criminal law sanctions over individual tourists and extend the regulation to some form of regulation of corporate behaviour (e.g. a civil liability regime) with respect to those national travel agencies and national tour operators that deliberately facilitate such tourism. The second variation would be to see if regulation, whether criminal or corporate, could be justified beyond a nationality basis for that jurisdiction. That is to say, if Australia began to allow civil suits against Japanese corporate sex-tour operators organising trips to Bangkok or Phuket, would Japan and Thailand accept this as a reasonable exercise of extraterritorial jurisdiction?

8 THE EXISTENCE OF RIGHTS : NORMATIVE VERSUS INSTITUTIONAL

The issue of jurisdiction shades into the question of the relationship between the “existence” of legal norms and the authority (or appropriateness) of a given institution giving some legal effect to those norms. Depending on one’s view of the sources of international law and how one interprets those sources, it may well be, for example, that as a normative matter, international law makes it a (civil) legal wrong for one private actor to torture another private actor. This does not, without more, give a foreign state legislative authority to sanction that conduct through its law, at least when there is no connection (other than humanity itself) to the specific wrong. International law may conceivably create a wrong but not assign universal jurisdiction to its enforcement. It may do so knowing, as it were, that politics and moral suasion are the ways in which the norm is to be enforced, or leaving enforcement to the creation, in time, of an international institution capable of supervising implementation of the norm in some juridical fashion. The basic point is that normative wrongs can exist without institutions to sanction those wrongs or, more commonly, wrongs can exist where political institutions and processes are the preferred avenues for resolution. Two further examples are relevant. First of all, it may be that the norm/institution tension does not manifest itself in all-or-nothing fashion. It may be, for instance, that a foreign court should have authority to adjudicate, as a delict, torture occurring anywhere in the world, but it may also be that the court should assume a rather positivistic stance to its interpretive function. That is to say, whereas the court might be well within its institutional authority to adopt a creative interpretive role in interpreting what torture means in its application to conduct within its own society, it may only be justifiable for a court to act on a solid core of consensus in the “international community” when applying that norm to conduct that occurs abroad. A second example takes us to the international realm. There may come a time when some of the UN human rights treaty bodies begin to interpret the rights in the various UN human rights treaties as placing direct obligations, as a matter

Translating Torture into Transnational Tort 57 of treaty law, on non-state actors. This does not, in and of itself, give those bodies institutional authority to exercise direct review over those actors, such as corporations. They may have to filter assessment of corporate liability through indirect state responsibility until such time as an interstate treaty process expressly assigns them a new monitoring role, or, with time, states come to recognize that the treaty bodies’ institutional authority has evolved, as a matter of necessary implication, to include the power to assess non-state conformity with the treaty norms. Again, the point is that the existence of norms within a legal order does not in and of itself decide the question of allocation of institutional authority to judge those norms.

9 INSTITUTIONAL POWER OF NORMATIVE DEVELOPMENT : LEGISLATIVE VERSUS JUDICIAL

Directly related to section 8, a central tension stems from the necessity to take seriously the following question: given the state of international law on the applicability of human rights norms to corporate actors, should judges assume the authority to develop such accountability without express or at least clear authorisation from the relevant legislature? This question is bound up with the general question of justiciability and the associated debates on the relative competence and legitimacy of courts and legislatures in relation to law-creation activity. When one grafts onto the general issue of the appropriate function of courts the fact of an interstate context with attendant foreign policy sensitivities, it can be seen that there will be a constant tug-of-war between the peculiarly judicial responsibility to take human rights law seriously (and corresponding degree of authority to be interpretively generous in so doing) and the peculiarly non-judicial nature of responsibility for conducting foreign relations. In civil law jurisdictions, the role of the judiciary will present itself as a question of how broadly and assertively to interpret statutes (on both jurisdiction and on applicable law) while, in common law jurisdictions, this question will be supplemented by the question of whether the historical role of judges in forging non-statute-based private law should continue into the transnational realm. A corollary will exist at the international level with respect to the issue of how broadly international human rights treaty bodies can interpret both the applicability of treaty norms and the scope of their institutional powers of review. If it is a given that the original conception of the treaty regime, the ongoing mainstream understanding and the bulk of textual indicators point to the human rights treaties as being concerned mostly about state obligations and review of state compliance, what degree of acquiescence do treaty bodies need from states parties to the treaties before venturing in a robust way into the realm of corporate accountability for human rights harms? Or, is no degree of implicit acquiescence acceptable, such that the bodies may only take this route when they have the equivalent of interstate legislative approval in the form of treaty protocols

58 C Scott or parallel agreements? Would consensus (or large-majority) resolutions of the UN Commission on Human Rights or the UN General Assembly help bridge the gap between implied authority and express authorisation?

10 NORMATIVE AUTHORITY OF DECISION : BINDING VERSUS NON - BINDING

The final relevant distinction deals with the formal authority of normative acts of a given institution. Does an evaluation of compliance or an interpretive statement in some dispositive manner bind the actor(s) to whom it is addressed, or does it have non-binding force? Take for example the “views” that the Human Rights Committee may issue under the ICCPR’s communications procedures or the “concluding observations” of all the UN committees. If these juridical acts are formally non-binding, such that their authority is persuasive only, what degree of persuasiveness attaches to them? It is arguable that the non-binding status of certain forms of judgment may be a blessing in disguise, as it allows the special qualities of judicial or quasijudicial scrutiny or reasoning to be brought to bear on conduct while accommodating the political dimension to given subject matter by not closing down the possibility that governmental bodies may take a different view. It also, arguably, allows for a greater degree of interpretive boldness because matters of principle can be embraced while leaving pragmatic and “political” issues to the operation of give-and-take in the aftermath of a body’s appraisal. Nonbindingness allows principled prodding of the political process while avoiding pre-emption of that process. Apart from the example of the UN human rights bodies, several other contexts help cast light on the necessity to take into account the status of juridical acts in order to assess the acceptability of a given institution engaging in that juridical act. Firstly, under the European Convention on Human Rights, states “undertake to abide by” (article 53) European Court judgments when they are parties to a case. This includes abiding by Court orders under article 50 whereby the Court may “afford just satisfaction to the injured party”. To date, the Court has insisted that these words only give it authority to order financial compensation and has eschewed any form of injunction that would tell states what they must do to stop the violation or fully remedy it. One reason for this reluctance is almost certainly the assumption that any article 50 order will, by definition, be binding under article 53, and this raises the normative stakes of the Court going beyond ordering compensation. However, were the Court to read a soft authority into article 50’s wording, permitting it to recommend to states what measures need to be considered by them, this could add immeasurably to the impact of Court judgments. In other words, the possibility of a non-binding aspect to the Court’s judgments could increase the normative engagement of the Court with states. A second context is the potential for an evolving relationship between different international regimes, such as between the ILO and the WTO with respect

Translating Torture into Transnational Tort 59 to core labour standards. Rather than insert a binding social clause into the WTO, states parties to the GATT/WTO regime have shunted the matter to the “appropriate” institution, the ILO. Regardless of whether a formal link gets drawn between the regimes in the form of an interstate agreement, the ILO could interpret these events as already giving it the authority to begin to assess states’ compliance with ILO core labour norms in a way that directly insists on harmonised minimal standards in each state’s trade and investment laws and policies. Not only would this provide an external form of review of the WTO order, albeit very indirectly, it could also seek to harness the recommendatory power of the ILO to put forward comprehensive remedial suggestions in situations where the negative and blunt effect of the trade order’s ultimate tool, trade sanctions, will do little to better certain labour situations and may easily exacerbate them. Finding solutions to child labour and forms of bonded labour come to mind as situations that would benefit from complex, non-impositional remedial processes engaging states, corporate actors and affected groups in some cooperative problem-solving enterprise. A third context involves cases in which domestic courts are called upon to resolve civil liability disputes between private actors, but the defendant seeks to raise what is known as the “act of state” doctrine as a defence. Put in very simplified terms and ignoring many debates over its precise contours, that doctrine tells domestic courts that, where a foreign state’s legal system has determined rights or obligations of private actors or goods in an exercise of territorial jurisdiction, then the foreign court must, as a measure of respect for the sovereignty of the foreign state, treat the legal effects of the foreign law’s determination as valid. So, for example, victims of forced labour on an oil pipeline project in Dystopia sue a Utopian oil company in Utopian courts. The forced labour is organised by Dystopia, acquiesced in by the company, and used to forward a joint venture in which the government and the company are equal partners. The government of Dystopia issues a decree, lawful within its own (repressive) legal system, saying: “No company, or corporate officers, shall incur any civil liability for acts performed by it or by the government of Dystopia on its behalf where such acts are done within the context of a joint venture project involving Dystopia and the company.” When the Utopian company invokes the decree, it will say to the courts of Utopia that the act of state doctrine should apply such that the Dystopia decree must be treated as valid in Utopia; given that the decree determines the issue of liability, the plaintiff victims’ claim must fail, it would be argued. Apart from any substantive reasons that the court may be willing to invoke to decline to respect the foreign act of state (notably what is called a “public policy exception” the content of which can be informed by international human rights values), the court may also take note of the formal legal effects of any determination of liability on the part of the Utopian company. That is to say, by refusing to treat the foreign decree as valid, the Utopian court is neither adjudicating the liability of the state of Dystopia (it is a Utopian company’s liability that is at issue, only) nor is it purporting to say that the decree is invalid

60 C Scott within Dystopia’s own legal order. The binding effects of the decision to ignore the decree are limited to a Utopian company being sued before a Utopian court; of course, the judgment does affect activities that go on in Dystopia,but it does not do so in a way that has binding force either over Dystopia or within the Dystopian legal system. The final context is rapidly becoming the most prevalent form of transnational “regulation” of corporate actors for human rights abuses and other kinds of problematic conduct (environmental harm, bribery, money laundering and so on): codes of conduct. There are some codes of conduct that apply to corporate actors which have been issued by international organisations—for example, the 1977 ILO Tripartite Declaration of Principles Concerning MNEs and Social Policy and the 1976 OECD Guidelines for Multinational Enterprises. These documents are intended to affect conduct by suasion, but, to date, have had limited discernible impact because their formally non-binding status has not been complemented by any kind of rigorous monitoring or conduct-spotlighting procedures. Another phenomenon has been the issuance by unofficial organisations of principles applying to conduct in specific countries, with corporations invited and pressured to endorse and internalize the principles; notable examples are the Sullivan Principles for the former apartheid South Africa and the MacBride Principles for Northern Ireland. In the 1990s, indeed primarily in the past few years, there has been a profusion of initiatives by individual corporations and industry-wide associations to promulgate codes of conduct. The vast majority of these codes are based on a model of self-regulation, with some companies being more serious than others about bringing the codes in as standard operating premises of the company and about permitting credible external auditing of compliance with the codes. Whether or not corporate self-regulation is largely an effort to fend off binding state and interstate regulation, it has had the distinct advantage, by virtue of its very non-bindingness, of having put a small army of Trojan horses onto the field of ideological struggle over corporate social responsibility, an army which NGOs are rapidly becoming adept at pushing through the corporate gates.

11 CONCLUSION : THE IDEA OF MUTUAL TRANSLATION OF TORTURE AND TORT

“[W]e might say . . . that in the literary text an awful lot of things may be happening at once, perhaps contradicting each other, perhaps qualifying each other, and that as a result the translator may find that it is not possible to express all of these complications in the target language. The idea . . . is that by looking at the original and translation side by side and identifying those areas where translation turned out to be problematic, we can achieve a better appreciation of the original’s qualities and complexities . . .”2 2 T Parks, Translating Style: The English Modernists and their Italian Translations (London and Washington, Cassell, 1998) at 12–13.

Translating Torture into Transnational Tort 61 I started this chapter with the suggestion that the relationship between tort and torture may be usefully approached in terms of a series of problems of translation, and also noted that the notion of a hybrid mediating space between poles may be a more congenial way of understanding than thinking in terms of conversion, that is, of a movement from tort to torture and, similarly, from torture to tort. To this extent, translation has its limitations as a metaphor because it risks conjuring up the image of a one-way rendering of an original into its mirror in another language—the pubic international right of torture becomes a domestic private law tort, and the law of tort becomes part of the apparatus of international law—and thus obscuring the fact that the “torture as tort” issue involves two-way normative traffic, a conversation between two originals neither of which can be understood entirely within the existing language of the other. While virtually no translator would contend that it is possible fully to render a literary work in the existing vocabulary, syntax, and embedded structural context of the receiving language, it is nevertheless generally accepted that the translation enterprise involves striving to be as faithful as possible to an ideal of reproduction of the original in another form rather than to a goal of creating something substantively new. Without begging the very set of questions being asked in this volume about the relationship between tort and torture, the same cannot be said to be the very purpose of the interaction of two legal discourses unless one, again, simply assumes a normative priority of one and a kind of blank-slate receptivity of the other. Rather, it would seem more useful to think of a process of mutual modification in which we must be constantly open to the possibility of a new text emerging which is a reflection of neither one original nor the other original. Yet, something remains to be said for the image of translation, if we understand it less as an outcome or product than as a process or method in which, even when the movement is towards the forging of a normative betweenness, we will inevitably approach that space initially by seeking to understand a comparatively unfamiliar discourse in terms of a discourse with which we are most at home. In the course of this, we may come to better understand the limits and potentialities of our home discourse. Here, there is much of value in Tim Parks’ conclusions, as set out in the quotation starting this sub-section, about how the process of translation creates the opportunity to view an original text (that being translated) more reflectively and with greater insight, to achieve “a better appreciation of the original’s qualities and complexities.” When adapted to the normative realm of law and human rights, we could do worse than to take the cue from Parks’ view of this key value of translation and think in terms of a process of mutual translation in which we seek not only to grasp the strange text but also to more fully grasp the home text. With this conceptual backdrop in mind, it may be helpful to take one example in the “torture as tort” dynamic of a doctrinal issue in which a multiplicity of options are potentially open for choice in the interplay of categories that goes on within a process of mutual translation. In view of the dual focus of this

62 C Scott chapter on both general framework and the specifics of corporate accountability, the example will be that of the characterisation of the causes of action that might be recognised in proceedings in domestic courts brought against multinational corporations, where both “tort” and “torture” figure in some inchoate way in the understanding of that cause of action. Here, an analytical scheme suggestive of the aforementioned multiplicity is offered. In broad terms, it could be said that there are two ways in which a humanrights-related claim could be characterised in formulating a private law cause of action. First of all, human rights could be cited as the direct cause of action such that, for instance, a company could be sued for a violation of the human right not to be tortured. Secondly, human rights could be indirectly pleaded in that, while they could be the object or purpose of the litigation, other legal categories would be invoked in order to vindicate the substance of human rights protections; for example, rather than a human right of torture providing the direct cause of action, a plaintiff might choose to sue a corporation for a recognised cause of action such as the tort of battery. In terms of a direct human rights action, it is conceptually possible that the plaintiff could point to four different sources for the human right in question that the plaintiff alleges to have been violated: local law (the law of the forum); foreign law (notably the law of the place where the harm occurred); public international law (either international treaty law, international customary law, or general principles of law); or transnational law (a blend of norms that do not derive from any specific legal system).3 As for indirect human rights claims, it may be analytically helpful to make a further distinction between “surrogate” claims and “instrumental” claims. By surrogate claims is meant existing causes of action that come very close to capturing, in an intrinsic sense, the kinds of harms that a human rights tort would seek to tackle.4 By “instrumental” is meant a claim in which a recognized legal interest is used as the basis for a law suit with some recognition that protection of that legal interest will instrumentally benefit the protection of human rights interests.5 3 As for the invocation of international law as the juridical source of the alleged human rights violation, it will often be the case, formally, that it will be one national legal system or another that is the source of the human rights claim by virtue of that legal system having received international human rights law in one way or another. 4 The example has already been given of the connection between the existing tort of assault and a potential human rights tort of torture. Another example might be the existing tort of false imprisonment and torts related to arbitrary detention and disappearances. That is not to assume, it bears adding, that reliance on existing causes of action will do justice to the specific principles that already are, or should be, in place with respect to the human rights norm: see, e.g., the discussions of the adequacy of existing tort law in G Virgo, “Characterization, Choice of Law, and Human Rights”, chapter 12 in this volume; and S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, chapter 14 in this volume. 5 Probably the best and most relevant example would be tort and other civil liability claims related to harms to the environment in which rights related to the right to health, the right to inadequate standards of living (including the right to earn a livelihood) and rights to nutrition can be shown to have been detrimentally affected by impairment of the environment in which persons or whole communities live.

Translating Torture into Transnational Tort 63 A final word, a caution. Nothing in the foregoing is meant to suggest a onceand-for-all choice. Temporal, contextual and pragmatic variables will, and should, influence the approach a given court decides to take. What made sense in the United States given the presence of ATCA and later the TVPA may not make sense today in Australia or France or Egypt. Tentative first steps, based on existing choice of law rules, tort law and jurisdictional principles, may be all that is necessary for a first transnational torture case to be decided in Ontario, but by 2005 this may be an unduly constrained way of understanding even cases in which no Canadian actor is either victim or perpetrator. And so on. Finally, it should be noted that even the foregoing analytical scheme has its dangers if it is mistakenly interpreted as suggesting that multiplicity means only a range of categorisation choices from which a judge must, in the end, select only one for purposes of a given case. This is not necessarily the best way to understand multiplicity. Rather, a notion of simultaneous or co-existent multiplicity may do greater justice to the issues and complexity at stake. By simultaneous multiplicity, I mean to say that courts, and legislators, should keep their minds open to plural characterisations of a claim, such that different characterisations might be chosen depending on what specific issue is dependent on that characterisation. One illustration, situated toward the more conservative end of the spectrum of possibilities, conveys something of what I mean by plural characterisations. Assume a judge declines to characterise a cause of action as an action for torture per se, a kind of auxiliary characterisation as torture might still be used to trigger the application of the “public policy” exception in choice of law in order to justify applying the lex fori when there are barriers (such as an unjust amnesty) in the lex loci delicti which are illegitimate under the public international law on torture, especially when the state of the lex fori and the state of the lex loci delicti are both parties to the Convention against Torture. This way, even if the tort is initially characterised as “battery”, an “auxiliary” characterisation as “torture” would allow public policy to be used in a more principled way than presently tends to occur with the unelaborated and gut-level ad hocery that goes on under the guise of “public policy”. Such a notion of auxiliary characterisation does justice to the fact that, while certainly constituting battery within current tort law doctrine, the elements of this tort also constitute torture under international human rights law. On such an approach, simultaneous multiplicity might have a conceptually liberating influence on judicial reasoning, while remaining principled and firmly within the domain of the appropriate role(s) for judges to play in working justice through law.

3

International Human Rights Tort Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines MICHAEL SWAN 1

1 INTRODUCTION

the present chapter is to provide a very general overview of the statutes grounding, and case law resulting from, “international human rights tort” claims before United States federal courts. The experience up until the end of the 1990s has been reviewed, with somewhere approaching one hundred cases having been consulted. In the course of creating a synthesis for readers not familiar with the US legal system, let alone with these kinds of transnational tort claims, many nuances debated in the now-quite-voluminous secondary literature will undoubtedly be blunted. The present author does not wish, on any account, to have this chapter treated as more than a starting point for those seeking basic familiarity with the state of US law in this field and with the basic normative controversies. Its modest goal is to help create a context for the more detailed discussions that follow in the other chapters in this volume, in order to assist the reader in fruitfully engaging with the arguments of the other authors without the necessity for the reader to have background knowledge in each of the various intersecting fields implicated by this still-unique American legal phenomenon.

T

HE PURPOSE OF

1 Bar of Ontario, Province of Ontario, Canada. Some editorial work on this chapter has had to be done without the author’s final feedback due to circumstances resulting in the unavailability of the author. All responsibility for any errors is thus assumed by this volume’s editor, Craig Scott.

66 M Swan

2 THE FEDERAL COURTS , STATE COURTS AND THE UNITED STATES SUPREME COURT

Federal Courts The federal courts are arranged into thirteen regional “circuits”. Each circuit is generally free to interpret the law independently, and has its own Court of Appeals, but all are subject to the decisions of the United States Supreme Court. The federal courts (“circuit courts”) have taken the lead in the adjudication of international human rights tort (IHRT) claims in the US. These claims were first given a sympathetic hearing in the federal court system in Filártiga v. Peña-Irala, in 1980.2 The federal courts are courts of limited jurisdiction. They do, however, have original jurisdiction over broad areas of law, including constitutional matters and suits between residents of different US states. The US federal courts thus enjoy a high profile within the overall US legal system.

State Courts The state courts are the courts of general jurisdiction in the United States. As yet, no American court has foreclosed the possibility of bringing an IHRT-style action as a state law claim in a state court forum.3 While at least two federal court cases have discussed the possibility, only one seems to have allowed such an action to actually proceed in a state court.4 In light of this, the possibility of 2 Filártiga v. Peña-Irala, 630 F.2d 876 (C.A., 2d Cir. 1980). The facts that gave rise to the lawsuit in the United States were, that: Filártiga, a Paraguayan, was tortured to death in Paraguay by PeñaIrala, also a Paraguayan, a member of the Paraguayan police forces; Peña-Irala ended up in New York where he was spied in deportation detention by a member of the Filártiga family who also happened to be in the US; a lawyer was retained and process was served on Peña-Irala for an Alien Tort Claims Act (ATCA), 28 U.S.C. §1350, action before Peña-Irala was deported back to Paraguay; and the case, from the preliminary rulings to the judgment on the merits and resulting award of damages, then proceeded in Peña-Irala’s absence. 3 For an elaboration of the law here see the section on “Federal Question Jurisdiction”, Section 3 infra. One would plead the case according to state laws, and allege no violation of the law of nations, to avoid §1350, ibid; discussed further in Section 3. 4 In Hilao v. Estate Of Ferdinand Marcos, 25 F.3d 1467 (C.A., 9th Cir. 1994), also referred to as Hilao and Estate II, the Court of Appeals seemed to suggest that it might be possible to bring such an action in a state court, while noting that the claims in Trajano v. Marcos 978 F.2d 493 (1992) (Trajano or Estate I) were brought both as IHRT claims and as state law claims. The latter suit was tried in federal court, however, and the Court of Appeals expressly declined to decide this issue. To avoid confusion, I will note here that Trajano involved a claim brought against Ferdinand Marcos’s daughter, which was tried separately from the Hilao class action suit, in 1992. An important preliminary injunction was brought in the Hilao suit in 1994 to restrain the transfer or dispersal of the Marcos Estate’s assets by surviving family members, which is also an important part of the Estate II case history (25 F.3d 1467). The Appeal from final judgment in the Hilao action was heard in 1996 (103 F.3d 767; Estate II, 1996). Though the Trajano and Hilao actions were separate matters, all

The Experience of United States Courts 67 state court jurisprudence in the area will not be addressed any further in this chapter. The law produced by the state courts and legislatures is an important source of law for the federal courts, however, as will be discussed in Section 4 below.

The United States Supreme Court The US Supreme Court has taken a very “hands-off” approach to the adjudication of IHRT actions in US courts, despite the number of very high-profile actions that have arisen since 1980. The Supreme Court has never spoken directly about the federal courts’ handling of IHRT claims, but only indirectly with respect to matters concerning the implications of these suits for the application of the Foreign Sovereign Immunity Act.5 The federal courts have thus far been very free to develop IHRT jurisprudence. While the federal courts have confidently developed the jurisprudence, however, they have been careful to heed US Department of State submissions on potential foreign relations matters (while carefully asserting their independence).6 three Court of Appeals judgments are referred to as In re Estate of Ferdinand Marcos, presumably because these related matters dealt with many of the same facts and issues, and therefore the rulings in each were important to the others. In Yosofa Alomang v. Freeport-McMoran, 718 So. 2d 971(1996), an action involving an American mining company’s involvement in Burma was remanded to state court by a judge of the Federal District of East Lousiana (5th Circuit). Yosofa is a companion case to Beanal v. Freeport-McMoran, Inc., 969 F. Supp. 362 (E. Louisiana, 1997), in which the same group of companies is alleged to have acted in concert with the government of Burma in perpetrating various international human rights violations and cultural genocide. The companion cases give the appearance that it was a key part of the litigation strategy on the part of the various plaintiffs involved in both cases (individuals, union officials, and persons who claim to represent the rightful government of Burma) to bring separate actions in federal and state courts with respect to, broadly speaking, the same “set” of events. By coaching the claim in terms of existing private law categories of personal injury, the plaintiffs in Yosofa were able to proceed at the state court level for an action that was simultaneously characterizable as an IHRT tort action. 5 28 U.S.C. §1330. In Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480, 493, 103 S. Ct. 1962 (1983), discussed in G B Born and D Westin, International Civil Litigation in United States Courts— Commentary and Materials (Kluwer Law and Taxation Publishers, Deventer and Boston, 1988) at 420–1, the Supreme Court “clarified” the different scopes of the two “arising under” provisions (in Article III of the Constitution, and §1331 of the Official Code of the United States (Federal Question Jurisdiction) respectively), and how these are affected by questions of sovereign immunity. Verlinden was not an IHRT case, however. In Argentine Republic v. Amerada Hess Shipping Corp. et al., 488 U.S. 428; 109 S. Ct. 683 (1989) the Supreme Court dealt with jurisdiction under the Foreign Sovereign Immunity Act (“FSIA”), and its relation to jurisdiction under the Alien Tort Claims Act (“ATCA”) (28 U.S.C. §1350; also referred to as the Alien Tort Act and the Alien Tort Statute). The Supreme Court saw it as necessary to protect the integrity of the FSIA against an impinging expression of the scope of the ATCA. While the decision in Amerada Hess also contained important dicta on the issue of possible waivers of sovereign immunity, the court was very careful not to otherwise undermine the various possibilities for IHRT actions under the ATCA. All of these issues are discussed, infra. 6 See especially Kadic v. Karadzic, 70 F.3d 232, at 250 (C.A., 1995 2d Circ).

68 M Swan As things now stand, it may be said that there is no matter critical to the “survival” of the jurisprudence which is any longer in serious issue despite some vigorous judicial challenges early on.7 That being said, the federal courts’ jurisdiction to hear these cases has been settled as a discussion amongst the circuit courts with virtually no guidance from the US Supreme Court. Congress assisted, however, with the passage of the Torture Victim Protection Act (TVPA) in 1992.8 The courts have trod carefully (and creatively) around the difficult choice of law issues. The result is that one can consider the jurisprudence as a whole, rather than by way of a circuit-by-circuit account.

3 JURISDICTIONAL CONCEPTS

To be able to hear a claim, an American court must have the power to hear the category of disputes within which the claim is properly founded. This is referred to as “subject-matter jurisdiction”. In addition, the court must have adjudicative power over the persons or property involved in the litigation. This can take the form of jurisdiction that is either in personam, in rem, or quasi in rem in nature. It must be kept in mind that jurisdiction over a category of disputes, and over the litigants or property involved, does not speak to the matter of the existence of a cause of action. A cause of action provides the court with the authority to apply substantive rules of law to a dispute, and grant remedies where appropriate. One may plead a state law cause of action for “negligence”, for example, and the forum court will ordinarily apply that state’s substantive law to the dispute. There has been much discussion as to the source of the cause of action that may be said to arise in IHRT cases, despite the fact that it is clear that the federal courts may take subject-matter jurisdiction over a tort claim involving non-residents (“aliens”) disputing a violation of the law of nations. The latter italicised phrase describes a jurisdictional “category” of disputes under the Alien Tort Claims Act (ATCA), but may not necessarily speak to the existence of a cause of action. This will be discussed again, and clarified, in later sections. As courts of limited jurisdiction, the federal courts must have both a constitutional and a statutory grant of subject-matter jurisdiction to enable them to hear a claim.

7 Notably in Tel-Oren v. Libyan Arab Republic, 233 U.S. App. D.C. 384; 726 F.2d 774 (1984), a case in which the ATCA’s scope was vigorously questioned, notably by Judge Bork. 8 28 U.S.C. §1350; see the sections on the TVPA and ATCA, in Section 3 infra.

The Experience of United States Courts 69 Subject-Matter Jurisdiction and Cause of Action The constitutional subject-matter jurisdiction of the federal courts in IHRT cases The constitutional grant of subject-matter jurisdiction to federal courts adjudicating IHRT claims flows from Article III of the Constitution of the United States: “§1. The judicial power of the United States shall be vested in one Supreme Court, and in such inferior courts as the Congress may, from time to time, ordain and establish. ... §2. The judicial power shall extend to all cases, in law and equity, arising under this Constitution, the laws of the United States, and treaties made, or which shall be made, under their authority; . . .”9

The “judicial power of the United States”, of course, refers to that power which is not reserved to the state legislatures and state courts in the federal system. This grant of constitutional subject-matter jurisdiction is referred to as the Article III “arising under” jurisdiction. Of central importance to this constitutional grant of jurisdiction over IHRT claims is that public international law (the “law of nations”) is part of US federal common law through the vehicle of the Constitution.10 It is said that “[i]t is an ancient and a salutary feature of the Anglo-American legal tradition that the Law of Nations is a part of the law of the land to be ascertained and administered, like any other, in the appropriate case.”11 A claim involving international human rights thus “arises under” the “laws of the United States”. The federal courts may exercise constitutional subject-matter jurisdiction with respect to such subject matter, and Congress may exercise legislative jurisdiction with respect to it.12 The statutory subject-matter jurisdiction of the federal courts in IHRT cases There are (at least) five possible sources for statutory subject-matter jurisdiction in IHRT cases. The ATCA is by far the most important, but a brief discussion 9

U.S. Const. art III, §1 & § 2, cl.1.(emphasis added). Filártiga, supra n. 2, at 886, citing E D Dickenson, “The Law of Nations as Part of the National Law of the United States,” 101 U.Pa.L.Rev. 26, 27 (1952). All other circuits have agreed. Note that it is the “law of nations” that is referred to, and not “customary international law”. The courts, however, seem to discuss the adoption of “customary international law” by federal common law, rather than using the archaic language of the “law of nations”. Here, the courts draw heavily on the discourse and conclusions of the American Law Institute, Third Restatement of the Foreign Relations Law of the United States (St. Paul, Minn., American Law Institute Publishers, 1987) (“Restatement (Third)”) esp. at §§701–2. 11 Ibid. 12 This grant of jurisdiction must be kept distinct from the “arising under” jurisdiction of the “federal question” grant of jurisdiction, dealt with in the “Federal Question Jurisdiction” section, infra. 10

70 M Swan of the others will assist in putting that Act into perspective and raise important questions for subsequent analysis.

(i) The Foreign Sovereign Immunity Act of 1976 (FSIA): 28 USC §1330 Section 1330(a) of Title 28 of the United States Code reads, in part: “The District courts shall have original jurisdiction without regard to amount in controversy of any nonjury civil action against a foreign state . . .” But, section 1604 goes on to provide, in part, that “subject to existing international agreements to which the United States is a party at the time of enactment of this Act a foreign state shall be immune from the jurisdiction of the courts of the United States and of the States except as provided in sections 1605 to 1607 of this chapter.” The US Supreme Court made it clear in Amerada Hess13 that a suit against a foreign sovereign must be brought under this section, if it can be brought at all, through one of the exemptions identified in sections 1605 to 1607. One cannot avoid having to come to terms with the exemptions to sovereign immunity therein by trying to bring suit under another act, such as under the ATCA. Sovereign immunity is an affirmative defence which must be raised, and proved, by a defendant.14 A “state” includes its political subdivisions (provinces, states, et al.),15 and its “agencies or instrumentalities”.16 Those (federal) courts which have dealt with the issue have all held that a state official whose impugned acts were not committed within the scope of his or her official authority cannot raise the defence of sovereign immunity.17 In these and other cases, it has been impossible for former state officials to successfully maintain that acts of torture or extrajudicial killing, for example, were carried out within the proper scope of their authority, so that a grant of immunity could be said to be “appropriate”. The FSIA, therefore, has presented no impediment to US courts taking jurisdiction in suits against such former state officials, per se. It might be helpful, however, to provide some further background, and a general outline of the FSIA. With the “Tate Letter” of 1952, the US Department of State signalled its desire to have the primary responsibility for matters of sovereign immunity before the judiciary.18 Before that date, matters of sovereign immunity were decided by the judiciary, but according to initial determinations made by the State Department, and through common law doctrines which construed sovereign immunity in a blanket fashion. Blanket immunity arose from old notions such as that of the “inherent dignity” of nation-states. The inconsistent results 13

Amerada Hess, supra n. 5, at 434. Ibid. 15 Section 1603(a), as cited in Born and Westin, supra n. 5, at 340. 16 Section 1603(b) as cited in ibid. at 342. 17 Trajano, supra n. 4, at 496–7 (9th Cir. 1992); Kline v. Kaneko, 685 F. Supp. 386 (S.D.N.Y. 1988); Rios v. Marshall, 530 F. Supp. 351, 371–2 (S.D.N.Y. 1981). 18 See Born and Westin, supra n. 5, at 338–9 14

The Experience of United States Courts 71 achieved by initial executive determinations of these matters was one reason for the change signalled by the Tate Letter. The desire to move, as other states were doing, to a “restrictive” theory of the immunity of states was another.19 The need to balance public and private interests more closely, and recognize that governments were not always implementing (often controversial) public policy but instead often acting like private actors, provided the rationale for the “restrictive theory”. The comprehensive statutory scheme in the FSIA embodied these shifts in perspective and inclination. It is not surprising, therefore, that most of the exceptions have to do with very straightforward commercial transactions, or the protection of US citizens and interests. There are currently five main FSIA categories of exceptions to sovereign immunity. Four do not seem relevant to the bringing of IHRT claims. Even if the FSIA’s “Commercial Activity” exception could be viewed as relevant especially where human rights abuses occur in the context of a joint venture commercial project between a foreign state and corporations, that exception applies only if the activity occurred in, or has direct effects in, the US.20 The “Noncommercial Torts” exception applies only if both the “act” which constitutes the tort and the harm suffered occur in the US.21 The “Expropriations” exception has to do with the nationalisation of foreign-owned private property.22 The recent “Flatow Amendment” was passed explicitly to facilitate tort actions by US citizens (or their next-of-kin) only against officially designated “terrorist states”, and then, only in connection with state-sponsored acts of international terrorism.23 Only the “Waiver” exception seems to have had potential relevance to IHRT claims taken generally.24 The Act itself states only that a waiver may be either express or implied. The House Report which accompanied the passage of the FSIA, however, gave three examples of waivers.25 It was said that they could be implied where a sovereign had agreed to the arbitration of a dispute in another country, had agreed that the law of another country would govern a dispute, or had entered an appearance in a US court, without raising the sovereign immunity defence.26 The Supreme Court seems to have foreclosed the possibility that the signing of an international agreement can act as 19

Ibid., generally, at 336–40. §1650(a)(2), discussed in Born and Westin, supra n. 5, at 370–1. See R Wai, “The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legislation”, chapter 8 of this volume. 21 §1605(a)(5), discussed in Born and Westin, supra note 5, at 383–97. 22 §1605(a)(3), discussed in ibid., supra n. 5, at 397. 23 §1605(a)(7), following as a response to Flatow v. The Islamic Republic of Iran, 999 F. Supp. 1 (D.D.C., 1998) at 12–13. The foreign states designated as sponsors of terrorism pursuant to 50 U.S.C. App. §2405(j), as of 31 October 1997 were: the Islamic Republic of Iran, Cuba, Syria, Iraq, Libya, Sudan and North Korea. See 22 C.F.R. §126.1(d). 24 §1605(a)(1). 25 As referred to in Siderman de Blake v. The Republic of Argentina, 965 F.2d 699 (C.A., 9th Cir., 1992) at 721. Infra n. 37 for the citation to the House report. 26 Ibid. 20

72 M Swan a waiver in the event of a breach of its terms, unless this is specifically included as a term of the treaty.27 As mentioned above, the FSIA will ordinarily present no impediment to bringing suit against state officials who have engaged in IHR violations. Concerns having to do with the implications of such a suit for US foreign policy are dealt with through federal common law under the doctrines of “Head of State Immunity”, the “Political Questions Doctrine” and the “Act of State” doctrine. All of these are discussed separately below.

(ii) Federal Question Jurisdiction: 28 USC §1331 The US Code’s section on federal question jurisdiction reads: “The district courts shall have original jurisdiction of all civil actions arising under the Constitution, laws, or treaties of the United States.” Here the grant of subjectmatter jurisdiction turns solely on the federal character of the parties’ claims.28 However, the section 1331 statutory “arising under” grant of jurisdiction is construed less broadly than the Article III constitutional “arising under” grant.29 Whether a case “arises under” section 1331 is determined by the “wellpleaded complaint rule”—which provides that federal question jurisdiction exists only when a federal question is presented on the face of a plaintiff’s properly pleaded complaint.30 Federal law must create the action, or the plaintiff’s relief must depend on the resolution of a “substantial” question of federal law.31 27 In Von Dardel in 1985, an alternative holding was that a party to an international agreement or treaty could act as a waiver of sovereign immunity on the part of a state which subsequently breaches the terms of that agreement or treaty: Von Dardel v. Union of Soviet Socialist Republics, 623 F.Supp. 246 (D.D.C., 1985) at 262–3. Subsequently, however, the Supreme Court held in Amereda Hess (supra, n. 5) that for this to be the case, the treaty must specifically mention the waiver of immunity by a signatory to a suit in the United States. The (mere) mention that remedies were available under a treaty, without this specificity, would not be enough. Nor can the violation of jus cogens norms operate as a waiver of sovereignty: see Siderman de Blake, supra n. 25 at 718. (Jus cogens norms, which are mandatory international law prohibitions, are discussed at length below.) While agreeing that waivers should be implied on a limited basis, however, the Siderman court did find that an attempt by Argentine officials to enlist the aid of US courts, in extraditing the plaintiff for criminal charges stemming from the same set of facts as those pleaded by the plaintiff in the US suit, could act as an implied waiver. 28 Born and Westin, supra n. 5, at 416. 29 This is only true when considering a statutory grant of jurisdiction other than the § 1331 grant: see K C Randall, Federal Courts and the International Human Rights Paradigm (Durham, Duke University Press, 1990) at 60–1, n. 3. Randall notes that these different interpretations of the two “arising under” grants are the result of the Supreme Court’s decisions in the companion cases Osborn v. Bank of the United States, 22 U.S. (9 Wheat.) 738 (1824), and Bank of the United States v. Planter’s Bank, 22 U.S. (9 Wheat.) 904 (1824). 30 United States Code Annotated Title 28 (Judiciary and Judicial Procedure), §1331, n. 107; see also Born and Westin, supra n. 5, at 417–19. 31 Born and Westin, supra n. 5, at 417, citing Republic of the Philippines v. Marcos, 806 F.2d 344 (2d Cir. 1986). In this case the government of Corazon Acquino launched an action which contended that a constructive trust had arisen under state law in the US over the assets that former ruler Ferdinand Marcos had transferred in the U.S. Although it was not an IHRT case, the 9th Circuit

The Experience of United States Courts 73 Most actions under section 1331 are based on specific federal statutes that set forth substantive rules of law and create federal subject-matter jurisdiction.32 Cases do arise, however, where the necessity of determining the application of a federal common law doctrine results in federal court jurisdiction. The reason for the judicial discussion of potential federal court jurisdiction over human rights torts under this section seems to have had to do with the initially narrow interpretation given to the ATCA by the Ninth Circuit (and the DC Court of Appeals) early on. It had held that the ATCA was a solely jurisdictional statute and did not create a cause of action under federal substantive law, thereby finding it necessary to found federal subject-matter jurisdiction upon section 1331, to make sure that the Article III “arising under” jurisdictional grant was complied with.33 As the Ninth Circuit has since held that the ATCA does create a cause of action, the need for such innovations seems to have lapsed.34 There would seem to be no advantage to founding jurisdiction in section 1331, as compared to the more natural ATCA source.

(iii) Diversity Jurisdiction: 28 USC §1332 Section 1332(a) provides, in part: “The [federal] district courts shall have original jurisdiction of all civil actions where the matter in controversy exceeds the sum or value of $50,000, exclusive of interest and costs, and is between— (1) citizens of different States; (2) citizens of a State and citizens or subjects of a foreign state; (3) citizens of different States and in which citizens or subjects of a foreign state are additional parties . . .”

Diversity jurisdiction is sometimes referred to as “alienage jurisdiction”, and the grant clearly turns on the residence or citizenship of the parties. An IHRT claim can be brought under this grant of subject-matter jurisdiction only if an American citizen is involved in it, as was the case in Linder v. Portocarrero,35 which is discussed in Section 3 below in connection with the “Political Question” doctrine.

Court of Appeals held that the foreign relations implications justified the displacement of the purely state law cause of action so that the action “arose under” federal common law for the purposes of §1331. 32 Born and Westin, supra n. 5, at 417. 33 Republic of Philippines v. Marcos, supra n. 31. See also Forti v. Suarez Mason, 672 F. Supp. 1531 (N. Cal., 1987), (Forti I). 34 Hilao, supra n. 4, at 1473–4. 35 963 F.2d 332, (C.A. 9th Cir. 1992).

74 M Swan (iv) The Torture Victim Protection Act (TVPA): 28 USC §1350 Section 2 of this Act reads, in part: “(a) Liability.—An individual who, under actual or apparent authority, or colour of law, of any foreign nation— (1) subjects an individual to torture shall, in a civil action, be liable for damages to that individual; or (2) subjects an individual to extrajudicial killing shall, in a civil action, be liable for damages to that individual’s legal representative, or to any person who may be a claimant in an action for wrongful death.”

With this Act, Congress extended the protection afforded to alien victims of torture and extrajudicial killings in the ATCA to US citizens. The TVPA explicitly provides a cause of action for “official torture” and “extrajudicial killing”. The House Report which accompanied the passage of the TVPA expressed Congress’s view that the ATCA should be left alone in order to allow the jurisprudence to continue to develop, while extending its clearest protections to US citizens.36 Congress thus gave its blessing to the line of ATCA jurisprudence which had held that the ATCA recognised a cause of action for victims of violations of international law and not only served as a subject-matter jurisdictional grant to the federal courts.37

(v) The Alien Tort Claims Act (ATCA): 28 USC §1350 The Alien Tort Claims Act was actually enacted very shortly after the American Revolution. The ATCA is thus not a statute that was enacted against the backdrop of modern international human rights law. Rather it is a re-codification of the first Judiciary Act of 1789 which was enacted against a backdrop of a concern with a more limited range of conduct, such as piracy. The Judiciary Act then and the ATCA now contains a federal court grant of statutory subject-matter jurisdiction in the following terms: “The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.”38 Most IHRT actions brought in the US are brought under this statute. It had received scant judicial attention until the Second Circuit Court of Appeals gave it new life nearly two hundred years after its passage in the Filártiga case in 1980.39 The origins of the statute, and the intentions of its drafters, being some36

As noted in Kadic, supra n. 6, at 241. Infra n. 37 for citation to the House Report. Beanal supra n. 4, at 380, citing and discussing at length the House Report which accompanied the passage of the TVPA: H.R. Rep. 367 (I), 102nd Cong., 1st Sess. 1991, 1992 U.S.S.C.A.N. 84, 1991 WL 255964. 38 Judiciary Act of 1789, ch. 20 §9(b), 1 Stat. 73, 77. The ATCA is now codified at 28 U.S.C. §1350, (1982). 39 See Born and Westin, supra n. 5, at 421–2. 37

The Experience of United States Courts 75 what obscure, the courts starting with Filártiga had to give meaning to the three basic conditions for hearing a lawsuit set out in the short ATCA provision in the context of modern legal understandings.40 First of all, that the suit must be brought by an “alien” is not problematic.41 The two other conditions—that the suit must involve a “tort only”, and that tort must in some sense be “committed in violation of the law of nations”—have raised important questions. The initial question concerns whether the “tort” and “law of nations” aspects of the §1350 grant of jurisdiction are separate conditions. Must a separate “tort” (from some or other municipal law, or international law) be pleaded, as well as a “violation of the law of nations”? Or will the pleading of a single “international human rights tort” cause of action, whereby the fact of a violation of an international human rights norm by the sued actor is sufficient to style the claim as a tort, satisfy the grant of jurisdiction? After some initial difficulties, the courts have now settled on the latter reading of the statute. The courts still speak of there being three conditions for the taking of jurisdiction under §1350, but the “tort” condition is now purely formal, having been folded into the “violation of the law of nations” condition— there is no separate analysis required with respect to it. In other words, one need only plead a recognisable “violation of the law of nations” to successfully plead the (federal) “tort”. These initial difficulties arose when the D.C. Court of Appeals held, in the Tel Oren case, that §1350 was a purely jurisdictional statute. Neither the statute, federal common law, nor international law, therefore, could be said to create a “tort” cause of action in an IHRT case. As such, the court held that it could not exercise its jurisdiction in such a case (although Judge Edwards suggested that a cause of action might be borrowed from the municipal law of the forum state).42 The concern was, among other things, that if international law were held to be the source of the tort cause of action, in the way the Filártiga court had seemed to analyse, any distinction between “self-executing” and “non-self-executing” international treaties would disappear.43 This is to say that a plaintiff would be able to bring an action, based on international law sources, whether or not the international law instruments as embodying the “law of nations” relied upon had specifically provided for the creation of a private right of action which would then, under existing US doctrine on the direct effect of treaties in US law, be recognised as a cause of action by US courts.44 In contrast to the Tel-Oren approach, Filártiga has since been cited, with approval, by all subsequent courts (except the DC Circuit), as standing for the 40

Kadic, supra n. 6, at 238. The term is taken to mean “any person not a citizen of the U.S.” according to the Immigration and Nationality Act, 8 U.S.C. §1101(a)(3)(e) (1982), as cited in Randall, supra n. 29, at 50. 42 Tel-Oren, supra n. 7, per Bork, J. 43 Ibid. 44 Ibid. See also Born and Westin, supra n. 5, at 428, and n. 3. 41

76 M Swan proposition that the ATCA creates a federal (statutory “tort”) cause of action even if, as noted in Filártiga, the governing law for the tort does still need a choice of law inquiry as the ATCA itself does not, as such, provide for applicable law.45 So, in terms of the ATCA, the fact that customary international law (CIL) is part of federal common law is irrelevant to the jurisdictional question which can rely simply on CIL’s existence at the international level as the trigger for ATCA’s application. Similarly, from the Filártiga and post-Filártiga perspectives, CIL does not create the cause of action any more than a non-selfexecuting treaty does—the statute itself does so. The question of which substantive law will be applied when adjudicating this federal cause of action, therefore, is (only) a choice of laws question. It will turn out, as we will see, that the Filártiga court did also turn to CIL at the stage of substantive law application, based on CIL being part of federal law “as it has evolved”, although it is possible to interpret the courts as applying a transnational mixture of general international law and federal common law, developed specifically for the IHRT tort context.46

Judicial Jurisdiction over the Parties and their Particular Dispute To be able to hear a claim, an American court must also have judicial power to deal with the litigation itself in terms of the right to adjudicate the contending claims of these parties in reference to this particular dispute. There are essentially three types of judicial jurisdiction in American law: in rem, quasi in rem 45 See, for example, Xuncax v. Gramajo, 886 F. Supp. 162 (D.C. Mass., 1st Circ, 1995) at 179; Beanal (5th), supra n. 4, at 370; Trajano (9th), supra n. 4, at 1475; Abebe-Jira v. Negewo, 72 F.3d 844 (C.A., 11th Cir. 1996). The Ninth Circuit initially accepted the view that the ATCA did not create a cause of action, but also held that a cause of action could be borrowed from municipal law. After the TVPA was passed, however, the Ninth Circuit joined the other courts in adopting the Filártiga view that the statute created a federal cause of action. The D.C. Circuit does not seem to have had the opportunity to reconsider its view of the matter since the passage of the TVPA. The only IHRT cases it has adjudicated turned solely on the availability of jurisdiction under the FSIA (see Princz v. Federal Republic of Germany, 307 U.S. App. D.C. 102 (1994) ) and the special provisions of the Flatow amendment, supra n. 23. The significance of the TVPA to this question is discussed above. 46 Filártiga, supra n. 2, 881. Importantly, however, there is the possibility that the courts may be said to be applying a mixture of international law and federal common law, developed specially for IHRT claims. See infra, and M Moran, “An Uncivil Action: The Tort of Torture and Cosmopolitan Private Law”, chapter 25 of this volume. Do note, however, that proving the content of international law can become a point of litigation at the stage of the jurisdictional inquiry as there is a requirement that a “violation of the law of nations” must have been sufficiently pleaded. The sources of proof involve substantially the same inquiry as to whether or not a defendant has committed an IHRT. The inquiry into whether or not a “violation of the law of nations” has been sufficiently pleaded is a preliminary one, which may also be raised separately from a motion to dismiss the suit for lack of jurisdiction, for example. Whether there is a sufficiently pleaded “violation of the law of nations” has been said to involve a more stringent preliminary inquiry than is appropriate when seeking to determine whether a claim can be said to “arise under” Article III. A more searching review of the merits of the pleaded claim may be involved: see Filártiga, supra n. 2, 887–8.

The Experience of United States Courts 77 and in personam jurisdiction.47 In rem jurisdiction relates to a court’s ability to adjudicate pre-existing claims of ownership or other rights in specific property.48 Quasi in rem jurisdiction generally involves the seizure of specific property for the purpose of providing security, pending the adjudication of unrelated claims against the owner of the property.49 In personam jurisdiction concerns a court’s power over a person (natural or not), and the court’s ability to render judgment against the person’s assets.50 It is in personam jurisdiction that is typically at issue in IHRT claims. The plaintiff has the burden of establishing the court’s jurisdiction,51 and must meet two requirements. Firstly, the court must have statutory authority for the exercise of jurisdiction and, secondly, the exercise of jurisdiction must be consistent with the “Due Process” requirements of the Fourteenth Amendment of the Constitution.52 “Due Process” allows a court to exercise “general” judicial jurisdiction over any claim involving a defendant as long as the person has “sufficiently continuous and systematic” contacts with the forum state.53 “Specific” jurisdiction may be taken as to a defendant, generally speaking, if the claim is related to, or arises out of, the defendant’s contacts with the forum state.54 The test for “specific” jurisdiction involves at least two questions. The defendant must have purposely availed herself of the protections and benefits of the forum state’s law, and the exercise of jurisdiction must be “reasonable”.55 To obtain in personam jurisdiction over a defendant, the plaintiff must serve her or him with originating process. State legislatures have all passed “long-arm” statutes (the American equivalent of Canadian rules for service ex juris) which authorise state courts to exercise in personam jurisdiction over “foreign” defendants.56 These statutes tend either to directly incorporate the 47

Born and Westin, supra n. 5, 20. Ibid. 49 Ibid. 50 Ibid. 51 Byung Wha An v. Doo-Hwan Chun, 525 U.S. 816; 119 S. Ct. 53 (1998). 52 Born and Westin, supra n. 5, at 21. 53 Ibid. at 25, citing Helicopteros Nacionales de Colombia, S.A. v. Hall, 466 U.S. 408, 415–16 (1984). 54 Ibid. at 27, citing Helicopteros Nacionales, 414–15. 55 Ibid. 56 The rules for securing service abroad are quite complicated, and involve practical difficulties at the best of times. Service is accomplished with the assistance of foreign governments, and in accordance with foreign law: see Born and Westin, supra n. 5, at 124 et seq. In Flatow, supra n. 24, at p. 19 et seq., there is a lengthy discussion of the requirements of service, but in the context of a case where the defendants were a sovereign state (and its officials) acting in its sovereign capacity. The case was brought under the “Flatow amendment” discussed above, which concerns terrorist acts undertaken by officially designated “terrorist states”. The case discusses the possibility that a state may not be a “person” for “due process” purposes, as the concern in the XIVth amendment is as to the liberty interest of persons. The case notes too that aliens outside the United States are not entitled to US constitutional protections. The discussion of the attempts to secure service upon the highest ranking officials of the Iranian government, with the help of the Swiss embassy in Tehran, and the efforts made to evade service are instructive as to the difficulties here. There is a brief discussion of the issue of the appropriateness of the exercise of “long-arm” jurisdiction in Byung (supra n. 51), in which the court refused to take personal jurisdiction over military 48

78 M Swan constitutional due process requirements directly, or provide brief (although sometimes lengthy) statements as to the circumstances within which the courts may exercise jurisdiction. However, the cases involving IHRT claims have usually involved claimants who have been served in juris while in the forum state. While it seems to be the case that this form of service must also meet the requirements of due process, it is said that the Federal Rules of Civil Procedure specifically authorise personal service upon even those individuals who are only temporarily present within a judicial district of the United States, and that such personal service comports with the requirements of due process for the assertion of personal jurisdiction.57 The exercise of personal jurisdiction in Kadic might seem controversial, however, as service was accomplished upon Radovan Karadzic while he was in New York as an invitee of the United Nations.58

Common Law Doctrines which Act as Exceptions to Jurisdiction There are four common law doctrines which might defeat the taking of jurisdiction by a court. In the result, they seem to have had very little applicability to IHRT actions. The “act of state” doctrine The act of state doctrine involves the principle that the courts of one country should not pass judgment on the acts of another in its own territory.59 This doctrine is a municipal law doctrine, to some extent shared across some leading jurisdictions, although it is premised on principles and structural premises of the international legal order. The doctrine has been fully re-articulated since its major modern statement in Sabbatino60 and is now said to be “a flexible leaders and the president of North Korea. The court noted that the only visit by any of the potential defendants had been an unofficial visit in the form of a one-week vacation in Hawaii taken by General Roh. 57 Kadic, supra n. 6, 246–7, citing the Federal Rules of Civil Procedure R. 4(e)(2), and Burnham v. Superior Court, 495 U.S. 604, 109 L. Ed. 2d 631, 110 S. Ct. 2105 (1990). The Supreme Court has also said that “reasonableness considerations sometimes serve to establish the reasonableness of jurisdiction upon a lesser showing of minimum contacts than would otherwise be required.”: Burger King v. Rudzewicz, 471 U.S. 462, 483–4, 105 S. Ct. 2174 (1985), as cited in Flatow, supra n. 23, at 22. In Palestine Information Office v. Shultz, 853 F.2d 932, 942 (D.C. Cir. 1988), as cited in Flatow, ibid., the Court held that policy considerations could significantly raise the hurdle of constitutional requirements where an area of policy was committed to the political branches of government—such as in the area of foreign relations. See n. 74, and the section on the political question doctrine, and note 70, infra. 58 Service may have been said to have been inappropriate according to the Headquarters Agreement, which generally prohibits service within a prescribed area, and which prohibits US officials from impeding the progress of invitees to and from the headquarters. The court noted that Karadzic was not within the prescribed area when served (he was in his hotel), and interpreted the second provision quite narrowly (perhaps laudably so): Kadic, supra n. 6, at 247–8. 59 Ibid. at 114. The author refers to the decision in Underhill v. Hernandez, 168 U.S. 250 (1897). 60 Banco Nacional de Cuba v. Sabbatino, 376 US 398 (1964).

The Experience of United States Courts 79 [doctrine] designed to prevent judicial pronouncements on the legality of the acts of foreign states [within their own borders] which could embarrass the Executive Branch in the conduct of foreign affairs”.61 As now articulated by the US Supreme Court in the Kirkpatrick case, it operates as a choice of law rule, rather than as a jurisdictional abstention doctrine as such.62 However, its effect may be said to be almost identical to a jurisdictional doctrine, as successful pleading of the doctrine results in a court knowing in advance that another state’s law governs and that the claim, being futile, must therefore be dismissed. In any event and however characterised, the re-articulation of the doctrine has narrowed its scope, and enlarged the scope for judicial activity. In Liu v. The Republic of China, it was said that modern American courts consider three factors in their assessment of the doctrine’s applicability.63 First, they consider whether the foreign state was acting in the public interest.64 Second, the courts consider the “degree of international consensus regarding an activity”. The notion is that, if a foreign state’s act of state falls clearly afoul of this consensus, then another state’s courts will be justified in not giving legal effect to the foreign act of state. In Sabbatino, the words of the Supreme Court

61 Liu v. The Republic of China, 892 F.2d 1419, at 1432 (C.A., 9th Cir. 1989) (emphasis added). This understanding of the doctrine has been confirmed by the Supreme Court: see Siderman de Blake, supra n. 25, at 707, relying on W.S. Kirkpatrick & Co. v. Environmental Tectonics Corp., 493 U.S. 400, 404; 110 S. Ct. 701 (1990). The Liu case was brought under the domestic tort exception in the FSIA, as it dealt with an assassination in California of a man who had been critical of the Taiwanese government. As in the 9th Circuit cases cited in n. 4, supra, the case was pleaded as a wrongful death action, prior to that circuit joining the Filártiga line of jurisprudence. 62 Kirkpatrick v. Environmental Tectonics Corp, supra n. 61; Randall, supra n. 30, at 206, n. 223. According to the choice of law understanding of the doctrine, once choice of law analysis selects a foreign legal system as providing the applicable law for a determinative issue at stake in the case, it follows from the basic structure of applying the foreign governing law that the American court must give legal effect to the foreign state’s “act” which is lawful according to that state’s own constitutional system. Understood this way, the real ground of debate for the act of state doctrine is not whether there is a presumption of respect for the foreign law, but, rather, the conditions under which foreign legality will be critically evaluated. The way in which Anglo-Canadian law approaches this critical evaluation is, doctrinally, quite different from the US approach, in that the traditional conflict of laws “escape device” of public policy is key to Anglo-Canadian doctrine while, as will be seen in the upcoming paragraph, public policy is much less of a referent in the US: see M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of the ‘Act of State’ in AngloCanadian Law”, chapter 13 of this volume, for a discussion of the differences between the act of state doctrine as amongst the US, Canada and the UK. 63 Liu, supra n. 61, at 1432–3. 64 Ibid. See International Ass’n of Machinists & Aerospace Workers v. OPEC, 649 F.2d 1354, 1359 (9th Cir. 1981), as cited in Siderman de Blake, supra n. 25, at 778. This case asserted: “While the FSIA ignores the underlying purpose of a state’s action, the act of state doctrine does not.” Ibid. at 1360. This, presumably, refers to the fact that what might be a matter of “private contracting” for the purposes of the commercial exemptions to the FSIA may be a matter of public purpose for the act of state doctrine. Note that the OPEC court was drawing on the reasoning of a decision which upheld the invocation of the doctrine in a case involving OPEC’s price-fixing activities. The concern with respect to an injunction “instructing a foreign sovereign to alter its chosen means of allocating and profiting from its own valuable natural resources” seems qualitatively different from the kinds of concerns raised in IHRT cases: OPEC, ibid.

80 M Swan still provide a cogent statement of the balance that is sought between respect for foreign sovereignty and respect for international law: “[T]he greater the degree of codification or consensus concerning a particular area of international law, the more appropriate it is for the judiciary to render decisions regarding it, since the courts can then focus on the application of an agreed principle to circumstances of facts rather than on the sensitive task of establishing a principle not inconsistent with the national interest or with international justice.”65

The Filártiga trial court, after the case on the merits was give the green light by the landmark Court of Appeal’s decision in 1980, focused on this passage in pointing out that the strength of the international law norm prohibiting official torture argued against the invocation of the doctrine.66 Third, the court must consider whether the case has the potential to embarrass the Executive Branch of government. One can see that, although this is at the heart of the doctrine, negative determinations of the previous two criteria will substantially lessen the potential for this to be so. The burden of proving an act of state rests on the party asserting the applicability of the doctrine. At a minimum, this burden requires that a party offer some evidence that the government acted in its sovereign capacity and some indication of the depth and nature of the government’s interest.67 The Filártiga Appeals Court had mentioned, as a practical matter, that “[w]here reports of torture elicit some credence, a state usually responds by denial or, less frequently, by asserting that the conduct was unauthorised or constituted rough treatment short of torture.” Thus, it should come as no surprise that the Kadic court noted that: “. . . [T]he appellee has not had the temerity to assert in this Court that the acts he allegedly committed are the officially approved policy of a state. Finally, as noted, we think it would be a rare case in which the act of state doctrine precluded suit under section 1350.”68

A final observation should be made about the stage at which these three elements are considered by American courts. The foregoing analyses only proceed on the assumption that the sanctioning of the alleged human rights violation by the foreign state is properly characterizable as an “act of state”. One need not even get to the analysis of whether to respect the foreign act if it is possible to characterize it as not being an act of state, due to the fact it is not in conformity with the foreign state’s own legal system. Thus, of considerable significance is the following statement by the Filártiga Court of Appeals when it said: “We note in passing, however, that we doubt whether action by a state official in violation of the Constitution and laws of the Republic of Paraguay, and wholly 65

Sabbatino, supra n. 60 at 428. Filártiga v. Peña-Irala (trial decision), 577 F. Supp. 860, 862 (E.D.N.Y. 1984). Filártiga (C.A.), supra n. 2. 67 Liu, supra n. 61. 68 Kadic, supra n. 6, at 250. 66

The Experience of United States Courts 81 unratified by that nation’s government, could properly be characterized as an act of state.”69

Thus, it is arguable that legality according to the foreign state’s own system is a sine qua non for the triggering of the American act of state doctrine. The “political question” doctrine With the shifting of the juridical basis of the act of state doctrine away from the notion of deferential respect for the inherent dignity of the foreign sovereign, it has come to overlap significantly with the political question doctrine. This doctrine was given its classic (if serpentine) definition in Baker v. Carr in 1962:70 “A nonjusticiable political question would ordinarily involve one or more of the following factors: [1] a textually demonstrable constitutional commitment of the issue to a coordinate political department; or [2] a lack of judicially discoverable and manageable standards for resolving it; or [3] the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion; or [4] the impossibility of a court’s undertaking independent resolution without expressing lack of the respect due coordinate branches of government; or [5] an unusual need for unquestioning adherence to a political decision already made; or [6] the potentiality of embarrassment from multifarious pronouncements by various departments on one question.”

There are two broad concerns identified here.71 One is the “manageable standards” concern identified in factor 2. This concern is reflected in some courts’ caution in identifying violations of international law, which is discussed below. The other is the “separation of powers” concern found in factors 1, 3, 4, 5 and 6. The legislative and executive branches are said to have exclusive control over the conduct of US foreign relations.72 Although this aspect of the doctrine might potentially have been very broadly interpreted, it has only been invoked successfully in IHRT cases where (direct) American involvement in the (broad) factual context was said to present adjudicative difficulties.73 The rationale for the narrowness of the courts’ interpretation of the doctrine is found in the oft-cited 69

Filártiga, supra n. 2, at 889. Baker v. Carr, 369 U.S. at 217, as cited in Kadic, supra n. 6, at 249. 71 See Randall, supra n. 29, at 105, where these aspects are discussed as the “topical” and “functional” categories. 72 This is discussed in detail in the trial judgment in Linder (reversed by the Court of Appeals, supra n. 35, on the question of the applicability of the political question doctrine): Linder v. Portocarrero, 747 F. Supp. 1452, at 1456–9 (S.D. Fla., 1990). The Linder trial judgment cited Oetjen v. Central Leather Co., 246 U.S. 297, 302, 38 S. Ct. 309, 310, (1918) for the proposition mentioned in the text: “The conduct of foreign relations of our government is committed by the constitution to the executive and legislative—‘the political’—departments of the government, and the propriety of what may be done in the exercise of this political power is not subject to judicial inquiry or decision.” 73 See Randall, supra n. 29, at 109, which discusses cases which touched upon American involvement in Vietnam and Nicaragua. 70

82 M Swan phrase that “the doctrine is one of ‘political questions’, not ‘political cases’ ”.74 In Linder v. Portocarrero, for example, claims against individual Nicaraguan Contra leaders were sustained in the face of the doctrine—although broader claims against the Nicaraguan opposition’s leadership were not. The Linder case involved the torture and murder by Contra forces of a US citizen who was assisting as an engineer in the building of a hydroelectric dam in Nicaragua. The Court of Appeals said that the complaint: “challenges neither the legitimacy of the United States foreign policy toward the contras, nor does it require the court to pronounce who was right and who was wrong in the Nicaraguan civil war . . . [It] is narrowly focused on the lawfulness of the defendants’ conduct in a single incident.”75

However, the court also held: “[T]he broad allegations of the claims in the amended complaint against the defendant organizations, FDN, UNO and NR, which comprise the entire military and political opposition in Nicaragua, are non-justiciable, and the motion to dismiss them was properly granted.”76

It would seem that the adjudication of these allegations would have involved the very “political questions” which were brushed aside in the previous quotation.77 One might think that the involvement of a US victim could have influenced the judicial reasoning in the case, but, when one considers the kinds of defendants and circumstances to which the political question doctrine has been said to be inapplicable, that seems much less likely.78 As was suggested above in the discussion of the act of state doctrine, however, the courts have entertained submissions from the US Department of State as to its opinion on the Executive Branch’s potential concerns with respect to a given case—but the Executive’s submissions have not, perhaps surprisingly, expressed objection to the case being heard, despite the difficulties which might arise in consequence of suits against such defendants as Radovan Karadzic, Ferdinand Marcos and the de facto government of Burma.79 In any case, the Kadic court noted that such submissions do not bind the court.80 74

Kadic, supra n. 6, at 249, 217, citing the leading case of Baker v. Carr, supra n. 70. Linder (C.A.), supra n. 35, at 337 (emphasis added). The court also stated that “there is no foreign civil war exception to the right to sue for tortious conduct that violates the fundamental norms of the customary laws of war.” See the section on “War crimes”, Section 4 infra. 76 Linder, ibid. 77 The Linder trial court, supra n. 72, at 1467, had quite openly and unabashedly discussed the questions that could be raised here with respect to the U.S. government’s support of the Contras, and raised the concerns which seem to be raised so frequently—those of national security and official secrecy. 78 For example Radovan Karadzic. Another good example is National Coalition Government of the Union of Burma v. Unocal, (1997) 176 F.R.D. 329 (C.D. Cal. (9th Cir.) ) (Unocal). Also see Trajano, supra n. 4, at where it is noted that the Department of Justice had pressed the Filártiga court to read §1350 expansively, but that its position was less enthusiastic in the Trajano case. The Hilao and Trajano cases involved suits against Ferdinand Marcos and his daughter, respectively (see note 4, supra). 79 Kadic, supra n. 6, at 250. 80 Ibid. 75

The Experience of United States Courts 83 Forum non conveniens A court may decline jurisdiction if it can be said that another forum exists which will serve the ends of justice, and where the trial can be held more conveniently.81 Both the private interests of the litigants and the public interest are relevant to this determination.82 Unless these factors weigh heavily in favour of an alternative forum, however, the strong presumption in favour of a plaintiff’s choice will prevail.83 The plaintiff’s choice of forum is entitled to less deference, however, where she has not chosen her “home” forum, as the assumption of convenience is less strong.84 Finally the court may conclude that dismissal is not appropriate if: “the remedy provided by the alternative forum is so clearly inadequate or unsatisfactory that it is no remedy at all . . . [or] if the plaintiff shows that foreign law is inadequate, or that conditions in the foreign forum plainly demonstrate that the plaintiffs are highly unlikely to obtain basic justice therein.”85

The private interest factors to be considered in the analysis include the ease of access to sources of proof, the availability and compellability of witnesses, the costs involved, and “all other practical problems that make a trial easy, expeditious and inexpensive”.86 The public interest factors include considerations of administrative difficulties or congestion in the courts, the policy of having localized controversies decided at home, and the view that, all things being equal, questions of foreign law will be better decided by a foreign tribunal expert in that body of law.87 However, this analysis has limited relevance to IHRT cases, for obvious reasons. The first is that, in many of the cases, both plaintiff and defendant are present in the United States, as noted above, in the section on judicial jurisdiction. Second, as we will see later, these claims must either allege the involvement of a state in the commission of the impugned acts, or acts of genocide or war crimes. Any such allegations will at least raise a presumption that the plaintiffs may have difficulty “obtaining justice therein”. Third, as the initial inquiry into §1350 allegations involves a more “searching” review than the usual consideration of the formal sufficiency of the pleadings themselves, the court is in a good position to make the determination of the likelihood of a plaintiff obtaining justice in an alternative forum. Fourth, the TVPA exhaustion-of-foreign-local81 Cabiri v. Assasie-Gyimah, 921 F. Supp. 1189 (S.D.N.Y., 1996), at 1198–9, citing R. Maganlal & Co. v. M.G. Chemical Co., Inc., 942 F.2d 164, 167 (2d Cir. 1991). 82 The modern source of the doctrine is Gulf Oil Corp. v. Gilbert, 330 U.S. 501 (1947), as discussed in Born and Westin, supra n. 5, at 209 et seq. 83 This is drawn from the Supreme Court’s elucidation of the doctrine in Piper Aircraft Co. v. Reyno, 454 U.S. 235 (1981), as discussed in Born and Westin, supra n. 5, at 214 et seq. 84 Ibid. 85 Rasoulzadeh v. Associated Press, 574 F. Supp. 854 (S.D.N.Y. 1983), aff’d, 767 F.2d 908 (2d Cir. 1985), as cited in Cabiri, supra n. 81, at 1199. 86 Cabiri, ibid., citing Gulf Oil, supra n. 82. 87 Cabiri, supra n. 81.

84 M Swan remedies requirement will mean that, in some cases, the forum non conveniens inquiry will be more than redundant, as an exhaustion analysis will ask many of the same questions related to the degree of accessibility of the plaintiff to justice if the plaintiff were to sue in the foreign court.88 Lastly, as international law is applied, to one extent or another, to the merits in such cases, the force of the forum non conveniens argument is lessened at the outset. From a comparative perspective, it is significant that no IHRT case seems to have been dismissed on the basis of forum non conveniens. In Cabiri, for example, the plaintiff had applied for asylum in the US. In Kadic, it was a simple matter for the court to point out that the courts of the former Yugoslavia offered no viable alternative forum.89 In Filártiga, the court had evidence before it that the government of Paraguay would not prosecute the defendant criminally, and that no civil suit could be launched until a criminal trial had been concluded.90 Head of state immunity This separate common law immunity is a privilege extended at the pleasure of the Executive Branch, and is not a matter for judicial determination. This doctrine seems to have survived intact the re-articulation of sovereign immunity under the FSIA and of the act of state doctrine, and is thus grounded in older notions of the unity of the head of state and the sovereign. The grant of immunity entitles the foreign head of state to absolute immunity from the jurisdiction of US courts—the ultimate “get out of jail free” card, as it were. The immunity was successfully invoked in the Aristide case, in which the doctrine was discussed at length.91 The situation with respect to immunity of former heads of state is more complex from the perspective of existing positive law in part because of the inconsistent ways in which various states formulate this immunity in their immunity statutes. However, former head of state immunity is certainly less than absolute as revealed by the comparative law discussion and reasoning in the recent British House of Lords Pinochet (No. 3) judgment. As the Pinochet case will now be far more on point as a source of persuasive authority for many courts outside the US, US case law on former head of state immunity will not be further explored here.

88 The TVPA requires that any remedies available in the place where the conduct giving rise to the claim occurred be exhausted, in order for the court to take jurisdiction in the matter: § 2(b). 89 Cabiri, supra n. 81, held that the TVPA requirement that the plaintiff exhaust the remedies of a foreign forum is generally not required “when foreign remedies are unobtainable, ineffective, inadequate, or obviously futile” (at 1199, citing Xuncax, supra n. 45, at 178); Kadic, supra n. 6. 90 Filártiga, supra n. 6 (trial decision, 1984). 91 Lafontant v. Aristide, 844 F. Supp. 128 (E.D.N.Y, 1994). It was claimed that Mr. Aristide ordered the murder of a man who had opposed him and aided in the planning of the coup which had deposed him.

The Experience of United States Courts 85

4 CHOICE OF LAW ISSUES AND APPLICATIONS OF SUBSTANTIVE LAW

Overview of the Choice of Law As was noted above, one need only plead a “violation of the law of nations” under §1350 and, in effect, one does not need to plead a separate “tort” cause of action. The cause of action comes from federal law, the statute itself. This still leaves open the question, however, of which substantive rules of decision will be applied in giving effect to this federal cause of action. This is the choice of law question. The bottom-line practical “answer” to the question is as follows. The courts apply international law (as “guided by” domestic municipal law for matters of detail) to questions of the existence of a norm, and whether it has been violated. The courts presently apply domestic law to questions of standing and the application of statutes of limitations. With respect to questions of remedy, the courts are developing a federal common law response, unique to IHRT jurisprudence. The “identity”, if you will, of the choice of law rules which direct the courts to apply these multiple sources of law seems somewhat unclear, however.92 Ordinarily, in private international law cases, the federal courts are bound to apply the substantive law of the state in which they sit, unless otherwise directed by the federal statute giving rise to the grant of jurisdiction, or the Constitution.93 This includes the use of that state’s choice of law rule.94 This is the case unless the state law at hand would defeat the purpose of the federal enactment.95 The Supreme Court has also held that there is no “general” federal common law, which the federal courts might have developed to guide them in such questions.96

92 For further discussion of this theme of hybrid and unclear sources, see, in this volume, M Moran, supra n. 46. 93 The Supreme Court, in Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938) thus construed the Rules of Decision Act, 28 U.S.C. 1653 (1982), as cited in D C Chow, “Limiting Erie in a New Age of International Law: Toward a Federal Law of International Choice of Law” (1988) 74 Iowa Law Review 165–225:

“Except in matters governed by the Federal Constitution or by Acts of Congress, the law to be applied in any case is the law of the State. And whether the law of the State shall be declared by its Legislature or by its highest court in a decision is not a matter of federal concern. There is no federal general common law.” (emphasis added) 94 The federal courts had been using a federal common law choice of law rule, and continued to do so in some cases after Erie, until the Supreme Court explicitly foreclosed this possibility in Klaxon v. Stentor Electric Manufacturing Co. 313 U.S. 487 (1941): see Chow, supra n. 93 at 179 95 Xuncax, supra n. 45, at 189–90, citing Bell v. City of Milwaukee, 746 F.2d 1205 for the general proposition and as to the specific inapplicability of wrongful death statutes to standing issues. 96 Erie, supra n. 93.

86 M Swan That being said, there is such a thing as “specialised” federal common law. It has been said that the courts have developed “enclaves” of “specialised” federal common law in three principal contexts:97 1) to protect uniquely federal interests, i.e., where the proprietary interests of the government are involved;98 2) to fill in the interstices of a pervasive scheme of federal legislation;99 and 3) in cases touching upon issues of national sovereignty, i.e. involving controversies between states, admiralty and foreign relations.100 In IHRT cases, the courts have clearly chosen, wherever possible, to turn to rules of decision provided by international law, as will be discussed below. No court, however, speaks, per se, of the choice of law analysis which allows this. One way to understand this might be to say that it is simply done on the assumption that the statutory language of §1350 mandates it, as a special choice of law rule. Another way to understand it, however, is to conclude that the courts are (implicitly) fashioning choice of law rules as a “specialized” federal common law response to the cause of action granted by §1350. The courts do speak of developing federal common law in the area of IHRT claims, but, in so doing, make virtually no mention of the jurisprudence surrounding issues of “specialized” versus “general” federal common law. One might suggest three reasons for this. First, the ATCA itself cannot be described as a “pervasive” scheme of legislation. With the recent passage of the TVPA and the Flatow amendments to it, however, the courts might be biding their time before making a judicial pronouncement of this sort, at which time the possible presence of a sufficiently pervasive scheme can perhaps be referenced not only to the ATCA alone, but to the ATCA understood in relation to the TVPA, the FSIA and the adoption of CIL by federal common law. Second, the courts have argued often and at length that the cases do not generally have important implications for the conduct of foreign relations, so as to avoid the operation of the political questions doctrine. In so arguing, they have made problematic recourse to the third of the above-noted enclaves of existing federal common law. Third, although the courts have had much success in forging a consensus on questions relating to the existence of a “cause of action” and with the sufficiency of international law concepts, the area of IHRT jurisprudence is still young and quite fluid. The language the courts use in affirming the appropriateness of developing federal common law in IHRT cases is exemplified by the following quotation:

97 Chow, supra n. 93, at 197–8. As this article’s title would indicate, the author is in favour of the federal courts development of a “specialised” federal common law in this area, according to its practice of doing so in the ways mentioned in section numbers 2 and 3 supra 84 and 85. 98 Ibid. at 197, citing Sabbatino, supra, n. 60. 99 Chow, supra n. 93 at 198, citing Sola Electric Co. v. Jefferson, 317 U.S. 173 (1942) at 176. 100 Ibid., citing Sabbatino, supra n. 60.

The Experience of United States Courts 87 “Congress, of course, may enact a statute that confers on the federal courts jurisdiction over a particular class of cases while delegating to the courts the task of fashioning remedies that give effect to the federal policies underlying the statute.”101

Although all the courts are clearly applying international principles to the merits of the cases, they do not usually say even this much about the matter. The trial court in Filártiga had stated: “The court would . . . in accordance with traditional conflict of laws principles, apply the substantive law of Paraguay. If the ‘tort’ to which the statute refers is the violation of international law, the court must look to that body of law to determine what substantive principles to apply. ... The court concludes that it should determine the substantive principles to be applied by looking to international law, which, as the Court of Appeals stated, ‘became a part of the common law of the United States upon the adoption of the Constitution’.”102

The seeming directness of this statement, however, is accounted for by the fact that this case was decided before the “cause of action” controversy raised its head in Tel-Oren. While international law is looked to as a generator of relevant substantive principles of law, it is nowhere actually held that it actually provides choice of law rules, despite what may, at first glance, be inferred from the justquoted passage. The “location” of the choice of law rule providing for the application of international law must be said to be either implicit in §1350 or come from federal common law. The choice of law question is made more complex by the fact that, even if it is said that international law “applies” in some fashion, it does not provide a complete package of substantive rules of decision on all the civil liability issues raised by IHRT cases.103 Possibly with this in mind, the courts have split the question of which substantive law is applicable to the merits of a case into three parts. In the first part of the inquiry the courts look to international law sources to affirm the existence of a legal norm which will provide the essential foundation for the adjudication of IHRT claims. The norm speaks to the existence of an individual’s “right” not to be treated in certain ways, by certain people. Thus, the norm must identify two things. It must identify the nature of the behaviour which international law prohibits or requires—the “what”, whether it be

101 Abebe-Jira, supra n. 5, at 848, Xuncax, supra n. 5, at 179–83 and Kadic, supra n. 6, at 236. For the proposition, all cases cite Textile Workers of America v. Lincoln Mills, 353 U.S. 448, 77 S. Ct. 912 (1957) (jurisdictional statute permitting judicial explication of federal common law). 102 Filártiga (trial decision, 1984), supra n. 66, at 862–3. 103 This is a view derived entirely from my reading of the American legal sources, however. It is entirely possible that the courts of other jurisdictions might disagree with the American courts on this point, either generally, or with respect to the existence of given elements of a cause of action in international law.

88 M Swan torture, genocide, slave-trading or something else.104 It must also identify the subject of the international legal prohibitions—the “who”. If the alleged violator of the right is an individual, the analysis starts with the international law premise that the defendant must have acted in concert with “state action” or “under colour of law”, rather than as a private individual. This international law requirement is interpreted using concepts drawn from US municipal law, as a “guide”, where necessary.105 With respect to a few IHRT norms, however, international law itself is said to allow suits against private individuals per se, and international law rules seem sufficient to answer the “who” question. The extent to which US courts have correctly interpreted international law on its scope of applicability would need more extensive analysis than can be provided here. For example, the degree and kind of state nexus needed with respect to article 1 of the UN Convention against Torture (CAT) is a complex interpretive question.106 Similarly, it is not unlikely that CAT’s definition of torture is more conservative in some respects than that of general international law or that of the evolving jurisprudence of the Human Rights Committee under the International Covenant on Civil and Political Rights (ICCPR), and the degree to which this is the case is also an important question.107 As to questions regarding standing rules and statutes of limitations, the courts appear to be applying generic federal choice of law rules, which have been developed independently of the IHRT context.108 When it comes to remedies, the courts appear to be granting monetary awards strictly (and impressively) with 104 As the rules of customary international law are said to form part of federal common law, this is arguably a lex fori approach, but to construe it so might be said to miss the point of the exercise: see the subsection following. 105 The term “guide” is used in Kadic, supra n. 6, at 245. The phrase itself is suggestive of the caution and sensitivity which the Second Circuit feels is necessary to maintain the sense that it is substantive international law which is being applied. 106 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.C.M. 1027 [hereinafter “CAT”]. Note that nowhere does article 1 say the actor who tortures must be a member of the state or even acting as agent of the state: see Appendix 1. Rather, acquiescence by state officials is sufficient to trigger the legal presence of torture for purposes of CAT, such that torture by private actors can count as torture. Thus, the inaction of the state in the face of sufficient knowledge of a specific situation or general practice is enough to surmount the public/private barrier. 107 The ICCPR’s general formulation leaves much more room for expansive interpretation than the more carefully tailored CAT definition International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. (entered into force 23 Mar.1976)[hereinafter “ICCPR”]. Although it must be that CAT has interpretive relevance for the ICCPR, two overriding normative principles are of greater relevance. One, the ICCPR is one of the two pillars of the UN human rights treaty order, the other pillar being the International Covenant on Economic, Social and Cultural Rights; as such, its normative priority means that the ICCPR has greater interpretive force vis-à-vis the CAT than the other way around. Two, most human rights treaties have an important savings clause intended to make sure that lower protection under one treaty does not undermine greater protection under another, and the CAT is no exception. Its article 1(3) reads: “This article is without prejudice to any international instrument or national legislation, which does or may contain provisions of wider application.” 108 While the operation of statutes of limitations and standing rules may be characterised as either a substantive or a procedural matter, the American IHRT cases which I have read have not

The Experience of United States Courts 89 respect to their intuitions as to what is appropriate to the IHRT in question. This is done as a matter of the provision of federal common law remedies for the violations of international human rights, as was noted eralier. They do not yet seem to have concerned themselves with the possibility of injunctive relief. Finding Violations of International Law: “Who has done What . . . (Exactly)?” The rationales for the application of international law to substantive questions of liability In the trial court decision in Filártiga, the court suggested that traditional choice of law principles in international matters are concerned with comity and the interests of particular states. Where the court is dealing with “hosti humani generis” (enemies of all humankind), however, the courts should recognise that the interests of the global community transcend these issues.109 General acknowledgement of this position would allow the US courts to give effect to the consensus that established the existence of the violations, and thus fashion remedies which recognise the “monstrous” nature of the acts involved.110 It has also been suggested by the court in Xuncax v. Gramajo that taking the “tort” action out of international law and making it a matter of a traditional choice from amongst different municipal laws would invite governments to enact immunities, exemptions and statutes of limitations for government officials.111 The Xuncax court added that an international law approach (as developed under federal common law) would leave the courts freer to incorporate the diverse elements, including but not limited to the municipal law of other interested countries, necessary to resolve international legal issues, so long as these were not inconsistent with international or US law.112 I take this to mean that the courts could develop a federal choice of law rule that could take account of these sources, rather than that the federal common law would be said to be developing law that is purely “American”. The court also suggested that there might be novel problems associated with applying foreign municipal law, but the suggestion was weakly asserted. I note in passing that the case also stands for an important evidentiary point regarding proof of foreign law: that a federal court “may look to any relevant material or source, whether or not discussed the distinction (directly, at least). I will infer, with great trepidation, that the matter is generally considered as a matter of substantive law, and include it here. 109 Filártiga (trial decision), supra n. 66, at 863–4. The court noted that foreign municipal law could be taken into account where it did not conflict with either the pursuit of the objectives of international law, or matters of US public policy, but these seem not to be discussed in later cases—insofar as there is any serious discussion at all about “choice of laws” beyond that in the Xuncax case, supra n. 45 see also infra, n. 125 (and accompanying text) and n. 142. 110 The Xuncax court also voiced this concern. Xuncax, ibid. at 182–3. 111 Ibid. 112 Xuncax, supra n. 45 at 187, where the court used municipal law, in supplementary fashion only, to flesh out the terms of the prohibition against “cruel, inhuman, or degrading treatment”: see infra n. 116 (and accompanying text) and nn. 121, 139 and 142.

90 M Swan submitted by a party or admissible under the Federal Rules of Evidence” to determine the content of the laws of a foreign country. The application of international law to substantive questions of liability (i) Is What Has Been Alleged an International Human Rights Tort? (1) Treaty law While a violation of international treaty law is the most evident, and prima-facie accessible kind of violation covered by the ATCA, the focus of the cases has not been on the phrase “violation of . . . a treaty of the United States”. Rather, the focus has been on the words “violation of the law of nations” with the courts generally not, it seems, seeking to derive a tort directly from the fact of a violation of a multilateral international human rights treaty. They appear, in general, to prefer a more indirect recourse to widely ratified treaties as a centrepiece in determining whether general international law applicable to all states has been violated. Here, the language of the Filártiga court is of special interest: “It is only where the nations of the world have demonstrated that the wrong is of mutual, and not merely several, concern, by means of express international accords, that a wrong generally recognized becomes an international law violation within the meaning of the statute.”113

The context was such that it is clear the court was not saying that treaties themselves are being used as the ATCA source, but, rather, that multilateral human rights treaties are to be used as evidence of the general “law of the nations”, with there appearing to be no requirement that the US itself need necessarily be party to the treaties most relevant to the claim. Indeed, at the time of Filártiga and until very recently, the US was not party to the ICCPR, perhaps the most relevant UN instrument with respect to the subject-matter of the bulk of the IHRT claims brought under the ATCA and the TVPA. Perhaps it is for this reason that the US courts have not drawn a distinction between the relevance of “selfexecuting” and “non-self-executing” treaties for purposes of their relevance to helping prove the existence and content of the general “law of nations”. On the whole, then, it cannot be said that the US IHRT cases have grappled with international treaties as the immediate formal source of tort-grounding norms, preferring instead the more open and ambiguous recourse to the “law of nations”. Here, it might further be noted that one major axis of analysis has been with respect to how fundamental a given “law of nations” norm must be in order for it to ground an ATCA tort claim, which suggests that, even if treaty law as such can contain tort-generating norms, a qualitative test for the kinds of treaty norms still needs to be satisfied. An analysis of how fundamental a norm is may be viewed as a surrogate for satisfying the court that a norm enjoys the 113

Filártiga, supra n. 2, at 888.

The Experience of United States Courts 91 degree of general consensus necessary for US courts to adjudicate claims centred in other states. It is also possible that the courts are concerned not just with a requirement of breadth of support for a norm, but are tending to the view that it is the special quality of a norm that justifies some “law of nations” norms grounding an ATCA claim while not others. Invocation of suppression of freedom of expression in a foreign state is of a different order from invocation of torture. It is to this hierarchy of international human rights norms that I now turn. (2) Violations of customary international law (CIL), or must the norms have jus cogens status? As hinted at at the end of the previous section, there seems to be some confusion as to whether (mere) violations of CIL are all that is required under §1350, or whether jus cogens (“peremptory”) norm violations are called for. While Filártiga discussed only the demands of CIL, it is now generally the practice of the courts to ensure that the norm violated is also a jus cogens norm. As jus cogens norms are said not to depend upon the consent of states for their legal force, and are thus binding upon all states regardless of whether they object to the norm (“non-derogable”), they offer a firmer basis for adjudication—a jurisdiction more plausibly justified as a universal jurisdiction.114 Regular CIL norms (jus dispositivum) are said not to be binding upon a state which persistently expresses its lack of accord with them.115 Most courts have explicitly said that the applicable norm must be (1) universal, (2) “binding upon all states” or “obligatory”, and (3) justiciable in the sense of presenting standards sufficient to permit adjudication of prohibited conduct.116 The courts tend to cite Filártiga for authority for these propositions, but while the court in Filártiga did note that official torture is, in fact, the subject of the strongest universal condemnation, it did not make jus cogens status a condition for §1350’s applicability.117 In some contrast to the tendency of courts to also look to see if a norm is jus cogens, the Second Circuit courts, since Filártiga, have been consistent in suggesting instead that the defendant’s conduct must do no more than violate “well established, universally recognized norms of international law”—presumably to leave open the possibility of drawing upon custom alone as a source of international law 114 Siderman de Blake v. The Republic of Argentina, supra n. 25 citing the definition of jus cogens given in the Vienna Convention on the Law of Treaties, art. 53, May 23, 1969, 1155 U.N.T.S. 332, (1969) 8 I.L.M. 679. 115 116 Xuncax, supra n. 45 (D. Mass. 1st Cir.), at 184, found that the norm must be “non-derogable” and that there must be “sufficient criteria to determine whether a given action amounts to the prohibited act”; Hilao, supra n. 4, (9th C.A.) at 1475, found that the norm must be “specific, universal and obligatory”. The latter Hilao language was adopted in Abebe-Jira, supra n. 45, at 847. Beanal v. Freeport-McMoran, supra n. 4, (5th Cir) at 370, used a similar formulation: “The alleged violation must be definable, obligatory (rather than hortatory), and universally condemned.” 117 See Kadic, supra n. 6 at 239, and Cabiri, supra n. 81, at 1196.

92 M Swan norms.118 This is a potentially expansive approach given that CIL is a much broader category of international law norms than jus cogens.119 That being said, all the norms that have thus far been applied by the courts, however, have in fact been jus cogens norms—an “elite sub-set” of customary norms, to borrow the language of the court in Nicaragua v. Reagan.120 The third condition—justiciability—seems eminently sensible albeit not a little circular, begging as it does the question of the best theory of the appropriateness of an active judicial role in concretising fairly abstract, yet clear, principles of international human rights law.121 (ii) The Subjects of International Law Prohibitions: The “State Action” Requirement (1) General Overview As noted above, while this “who” inquiry is not really severable from the “what” inquiry, it is the subject of separate legal analysis. Here, the courts have felt it necessary to flesh out the dictates of the international law requirement with concepts borrowed from analogous municipal (American) law. They have held that private individuals are not generally the subjects of international law, and it must ordinarily be shown that some “state action” was present within the context in which the individuals acted, so that they may be treated as a subject of international law.122 This proposition has been the subject of detailed discussion recently, as cases have come before the courts in which the defendant was not (or may not have been) a state official. In Beanal, for example, it was alleged that an American corporation (Freeport-McMoran) was involved in the commission of human rights abuses with respect to its mining operations in Indonesia.123 It was further alleged that 118 See for example Kadic, supra n. 6 at 238–9, and Cabiri, supra n. 81 at 1196 and Filártiga, supra n. 2, at 881, 884 and 8. 119 There are many more human rights, including political, social and economic rights, which are recognized by the Universal Declaration of Human Rights, for example, which might achieve this status—such as the right to equal pay for equal work, the right to marry, the right to join a trade union—but which might never (or not soon) achieve the status of jus cogens norms: see Restatement (Third) supra n. 10, §701, reporter’s note 6 for a list of other CIL rights. And see the list of customary human rights norms set out by the UN Human Rights Committee in its General Comment on reservations to the ICCPR: Human Rights Committee, General Comment 24(52), General comment on issues relating to reservations, UN Doc. CPR/C/21/Rev.1/Add.6 (1994). 120 Siderman de Blake, supra n. 25 at 715, citing Nicaragua v. Reagan, 859 F.2d 929, 940 (D.C. Cir. 1988). 121 Compare Xuncax, supra n. 45, at 187, where it was said here that not every aspect of a prohibition need be fully defined and that municipal law can be drawn upon in order to help achieve definitional clarity. Contrast with the 9th Circuit declining to find an international law violation on the basis of a norm lacking, in its view, sufficient concreteness: see Forti v. Suarez-Mason, 694 F. Supp. 711 at 712, n. 33 (“Forti II”). 122 Kadic, supra n. 7, 239–40. Beanal, supra n. 4, at 374. This requirement is explicit in the TVPA, §2(a), supra n.8 reproduced in Appendix 2 of this volume. The question of which norms require some form of “state action” is dealt with in the following section. 123 Beanal, supra n. 4.

The Experience of United States Courts 93 the Indonesian government was a major shareholder of the corporation, and had supplied military personnel to the mining sites, where Freeport’s own security personnel had engaged in, inter alia, summary executions of indigenous people. The court, in a detailed treatment of the issue, referred to the “under color of authority” test contained in the comments to Restatement of the Foreign Relations Law of the United States (Third) §207: “. . . one must consider all the circumstances, including whether the affected parties reasonably considered the action to be official, whether the action was for public purpose or for private gain, and whether the persons acting wore official uniforms or used official equipment.”124

After some discussion, however, the court found that this test yielded inconclusive results, and so turned to jurisprudence under USC §1983, which concerns government involvement in deprivations of civil rights sufficient to elevate conduct by private actors to the status of constitutional torts. Both Kadic and Forti have also held that this jurisprudence provides a “guide” by which the international law requirement may be interpreted.125 The Beanal court listed four different §1983 tests, any one of which may suffice to show “state action”. “[1] . . . whether there is a sufficiently close nexus between the State and the challenged action of the regulated entity so that the action of the latter may be fairly treated as that of the State itself. [2] . . . whether the state has so far insinuated itself into a position of interdependence with the private party, that there is a symbiotic relationship between them. [3] . . . [whether] a private party is a wilful participant in joint activity with the State or its agents [so that] state action is present. [4] . . . [whether] a private entity . . . exercises powers traditionally exclusively reserved to the State.”126

These are often referred to as the “nexus”, the “symbiotic relationship”, the “joint action” and the “public function” tests, and are discussed in detail in Beanal. The Beanal court dismissed the suit for want of the allegation of sufficient facts in light of the court’s view of the law on liability of private actors for international human rights violations.127 If one considers the reasoning 124

Beanal, supra n. 4 at 375, citing the Restatement (Third), supra n. 10, §207, Comment d. Kadic, supra n. 6, at 245. Beanal, supra n. 4, at 374. It appears that the suggestion that courts may be “guided by” concepts found in municipal law is used consciously in Kadic to underscore the fact that the norm may reasonably be said to have been created by international law, where such detailed elements of the “what” of the prohibition have been provided, along with substantial, if preliminary, indications of the “who”. The assertion may be harder to maintain in a situation where the “what” is as vague as it is with the prohibition against “cruel, inhuman and degrading treatment”. This does not mean that the move made by Xuncax with respect to the use of municipal law to “supplement” international law is necessarily illegitimate, though. It is important to remember that the courts have said from the outset that they are interpreting customary international law as it evolves. It may be the case that the more complex problems of definition that this norm presents argue for a kind of “international common law” response. One would want the courts to be careful, but one would also not want them to be too conservative about such niceties, however, if one wishes to see such a novel, and involved, body of law to develop. 126 Beanal, supra n. 4, at 376, citing Gallagher v. Neil Young Freedom Concert, 49 F.3d 1442, 1446 (10th Cir. 1995). 127 However, the court did grant the plaintiff leave to amend his claim. The case was finally dismissed when the court held that the plaintiff had failed to avail himself of this opportunity to better 125

94 M Swan offered in the Beanal case, however, it seems to be an example of how mechanical applications of municipal law concepts to IHRT claims can yield unsatisfactory results. The Beanal court relied on a case in which “pat-down” searches had been conducted by security personnel in the presence of police officers,128 in which it was held that the police could not have been said to have adopted the behaviour of the security personnel by their “mere presence”.129 One has to wonder how much this situation really has to say about one in which military personnel observe (at the least) the carrying out of summary executions, in their country, by the security personnel of a corporation. By reason of the very seriousness of the abuse of rights here, it would seem fair and reasonable to infer an adoption of (by acquiescence in) the behaviour in the one situation, if not in the other. This is not to suggest, however, that it is inappropriate to use the categories developed under §1983 as a guide to the interpretation of the international law principle. The Kadic court stated that “[a] private individual acts under color of law within the meaning of §1983 when he acts together with state officials or with significant state aid”.130 This simpler articulation of the jurisprudence might indicate a desire to rework and adapt the principles found in §1983 to the IHRT context. (2) Command responsibility The Senate report accompanying the TVPA stated that: “Under international law, responsibility for torture, summary execution, or disappearances extends beyond the person or persons who actually committed those acts— anyone with higher authority who authorised, tolerated or knowingly ignored those acts is liable for them.”131

The court in Xuncax awarded damages for summary execution, disappearance torture, arbitrary detention and cruel, inhuman, or degrading treatment against the sole defendant in the case on this basis.132 The court in Forti I reached a similar result.133

state the claim so as to allege the requisite degree of state action required by the court’s understanding of the law: Beanal v. McMoran, Inc., 197 F.3d. 161; 30 Envtl. L. Rep. 20 (1999). 128

Beanal, supra n. 4. Ibid., at 377. 130 Kadic, supra n. 6, at 245, citing Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S. Ct. 2744 (1982). 131 S. Rep. No. 249, 102d Cong., 1st Sess. 9 (1991), as cited in Xuncax, supra n. 45, at 172. 132 Xuncax, ibid. 133 Forti I, supra n. 33, at 1537–8 and 1541–2. Both cases discuss the proposition at length, and draw on American case law and war crimes concepts. 129

The Experience of United States Courts 95 (iii) Jus Cogens Norms Identified as Actionable Under §1350 by the US Courts When discussing the norms that have been identified, the courts generally make reference to the Third Restatement of the Foreign Relations Law of the United States (Restatement (Third) ), §§701 and 702. Section 702 identifies seven customary international law violations, and states that the first six of these are jus cogens norms.134 The Comments and Reporters Notes which follow provide further specifics, but not all of the prohibitions are defined in §702 with the definitional specificity required by the courts. It is important, therefore, to look for cases which deal with the prohibition as a specific allegation in the case, and to look at the courts’ independent analysis of the prohibition’s detail. The discussion in the Restatement (Third) is also focused on IHR torts as a function of “state policy”, and while it does discuss individual responsibility for some IHR torts such as genocide, it does not allude to the more developed “state action” analysis that the courts have embarked upon by borrowing from US constitutional torts jurisprudence. I will not attempt to provide authoritative or exhaustive detail in my discussion of the courts’ identification of these norms, but will only provide more general remarks as to the level of success the courts have had in identifying the norms, as well as identify where the discussions have taken place, and where the juridical concepts have come from. (1) Norms generally actionable only when “state action” is present (a) Summary Execution, Extrajudicial Killing, Murder or Wrongful Death These four terms seem to be used interchangeably, although the first two are more suggestive of a formal inquiry into the paucity of existent procedural protections before an execution. The Filártiga case dealt with these allegations, but the legal analysis focused exclusively on the torture allegations, until the court reached the discussion of damages. In Forti I the court cited several international agreements, the Restatement (Third), and other federal court judgments in holding that the prohibition against summary execution (or murder), is “universal[,] . . . readily definable, and . . . obligatory”, although it is worth noting that the prohibition was

134 A state violates international law if, as a matter of state policy, it practices, encourages, or condones: (a) genocide; (b) slavery or slave trade; (c) the murder or causing the disappearance of individuals; (d) torture or other cruel, inhuman, or degrading treatment or punishment; (e) prolonged arbitrary detention; (f) systematic racial discrimination; or (g) a consistent pattern of gross violation of internationally recognised human rights. See also Comment n, and Reporter’s Note 11, to the effect that (a) to (f) are jus cogens norms, and, in addition, are not subject to derogation even in times of emergency. See e.g. J Charney, “The Persistent Objector Rule and the Development of Customary International Law,” (1985) 56 BYIL 1.

96 M Swan there referred to as a customary norm without being additionally styled as jus cogens.135 No specific definition of the norm was given. With the passage of the TVPA, however, it is likely that the detailed definition of “extrajudicial killing” given therein will provide the judicial standard, where necessary, although, as was the case in Filártiga and Forti I, the courts do not appear to feel a need to engage in a detailed inquiry.136 This is not simple carelessness, however, as the courts do separately analyse whether or not a state official undertook tortious acts within the proper scope of official authority, both to assess the potential applicability of the FSIA, and to decide whether the defendant is properly the subject of international law norms. Thus it seems reasonable not to spill much ink over questions as to the “judicial nature” or the “rightfulness” of a killing, if it has been shown that a defendant’s acts were taken entirely outside of his or her lawful authority. As suggested above, it is also seems that the TVPA definition is derived from international law instruments.137 (b) Disappearance On reconsideration of Forti I, in which the court had not found it possible to adjudicate a claim based on a “disappearance”, the Forti II court held that the claim was actionable and that the prohibition: “. . . has two essential elements: (a) abduction by a state official or by persons acting under state approval or authority; and (b) refusal by the state to acknowledge the abduction and detention.”138 135 Forti I, supra n. 33, at 1542. Forti I cited the following legal sources: Universal Declaration of Human Rights, art. 3, G.A. Res. 217A, U.N. Doc. A/810 (1948); International Covenant on Civil and Political Rights, art. 6, G.A. Res. 2200, 21 U.N. GAOR Supp. (No. 16), U.N. Doc. A/6316 (1966); American Convention on Human Rights, art. 5, OAS Treaty Series No. 36, OAS Off. Rec. OEA/Ser. 4 v/II 23, doc. 21, rev. 2 (English ed. 1975); Tel-Oren, supra n. 7 at 781 (quoting Restatement (Revised) of the Foreign Relations Law of the United States §702 (Tent. Draft No. 3, 1982) ); Guinto v. Marcos, 654 F. Supp (S.D. Cal., 1986) at 280; and De Sanchez v. Banco Central de Nicaragua, 770 F.2d 1385, 1396 and n. 15 (5th Cir. 1985) at 1397. 136 The TVPA, supra n. 8, § 3(a) reads:

“Extrajudicial Killing.—For the purposes of this Act, the term ‘extrajudicial killing’ means a deliberated killing not authorized by a previous judgment pronounced by a regularly constituted court affording all the judicial guarantees which are recognised as indispensable by civilised peoples. Such term, however, does not include any such killing that, under international law, is lawfully carried out under the authority of a foreign nation.” In addition to the Filártiga and Forti cases, one can look to the second Marcos estate case, Trajano, supra n. 4, as a further example of the cursory nature of these discussions. 137 For an answer to this question one might look to the report of the House of Representatives which accompanied the passage of the TVPA: H.R. Rep. 367(I), 102nd Cong., 1st Sess. 1991, 1992 U.S.S.C.A.N. 84, 1991 WL 255964. The Restatement (Third) mentions that a killing “necessary under exigent circumstances, for example by police officials in line of duty in defence of themselves or of other innocent persons, or to prevent serious crimes” will not be “murder” for CIL purposes: §702, Comment f. This is a slightly more detailed discussion of “lawful killings” than is given under the TVPA, which says that an “extrajudicial killing” “does not include any such killing that, under international law, is lawfully carried out under the authority of a foreign nation” (Section 3(a) ). 138 Forti II, supra n. 124, at 711.

The Experience of United States Courts 97 The court in Xuncax adopted this definition, and both cases cite a number of international sources for the norm, as well as briefs on the subject submitted to the court by panels of international law scholars.139 (c) “Official” Torture This was the first norm identified in the modern line of ATCA cases, in Filártiga. That case cites a number of international law sources for its existence including the forerunner to CAT, the UN General Assembly’s Declaration Against Torture.140 The TVPA also sets out a definition of the norm and standards for its adjudication which, while closely related to international law, is more expansive than the definition in CAT.141 (d) Prolonged Arbitrary Detention It has been accepted by the courts that have had to consider the question that prolonged arbitrary detention is an actionable jus cogens norm, but its content is not expressly developed in the cases.142 The Forti I court held that the term is readily definable, without actually doing so.143 The Xuncax court awarded damages for arbitrary detention without giving more than a detailed recital of the underlying facts of the claim. The Restatement (Third) §702, however, gives a number of specific definitions of the term “arbitrary”.144 139

Forti II, supra n.124, and Xuncax, supra n. 45, at 185. Filártiga, supra n. 2, at 882–3, citing UNGA Declaration on the Protection of All Persons from Being Subjected to Torture, GA Res. 3452, 30 UN GAOR Supp. (No. 34) 91, UN Doc. A/1304 (1975). 141 The TVPA, § 3(b) reads: 140

“Torture.—For the purpose of this Act— (1) the term ‘torture’ means any act, directed against an individual in the offender’s custody or physical control, by which severe pain or suffering (other than pain or suffering arising only from or inherent in, or incidental to, lawful sanctions), whether physical or mental, is intentionally inflicted on that individual for such purposes as obtaining from that individual or a third person a confession, or punishing that individual for an act that individual or a third person has committed, intimidating or coercing that individual or a third person, or for any reason based on discrimination of any kind. (2) mental pain or suffering refers to prolonged mental harm caused by or resulting from— (A) the intentional infliction or threatened infliction of severe pain or suffering; (B) the administration or application, or threatened administration or application, of mind altering substances or other procedures calculated to disrupt profoundly the senses or personality; (C) the threat of imminent death; (D) the threat that another individual will imminently be subjected to death, severe physical pain or suffering, or the administration or application of mind altering substances or other procedures calculated to disrupt profoundly the senses or personality.” 142

See Xuncax, supra n. 45 (in which damages were awarded for the breach of the norm). Forti I, supra n. 33, at 1541–2. 144 Comment g. The Comment suggests, inter alia, that a detention is “arbitrary” if not pursuant to law, if supported only by a general warrant, if charges are not specified, if no early opportunity to consult with family or counsel is afforded, and if a detainee is not brought to trial in a reasonable time. 143

98 M Swan (e) Cruel, Inhuman or Degrading Treatment or Punishment It seems that this norm clearly exists in international law even if its content is subject to wide interpretive differences of view.145 It was said in Xuncax: “[T]he international legal scholars assert that the major international agreements on human rights generally treat the norm proscribing cruel, inhuman, or degrading treatment in parity with the prohibition against official torture.”146

In relation to the problem of nailing down the content of this very open-textured norm, the court in Forti I noted: “[I]t is unclear what behavior falls within the proscription—beyond such obvious torts as are already encompassed by the proscriptions of torture, summary execution and prolonged arbitrary detention.”147

Here it is important how the Xuncax court dealt with the relation between the undoubted existence of the general norm and the concrete content that could be given to it. That court held that not every element of a tort need be defined in order to be actionable (while finessing the point that it seems completely undefined as a matter of international law). It was held that the norm could, in part, be defined as “any act by the defendant which is proscribed by the Constitution of the United States and by a cognisable principle of international law”, and it adopted this view of the need for an overlap of content with US law in light of a US Senate reservation to the Convention against Torture to this effect.148 The Xuncax court was fully aware that it was supposed to be applying international law, and that the interpretive “move” just described here might be objectionable. The court justified its decision with an assertion that the international law norm and the domestic jurisprudence “overlap”, and noted that the United States, which had participated in the achievement of the consensus establishing the rule, had also embodied the principle in its domestic jurisprudence.149 The problem that this renvoi to municipal law may be a form of normative projection of one state’s own law in the guise of giving content to an abstract international norm was not discussed. Plaintiffs in Xuncax asserted violations of the norm by a former Guatemalan military official, based on the allegations that his conduct “ ‘had the intent and 145 The Eleventh Circuit Court of Appeals has affirmed a trial judgment which had awarded damages for torture and cruel, inhuman and degrading treatment: see Abebe-Jira v. Negewo, supra n. 45, at 846–8; The Xuncax case is discussed below. The norm is identified as a jus cogens norm in the section of the Restatement (Third) which is excerpted in n. 134, supra. Many cases acknowledge its existence, but only in passing and without detailed discussion. 146 Xuncax, supra n. 45, at 186. 147 Forti I, supra n. 33, at 1543. The Forti II court declined to find the norm actionable, despite the fact that it reversed Forti I’s decision with respect to the question of the non-justiciability of the norm prohibiting disappearances. 148 Xuncax, supra n. 45, at 187. The Senate Reservation had cited the Fifth, Eighth, and Fourteenth Amendments specifically: ibid. 149 Ibid.

The Experience of United States Courts 99 the effect of grossly humiliating and debasing the plaintiffs, forcing them to act against their will and conscience, inciting fear and anguish, breaking physical or moral resistance, and/or forcing them to leave their homes and country and flee into exile[.]’ ”150 The court then suggested that this “general complaint” could be divided into two categories of actions: “The first category includes acts by soldiers under defendant’s command that caused a plaintiff to: (1) witness the torture . . . or severe mistreatment . . . of an immediate relative; (2) watch soldiers ransack their home and threaten their family; (3) be bombed from the air; or (4) have a grenade thrown at them. . . . The second category consists of the claim that, as a consequence of Gramajo’s acts, plaintiffs ‘were placed in great fear for their lives . . . and were forced to leave their homes and country and flee into exile’.”151

Actions in the first category were held to be actionable under the norm, while actions in the second were not.152 The distinction seems to be based on the directness of the link between the impugned acts and their alleged consequences. While it may be true, as suggested in Forti I, that this norm poses more complex problems of definition than the other norms, the kinds of acts accepted in Xuncax as instantiating the norm seem to provide illustration of the fact that such a “catch-all” category of human rights abuses is necessary. (f) Systemic Racial Discrimination While systemic racial discrimination is identified as a jus cogens norm by the Restatement (Third), it is not discussed in any detail in the cases, and does not seem to have been the subject of a (reported) ATCA claim. The question of racism and human rights is discussed briefly and more indirectly in Xuncax, however.153 (2) Norms not requiring “state action” to be actionable (a) War Crimes and Genocide The Kadic court, in 1995, held that a private individual may be the subject of international law prohibitions without it being necessary to allege that such individual acted under “colour of law”, or in concert with a state. The Kadic court started from the existence of certain international criminal law offences 150

Xuncax, supra n. 45, quoting the Xuncax Complaint at p. 76. Ibid. 152 The court made the following reference to one case decided under the European Convention on Human Rights: The Greek Case, Y.B. Eur. Conv. on H.R. 186, 461–5 (1969) (describing cases where political detainees were subjected to acts of intimidation, humiliation, threats of reprisal against relatives, presence at torture of another, and interference with family life in violation of article 3 of the European Convention on the Protection of Human Rights and Fundamental Freedom). 153 Xuncax, supra n. 45, at 189. 151

100 M Swan according to which, it was said, the courts of any nation may punish individuals “without regard to territoriality or the nationality of the offenders”.154 This is possible for offences which are identified as of “universal concern” in §§402 to 404 of the Restatement (Third).155 This latter category of offences includes some which have traditionally been actionable against non-state actors, such as piracy, the slave trade and aircraft hijacking.156 The court asserted that international law permits the establishment of civil law remedies by states for these offences, such as the kinds of actions which are available under the ATCA.157 The Kadic court then went on to establish that there were criminal law norms prohibiting “genocide” and “war crimes”, which, according to the international instruments that created the prohibitions, are also punishable (and therefore actionable, according to the reasoning above) when the perpetrators are private individuals. Both of the norms have been given detailed definitions in the instruments which created them.158 However, a brief look at the definitions of the 154 155

Kadic, supra n. 6, at 241–4. Section 404 provides:

“A state has jurisdiction to define and prescribe punishment for certain offences recognised by the community of nations as of universal concern, such as piracy, slave trade, attacks on or hijacking of aircraft, genocide, war crimes, and perhaps certain acts of terrorism, even where [no other basis of jurisdiction] is present” (as cited in Kadic, supra n. 6 at 240). 156

Kadic, supra n. 6 at 239–41. Ibid. at 240. 158 Ibid. at at 241–2, citing the Convention on the Prevention and Punishment of the Crime of Genocide, art. II., 78 U.N.T.S. 277 and noting its ratification by more than 120 nations. The Genocide Convention defines “genocide” to mean: 157

“any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such: (a) Killing members of the group; (b) Causing serious bodily or mental harm to members of the group; (c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) Imposing measures intended to prevent births with the group; (e) Forcibly transferring children of the group to another group.” The Convention also states that: “persons committing genocide . . . shall be punished, whether they are constitutionally responsible rulers, public officials or private individuals.” Ibid., art. IV. As to war crimes, Kadic cites the four Geneva Conventions (noting their ratification by over 180 nations) which prohibit war crimes. (These are “Geneva Conventions I, II, III, IV”, respectively): The Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field, 75 U.N.T.S. 31; The Convention for the Amelioration of the Condition of the Wounded, Sick, and Shipwrecked Members of Armed Forces at Sea, 75 U.N.T.S. 85; The Convention Relative to the Treatment of Prisoners of War, 75 U.N.T.S. 135; The Convention Relative to the Protection of Civilian Persons in Time of War, 75 U.N.T.S. 287. The court then cites common article 3, which it said is “substantially identical” in each of the four Conventions, and which applies to “armed conflicts not of an international character” and binds “each Party to the conflict . . . to apply, as a minimum, the following provisions”: “Persons taking no active part in the hostilities . . . shall in all circumstances be treated humanely, without any adverse distinction founded on race, colour, religion or faith, sex, birth or wealth, or any other similar criteria. To this end, the following acts are and shall remain prohibited at any time and in any place whatsoever with respect to the above-mentioned persons:

The Experience of United States Courts 101 norms cited in Kadic as explicit as they are, reveals that various terms used in the definitions will need clarification of the sort discussed above with respect to the prohibition against “cruel, inhuman or degrading treatment or punishment”. Like the latter norm, “genocide” and “war crimes” are “umbrella” categories which group various (tortious or criminal) acts together under the norms according to common linking elements. Of great importance here is that the acts for which individuals may be held accountable can include those listed in the previous section as requiring some state nexus, if they can be fitted within the terms of the norm. Summary execution and torture (that is to say “unofficial torture”, as it were) are prohibited by both norms.159 In the case of genocide, the factor which links these acts is that they have been committed “with [specific] intent to destroy, in whole or in part, a national, ethnical, racial, or religious group . . .”. The linking factor with respect to “war crimes”, obviously, is that the actions must have been committed within the context of “hostilities” of a certain character.160 At least one case has followed Kadic, and two other trial courts appear to have indicated that they are sympathetic to the line of reasoning presented in the case.161 The Supreme Court denied an application for certiorari in the case.162

(a) violence to life and person, in particular murder of all kinds, mutilation, cruel treatment and torture; (b) taking of hostages; (c) outrages upon personal dignity, in particular humiliating and degrading treatment; (d) the passing of sentences and carrying out of executions without previous judgment pronounced by a regularly constituted court . . .” Geneva Conventions I–IV, common art. 3(1), as cited in Kadic, supra n. 6. Finally, the Kadic court also noted that the plaintiff was alleging that the defendant Radovan Karadzic was bound by the Protocol Additional to the Geneva Conventions of 12 August 1949, Relating to the Protection of Victims of Non-International Armed Conflicts, 16 I.L.M. 1442 (1977) (“Protocol II”), which has been signed but not ratified by the United States: see International Committee of the Red Cross: Status of Four Geneva Conventions and Additional Protocols I and II, 30 I.L.M. 397 (1991). It was said by the Kadic court that Protocol II supplements the common article 3 for armed conflicts that “take place in the territory of a High Contracting Party between its armed forces and dissident armed forces or other organised armed groups which, under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol.” Ibid., art 1, as cited in Kadic, supra n. 6, at 243. 159

Kadic, ibid. at 243. Kadic, supra n. 6. I will only note here that there are important distinctions drawn between various kinds of conflicts. 161 Kadic was followed in Beanal, supra n. 4, (D.C. E. La (5th Cir.) ). The court in Unocal (supra, n. 78) referred to the analysis in Kadic by suggesting that it “provides a reasoned analysis of the scope of a private individual’s liability for violations of international law”. The Unocal court also noted that Judge Edwards, in his concurring opinion in Tel-Oren (supra n. 7, at 794), had also acknowledged this possibility. In Xuncax, supra n. 45, the court went so far as to suggest that a claim for genocide seemed supportable on the facts, without, however, discussing the possibility of liability for private individuals and while noting that the plaintiffs had chosen not to bring such a claim. 162 Certiorari denied 17 June 1996, reported at: 1996 U.S. LEXIS 3924. 160

102 M Swan (b) Other Claims Not Requiring State Action Of the other claims said not to require state action mentioned in Kadic, and in the previous section (piracy, the slave trade, aircraft hijacking, and possibly terrorism), I will only here remark upon the possibility of the existence of a prohibition against the slave trade.163 Several cases have included the prohibition as a recognized jus cogens norm, but in “litany” fashion without discussion of this norm specifically.164 The Restatement (Third) §702, as mentioned above, lists a prohibition against “slavery or slave trading” as a jus cogens norm.165 It also lists conventions which prohibit these practices, but offers no definition of them as it does do for “genocide”.166 Only the Unocal case seems to have discussed the norm specifically, in connection with allegations of, inter alia, forced labour against the government of Burma and against two oil companies, the American Unocal and the French Total.167 The case dealt with the status of the slaving norm only in terms of its slave-trading aspect, however, and did not need to decide the issue raised (on the motion to dismiss), as the court was otherwise able to found jurisdiction in the matter. The distinction between “slavery” and the “slave trade”, if the claim is maintained, will be crucial in Unocal, as the court noted there that there were no allegations of the sale of persons in the case. As to slavery, the case was not entirely silent. The defendant suggested that the forced labour alleged in the case amounted to something more akin to a civil service requirement, and the court did suggest the possibility that, at the merit stage, forced labour of the sort alleged might qualify as slavery.168 It might be fair to speculate that this norm will be sustained in the future, as it seems to involve less complex problems of definition, or “slippery slope” problems, than, for example, the norm prohibiting “cruel, inhuman, or degrading treatment”. It is also an international prohibition of older historical origin.169 These factors make it possible that the courts might be more appropriately “guided” by definitions available in municipal law, as was done in Xuncax with respect to “cruel, inhuman, or degrading treatment” (assuming there is, indeed, no definition available in international law). 163 I say “possibly” a prohibition of “terrorism” because of what appears to be a lack of consensus that such a catch-all category exists in international law as opposed to specific forms of terrorism on which states have reached agreement, such as hijacking. To the extent more and more reference is made to “terrorism” as a general category in official interstate discourse, such as some Security Council resolutions, and were “terrorist” acts to begin to be the subject of ATCA suits, the question of definitional concreteness will likely present an even more vexing challenge than for the open-textured norms discussed earlier. 164 United States v. Matta-Ballesteros, 71 F.3d 754, 763 n.5 (9th Cir. 1996), as amended, 98 F.3d 1100 (1996) (citing Siderman de Blake, supra n. 25, at 717). See also Kadic, supra n. 6, at 240. 165 See n. 134, supra; see also § 404 at n. 155, supra. 166 See Comment g. 167 The case against Total was dismissed and severed from the claim against Unocal on the basis of there being an insufficient connection between Total and business activities of related businesses in California: 168 Unocal, supra n. 78. 169 See House Report, supra n. 137.

The Experience of United States Courts 103 The Choice of Law Rule Concerning Questions of Standing and Statutes of Limitations in IHRT Actions While it has been held that the courts may apply international law rules of decision to the issues raised in the previous sections, it seems that these sources do not supply sufficient guidance in the areas of standing and limitations periods. The Restatement (Second) of Conflict of Laws approach to traditional conflicts problems is that of “interest analysis”, or the “most significant relationship”.170 The test must balance such factors as the situs of the injury, and the domicile of the parties, rather than simply applying the lex loci rule.171 The choice of law rule with respect to the issues of standing and limitations periods, however, is that: “If a federal statute provides a cause of action without specifying important details such as the limitation period or survivorship, courts look to analogous state statutes.”172 This choice of law rule does not seem to admit the possibility of selecting the law of the place of the occurrence of the tort.173 The federal courts have recently stretched this rule, so that the inquiry now seeks to identify the closest analogues found in both federal and state law.174 This latter “move” means that the courts can now easily identify the TVPA as the closest analogy to actions under the ATCA, and apply its 10-year limitations period. Nonetheless, the point that state law may be expected to put forward rules of adjudication that are not appropriate to IHRT cases is readily made, as can be seen, for example, by the analysis of the trial judge in the Arone case in which the Canadian judge did not even allude to the underlying international human rights dimension as being relevant to deciding whether a variety of limitation periods invoked by the defendant state, Canada, applied, or whether a separate non-statutory development of the relevant limitations principles was more appropriate.175 170 American Law Institute, 1 Restatement (Second) of Conflict of Laws §145 (1971), as cited in Xuncax, supra n. 45, at 195. 171 “1980 Survey of International Law in the Second Circuit”, (1980) 8 Syracuse Journal of International Law and Commerce 206 at 208. 172 Forti I, supra n. 33 at 1547. See, e.g., Wilson v. Garcia, 471 U.S. 261, 266–7, 85 L. Ed. 2d 254, 105 S. Ct. 1938 (1985). 173 The possibility exists that this might very well suggest that these are procedural matters to which, therefore, the law of the forum applies. 174 The district courts seem to be stretching the Supreme Court decision which, arguably, gives them authority to do this: see Agency Holding Corp. v. Malley-Duff & Associates, 483 U.S. 143, (1987), discussed in Forti I, supra n. 33,at 1547. 175 This case involved a lawsuit against the family of a young Somalian man, Shidane Arone, who was tortured to death in Somalia by Canadian peacekeeping troops operating under UN auspices. Although torture was mentioned in the style of cause in the claim brought before an Ontario court, it is not unlikely that the substantive pleadings by the plaintiffs did not adequately take the judge through an argument in which international human rights law is significant. Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. Canada (Attorney General) (Unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J.). For commentary on this case, see C. Scott, “Introduction to Torture as Tort: From Sudan to Canada to Somalia,” chapter 1 in this volume; and J Terry, “Taking Filártiga on the Road: Why Courts

104 M Swan To the extent local rules allow for judicial discretion to waive or extend limitation periods, the specific needs of claims grounded in IHR principles can be more easily accommodated. For example, in the Marcos cases, the defendant argued for the application of either the law of the Philippines or the law of the forum state (Hawaii), which provided for one- and two-year limitations periods, respectively.176 Faced with the choice between two very short limitation periods, the court noted that Ferdinand Marcos had engineered the passage of a constitutional amendment granting him and others immunity from suit during his tenure in office.177 In addition, it was thought that many IHR abuses had gone unreported while he was in power, for fear of reprisal.178 In that context, the court held that the actions were “equitably tolled” until he left office (in February of 1986).179 It seems unreasonable to expect that, in circumstances typical of IHR abuse situations, ordinary citizens can manoeuvre themselves into a situation to bring a suit, by leaving their home country (and locating their nemesis), for example, in accordance with the dictates of a particular statute of limitations. Time seems generally to be irrelevant to the prosecutions of war criminals, except as to one factor relevant in assessing whether a trial can be fair in view of the fading of memories of key witnesses. One would think that this is not only because a limitations period is inconvenient, but also because of the particularly heinous crimes involved.180 Even the TVPA ten-year period may prove harsh in the next case brought forward, or the one after that.181 The above-noted doctrine of “equitable tolling”, however, may relieve against harsh effects that this doctrine might lead to. Problems relating to the issue of a plaintiff having standing to bring an action have arisen in two forms. The first concerns the question of who has standing to bring an action on behalf of a deceased person. This must be the case, of course, in any IHRT claim for summary execution or extrajudicial killing, genocide, or Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad,” chapter 4 in this volume. 176 Hilao, supra n. 4, at 773. 177 Ibid. 178 Ibid. 179 Ibid. The actions had been launched in March of 1986. Equitable tolling “stops the clock” on the running of the statute of limitations. This can be done where the defendant’s conduct has led to the delay in bringing suit, for periods in which the defendant is absent from the jurisdiction or immune from lawsuits, and for periods in which the plaintiff is imprisoned or incapacitated—or otherwise where extraordinary circumstances have intervened. This is a federal common law doctrine which applies even if the state law provides the limitations period. 180 Prior to the enactment of the TVPA, the courts often analogized to 42 U.S.C. §1983, according to which plaintiffs must allege both deprivation of a federally protected right and action “under color of” state law. This federal civil rights statute assisted the court with respect to other problematic features of state law (discussed below), but mandated that the state statute of limitations be applied. Thus, state law was unavoidable when drawing on §1983—except by “equitable tolling:” see Forti I, supra n. 33. 181 Here, it is of interest to note the Arone claim was lodged only a day before the expiry of six years from the date of the torture. It could easily have been much later, in view of what it takes for a Somalian family to find a way to get a Canadian lawsuit going. Arone, supra n.174

The Experience of United States Courts 105 disappearance. Again the passage of the TVPA has been helpful. It directs that the victim of extrajudicial killing’s “legal representative, or any person who may be a claimant in a wrongful death” may bring an action.182 While the House and Senate reports on the passage of the TVPA had generally directed that state law should be looked to in order to determine who falls within this phrase, the Senate Committee Report had suggested in a footnote that: “[W]here application of Anglo-American law would result in no remedy whatsoever for an extrajudicial killing, however, application of foreign law recognising a claim by a more distant relation in a wrongful death action is appropriate.”183

There is more than a hint of a very interesting, human-rights-sensitive choice of law principle here: apply the most favourable law. This approach has been acted on. Massachusetts law on wrongful death suits was the applicable state law in Xuncax and would have denied recovery, inter alia, to a sibling if the deceased had left parents or issue.184 The court applied Guatemalan law, which was the law of the situs of the tort, to allow recovery, in light of the Senate report.185 In Hilao, the defendant argued that, under federal common law, an action for an intentional tort abates upon the death of either party.186 Instead of accepting the argument, the court analogised to actions under §1983 for cruel and unusual punishment, which do not abate upon either party’s death.187 The second kind of standing question relates to who may bring actions on behalf of victims of torts which do not (necessarily) cause the death of the victim, such as for “official torture”. This is especially relevant when an individual seeks standing on behalf of another individual (or class of individuals) who has in fact died. In Xuncax, both state law and Guatemalan law denied the possibility of third party recovery (by a parent or spouse) for a claim of arbitrary detention or torture. The court noted that the TVPA is “conspicuously silent on the matter”, in contrast to its language concerning extrajudicial killings, and therefore denied the possibility of recovery. The analogy to §1983 now seems to have been foreclosed by the passage of the more closely analogous TVPA, even assuming that it would have led to a different result here. This shows that available sources of municipal law are not perfectly manipulable by the courts where it might be felt that their principles are inappropriate. One can imagine scenarios where the successful recovery (or prospect of recovery) of money damages, in a suit by a family member, might provide the only substantial hope for the prisoner’s survival, for example. The unique problems raised by issues of “associational” standing, as they apply to IHRT claims, might present clearer examples of the inapplicability of municipal law applications to the cases. In Unocal, the court denied standing to a Burmese 182

TVPA, § 2(a)(2). Xuncax v. Gramajo, supra n. 45, at 191. Ibid. 185 Ibid. 186 Hilao, supra n. 4, citing Heikkila v. Barber, 308 F.2d 558, 560 (9th Cir. 1962). This does seem to support the inference that the issue of standing is a matter of substantive law, federally. 187 Ibid. 183 184

106 M Swan trade union organisation for claims on behalf of its members. The court said that it was bound by a Supreme Court decision which asserted that the potential “associational” plaintiff may represent persons where there is a need for them to appear individually in the action.188 The Supreme Court had also ruled that the condition could be waived by Congress, as it had done in labour legislation, but the Unocal court proceeded on the assumption that this could not be said to have been done.189 All of this is to say that it can be the case that municipal law may provide excellent “supplementary” rules of decision if thoughtfully applied—or it may put up roadblocks if applied unimaginatively or simply without regard to the principles at stake. The possibility still exists that the courts might assert their ability to develop “specialised” federal common law principles to overcome such difficulties. The Choice Of Law Rule Concerning Remedies in IHRT Actions Even when the courts were applying “wrongful death” statutes and other municipal sources of law to the merits of cases, the courts had sought to award damages according to their sense of the gravity of the underlying facts.190 The courts seem to feel free to award damages as they see fit, in the name of the fundamental principles underlying the IHRT cases, as a function of their power to develop common law remedies in order to give effect to the purposes of a statute and in the absence of statutory directives.191 Furthermore, they are free to award 188

Unocal, supra n. 78, at 343. Ibid. 190 See the section on “Rationales for the Application of International Law”, supra. In the Filártiga trial court decision, which took place after the “cause of action” question had been raised and resolved, the court had considered both Paraguayan municipal law and international law—but rejected their negative implications as to the issue of punitive damages and awarded them anyway. In the 9th Circuit decisions decided before that circuit’s approval of the 2nd Circuit’s interpretation of the “cause of action” question, the courts applied municipal law on these questions. 191 The court in Xuncax, supra n. 45, included the following catalogue of damage awards made under § 1350: Filartiga v. Peña-Irala, 577 F. Supp. 860 (E.D.N.Y. 1984) (for torture to death: $175,000 to sister, $200,000 to father in compensation, $5,000,000 to each as punitive damages); Martinez-Baca v. Suarez-Mason, No. 87-2057 SC (N.D. Cal., Apr. 22, 1988) (for systematic arbitrary detention and torture: $11,170,699 in compensation (including lost earnings), $10,000,000 in punitive to victim); Forti v. Suarez, No. 87-2058-DLJ (N.D. Cal. Apr. 25, 1990) (for first plaintiff, for arbitrary detention, torture, and witnessed abuse and execution of brother: $3,000,000 compensatory, $3,000,000 in punitive; for second plaintiff, for arbitrary detention, abuse and “disappeared” mother: $2,000,000 in compensation, $1,000,000 punitive); Trajano v. Marcos, No. 86-0207, (D. Hawaii, May 19, 1991) (for torture and summary execution: $236,000 in lost earnings, $175,000 moral damages, $1,250,000 exemplary damages to victim’s estate; $1,250,000 in compensation, $1,250,000 exemplary to victim’s mother); Siderman de Blake v. The Republic of Argentina, No. CV-82-1772RMT (MCx) (C.D. Cal. Sep. 28, 1984) (for torture: compensatory damages totalling $2,607,575.63 to victim), vacated on other grounds, No. CV-82-1772-RMT (MCx) (C.D. Cal. Mar. 7 1985), rev’d and remanded, 965 F.2d 699 (9th Cir. 1992); Quiros de Rapaport, et al. v. Suarez-Mason, No. C872266 JPV (N.D. Cal. Apr. 11, 1989) (for torture and murder of one victim, disappearance of another: $10,000,000 in compensation, $10,000,000 punitive to victims’ widows, $5,000,000 in compensation, $5,000,000 punitive to victims’ mother and sister, respectively). See also Todd v. Panjaitan, No. CV92-12255-PBS (D. Mass. Oct. 26, 1994) (awarding $2,000,000 in compensation to mother as administratrix of son’s estate, $2,000,000 in compensation to mother, and $10,000,000 in punitive 189

The Experience of United States Courts 107 punitive damages where recourse to municipal law might forestall this. The courts do on occasion discuss US Constitutional due process imperatives in this regard, however.192 The Ninth Circuit has also used statistical sampling methods to assist in the determination of the nearly ten thousand claims in the class action suit against Ferdinand Marcos.193 I conclude this sub-section on remedies, and the chapter as a whole, by noting that the provision of injunctive relief has by and large not arisen. It will be interesting to see whether such relief could become part of the next generation of ATCA claims where the conduct that is subject of the suit is ongoing, as in Unocal. This may turn out to be one area for some comparative law reciprocity. US lawyers, and consequently US courts, may well find instructive recent proceedings before an Australian court wherein a British national sought an injunction to prevent an Australian national from exercising a power granted the Australian by Saudi Arabian law, which the British national alleged would, if exercised, violate her human rights.194

damages); Paul v. Avril, No. 91-399-CIV (S.D. Fla. July 1, 1994) (awarding six victims of torture and arbitrary detention between $2,500,000 and $3,500,000 in compensatory damages each together with $4,000,000 each in punitive damages). It is also worth noting that a look at the citations given underscores the fact that, as this part of the trial often goes unreported, it can be difficult to get a sense of the activity on the damages front. 192 Xuncax, supra n. 45, and Hilao, supra n. 4. These have discussed the appropriateness of awarding punitive damages in civil suits, for purposes of deterrence, without the procedural safeguards available to the criminally accused. See also Filártiga (trial decision), supra n. 66. 193 See Hilao, supra n. 4. 194 See the chapter devoted to these proceedings: B Wells and M Burnett, “Cultural Challenges: Injunctions in Australian Courts and the Right to Demand the Death Penalty under Saudi Arabian Law,” chapter 23 in this volume.

4

Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad JOHN TERRY 1

1 INTRODUCTION N R E C E N T Y E A R S , the international human rights community’s focus has shifted from issues of norm creation to problems of human rights enforcement and the remedy and punishment of human rights violations. Given the limited mechanisms available at the intergovernmental level, human rights advocates have placed increased emphasis on what might be called “third country legal actions”—nationally-based remedies that attempt to hold human rights violators individually responsible for serious human rights violations committed abroad. Third country legal actions can involve criminal law or civil law. The criminal route is through the doctrine of universal jurisdiction, which enables a state to prosecute individuals for certain very serious offences even if the state has little or no connection to the offender, the victim or the offence. The civil remedy allows the victims or their dependants to bring a tort action against persons who have committed human rights violations abroad. In the case of torture, the criminal remedy has been implemented by means of the 1984 Convention against Torture (the Convention),2 now ratified by one hundred and eighteen states.3 The Convention requires, among other things,

I

1

Of Torys, Toronto, Canada. Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027 [hereinafter “CAT”]. 3 University of Minnesota Human Rights Law Library, http://www1.umn.edu/humanrts/, 29 March 2000. 2

110 J Terry that state parties ensure that all acts of torture are offences under its criminal law,4 extradite or prosecute alleged torturers found within their territory no matter where the alleged torture has occurred,5 and take the measures necessary to ensure that they have jurisdiction to do so.6 In accordance with these provisions, state parties have enacted laws that give their courts criminal jurisdiction over torture committed abroad. The legitimacy of these provisions has been confirmed by the House of Lords in the Pinochet case.7 The tort remedy, in contrast, has no such international law pedigree. The litigation engendered by the Torture Victim Protection Act (TVPA) and the Alien Tort Claims Act (ATCA) in the US has arisen from a set of circumstances peculiarly American—an influential US human rights bar, a Congress seeking to improve its own lacklustre record of ratifying international human rights treaties by appointing its courts as arbiters with respect to torture committed abroad, and a US justice system known for its generosity to the plaintiffs. As Lord Denning has said, “[a]s a moth is drawn to the light, so is a litigant drawn to the United States”,8 and that appears to be as true for victims of human rights violations as it is for any other personal injury litigant. Unlike the criminal remedy, there is no treaty that clearly obliges or even authorises courts to take jurisdiction over civil actions respecting torture committed abroad. With the exception of the US, there is no domestic legislation in any other country that expressly grants courts jurisdiction with respect to these matters. In these circumstances, it is easy to marginalise the tort remedy as a sui generis creation of the US legal system. My thesis in this chapter is that the tort remedy should not be marginalised. To the contrary, it should be embraced by courts in other jurisdictions even if the legislative bodies in those jurisdictions do not take steps to enact legislation equivalent to the TVPA. Courts have always taken jurisdiction, in appropriate circumstances, over torts committed in other countries. As the court in Filártiga v. Peña-Irala pointed out, “it is not extraordinary for a court to adjudicate a tort claim arising outside of its territorial jurisdiction.”9 It bears remembering that the well-known 1870 English conflict of laws case of Phillips v. Eyre can be viewed as an early example of a Filártiga-style tort action, in which the plaintiff sued the Governor of Jamaica in England for having arrested and imprisoned him after declaring martial law.10 I make my argument for the legitimacy of the tort remedy in two main parts. First, I explain why, in my view, the civil remedy is in many respects more useful and effective than the universal criminal remedy. Second, I argue that it is 4

CAT, Art. 4(1). CAT, Arts. 6, 7. 6 CAT, Art. 5(2). 7 R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte [1999] 2 All ER 97 [hereinafter the “Pinochet case”]. 8 Smith Kline & French Laboratories Ltd. v. Bloch [1983] 2 All ER 72 at 74. 9 Filártiga v. Peña-Irala, 630 F. 2d 876, 885 (1980). 10 (1869), LR 6 QB 1. 5

Taking Filártiga on the Road 111 legitimate for courts in countries outside the US to take jurisdiction over foreign torture tort claims, in appropriate circumstances, even if the taking of such jurisdiction is not sanctioned by statute. In my view, the key to courts taking jurisdiction over such actions in the absence of express legislative authority lies in the application of traditional principles of forum non conveniens and choice of law. The creation of a new tort of universal jurisdiction over torture is neither necessary nor likely to be seriously considered by the courts in the absence of a specific legislative mandate. I leave it to other contributors in this volume to examine how such a new tort might be established.

2 ROLE OF THE TORT REMEDY

Although the idea of pursuing torturers for damages rather than jailing them may strike some as a second-best solution, the tort remedy is an important complement to, and in many ways a more useful mechanism than, the criminal remedy.

In General The principle that victims of human rights offences should be compensated is widely accepted within domestic and international legal systems. In the US and other jurisdictions, it has become accepted practice for courts to award damages in tort to compensate individuals for violations of their constitutionallyguaranteed rights.11 Even in countries where amnesties have been enacted that grant impunity to human rights offenders, states have in many cases provided compensation to the victims of those human rights violations.12 Rights to fair and adequate compensation are contained in numerous human rights treaties and other UN instruments.13 The Committee against Torture has emphasised 11 See Monroe v. Pape, 365 US 167 (1960), Bivens v. Six Unknown Named Agents, 403 US 388 (1971) and subsequent cases. For a comparative analysis of damages remedies for civil rights violations see P Rishworth and G Huscroft, “Damages for Breach of Individual Rights in the United States of America, Canada, Ireland, the Caribbean, India, Sri Lanka, the European Union and under the European Convention on Human Rights”, an unpublished study prepared for and on file with the New Zealand Law Commission, 1996. 12 See Commission on Human Rights, Study concerning the right to restitution, compensation and rehabilitation for victims of gross violations of human rights and fundamental freedoms, Final report submitted by Mr. Theo van Boven, Special Rapporteur, E/CN.4/Sub.2/1993/8 at 46–48 [hereinafter “1993 Van Boven Report”]. See also the numerous examples of national legislation respecting the compensation and rehabilitation of victims of human rights abuses contained in N J Kritz (ed.), Transnational Justice: How Emerging Democracies Reckon with Former Regimes, Vol. 3 (Washington, D.C., United States Institute of Peace Press, 1995) at 661–818. 13 Examples of these treaty provisions include: art. 9(5) of the International Covenant on Civil and Political Rights and art. 5(5) of the European Convention for the Protection of Human Rights and Fundamental Freedoms (both guarantee an “enforceable right to compensation”); art. 14(1) of the Convention against Torture (“an enforceable right to fair and adequate compensation”); art. 10

112 J Terry the importance of compensation for victims of torture and observed that amnesties that do not provide for compensation are incompatible with the provisions of the Convention.14 International claims commissions have routinely awarded damages where the nationals of one state have been mistreated by the officials of another state.15 International human rights treaty bodies such as the Inter-American Court of Human Rights have developed principles for awarding damages to individuals whose rights have been violated by their own governments.16 That being said, it is important to remember that third country civil actions in general serve symbolic rather than truly compensatory objectives. In most of the actions that have been brought in the US under the ATCA and TVPA, the damages awarded have never been collected.17 Typically the torturer has left the jurisdiction, thereby making the judgment difficult to enforce, or the assets of the torturer potentially available to satisfy a large damages award are either insufficient or difficult to access.18 In truth, therefore, it is perhaps more accurate to describe the civil remedy not so much as a mechanism to fill a gap in “enforcement” under international law but as a means for providing a measure of self-respect, vindication and recognition for the victims of serious violations of international human rights. In the view of the Centre for Constitutional Rights, which has coordinated the initiation of many international human rights civil actions in the US, the typical plaintiff’s “satisfaction comes from the mere filing of a lawsuit, from of the American Convention on Human Rights (“right to be compensated in accordance with the law”); and art. 21(2) of the African Charter on Human and Peoples’ Rights (“right to adequate compensation”). See also the UN’s Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power, GA Res. 40/34, Report No. A/40/881, 29 Nov. 1985. The principles established under these and other treaties as well as proposed basic principles and guidelines respecting compensation to victims of gross violations of international human rights are set out in the 1993 Van Boven Report, supra, n. 12. 14 Annual General Assembly Report of the Committee against Torture, 3rd Sess. (Nov. 1989) A/45/44 (GAOR, 45th sess.), Suppl. No. 44, annex VI. 15 See, for example the Roberts Claim (U.S. v. Mexico) (1926) US–Mexican General Claims Commission, 4 RIAA 77 and the Quintanilla Claim (Mexico v. US) (1926) US–Mexican General Claims Commission, 4 RIAA 101. See also G T Yates III, “State Responsibility for Nonwealth Injuries to Aliens in the Postwar Era” in R B Lillich (ed.), International Law of State Responsibility (Charlottesville, University Press of Virginia, 1983) 213. 16 See Inter-American Court of Human Rights, Velasquez Rodriguez Case, Compensatory Damages, Judgment of 21 July 1989, 11 Hum. Rts. L.J. (1989) 127. 17 See B Stephens and M Ratner, International Human Rights Litigation in U.S. Courts (Irvington-on-Hudson, New York, Transnational Publishers Inc., 1996); R B Lillich, “Damages for Gross Violations of International Human Rights Awarded by US Courts”, 15 Human Rts. Q. (1993) 207 at 207–8. 18 See, for example, Todd v. Panjaitan, No. 92-12255, 1994 WL 827111 (D. Mass. 26 Oct. 1994), an action brought by the mother of a New Zealand student killed during the Dili massacre against the Indonesian military official responsible for the massacre. Todd obtained a default judgment of US $14 million, but Panjaitan returned to Indonesia and the judgment has not been enforced. Another example is Paul v. Avril, 901 F. Supp. 330 (S.D. Fl.a 1994), in which six leaders of the Haitian opposition who had been tortured sued Prosper Avril and were awarded damages in the amount of US $41 million, none of which has been collected. For other examples see generally Stephens and Ratner, supra n. 17.

Taking Filártiga on the Road 113 confronting the defendant in court or forcing him to flee from the U.S., and from obtaining judgment from a U.S. court which makes a formal record of the human rights violations and of the defendant’s responsibility”.19 A civil action and judgment against an individual human rights offender vindicates the personhood of the individual victims and the “peoplehood” of the victimized community. Such trials can serve the same purpose as war crimes trials, which, as Edward Morgan has said, “in their guise as repetitive and formalised dramatic histories, . . . not only educate [us] with respect to history and the extremes of human nature, but vindicate the freedom that lies in our various conceptions of ourselves”.20 Within the dramatic and personalised context of a trial, the story of the torturer and his victim carries a symbolic resonance greater than multiple non-governmental reports or intergovernmental resolutions. Unlike war crimes trials, in which there is a strong emphasis on the impartial presentation of facts, civil actions provide broad scope for the victims to “tell their story” and have it received with empathy and respect.21 Ideally, this kind of cathartic ritual should be provided by the state in which the torture occurred, by means of a truth commission, a trial or some other public event. But in cases where that has not and perhaps will not occur, a tort action in a third country provides a surrogate forum and means by which victims can bring publicity and political pressure to bear against the government of the state in which the events occurred. The multiple roles of third country tort actions as instruments of justice, catharsis and political activism are well illustrated in this excerpt from the declaration of Helen Todd filed in Todd v. Panjaitan, a TVPA and ATCA claim against an Indonesian military official responsible for the death of Ms. Todd’s son, a New Zealand national, during the Dili massacre: “22. I bring this case because those who killed him and those in power who set the policies that killed him have not even acknowledged that a crime has been committed. They lead privileged lives. The policy of repression continues. The military culture that systematically tramples on human rights still flourishes. 24. I bring this case not only as Kamal’s mother but on behalf of the hundreds of East Timor mothers who are forced to grieve in silence for their dead children. Our grief and anger is the same, but, unlike them, I can bring a case against a military officer without putting the rest of my family in danger. Whatever compensation is awarded by the court in this case will belong to the mothers of all the victims of the Dili massacre, and I will find a way to get it into their hands. 26. There must be some accounting for the unarmed young people shot to death by the military in Dili that morning simply because they dared to raise their voices 19 M Ratner and B Stephens, “The Centre for Constitutional Rights: Using Law and the Filártiga Principle in the Fight for Human Rights”, unpublished paper, Centre for Constitutional Rights, 1993. 20 E Morgan, “Retributory Theater” (1988) 3 Am. U.J. Int’l L. & Pol’y 1 at 4. 21 See J Llewellyn and R Howse, “Institutions for Restorative Justice: The South African Truth and Reconciliation Commission” (1999) 49 UTLJ 355 at 364; J Alvarez, “Rush to Closure: Lessons of the Tadic Judgment” (1998) 96 Mich. L.J. 2031; Stephens and Ratner, supra n. 19 at 233–8.

114 J Terry against sixteen years of organized military brutality against the people of East Timor.”22

Finally, the decisions that emerge from third party actions promote and develop the international law of human rights. Writing thirty years ago, Richard Falk made the following empirical claim: “It has been generally true that domestic courts have been far more important than international courts in developing and upholding international law.”23 More recently, Canada’s former Chief Justice Antonio Lamer has described the kind of institutional dialogue between domestic courts and international human rights bodies that can deepen this reality to the extent the reality exists and help bring it about where it as yet does not: “[B]y looking to international treaties and the jurisprudence of international human rights bodies in the interpretation of domestic human rights norms . . . judges raise the profile of those international treaties and further the creation of a human rights culture.”24

Indeed, in the field of international human rights, decisions such as Filártiga, Eichmann and now Pinochet serve as crucial precedents for the universal power to sue, prosecute and extradite certain kinds of human rights violators wherever they committed their acts and wherever they may be found. Inspired by the Pinochet case, Chadian plaintiffs, assisted by Chadian and international human rights NGOs, have recently persuaded a Senegalese judge to indict Chad’s former dictator, Hissein Habre, on charges related to torture and disappearances that occurred while Habre was in power in Chad from 1982 to 1990.25 Human Rights Watch has identified Uganda’s Idi Amin, currently living in Saudi Arabia, Haiti’s Raul Cedras in Panama, Paraguay’s Alfredo Stroessner in Brazil and Haiti death squad leader Emmanuel Constant in the United States as individuals who should be indicted in the countries in which they reside.26 As discussed above, these national proceedings appear to fulfil a “morality play” dimension that international legal processes, including the Hague trials respecting former Yugoslavia and Rwanda, seem unable to match. With time, one might expect international bodies to invoke and endorse these cases—and the general role they represent for international human rights protection—and see, in turn, domestic courts take their cue from international bodies’ embracing of the idea of domestic courts as agents of the international legal order.27 22

Declaration of Helen Todd in Todd v. Panjaitan, supra n. 18. R Falk, “Nuremberg: Past, Present and Future”, (1971) 80 Yale L.J. 1501 (emphasis in original). See also R Falk, The Role of Domestic Courts in the International Legal Order (1964) and R B Lillich, “The Proper Role of Domestic Courts in the International Legal Order”, (1970) 11 Va. J. Int’l L. 9. 24 Rt. Hon. Antonio Lamer, “Enforcing International Human Rights Law: The Treaty System in the 21st Century”, address at York University, 22 June 1997 at 6. 25 For reports on the status of these proceedings, see http://www.humanrightswatch.org. 26 Ibid. 27 On the idea of domestic courts as agents, see generally K Randall, Federal Courts and the International Human Rights Paradigm (Durham and London: Duke University Press, 1990). 23

Taking Filártiga on the Road 115 Advantages Over Criminal Law The tort remedy also offers significant advantages over criminal law prosecution with respect to the initiation and subsequent carriage of a claim against a human rights offender. Although many states now have the express authority to prosecute crimes of torture, war crimes and crimes of humanity committed abroad, very few states have done so. The universal criminal remedy has been applied most notably by Israel to prosecute Adolf Eichmann in 1961 and John Demjanjuk in 1993, but jurisdiction in both instances can also be justified on the basis of the protective and passive personality principles.28 Canada and Australia are among the few states to have relied exclusively on universal criminal jurisdiction to prosecute persons alleged to have committed war crimes during World War II, but none of these prosecutions has resulted in a conviction. The Pinochet case establishes an important modern precedent for the use of universal jurisdiction, although in that case the Spanish government also relied upon the fact that certain of Pinochet’s victims were nationals of Spain to assert jurisdiction to prosecute him, and especially in order to increase the likelihood of securing extradition from the UK. In contrast, the tort remedy in the US has been used by plaintiffs to bring more than twenty actions in respect of human rights violations that have occurred in more than a dozen different countries.29 In some respects, states are better placed than individual victims to bring legal proceedings against persons who have committed human rights violations abroad. They have access to far greater financial and institutional resources, as well as information through their diplomatic, immigration, law enforcement and intelligence-gathering agencies. But states have little incentive to investigate and prosecute another state’s human rights offences. Human rights violations such as torture are examples of “system criminality”—crimes committed on a large scale and encouraged or at least tolerated by government authorities. As Antonio Cassese has explained: “By and large, [within a state] repression of ‘individual criminality’ is a more frequent occurrence than that of ‘system criminality’, for the simple reason that the latter involves an appraisal and condemnation of a whole system of government, of misbehaviour involving the highest authorities of a country.”30

Similarly, a foreign state is highly unlikely to undertake a prosecution of system criminality in another state where doing so will be understood as an impugning of an entire state and even society. In this respect, it is telling that the Spanish 28 On these principles of jurisdiction, see A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality”, chapter 6 of this volume. 29 These countries include Paraguay, Libya, Argentina, Guatemala, the Philippines, Ethiopia, Haiti, Indonesia, Bosnia, Rwanda, Mexico and the United States. For a list and summary of key ATCA and TVPA cases as of 1996, see Stephens and Ratner, supra n. 17 at 239–44. 30 A Cassese, International Law in a Divided World (Oxford, Clarendon Press, New York, Oxford University Press, 1986) at 275–6.

116 J Terry magistrate, Baltasar Garzon, who initiated the Pinochet process did so against the strong resistance of the government of Spain and even the Spanish prosecutorial authorities. In fact, it can be argued that the notion that universal jurisdiction can be relied upon to enforce human rights violations like torture is based on a flawed assumption that states will have any interest or incentive to conduct such prosecutions. As Jack Donnelly has observed, human rights regimes are premised on a weak structure of moral rather than material interdependence. In such a normative environment, human rights violations occurring abroad seldom have an impact in another state that is direct enough to justify retaliation or any other serious action.31 Another important concern for states is the “shoe on the other foot” syndrome. States may calculate reasonably that, if they bring a complaint against another state, that state will in the future use this as a precedent for criticizing them. According to Martti Koskenniemi, “[s]tates simply do not, in any manner which could be termed ‘customary’, take up violations of human rights in other countries. . . . On the other hand, they do make political protests every now and then, and these are made selectively, based on such factors as political alliances and controversies, inclinations of domestic audiences, and the like.”32

The interstate complaint mechanisms included in the International Covenant on Civil and Political Rights (ICCPR),33 the American Convention on Human Rights,34 the Convention against Torture35 and the Convention on the Elimination of All Forms of Racial Discrimination36 have, to this author’s knowledge, never been invoked.37 Within the UN Human Rights Commission, although special country rapporteurs have been appointed more frequently under the Resolution 1235 procedure, this process is still generally reserved for states that are widely viewed as pariahs and that also have few strong allies. If the past several decades in the development of international human rights law demonstrate anything, it is that individuals, acting through NGOs, and not states that are the agents of change and the promoters of the rule of international law. The influence of these factors is heightened where one state is considering an investigation of alleged human rights abuses conducted by an official in another 31

J Donnelly, International Human Rights (Boulder, Colo., Westview, 1993) at 137–8. M Koskenniemi, “The Pull of the Mainstream,” (1990) 88 Mich. L. R. 1946 at 1959–60. 33 Art. 41, International Covenant on Civil and Political Righs, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1948), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976). 34 Art. 45, American Convention on Human Rights, 1969, 1144 U.N.T.S. 123; reprinted in 9 I.L.M. 673. 35 Art. 21, CAT. 36 Arts. 11 to 13, Convention on the Elimination of All Forms of Racial Discrimination, 660 U.N.T.S. 195, entered into force 4 Jan. 1969. 37 States will at times take other actions, such as making diplomatic intercessions, in certain instances. See generally M T Kamminga, Inter-State Accountability for Violations of Human Rights (Philadelphia, University of Pennsylvania Press, 1992). 32

Taking Filártiga on the Road 117 state. Such proceedings entail a searching, detailed forensic examination of the activities of a foreign government. Not surprisingly, many states, while authorising the exercise of universal jurisdiction over foreign human rights abuses, require the special consent of the Attorney General. In Canada, for example, prosecutions of war crimes and crimes against humanity can be initiated only by the federal Attorney General38 and prosecutions of persons who are not Canadian citizens who are alleged to have committed torture require the federal Attorney General’s consent.39 When it amended its laws to provide for universal jurisdiction of war crimes and crimes against humanity, the Canadian government declined to accept a proposal that would have allowed private citizens to initiate prosecutions.40 The Commission whose report led to those amendments noted in its report that the enforcement of such universal jurisdiction provisions “entails the constant monitoring of Canada’s relations with foreign countries”.41 A decision to prosecute a foreign human rights offender will inevitably be taken only by the highest levels of a state’s justice and foreign affairs ministries.42 The decision to initiate a tort action, in contrast, lies entirely with the individual plaintiff or plaintiffs. Unlike the state, a victim has powerful incentives for pursuing a claim for vindication and compensation. If the victim has access to the required information and legal resources, there is a good chance that an action will be pursued. In the US, where international human rights lawyers have actively sought to assist these victims and promote the use of the tort remedy, there have been a large number and variety of actions.43 Like the NGO within the UN system, the individual plaintiff is therefore far more likely than the state to invoke a third country legal remedy. Another reason mitigating against the use of universal criminal proceedings is the onerous nature of the procedural requirements that must be satisfied. The prosecution must first obtain evidence sufficient to provide a reasonable basis for charging and detaining a suspect and, subsequently, to prove guilt beyond a reasonable doubt. Evidence will normally have to be gathered from the country in which the offence was committed. Where the rights-abusing government remains in power, this may be impossible. Even when a foreign government is 38

Criminal Code, R.S.C. 1985, c. C-46, s. 7(3.75). Ibid., s. 7(7). 40 According to the Report of the Commission of Inquiry on War Criminals, which was headed by Quebec Justice Jules Deschenes (currently serving as a judge of the International Criminal Tribunal for the Former Yugoslavia), “[t]he experience of this Inquiry has shown how high emotions do run and how barely skin-deep feelings are buried. In the matter of war crimes, no private citizen should be allowed to put the wheels of justice in motion on his own initiative”. Commission of Inquiry on War Criminals, Report, Part I: Public (1986) 166. 41 Commission of Inquiry on War Criminals, Report, supra n. 40. 42 In contrast to the situation in Canada, some civil law systems, such as those of Spain, France and Italy, retain a role for “investigating magistrates” to initiate prosecutions in response to complaints. As discussed supra in the text between n. 30 and n. 31, this enabled the Spanish magistrate Garzon to initiate the Pinochet process against strong resistance from the Spanish government. 43 See supra n. 29. 39

118 J Terry cooperative, problems of distance, language, culture, and the passage of time make evidence-gathering extremely difficulty. In respect of the Canadian and Australian war crimes prosecutions, for example, although Eastern European authorities generally cooperated, difficulties in gathering reliable evidence played a major role in the failure of these prosecutions. Prosecution is also made more difficult by procedural protections such as the accused’s right to remain silent and the onus on the prosecution to disclose its evidence to the defence. The Convention against Torture guarantees that, where a state party exercises universal jurisdiction, “the standards of evidence required for prosecution and conviction shall in no way be less stringent” than those applied where the prosecution in based upon territorial or nationality jurisdiction.44 Finally, the standard for conviction of guilt beyond a reasonable doubt establishes a difficult evidentiary burden for the prosecution. Tort actions, in contrast, have less onerous procedural and evidentiary requirements since they do not threaten the liberty of the tortfeasor. A defendant need not be detained or even present for trial. In Filártiga, for example, the defendant was served while in the custody of the US Immigration and Naturalisation Service but deported before the action had reached the courts. The defendant does not have a right to counsel or to remain silent. The evidentiary standard is the balance of probabilities rather than a standard of guilt beyond a reasonable doubt. If the defendant decides not to respond to the action, judgment will be entered in default. Many of the actions brought in the US under the ATCA have resulted in large default judgments being awarded against the defendants.

3 COURTS CAN TAKE JURISDICTION WITHOUT EXPRESS STATUTORY AUTHORITY

Assuming that taking jurisdiction over civil actions in respect of torture committed abroad is a good thing, what then should be done to enable courts in countries outside the US to do so? In the best of all possible worlds, the third country civil remedy, like its criminal counterpart, would be expressly permitted or required by treaty and then implemented by means of domestic legislation comparable to the TVPA. But, in my view, even without such a treaty or such domestic legislation, courts can still take jurisdiction—in appropriate circumstances—over civil actions respecting torture committed abroad.

Jurisdiction Aided By Background Human Rights Norms My analysis begins with a basic, uncontroversial legal fact—courts do take jurisdiction, in appropriate circumstances, over torts committed abroad. To 44

Art. 7(2), CAT.

Taking Filártiga on the Road 119 determine whether they should take jurisdiction with respect to a foreign tort and what law they should apply, courts rely upon a well-established body of case law respecting matters such as service ex juris, forum conveniens and choice of law. Torture is primarily an intentional tort, which, depending upon its character, might be classified as assault, battery or intentional infliction of mental suffering. It could also fall within the scope of the law of negligence or breach of fiduciary duty to the extent the claim is for a breach of a positive duty of care of officials above the level of the hands-on torturer, officials who do not themselves participate in or directly order the torture.45 There is no reason why a court in a third country cannot take jurisdiction over one or more of these torts committed abroad in the ordinary way in which it assumes jurisdiction within its private international law framework—that is, so long as the appropriate factors connecting the tort to the forum are present. In the case of torture, the legitimacy of a court taking jurisdiction is enhanced by the strength of the prohibition against torture under international law. The status of the prohibition against torture as a jus cogens norm has been affirmed by the International Criminal Tribunal for the Former Yugoslavia. In Prosecutor v. Anto Furundzija, the Tribunal stated: “Because of the importance of the values it protects, [the prohibition against torture] has evolved into a peremptory norm or jus cogens, that is, a norm that enjoys a higher rank in the international hierarchy than treaty law and even ‘ordinary’ customary rules. The most conspicuous consequence of this higher rank is that the principle at issue cannot be derogated from by States through international treaties or local or special customs or even general customary rules not endowed with the same normative force. Clearly, the jus cogens nature of the prohibition against torture articulates the notion that the prohibition has now become one of the most fundamental standards of the international community. Furthermore, this prohibition is designed to produce a deterrent effect, in that it signals to all members of the international community and the individuals over whom they wield authority that the prohibition of torture is an absolute value from which nobody must deviate.”46

In addition, although the Convention does not expressly authorise or require courts to take jurisdiction over civil actions respecting torture that has occurred in other countries, article 14 requires each state party “to ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable 45 Such responsibility of higher-level officials could also be conceptualised in terms of command responsibility and agency-like principles of the doctrine of respondeat superior: see V Oosterveld and A Flah, “Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability”, chapter 16 of this volume. 46 Prosecutor v. Anto Furundzija (10 December 1998, para. 153), reported at http://www.un.org/ icty/furundzija/trialc2/judgment/main.htm. This excerpt was also relied upon by Lord BrowneWilkinson in his reasons in the Pinochet case, supra n. 7. See also the General Comment No. 24 on “Issues relating to reservations made upon ratification or accession to the Covenant [on Civil and Political Rights] or the Optional Protocol thereto, or in relation to declarations under Article 41 of the Covenant”, issued on 4 Nov. 1994 by the United Nations Human Rights Committee, paras 8, 10, which states that the prohibition of torture has the status of a peremptory norm.

120 J Terry right to fair and adequate compensation, including the means for as full rehabilitation as possible”. This wording echoes the fundamental requirement in article 2(3) of the ICCPR that each state party ensure that any person whose rights or freedoms are violated shall have an effective remedy. There is nothing in the language of either of these provisions that limits this obligation to the state in which the alleged act of torture has taken place. The international legal prohibitions respecting torture and the provisions of the Convention are likely not sufficient—in and of themselves—to elevate the tort to one of universal jurisdiction under international law itself. However, they help to solidify the basis upon which a foreign court can take jurisdiction over torture committed abroad. In the rest of this part, I discuss the main issues that would likely arise were a court in a country such as Canada to be asked to take jurisdiction over a civil action respecting torture committed in another country. My conclusion is that in many, and perhaps most, circumstances a strong argument can be made that the court should take jurisdiction, even if there is no domestic legislation expressly granting it the right to do so.

Service From a practitioner’s perspective, a plaintiff’s first task in a transnational tort action is to ensure that her claim is properly served upon the defendant. Where the defendant resides in the territory of the forum court—like Peña-Irala or Marcos, in the US—service can be effected without difficulty. Where the defendant resides outside the territory of the forum, service is more difficult. In the US, the courts will take transient jurisdiction over a non-resident defendant who is served while temporarily present in the territory. It was on this basis that the plaintiffs in the US sued Radovan Karadzic when he was present in New York for a United Nations meeting.47 In other countries, courts are less likely to recognize transient jurisdiction. For example, the common rules for European Union member states found in the Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters exclude transient jurisdiction as an acceptable basis for taking jurisdiction over a defendant.48 In Canada, while transient jurisdiction has been accepted in the past, recent jurisprudence has thrown its validity into doubt.49 If a plaintiff is unable to effect service within the forum state, the plaintiff can attempt to serve the defendant outside of the jurisdiction by means of service ex juris. In Ontario, for example, the provincial rules of civil procedure expressly

47

Doe v. Karadzic, Kadic v. Karadzic, 70 F. 3d 232 (2d Cir. 1995). Arts 2, 3, Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters, done at Brussels, 27 September 1968. 49 Hunt v. T&N plc. [1993] 4 S.C.R. 289. See McConville (Chap. 6 infra) for discussion of the insufficiency of transient presence as a connection grounding judicial jurisdiction. 48

Taking Filártiga on the Road 121 allow service ex juris where damage has been sustained in Ontario.50 These rules have been relied upon, for example, to allow service ex juris in an action for damages arising from alleged sexual assaults perpetrated by the defendants in the Netherlands and in Ontario, in part because harm attributable to the Netherlands assaults was sustained in Ontario.51 In addition, the plaintiff can also move—ex parte—for leave to serve ex juris with the goal of persuading the judge of the sufficiency of the action’s connections to Ontario and simultaneously that Ontario is an appropriate forum for the action.52 This is done under Rule 17.03, which provides a residual discretion to a court to grant leave to serve ex juris for those situations not specifically provided for under the Rules.53 It can be argued (and has been accepted by certain judges54) that courts should uphold or authorise service ex juris only in those instances in which there is a real and substantial connection between the forum and the events at issue. Such a position is based on the principle of comity, which holds that territorial limits to jurisdiction must be respected to preserve international legal order and to avoid unfairness to defendants.55 But this principle has little application in cases of alleged torture committed abroad. Because the prohibition against torture is a jus cogens norm, there would appear to be little or no risk of illegitimate, let alone unlawful, foreign interference with the local policies of the state in which the alleged torture occurred. The jus cogens status of the norm similarly speaks against there being any unfairness to a defendant. Indeed, by accepting under the Convention that torture is a crime which attracts universal jurisdiction, the international community has already invited—in fact, required—states to exercise their investigatory and adjudicative jurisdiction in a manner that, as described above, is far more intrusive than the acceptance of jurisdiction over a civil action.

Forum Non Conveniens Assuming the defendant is properly served, the defendant can still move to stay the action on the grounds that the third country in which the action is brought is not an appropriate forum in which to try the action. To do so, the defendant has to establish that there is a more appropriate forum than that selected by the 50 Rule 17.02(h), (Ontario) Rules of Civil Procedure, R.R.O. 1990, Reg. 194, as amended: “A party to a proceeding may, without a court order, be served outside Ontario with an originating process or notice of a reference where the proceeding against the party consists of a claim or claims, . . . in respect of damage sustained in Ontario arising from a tort, breach of contract, breach of fiduciary duty or breach of confidence, wherever committed;” (emphasis added). 51 B.(J.) v. D.(E.) (1993), 16 C.P.C. 242 (Ont. Gen. Div.). 52 For a detailed discussion of these issues see McConville, Chap. 6 infra at n. 28. 53 Rule 17.03, Ontario Rules, supra n. 50. 54 See MacDonald v. Lasnier (1994), 21 O.R. (3d) 177 (Ont. Gen. Div.), discussed in McConville, Chap. 6 infra at n.28. 55 See, generally, De Savoye v. Morguard Investments Ltd. et al. [1990] 3 S.C.R. 1077; Hunt, supra n. 49.

122 J Terry plaintiffs. The existence of a more appropriate forum must be clearly established to displace the forum selected by the plaintiff.56 Where no one forum is proved to be more appropriate, the domestic forum wins out by default and the defendant’s motion must fail.57 To make this determination, courts consider factors such as the residence of the parties, the location of the core of the action, the location of witnesses and documentary evidence, the applicable law and any juridical advantage that the plaintiff receives by bringing its action in the forum he or she has chosen. In an action respecting alleged torture that was committed abroad, the fact that the events took place in another jurisdiction and that much of the evidence, such as it is, will likely have to come from persons or documents located in that jurisdiction will weigh against the plaintiff. However, the plaintiff’s case will be strengthened if the plaintiff or defendant (or ideally, both) reside in or have significant connections to the third country and if the plaintiff can demonstrate that he or she has continued to suffer from the effects of the torture while living in the third country. There may also be documentary evidence, such as medical reports, and possibly some witnesses located in the third country. Given the nature of the tort of torture, a plaintiff’s best response to a forum non conveniens motion may be to argue that the state in which the torture took place is not an alternative forum. The English Court of Appeal, in Mohammed v. Bank of Kuwait and the Middle East, for example, held that Kuwait was not available to the plaintiff in a practical sense as an alternative forum for resolution of this dispute.58 If a plaintiff is required to sue the government or official responsible for the torture in the country in which it took place, she risks not only inadequate injustice but further detention and even torture or death. As Lord Browne-Wilkinson stated in the Pinochet case, in discussing the crime of torture, “a totalitarian regime will not permit adjudication by its own courts on its own shortcomings.”59 Even if the government allegedly responsible for the torture and still in power falls short of being “totalitarian”, there may still be insurmountable barriers—for example, police complicity with the torturers, judicial bias, or intimidation by government officials or non-governmental actors—that make it impossible for the plaintiff to bring her claim in that country. In that situation, there is no alternative forum,60 a fact which surely is rele56 Amchem Products Inc. v. B.C. (W.C.B.) [1993], 1 S.C.R. 897 at 921; Frymer v. Brettschneider (1994), 19 O.R. (3d) 60 at 79 (C.A.). 57 Upper Lakes Shipping Ltd. v. Foster Yeoman Ltd. (1993), 14 O.R. (3d) 548 at 570–1 (Ont. Gen. Div.); Gilmour v. Barry-Wehmiller Co. (1996), 34 O.R. (3d) 304 at 309 (Ont. Gen. Div.). 58 Mohammed v. Bank of Kuwait and the Middle East [1996] 1 W.L.R. 1483 at 1496 (C.A.). See also Connelly v. RTZ Corporation PLC [1998] AC 854 in which the House of Lords held that because the plaintiff would be entitled to legal aid in the UK but not Namibia, Namibia was not a forum in which the case could be tried more suitably for the interests of all the parties and for the ends of justice. 59 Pinochet case supra n. 7 at 109. 60 Note, however, that courts are often reluctant to comment upon unfairness in the courts of another jurisdiction. In Mohammed, supra n. 58, for example, Evans LJ emphasised that while Kuwait was not a practical alternative for the plaintiff “I do not find any justification for holding

Taking Filártiga on the Road 123 vant to whether or not the forum state has met its obligation under article 14 of the Convention to ensure “that the victim of an act of torture obtains redress”. If, on the other hand, there has been a transition of government in the state in which the alleged torture occurred, a court may have to be persuaded on the evidence that the plaintiff will not now be able to have her action fairly and impartially resolved in the country in which it occurred.61

Choice of Law Assuming a third party court will uphold service and take jurisdiction over a civil action respecting torture that was committed abroad, the next issue for the court to determine will be what law should be applied. Until recently, in both the United Kingdom and Canada, the double actionability rule—as elaborated in Phillips v. Eyre,62 Chaplin v. Boys,63 and Red Sea Insurance Co Ltd v. Bouygues SA64—applied. Those cases established a general rule that the tort must be actionable by both the law of the forum and the law of the place of the tort. Now, as a result of statutory reform in the United Kingdom65 and changes in case law in Canada,66 both jurisdictions have adopted, with certain exceptions, the rule that prevails generally in the US and Europe—that the applicable law is the place of the wrong. These developments towards lex loci delicti could be taken as an obstacle for transnational tort claims that seek to apply some set of rules other than those of the state in which the torture occurred. Working on the assumption that the lex loci delicti is a barrier to successful claims, Jennifer Orange, in her contribution to this volume, argues persuasively that in the case of the tort of torture, the lex loci delicti should not be applied.67 Indeed, in many cases, for reasons of public policy or because of other connections between the tort and the forum state, there may be justifiable reasons for the court to apply the law of the forum instead of the law of the place where the torture occurred. But I would argue that, even if the lex loci delicti is applied, a court will often still be able to determine that there is a viable cause of action in respect of the tort of torture. that Kuwaiti judges would not have acted fairly if an Iraqi citizen appeared as plaintiff before them whenever that might have occurred, whether in 1991 or 1995 or now.” (p. 1496). 61 See e.g. Recherches Internationales du Québec v. Cambior [1998] QJ No. 2554 (Que SC). In this case, the trial judge was not persuaded that there was a significant risk that the judiciary in the post-dictatorship Guyana would not give a fair hearing to plaintiffs suing a Quebec-based mining company and its Guyana-incorporated subsidiary for a large cyanide spill into a major river in Guyana. The plaintiffs had initiated suit against Cambior in Quebec, but the judge stayed the proceedings on the basis of Guyana being the more appropriate forum. 62 Phillips v. Eyre, supra n. 10. 63 [1971] AC 356. 64 [1995] 1 AC 190. 65 Private International Law (Miscellaneous Provisions) Act 1995. 66 Tolofson v. Jensen [1994] 3 S.C.R. 1022. 67 J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11 of this volume.

124 J Terry Although many governments practise or condone torture, virtually no government will sanction its legality.68 In most states, there are constitutional, statutory or common law prohibitions against torture or related acts, such as assault and battery, as well as parallel civil remedies for victims of such acts. In Filártiga, for example, even though the judge applied the law of nations as the substantive law, for the most part, he placed some reliance, in assessing damages, on the fact that Paraguayan law forbids torture and provided for “moral damages” in such cases, which would include emotional pain and suffering, loss of companionship and disruption of family life.69 One hundred and eighteen states, including many that have been or still are regarded as states that practise torture, have ratified the Convention and have thus committed themselves to criminalising torture under their law as well as providing, per article 14, civil remedies for victims of torture at least within their own territorial jurisdiction.70 One hundred and forty-four states have ratified the ICCPR, which provides that no one shall be subjected to torture and which requires states to provide an effective remedy to any person whose rights under the treaty have been violated. All but two states in the world have ratified the Convention on the Rights of the Child, which guarantees that no child (a person eighteen years or under) shall be subjected to torture. It is impossible to avoid the conclusion that all states are governed by the international prohibition against torture, which has the status of a norm of jus cogens. As a result, even in a state that is known to condone torture—like Stroessner’s Paraguay—resort to that state’s law will often result in a determination that torture is an actionable wrong under the lex loci delicti. In these cases, by turning to the law of the place where the wrong occurred, a third country court can vindicate the victim’s constitutional, statutory or common law rights as they are formally protected under the law where the torture occurred. In some cases, there may be no cause of action in the lex loci delicti because of laws that immunise state officials from civil liability or protect them with very short limitation periods. In addition, the state where the torture occurred may have enacted an amnesty precluding civil actions respecting allegations of tor68 One exception is Israel, whose government and courts have taken the position that the techniques used to interrogate Palestinians are justified on grounds of necessity, even though they are regarded as torture by the UN Torture Committee and the UN’s Special Rapporteur on Torture. See, for example, the following Israeli decisions: ‘Abd al-Halim Bilbeisi v. The General Security Service, HCJ 7964/95, decision of 11.1.96; Muhammad ‘Abd al-Aziz Hamdan v. The General Security Service, HCJ 8049/96, decision of 14.11.96; and Khader Mubarak et al. v. The General Security Service, HCJ 3124/96, decision of 17.11.96. Since these cases, the Israeli Supreme Court seems to have taken a new course: see H.C. 5100/94 Public Committee against Torture in Israel et al. v. The State of Israel and the General Security Service [1999] (as yet unpublished), discussed in A Reichman and T Khana “Israel and the Recognition of Torture: Domestic and International Aspects”, chapter 24 in this volume. 69 Filártiga v. Peña-Irala, 577 F. Supp. 860 at 864 (E.D.N.Y. 1984). In this way, even though Dr. Filártiga’s Paraguayan lawyer had reportedly been shackled to a wall, threatened with death and later disbarred without cause, the US judge could rely in part on the “law in the books” to justify the validity of his damages remedy. 70 Supra n. 3.

Taking Filártiga on the Road 125 ture in order to facilitate a peaceful transition to democracy. Here the analysis becomes more complicated. An amnesty founded on a genuine policy of compromise and reconciliation, such as the recent amnesty in South Africa, may merit deference from the forum court. However, where the immunity—like Pinochet’s amnesty in Chile—is unilateral and provides no or inadequate compensation for torture victims, the forum court could determine, on the basis of the lex loci delicti, that such immunity is contrary to the jus cogens prohibition against torture and, if the state has ratified the Convention, to that state’s obligation to provide compensation to torture victims. In this way, the forum court can apply reasons grounded in international public policy not to apply the lex loci delicti. In considering a communication from relatives of deceased victims of torture in Argentina, for example, the Committee against Torture stated that Argentinian laws that eliminated their rights of civil action were incompatible with the spirit and purpose of the Convention.71

Sovereign Shields: State Immunity and Act of State Two “defences” that may be raised in a transnational tort case are state immunity and act of state. They are not the same. State immunity only arises in cases where a state or state entity of some kind is sued and is the putative defendant; it is a jurisdictional bar to one state’s court hearing a case involving a foreign sovereign, unless one of a few exceptions apply.72 State immunity is a personal jurisdictional bar: sovereigns and their agents cannot generally be sued abroad. The act of state “defence”, on the other hand, is not triggered by the presence of a state actor as defendant but by the presence of legal issues that require a foreign court to assess the validity of a foreign law in order to determine the rights and obligations of the parties.73 Put differently, the act of state doctrine creates a form of of subject matter bar to a foreign court’s role.74

71 Report of the Committee against Torture, UN GAOR, 45th Sess., Supp. No. 44, Annex V at 109–13, UN Doc. A/45/44 (1990). 72 For discussion of the widest exception in modern state immunity law, see R Wai, “The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legalism,” chapter 8 of this volume. 73 For example, a classic act of state situation involves goods being present in a court’s jurisdiction and one party suing another party in the tort of conversion, claiming that any title to the goods the defendant has is invalid and thus the defendant is illegally in possession of the plaintiff’s property. Where the defendant’s title is grounded in a transfer of ownership effected by some kind of expropriation decree or statute of a foreign state, the court has no choice but to decide whether or not to give effect to the foreign “act of state”. If the court does, the defendant wins; if the court instead decides the foreign “official” transfer of title is invalid for some reason, then the plaintiff wins. See for example Banco Nacional de Cuba v. Sabbatino, (1964) 376 U.S. 398; Hesperides v. Muftizade [1978] 2 All ER (H.L.); Laane v. Estonian State Cargo [1949] 2 D.L.R. 641 (S.C.C.); Juelle v. Trudeau (1968) 7 D.L.R. (3d) 82 (Que. Sup. Ct.). 74 It may be an important point to determine whether act of state operates as a choice of law rule or as a jurisdictional barrier, but this analytical distinction need not detain us here.

126 J Terry Each defence should be kept analytically distinct as they involve a different form of prima-facie interference of a foreign court into the sovereignty of a foreign state. However, the two doctrines can come together in suits in which individuals are sued and those individuals claim not act of state but state immunity by virtue of the official nature of the impleaded acts carried out on behalf of a foreign state—even when the individuals are no longer officials or agents of the foreign state. It is the claimed sovereign nature of their acts which—they claim—entitles them to claim immunity.75 It can be seen that, if courts fail to see the overlap in the two doctrines in these kinds of cases, they may narrow the scope of the act of state defence while simultaneously expanding the sovereign immunity defence, the latter outflanking the former. In the following two sections, it is assumed that the same logic should apply to limit both act of state and state immunity defences.76 (a) US court approach to sovereign shield defences under ATCA and TVPA When defendants to alien tort claims in the US have raised sovereign shield defences, the courts have held in general that these defences do not apply because the actions in question—human rights violations of one sort or another—are “unofficial”. However, the fact that torture is—legally—a public act, authorised explicitly or implicitly by a state, complicates the question of whether third party courts should be able to pass judgment over foreign acts of torture. Avoiding the application of the act of state doctrine by characterising conduct as “unofficial” is problematic when applied to torture, where the very definition of torture requires that the torture be “inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity”. A better approach is to focus on illegality under international law as a reason for not applying sovereign shield defences. In other words, the approach adopted by the House of Lords in Pinochet should be no less applicable in the civil context than in the criminal context. As Lord BrowneWilkinson asked in that case, “[h]ow can it be for international law purposes an official function to do something which international law itself prohibits and criminalises?”77 A leading example of the above-mentioned tensions is Forti v. SuarezMason,78 an ATCA case which involved human rights violations authorised by the highest levels of the Argentine government. The defendant, General SuarezMason, raised the act of state defence. He challenged the court to resolve the “Catch 22” of finding that he had committed an “official” act of torture without 75 For a discussion of these cases of personal sovereign immunity, see M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of ‘Act of State’ in Anglo-Canadian Law”, chapter 13 of this volume. 76 See pp. 126–9 of this chapter 77 Pinochet case supra n. 7 at 114. 78 672 F. Supp. 1531 (N.D. Cal. 1987).

Taking Filártiga on the Road 127 simultaneously ruling that he had been carrying out an “official” act of state.79 Instead of holding that the act of state doctrine did not apply to state violations of fundamental human rights, the court decided, contrary to all the evidence, that the General’s acts occurred under colour of authority but did not necessarily arise out of official government policy.80 The extensive human rights litigation involving former Philippines President Ferdinand Marcos provides another example, where the defendants this time raised both state immunity and act of state defences. During the late 1980s, various alien tort actions were brought against Marcos as well as General Fabian Ver and Imee Marcos-Manotoc (Marcos’s daughter). Together, these three individuals controlled the military intelligence personnel responsible for the torture and summary execution of thousands of persons. In disposing of the defendants’ reliance on foreign sovereign immunity and act of state defences, the Ninth Circuit Court of Appeals held that the defendants’ actions were not “official” or “public” acts, but were the acts of individual officials engaging in activities outside the scope of their official mandate.81 The approach taken in these cases has in effect allowed the US courts to rule upon actions committed by another foreign government only by “privatising” the actions of the perpetrators. But this fiction, although useful, is problematic. First, as noted above, it is inconsistent with the definition of torture in the Convention as being an official act in the sense of involving at least acquiescence on the part of a state official. Second, it depends ultimately on the assumption that the state involved will deny that the human rights violation was committed pursuant to its official policy. If the state acknowledges that it authorised the torture, any action respecting such torture might well be dismissed on the basis of a sovereign shield defence. (b) Universal criminal jurisdiction approach to sovereign shield defences In my view, a better approach to sovereign shield doctrines like sovereign immunity and act of state is that taken by states and courts as they have expanded the doctrine of universal jurisdiction from crimes such as piracy to human rights offences. From its earliest origins as a basis for taking jurisdiction over piracy, universal jurisdiction was limited to persons acting for private rather than public motives. The label hostis humanis generis, applied to pirates and adopted by the Court in Filártiga to describe human rights offenders, reflected the fact that pirates owed no allegiance to any sovereign. Many piracy cases turned on whether the alleged pirates were acting solely for private gain or whether there 79

Ibid. at 1546. Ibid. For the Ninth Circuit’s dismissal of the foreign sovereign immunity defences, see In re Estate of Ferdinand Ed. Marcos, Human Rights Litigation, 978 F. 2d 493 at 497 (9th Cir. 1992), cert. denied, 113 S. Ct. 2960 (1993) ((No. 92-1559); In re Estate of Ferdinand Marcos, Human Rights Litigation, 25 F. 3d 1467 (9th Cir. 1994). For its dismissal of the act of state defence, see Trajano v. Marcos, 878 F.2d 1439 (9th Cir. 1989). 80 81

128 J Terry had been a public element to their action.82 The piracy offence was premised on the need to police the high seas, where no state had jurisdiction, and was crafted to avoid foreign political entanglements. For example, when a number of states attempted in a 1922 arms control treaty to extend universal jurisdiction to submarine attacks against merchant ships, critics complained that “submarine offences, unlike piracy, are very liable to be committed by the authority of a State, and consequently this universal jurisdiction, . . . if exercised, might well lead to serious disputes.”83 Modern codifications of the piracy offence still require that piracy be committed “for private ends”.84 However, the events of World War II and the subsequent Nuremberg and Tokyo prosecutions led to the abandonment of the public/private distinction in the definition of offences eligible for universal jurisdiction. In cases in which the international military tribunals could not exercise jurisdiction on the basis of territoriality or passive personality jurisdiction, they relied on the universality principle to justify their assertions of jurisdiction and in this way extended the scope of universal jurisdiction to include war crimes and crimes against humanity.85 Instead of assiduously avoiding involvement with state actions, the Nuremberg judges openly extended universal jurisdiction to individuals who had been acting in direct conformity with the policies of the German state. Judges and writers analogized the crimes of Nazi officers to the acts of pirates, with apparent unconcern for sovereign immunity implications that would previously have disqualified such state-related crimes as an appropriate subject of universal jurisdiction. As the Nuremberg Tribunal stated: “The principle of international law, which, under certain circumstances, protects the representatives of a State, cannot be applied to acts which are condemned as criminal by international law. The authors of these acts cannot shelter themselves behind their official position in order to be freed from punishment in appropriate proceedings.”86

This statement would appear to be even more directly applicable in the case of the Convention in which the very definition of the crime of torture requires that torture be committed by an official or someone in an official capacity. It is true that Nuremberg involved the exercise of an international institutional jurisdiction. However, more than a half-century later, no meaningful 82 For example, a privateer acting under a commission from a foreign sovereign could not be prosecuted unless he had exceeded the terms of the commission. See the British trial in 1701 of Captain William Kidd (14 Howell State Trials 147). The case, which turned in part on the issue of whether Kidd had exceeded the King’s Commission to take prize ships, is described in A Rubin, The Law of Piracy (1988) at 94–100. 83 Roxburgh, “Submarines at the Washington Conference”, [1922–3] Br. Yk. Int’l L. 154. 84 Art. 15, Convention on the High Seas, done 29 April 1958, U.N.T.S. 82. Art. 110, Convention on the Law of the Sea, done 10 Dec. 1982, reprinted in The Law of the Sea, U.N. Doc. A/Conf. 62/122. 85 As the United States Court of Appeals for the Sixth Circuit stated in Demjanjuk v. Petrovsky, 776 F. 2d 571, 582 (6th Cir. 1985), cert. denied, 475 U.S. 1015 (1986), “it is generally agreed that the establishment of these tribunals and their proceedings were based on universal jurisdiction”. 86 IMT Judgment, excerpted from Kritz (ed.), supra n. 12, Vol. III at 462.

Taking Filártiga on the Road 129 distinction can be drawn between a sovereignty-based defence to individual accountability as it applies at the domestic and international institutional levels. Let us assume that, under international law, sovereignty cannot act as a normative shield to prevent the enforcement of an international norm of jus cogens such as the prohibition of torture. If domestic courts have a duty to act in accordance with the rule of law to the greatest extent permitted by their domestic legal system, and if the limits of sovereignty defences for individuals are set out in international law (and deepening by the day in their normative force), then domestic courts should strive to avoid acceptance of sovereign-shields where at all possible. The reasoning in the above paragraph has implications for both the act of state and sovereign immunity doctrines. In the view of the authors of the Restatement (Third) of the Foreign Relations Law of the United States, the act of state doctrine should not be used by domestic courts to dismiss actions for serious violations of human rights, “since the accepted law of international human rights is well established and contemplates external scrutiny of such acts”.87 As for state immunity, it is important to recall that it has long been established as a matter of statutory interpretation that statutes should be interpreted, to the fullest extent possible, to comply with a state’s international legal obligations, whether customary or treaty-based.88 This principle, founded in the rule of law, is based on the presumption that the legislature intends to act in compliance with international law. In Baker v. Canada (Minister of Citizenship and Immigration), the Supreme Court of Canada has very recently underlined the importance of this interpretative principle, stating that, as part of contextualised statutory interpretation, courts should make reference to underlying values reflected in Canada’s human rights treaty obligations even if these obligations have not been implemented through domestic legislation.89 Sovereign immunity provisions in domestic statutes should therefore be interpreted in a manner that reflects the international consensus that the prohibition against torture is jus cogens and takes seriously that the Convention requires both its criminalisation and the provision of compensation to its victims. Since torture has been identified as both a public act and a jus cogens crime under international law, the principle of sovereign immunity has no place as a defence to actions respecting torture—and state immunity statutes must be read down accordingly.90 87 The American Law Institute, Restatement (Third) of the Foreign Relations Law of the United States (St. Paul, Minn., American Law Institute Publishers, 1987), s. 443, comment c. 88 Baker v. Canada (Minister of Citizenship and Immigration) [1999] 2 S.C.R. 817 at 861; Jacobs and Roberts (eds), The Effect of Treaties in Domestic Law (London, Sweet & Maxwell, 1987) at 32–3 (Italy), 60 (France), 69 (Germany), 100 (Italy), 135, 137 (UK), 160 (US); R Sullivan, Driedger on the Construction of Statutes, 3rd ed. (Toronto and Vancouver, Butterworths, 1994) at 330–3. 89 Baker v. Canada, ibid. 90 For a contrary view of the effect of jus cogens international law on interpretation of Canada’s State Immunity Act, see W Adams, “In Search Of A Defence Of The Transnational Human Rights Paradigm: May Jus Cogens Norms Be Invoked to Create Implied Exceptions in Domestic State Immunity Statutes?”, chapter 9 of this volume.

130 J Terry

4 THE ARONE CASE

The Arone case is an action brought by Abukar Arone Rage and Dahabo Omar Samow against the Attorney General of Canada.91 Rage and Samow are Somalians, the parents of a deceased teenage boy named Shidane Abukar Arone. In March 1993, Arone was tortured and murdered while in the custody of Canadian military personnel in Somalia. The Arone case provides a recent example of a claim brought before a Canadian court with respect to torture committed abroad. Unlike the ATCA and TVPA cases in the US, this is not an action brought by a foreign national against a foreign tortfeasor before a court in Canada. Although the plaintiffs are foreign nationals, the defendant is the Canadian government and the torture was committed by Canadian nationals. Nevertheless, the case fits the paradigm of the tort actions discussed in this chapter in that the torture occurred in a country other than Canada and the underlying acts prima facie constitute not only torture but the cause of action is in tort. Arone thus elucidates issues that will have to be dealt with if and when “third country” claims are brought in Canadian courts. In December 1992, members of the Canadian Armed Forces were dispatched to Somalia to serve in the UN peacekeeping mission there. On 16 March 1993, Arone slipped into the Canadian military base where he was detained by members of the Canadian Armed Forces. Between 16 March and 17 March 1993, Arone was blindfolded, bound, beaten and tortured. Following Arone’s death, criminal proceedings were brought in Canada. One of the Canadian soldiers was convicted of torture and manslaughter and three others were convicted of negligence as a result of the torture and death of Arone. Six years after the torture-killing, on 15 March 1999, Arone’s parents, by their litigation guardian Abdullahi Godah Barre, commenced an action in the province of Ontario against the Government of Canada. Pursuant to the Crown Liability and Proceedings Act, the Crown is vicariously liable in respect of torts committed by its servants to the same extent that any adult person would be liable.92 Given that the defendant in the Arone claim was the Government of Canada, the court did not have to deal with issues of service ex juris or forum non conveniens. Instead, the court focused on a procedural issue respecting the plaintiffs’ litigation guardian93 and, most importantly for the purposes of our analysis, on the nature of the cause of action and the applicable limitation 91 Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. The Attorney General of Canada (unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J). 92 R.S.C. 1985, c. C-50, section 36. 93 The court held that under Ontario law, a Litigation Guardian could bring a claim on behalf of another person only if that person had a disability and the person had filed an affidavit appointing the Litigation Guardian and providing evidence respecting the nature and extent of the disability. The court found that there was no evidence of a disability in this case. Arone, supra n. 91, at 3.

Taking Filártiga on the Road 131 periods. On 6 July 1999, Mr. Justice Cunningham granted the Attorney General of Canada’s motion striking the action on all three of the just-mentioned grounds. He ruled that the action was improperly commenced, failed to disclose a cause of action and was statute-barred.94 That decision is under appeal. The plaintiffs, in their statement of claim, had claimed general and punitive damages for “torture, wrongful death, and/or murder, negligence, and breach of fiduciary duty”. The court found, however, that the plaintiffs had not made out a valid cause of action. With respect to the claim of negligence, the court held that the statement of claim did not set out facts to establish the Canadian Armed Forces owed a duty of care to Arone’s parents. On the claim for breach of fiduciary duty, the court held that the statement of claim had not set out the facts which give rise to the alleged fiduciary duty. Although the court acknowledged that the claim was also founded in assault and battery, it did not address these aspects of the claim. Nor did it mention the relevance of the reference to “torture” in the statement of claim to the establishment of a cause of action. The court also emphasised that the claim should have been brought by Arone’s estate rather than by Arone’s parents. In my view, the court erred in determining that a cause of action had not been made out. The statement of claim describes the torture of Arone in sufficient detail to make out a claim for assault and battery as well as wrongful death. On the claim of negligence, the duty of care at issue is that owed to Arone, not to his parents. Similarly, the facts alleged in the statement of claim—that Arone was tortured and killed while in the custody of the Canadian Armed Forces— are sufficient to establish a fiduciary duty owed by the Canadian Armed Forces to Arone while he was in custody. The court also did not appear to acknowledge that, because the events that gave rise to the cause of action occurred in Somalia, it is prima facie the substantive law of Somalia, supplemented by applicable international law, that must be applied. The court did not consider, for example, the facts that must be pleaded under Somalian law to prove torture, wrongful death, assault, battery, negligence, fiduciary duty, or equivalent causes of action under Somalian law. The court did not consider Somalian law concerning the ability of parents to bring a claim in respect of the injury or death of a child. The court did not consider article 14 of the Convention, which requires each party to ensure in its legal system that the victim of an act of torture has an enforceable right to fair and adequate compensation and the very clear provision that “his dependants shall be entitled to compensation”.95 In addition, the court made no reference to the provisions of the Geneva Conventions and their applicability in this case, particularly the provisions of common article 3 of the Geneva Conventions and

94

Ibid. See esp. A Byrnes, “Civil Remedies for Torture Committed Abroad: An Obligation under the Convention against Torture?”, chapter 20 of this volume, for a discussion of article 14. 95

132 J Terry Additional Protocol II, respecting the treatment of civilians in non-international armed conflicts.96 The court made the same error when it examined the limitation periods that apply to this action. The plaintiffs served their notice of action within six years after the events had taken place in Somalia, relying upon the six-year limitation period generally applicable in Canada for tort actions as well as the default limitation period set out in s. 32 of the Crown Liability and Proceedings Act.97 The court held that the claim was statute-barred for failure to comply with various limitation periods. These were: the two-year period under Ontario law for dependants to bring a tort claim in respect of the injury or death of a child;98 the two-year period under Ontario law for the legal representative of Arone’s estate to bring a claim;99 the six-month period under Ontario law for bringing a claim against a person for an act done in execution of a public duty or authority;100 and the six-month period under Canadian law for actions against members of the Canadian Armed Forces in respect of the execution of military duty or authority.101 With these findings, the court clearly erred in how it applied limitation periods under Ontario law, since the cause of action did not arise in Ontario. The court should have relied upon the six-year limitation period under s. 32 of the Crown Liability and Proceedings Act, which applies to causes of action that do not arise in any province of Canada.102 It is also arguable that the court should have had regard to the limitation periods under Somalian law that would pertain to this incident, although it bears emphasising that a specific statutory provision on limitation periods for actions against the state of Canada such as found in section 32 would still apply were it to turn out that Somalia law’s relevant limitation period is less than six years. As the Supreme Court of Canada established in Tolofson v. Jensen,103 a limitation period is a substantive matter that is governed by the lex loci delicti104—at least absent separate statutory direction to the courts by the legislature(s) of the forum state.105 96 The Geneva Conventions have the force of law in Canada by way of the Geneva Conventions Act, R.S.C. 1965, c. G-3 as amended. 97 R.S.C. 1985, c. C-50. The relevant part of section 32 reads: “. . . [P]roceedings by or against the Crown in respect of a cause of action arising otherwise than in a province shall be taken within six years after the cause of action arose.” 98 Family Law Act, R.S.O. 1990, c. F.3, s. 61(4). 99 Trustee Act, R.S.O. 1990, c. T.23, s. 38(3). 100 Public Authorities Protection Act, R.S.O. 1990, c. P.38 , s. 7. 101 National Defence Act, R.S.C. 1985, c. N-5, s. 269(1). 102 The plaintiffs in Arone have notice of the action one day short of six years from the day of Arone’s death. Arone, supra n. 91, at 6. 103 Supra n. 66 at 1066–74. 104 This question is complicated, however, by the fact that the defendant is the Crown and that the plaintiffs are relying on the permissive provisions of the Crown Liability and Proceedings Act to bring an action against the Crown that at common law was not available. 105 For further discussion of the Arone case, see C Scott, “Introduction to Torture as Tort: From Sudan to Canada to Somalia”, chapter 1 of this volume at 33–44. This discussion includes aspects not addressed in the present chapter, including the judge’s view that the Arone claim was covered by s. 269(1) the National Defence Act which bars actions brought after six months where the underlying

Taking Filártiga on the Road 133 In short, the Arone case presents the courts in Canada with an opportunity to consider seriously and, in my view, apply the analyses put forward in this chapter as well as others in this volume. It bears reiteration that, in contrast to the typical action under ATCA or TVPA, this claim is against the government of the forum state, so there are no issues of forum non conveniens or service ex juris. The facts, which have already been the subject of several judicial proceedings, are not in dispute. What is required by the court on appeal, assisted by arguments of counsel, is a careful choice of law analysis informed by the appropriate application of international law. With luck, the Arone case will establish the beginnings of a framework for the analysis and determination of ATCA- and TVPA-style actions in Canada.

5 CONCLUSION

The third country tort remedy for victims of torture has a useful role to play in the enforcement and development of international human rights law. Because it is founded on a private, individual right of action, it is far simpler to initiate and pursue than the universal criminal remedy. While it is premised on the principle that victims of torture must be compensated, its true role in many cases is admittedly symbolic: the third country tort remedy provides recognition for, and emotional vindication of, the victims of torture and places moral and political pressure on rights-abusing governments. Ideally, the third country tort remedy should be sanctioned by treaty or authorised by domestic legislation in the third country. Until such time, however, courts can, in appropriate circumstances, adjudicate such actions in accordance with long-standing principles respecting judicial jurisdiction and choice of law.

act is “done in pursuance or execution or intended execution of this Act or any regulations or military or departmental duty or authority”.

5

Torture: Prevention Versus Punishment? MALCOLM D. EVANS and ROD MORGAN 1

1 INTRODUCTION H E P R I M A R Y P U R P O S E of this volume is to explore the possibility of individual victims seeking redress through private law actions for acts of torture against both individuals and states before the courts of a third state. The underlying assumption is that, although the prohibition of torture has become part of customary international law, the practice of torture remains widespread. Torturers, and those who order or encourage torturers to ply their trade or acquiesce in their doing so, enjoy virtual impunity from prosecution within their own jurisdictions. This means that the individual goes unpunished and the responsibility of the state is itself engaged by its failure to punish. In the absence of any international judicial organs with jurisdiction to hear cases brought by individuals either against other individuals or against states, the most effective response is seen to lie in prising open the jurisdiction of domestic courts to allow such suits. Other chapters in this volume set out the successes and limitations of this strategy. The transnational litigation approach is premised on the implicit assumption that opening up the prospect of torture as tort will break down that impunity. But, transforming impunity into punishment cannot be the sole purpose of bringing such actions. Although it is important to provide effective means of redress for victims, it is equally—if not more—important to take steps to ensure that acts of torture occur as infrequently as possible. Successful criminal prosecutions and private tortious actions would almost certainly contribute to this by deterring the practice. But the purpose of this chapter is to reflect on how the promotion of ex post facto systems of redress will impact on a number of recent and rapidly developing mechanisms which aim to prevent torture ex ante through the construction of safeguards which make it less likely. In what follows we do four things. First, we give an introduction to various models of torture prevention. This enables us to locate the preventive mechanisms within

T

1 Respectively, Professor of Public International Law and Professor of Criminal Justice, Faculty of Law, University of Bristol, UK.

136 M Evans and R Morgan a range of possible responses to the phenomenon of torture, illuminating their differences from judicial responses. An outline of the paradigmatic preventive mechanism—the European Committee for the Prevention of Torture—is then given. With the advantage of this informational foundation, we pose two questions in the final sections. What contribution might these preventive mechanisms make to the pursuit of torture as tort? And, what might be the consequence of individuals pursuing private actions for the future operation of preventive mechanisms?

2 MODELS OF TORTURE PREVENTION

In the field of crime prevention it has become orthodox to distinguish initiatives at three levels.2 Primary prevention focuses on the general environment within which offences may occur without reference to criminals or potential criminals. That is to say, it seeks generally to reduce both the opportunities for offending and motivating factors. Such efforts range widely and can encompass an almost infinite range of activities. Examples include improving the security of property and siting vulnerable targets in places where they are more readily observable, thus increasing the difficulties and hence reducing the likelihood that they will be subject to illegal activity. Secondary prevention focuses on changing people, particularly those at risk of embarking on a criminal career. An obvious target is children growing up in deprived circumstances in high crime neighbourhoods: they may be made the subject of positive discriminatory interventions. Finally, tertiary prevention is the traditional preoccupation of the criminal justice system. It aims to truncate criminal careers already in progress. It involves policing, prosecution, sentencing and penal policy. The objectives range from incapacitation—that is, physically preventing offenders from repeating their offences, usually by means of incarceration—to punishment with a view to deterrence, to interventions designed to rehabilitate or address social or psychological deficits in order to reduce the likelihood of recidivism. It may be useful to apply these models of general crime prevention to the prevention of torture. The first, and most striking, feature of the torture prevention debate is that it is overwhelmingly focused on tertiary initiatives. In this respect, torture prevention discourse and activity is similar to crime prevention discourse. Crime prevention talk, expenditure and effort are overwhelmingly devoted to catching criminals, prosecuting them, convicting them and punishing them. Remarkably little thought, effort and public expenditure have until very recently been devoted to preventing crime from happening in the first place.3 The shortcom2 Following P J Brantingham and F L Faust, “A Conceptual Model of Crime Prevention”, (1996) 22 Crime and Delinquency 130–46. 3 See, for example, K Pease, “Crime Prevention” in M Maguire, R Morgan and R Reiner (eds), The Oxford Handbook of Criminology (Oxford, Oxford University Press, 1997).

Torture: Prevention Versus Punishment 137 ings of tertiary approaches are well known. A small minority of crimes are reported and cleared up, which means that most offenders get away with their offences most of the time and, perhaps not surprisingly, few of the horses typically caught after the stable door has been bolted show themselves to be amenable to significant change—not least because they are normally returned to precisely the same criminogenic environment from which they were plucked. The thematic focus of the present volume largely represents an extension of this preoccupation with punishment. The argument is that the culture of impunity regarding torture must be attacked and that this is best done by proscribing and condemning the practice, and by widening the opportunities for bringing torturers to book. If the prospects of torture being punished are increased, then the likelihood of recidivism will be reduced and potential torturers will be deterred. It is the orthodox criminal justice model onto which has been welded another recent criminal justice preoccupation—greater voice, support and compensation for victims. These tertiary model elements have been prominent in successive campaigns by the various human rights non-governmental organisations working against torture. Torture must officially be condemned, all reports of torture must be independently investigated, torture must be defined and prohibited in law; torturers must be prosecuted, and there should be parallel efforts within the United Nations and at a regional level to bring pressure to bear on governments accused of sponsoring torture.4 Moreover, victims of torture should be supported and compensated. This campaign has been highly successful in erecting an international prohibitionist framework.5 Following the prohibition contained in article 5 of the Universal Declaration of Human Rights (UDHR) and article 7 of the International Covenant on Civil and Political Rights (ICCPR), there is now the 1975 UN Declaration Against Torture and the 1984 Convention against Torture. The Convention against Torture obliges states parties, inter alia, to take “effective legislative, administrative, judicial and other measures to prevent acts of torture”.6 Such measures include making torture punishable as a crime 4 See, for example, Amnesty International, Torture in the Eighties (London, Amnesty International, 1984) and E Prokosch, “Amnesty International’s 12-Point Programme for the Prevention of Torture: An Example of NGO Standard Setting” in R Morgan and M D Evans (eds), Protecting Prisoners: The Standards of the European Committee for the Prevention of Torture in Context (Oxford, Oxford University Press, 1999), chapter 6 [hereinafter Morgan and Evans (1999)]. 5 For a general review of these developments, see M D Evans and R Morgan, Preventing Torture: A Study of the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment (Oxford, Clarendon Press, 1998), chapter 3 [hereinafter Evans and Morgan (1998)]; N Rodley, The Treatment of Prisoners Under International Law, 2nd ed. (Oxford, Oxford University Press, 1999), chapters 1 and 2. For a detailed commentary on the 1984 UN Convention against Torture, see J H Burgers and H Danelius, The UN Convention against Torture: A Handbook to the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (Dordrecht, Martinus Nijhoff, 1988). 6 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 [hereinafter “CAT”] at article 2.

138 M Evans and R Morgan of a “grave nature” and either extraditing alleged torturers or instituting criminal proceedings against them regardless of their nationality or where the crime was committed.7 The prohibition of torture is now an established part of customary international law, and is generally regarded as a norm of jus cogens. This has now been reinforced by the attempts made by the international community to construct mechanisms through which individuals who breach international law obligations attracting individual responsibility can be held to account before an international judicial process. In 1993 and 1994, the United Nations established the International Criminal Tribunals for the Former Yugoslavia and for Rwanda (ICTFY and ICTR). This was followed by feverish international activity culminating in the adoption of the Rome Statute of the International Criminal Court (ICC) in July 1998. The statutes of international instruments reinforce the prohibition on torture, as has the early jurisprudence under the ICTFY/ICTR, which has related to a considerable extent to acts of torture. The resources devoted to these activities have been immense. While the case for establishing international courts with criminal jurisdiction over alleged torturers is not, of course, dependent on narrow financial and numerical assessments, they nevertheless provide classic examples of tertiary interventions which are hideously expensive to erect and which deal with comparatively few offenders.8 During the last two decades, however, there has been growing interest in primary and secondary preventive initiatives. There have been campaigns to create safeguards for persons most vulnerable to torture, namely suspects detained in police custody. International standards for their treatment have been promulgated, and suspects’ putative rights have been laid out in various instruments. Thus, for example, incommunicado detention has been widely prohibited, and access to lawyers for persons in police custody has been made mandatory in many jurisdictions. Moreover, frameworks of accountability have been proposed, so that victims can have their complaints investigated and seek redress, and so as to ensure that police stations and other places of custody are independently inspected and supervised. Attention has also been given to the recruitment, training, management and terms of service of custodial staff so as to reduce the likelihood that they will resort to ill-treatment.9 The most recent and radical initiative designed to have primary and secondary preventive impact on the practice of torture was the European Convention for the Prevention of 7

CAT, arts 4 and 5–7. The 1999 budget for the ICTFY is $94,103,800 and it currently employs some 851 staff. Since its creation in 1993, the annual budgets for the ICTFY total some $280,000,000. To date (as of November 1999), the ICFTY has found eight individuals guilty. Of these, two pleaded guilty, and one has yet to be sentenced. Appeals are pending in the other six cases. One defendant has been acquitted on all charges brought; a prosecutor’s appeal of that acquittal is pending. The 1999 ICTR budget is $68,531,900, and it currently employs 688 staff. It has found five individuals guilty, two having pleaded guilty. 9 For a review of the UN Standard Minimum Rules for the Treatment of Prisoners, the European Prison Rules, the UN Code of Conduct for Law Enforcement Officials, and others, see Rodley, supra n. 5, chapters 9 and 12. 8

Torture: Prevention Versus Punishment 139 Torture and Inhuman or Degrading Treatment or Punishment (the ECPT or the Convention) which came into force on 1 February 1989.10 The ECPT creates a committee, the Committee for the Prevention of Torture (the CPT, or the Committee). In the section that follows, we briefly describe the mandate, composition, modus operandi and standards of the CPT.

3 THE EUROPEAN CONVENTION FOR THE PREVENTION OF TORTURE IN OPERATION

For the time being, the ECPT is unique among international human rights treaties.11 Rather than set out new—or, indeed, any—normative standards, it establishes a visit-based mechanism the purpose of which is to “examine the treatment of persons deprived of their liberty with a view to strengthening, if necessary, the protection of such persons from torture and from inhuman or degrading treatment or punishment”.12 The means by which the ECPT seeks to achieve its aims involves conducting visits to states party to it and drawing up a report that is sent to the state in question and which forms the basis for a dialogue between the CPT and the state concerned. Although reporting obligations are commonplace within the family of UN human rights instruments, these normally require a state to submit a report indicating the manner in which it is giving effect to the obligations contained in the relevant instrument. This is followed by an examination of the report in the course of a regular formal meeting of the treaty monitoring body, at which representatives of the state present their reports, answer questions, and receive “concluding observations” or “comments”. Although non-governmental organisations and others can feed material into the discussions, it is a relatively formal process over which the state exercises a considerable degree of control. This, for example, is the method of the UN’s Committee against Torture (CAT), as established under the UN Convention against Torture.13 The ECPT is very different. It is the CPT itself that is responsible for the production of the report upon which its dialogue with the state is conducted. In order to produce these reports, “[e]ach State shall permit visits . . . to any place within its jurisdiction where persons are deprived of their liberty by a public authority.”14 This is a much more wide-ranging authority than that exercised by 10 European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment, adopted 26 November 1987, ETS No. 126. 11 A Working Group of the UN Commission on Human Rights is currently working on a draft optional protocol to the United Nations Convention against Torture that would create a similar mechanism operative at the global level. 12 Supra n. 10, at article 1. 13 Supra n. 6. See, for example, A Byrnes, “The Committee against Torture” in P Alston (ed.) The United Nations and Human Rights (Oxford, Clarendon Press, 1992), chapter 13; R Morgan, “Preventing Torture and Protecting Prisoners”, (1998) 11 (4) Interights Bulletin 178–80. 14 Supra n. 10, article 2.

140 M Evans and R Morgan other human rights treaty bodies, which can only conduct visits subject to requirements of prior consent or other restrictions, or are limited to the investigation of particular situations of which they are seized.15 The CPT, by contrast, is free to roam where it pleases and has the right to inspect without warning any part of any place where people are detained by public order.16 For example, the CPT can, and does, turn up without warning at police stations in the middle of the night. Moreover, should it in the course of its visits encounter sufficiently serious situations, “the Committee may immediately communicate observations to the competent authorities of the Party concerned.”17 As a whole, the burden of responsibility for the effective functioning of the Convention rests on the CPT rather than on the states visited. The principal obligation of states is not to initiate but to facilitate and respond.18 This is a major advance. The ECPT is now in force for many more states than was originally envisaged. Moreover, recent signatories have introduced problems not originally encountered. As of November 1999, forty of the forty-one Council of Europe member states have signed and ratified the Convention, Georgia being the exception. Among the latest signatories are the Ukraine and the Russian Federation, both geographically vast countries with massive prison populations—both in absolute terms and relative to their populations. Whereas the fifteen states for whom the ECPT came into force in November 1989 had an estimated total of 292,250 prisoners, the forty states that are now members incarcerate an estimated 1,881,500 prisoners.19 When the ECPT was being drafted, part of its rationale lay in the idea that the Convention would set an example to the rest of the world. Many of the countries for whom that example was being set have now become parties to the Convention.20 It is now clearly understood that “ratification of the Convention is a sine qua non of member15 See, for example UNCAT Article 20 and the mandate of the UN Special Rapporteur on Torture. For a comparative analysis of the approach of these two mechanisms with that of the CPT, see R Bank, “International Efforts to Combat Torture and Inhuman Treatment; Have the New Mechanisms Improved Protection?” (1997) 8 EJIL 613–37. 16 Supra n. 10, articles 2 and 8. Article 9 allows for representations to be made against a visit being carried out on a limited range of grounds. This can only delay and not prevent a visit. This has not posed any real practical problems to effective operation of the visiting mechanism. 17 ECPT, supra n. 10, article 8(5). 18 ECPT, supra n. 10, article 3 of the ECPT provides that “. . . the Committee and the competent national authorities of the Party Concerns shall co-operate with each other.” 19 See Council of Europe, 9th General Report on the CPT’s Activities, CPT/Inf (99)12 (Strasbourg, Council of Europe, 1999) Appendix 1 (B), quoting Council of Europe Annual Penal Statistics based on 1 February 1990 and 1 September 1997. The original 15 member states were: Austria, Cyprus, Denmark, France, Ireland, Italy, Luxembourg, Malta, the Netherlands, Norway, Spain, Sweden, Switzerland, Turkey and the United Kingdom. The additional states which have since joined the convention system are: Albania, Andorra, Belgium, Bulgaria, Croatia, the Czech Republic, Estonia, Finland, Germany, Greece, Hungary, Iceland, Latvia, Liechtenstein, Lithuania, Moldova, Poland, Portugal, Romania, Russia, San Marino, Slovakia, Slovenia, The Former Yugoslav Republic of Macedonia, and Ukraine. It should be noted that these figures are for prison populations. The CPT’s mandate extends well beyond this, to include police establishments, detention centres for juveniles, military detention facilities, holding centres for aliens, psychiatric hospitals and so on. 20 See Evans and Morgan (1998), supra n. 5, chapter 4.

Torture: Prevention Versus Punishment 141 ship” of the Council of Europe.21 Indeed, the geographic scope of the CPT’s work may soon spread beyond the confines of the organisation, since it is expected that the First Protocol to the Convention will soon enter into force, permitting the Council of Ministers to authorise invitations to non-Council members to ratify the Convention.22 It is probably unwise to speculate at length as to which countries might be invited to consider joining, but the four observer states at the Council—Canada, Israel, Japan and the United States of America—might seem obvious invitees, if, in some cases, possibly reluctant ones. The site-visiting programme lies at the heart of the CPT’s working practices and provides the means by which it is able to test the congruence of state practice with the standards it propounds. Visits are conducted by delegations of Committee members, are assisted by members of the Committee’s secretariat and interpreters, who are usually accompanied by a small number of experts called in to assist by filling gaps in the expertise of the particular delegation. A typical visit might comprise up to five Committee members, as well as two members of the secretariat and two additional ad hoc experts. There are three sorts of visits. Periodic visits are made to all member states from time to time “as far as possible . . . on an equitable basis”. They typically last up to two weeks, during the course of which the CPT visits a range of institutions—police stations, prisons, psychiatric hospitals, immigration detention centres, and so on—in a variety of locations. Ad hoc visits are those “required in the circumstances”, and can be triggered by any circumstance that gives rise to a concern requiring fairly immediate attention.23 Ad hoc visits are sometimes arranged at very short notice indeed. Follow-up visits represent a sort of hybrid, and involve the CPT returning to an institution or institutions giving cause for concern during a periodic or ad hoc visit. The purpose is to see whether the situation has improved. Following each visit the CPT adopts a report setting out the delegation’s findings and making a series of “recommendations”, “comments” and “requests for information”. The report takes in the region of six months to be prepared and is then adopted by the whole Committee. It is thereafter transmitted to the state in question, which is asked to produce a written response within six months and, in the case of periodic visits, a second report within twelve months. There may be further communication based on these documents. In sum, this comprises the “dialogue” through which the CPT ’s preventive mandate is carried out. The work of the CPT is almost entirely confidential. Its plenary sessions are held in camera, and country inspection reports are transmitted in confidence to governments. These reports, the governments’ responses, and any other 21

9th General Report on the CPT’s Activities, supra n. 19, para. 13. See also infra n. 48. Protocol No. 1 to the European Convention for the Prevention of Torture and Inhuman Degrading Treatment or Punishment, adopted 4 November 1993, ETS No. 151. The First Protocol was opened for signature in 1993 and must be ratified by all states party to the ECPT before it comes into force. As at November 1999, the ratifications of Andorra, Croatia and the Ukraine were still awaited. 23 ECPT, supra n. 10, article 7(1). 22

142 M Evans and R Morgan communication with the state concerned become public only with the consent of government concerned. Furthermore, the CPT does not press the case for publication directly with the states themselves. The established practice is, however, one of openness. Every member state that has had a CPT report for long enough to reflect on its contents has permitted at least one report to be published.24 Most state parties have published all of the reports they have received together with their formal responses, so there is now a substantial corpus of CPT documents.25 There is, however, an exception to this rule of confidentiality. If the CPT finds evidence of torture or inhuman or degrading treatment, and the “Party fails to co-operate or refuses to improve the situation in the light of the Committee’s recommendations”, it can deploy its only formal “sanction” and issue a “public statement”.26 Despite often being colloquially referred to as “the Torture Committee”, the CPT has only found evidence of torture in relatively few reports and only slightly more often described as inhuman and/or degrading the practices and conditions it has inspected. Though the CPT has found widespread evidence of casual violence or excessive force, the ill-treatment has seldom been deemed both purposive and of sufficient severity to call it torture (reports to date on Austria, Bulgaria, Cyprus, Romania, Spain and Turkey are exceptions). The terms inhuman and degrading, employed either separately or together, have been reserved for forms of environmental ill-treatment, chiefly concerning the conditions in which groups of prisoners are housed. These are cases in which the purposive element is lacking or obscure, at least in terms of a particular individual.27 The bulk of the CPT’s reports are taken up with the articulation of those standards which it recommends be put in place to reduce the likelihood of torture and inhuman or degrading treatment occurring. It follows that the CPT is largely taken up with what we have earlier described as primary and secondary torture-preventive efforts. This is not to say, however, that it provides no grist for the tertiary prevention mill. There is a two-way process developing between the CPT and the European Court of Human Rights (the European Court) despite the fact that the framers of the ECPT stipulated that the recommendations of the Committee “will not bind the State concerned and the Committee shall not express any view on the interpretation of legal terms. Its task is a purely preventive one.”28 This stipulation only 24 Turkey has been the least open member so far, authorizing publication of only one out of eight reports received. 25 By the close of 1998, fifty-three visit reports and sixty-nine responses had been published. 26 ECPT, supra n. 10, article 10(2). So far, the Committee has issued only two Public Statements, in 1992 and 1996, both concerning Turkey. The First Public Statement was issued on 15 December 1992 as CPT/Inf (93) 1 and appears as Appendix 4 in Council of Europe, 3rd General Report on the CPT’s Activities, CPT/Inf (93) 12 (Strasbourg, Council of Europe, 1993) and is reproduced in (1994) 14 HRLJ 49 and (1995) 2 IHRR 251. The Second Public Statement on Turkey was issued on 6 December 1996 as CPT/Inf (96) 34 and is reproduced in (1997) 18 HRLJ 291 and (1997) 4 IHRR 509. 27 For a detailed consideration of these findings see Evans and Morgan (1998), supra n. 5, Chapter 4; Morgan and Evans (1999), supra n. 4 at 33–40.

Torture: Prevention Versus Punishment 143 makes sense if one were to accept the fanciful proposition that European states do not subject detainees to torture or house them in conditions that are inhuman and degrading. The CPT characterises its task as practical “conflict avoidance”, to distinguish itself from the legal “conflict solution” role of the European Court, but the fact is that most readers of CPT reports pay closest attention to whether it has found evidence of what it is prepared to call torture or inhuman or degrading treatment.29 It follows that, although the CPT may not be bound by the jurisprudence of the European Court, instead using that jurisprudence and other relevant standards as sources of guidance and points of departure, it must obviously be highly attentive to it. Equally, when, for example, the CPT finds—as it has done on several occasions—that prison conditions are inhuman and degrading, it greatly increases the likelihood that prisoners subject to those same conditions will petition the European Court claiming a breach of article 3 of the European Convention on Human Rights (ECHR). The judicial and the non-judicial mechanisms impact upon each other and it is this process that provides the point of departure for the remainder of this chapter. 4 PREVENTIVE MECHANISMS AND TERTIARY PROTECTION : THE POSITIVE ASPECTS OF THE RELATIONSHIP

The essential feature of the ECPT as a preventive mechanism is that the CPT has an unimpeded right of access to any place of detention by public order at any time and without giving any notice, and that it can interview and examine detainees in private. This means that it can acquire first-hand evidence of physical or mental torture or of ill-treatment. Some of the members and ad hoc experts who make up visiting delegations are always medical specialists able to produce compelling corroboration of allegations of ill-treatment. Similarly, when passing comment on the conditions of detention, the authority of the CPT is unimpeachable. It follows that the material contained in CPT visit reports 28 Council of Europe, Explanatory Report to the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment CPT/Inf/C(89) 1 (Strasbourg, Council of Europe, 1989), para. 25. The Explanatory Report was drawn up and adopted alongside the Convention text. There has been some dispute as to its status. The first President of the CPT, Antonio Cassese, argued that it was “an indispensable element of the proper interpretation and application of the Convention” (Letter to the Secretary General of the Council of Europe) but the Committee of Ministers placed less weight on it, taking the view that “The Explanatory Report is of great value for the interpretation of the Convention but it does not have the same value as the text of the Convention” (Report of the Minister’s Deputies’ Rapporteurs Group on Legal Co-operation). Cassese subsequently modulated his position, saying that “The CPT’s position was, and remains, that the observations in the explanatory report are part of the context in which the meaning of the terms used in the Convention is to be ascertained” (letter of 10 July 1991 to the Chairman of the Ministers’ Deputies). This exchange is reproduced in Council of Europe, Some issues concerning the interpretation of the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment, CPT/Inf/C (93) 10 (Strasbourg, Council of Europe, 1993). 29 Council of Europe, 1st General Report on the CPT’s Activities, CPT (91) 3, (Strasbourg, Council of Europe, 1991) at paras 2–3.

144 M Evans and R Morgan provides an excellent factual basis for those who are seeking to bring civil claims before domestic courts. In the course of our research into the practical impact of the CPT’s work, we have often been told by local NGOs in different countries that CPT reports did not reveal very much not already generally known. However, whereas claims that such practices did occur, or that an individual was the victim of such treatment, had previously met with a bureaucratic denial, or some drawn-out investigation that, if it ever reached a conclusion, tended to exonerate those allegedly responsible, the response was different when such claims were supported by the findings of a CPT report. Outright denial was no longer an option. The state could dispute the evaluation of such findings—by claiming, for example, that the incidents of ill-treatment discovered were isolated aberrations rather than routine practice, or that conditions, although undesirable, were not so adverse as to qualify as acts of torture or of inhuman or degrading treatment.30 But, whatever legal interpretation be placed on them, the facts themselves were established beyond doubt and could no longer be denied. As well as the evidential value of CPT reports, there is inevitably a broader impact upon the general understanding of what comprises torture and ill-treatment. It is axiomatic that the CPT’s views on this subject are not legally binding. Furthermore, as a preventive mechanism that, to use the Committee’s phrase, is intended to construct a “cordon sanitaire” around those taken into detention in order to reduce the likelihood of ill-treatment, it is equally clear that it will be articulating standards in advance of those currently recognised as reflecting the legal requirement.31 It is equally clear that the preventive standards advanced by the CPT today provide markers for what potentially represent possible breaches of the legal standards tomorrow. It is to be expected that the assessments made by the CPT in the course of its preventive work will inspire claims that failure to meet those standards, particularly regarding environmental conditions of detention, amount to violations of Article 3 of the ECHR at the international level and of corresponding prohibitions at the domestic. There is already evidence of this phenomenon at the international level. The findings of the CPT with regard to Turkey (known from Public Statements,) have been drawn on by both applicants and by the European Commission and Court of Human Rights to set the context in which specific allegations are examined.32 The construction of “background noise” that suggests that a claim 30

See n. 33, infra. See, for example, 1st General Report on the CPT’s Activities, supra n. 29, para. 3. This phrase and paragraph are generally found in the Preface to the first Periodic report sent to a state. 32 See, for example, Aydin v. Turkey, App. No. 23178/94, Comm. Rep. 7 March 1996, para. 183, where the Commission referred to the CPT’s findings, along with those of other international mechanisms as disclosing, “a consistent pattern of torture in custody and official tolerance on the part of the Turkish authorities”. The Court also took note of the by then two Public Statements on Turkey, but drew a different conclusion; that the totality of evidence established by the Commission was “insufficient to allow it to reach a conclusion concerning the existence of any administrative practice of the violation of [the relevant Convention] Articles”. Judgment, 25 September 1997, paras 49–50 and 121–4, RJD 1997-VI, p. 1866; 25 EHRR 251. 31

Torture: Prevention Versus Punishment 145 is indeed credible can be of considerable significance. There are also examples of the CPT’s findings prompting, or at least refocusing, elements of applications brought under the ECHR. In what has now become a well-documented saga, the CPT expressed the view, in the report on its first visit to the United Kingdom, that the combination of overcrowding, absence of out-of-cell activities and the lack of integral sanitation within cells at three prisons (Brixton, Wandsworth and Leeds) amounted to inhuman and degrading treatment.33 The facts were common knowledge but, predictably, this evaluation of their cumulative impact was disputed by the UK government.34 The views of the CPT were, however, drawn upon in two applications brought under the ECHR concerning conditions at Wandsworth prison. Although the evaluation was eventually declared inadmissible, it pointed to the potential relevance of CPT findings in such cases, as well as the potential for tension between the CPT and the ECHR institutions.35 Two applications concerning Greece illustrate the potential and the problems. In 1993 the CPT had visited Larissa prison in central Greece, where the conditions of detention were described as “totally inappropriate”.36 Subsequently, in Tosunoglu v. Greece, the Commission, in declaring the application inadmissible under the ECHR, noted that the CPT Report had not gone so far as to describe the conditions as inhuman or degrading.37 During the same visit the Committee had visited the Korydallos prison complex outside Athens. Although this facility was the subject of strong criticism, the Committee fell short of describing it as inhuman or degrading. However, in Peers v. Greece, the Commission exercised an independent view, declaring admissible an application that argued that these conditions were in breach of article 3, 33 Council of Europe, Report to the United Kingdom Government on the Visit to the United Kingdom carried out by the CPT from 29 July 1990 to 10 August 1990, CPT/Inf (91) 15, (Strasbourg, Council of Europe, 1991), paras 47 and 57. 34 The UK Government agreed that the conditions at the three prisons “needed considerable improvement”, but argued that they “were not so poor that the prison authorities could be said to be treating in an ‘inhuman and degrading manner’ ”: Council of Europe, Response of the United Kingdom Government to the Report of the CPT on its Visit to the United Kingdom from 29 July 1990 to 10 August 1990, CPT/Inf (91)16 (Strasbourg, Council of Europe, 1991) para. 6. 35 In Raphaie v. UK, App. No. 20035/92, Comm. Dec. 2 December 1993 (unpublished), the Commission noted that the applicant’s factual allegations were “for the most part substantiated” by reports of both the Chief Inspector of Prisons and of the CPT. However, the application was declared inadmissible, since it had been lodged more than six months after the exhaustion of local remedies. In Delazarus v. UK, App. No. 17525/90, Comm. Dec. 16 February 1993 (unpublished), the Commission dismissed the application because the applicant had been held in a single cell (in the segregation unit) during his time at Wandsworth prison and so not been subjected to the overcrowding which comprised one of the elements which, in the CPT’s view, combined to produce the violation. Nevertheless, in Peers v. Greece, infra, n. 39, the Commission concluded that being held for a considerable portion of each day and all of the night for prolonged periods (at least two months) in a cramped shared cell which was hot and lacked ventilation and in which both prisoners had to use a toilet in each other’s presence, amounted to degrading treatment (para. 96), and quoted the CPT’s report on its first visit to the UK (supra n. 33) in support of this. 36 Council of Europe, Report to the Government of Greece on the Visit to Greece carried out by the CPT from 14–26 March 1993, CPT/Inf (94) 20 (Strasbourg, Council of Europe, 1994) para. 119. 37 Tosunoglu v. Greece, App. No. 21892/93, Comm. Dec. 12 April 1996 (unpublished).

146 M Evans and R Morgan notwithstanding the CPT’s failure to make a finding in terms of the standards in article 3.38 Delegates of the Commission then visited the relevant sections of the prison in order to make their own assessment of the conditions and concluded that article 3 had indeed been breached.39 It seems, then, that whilst they are not determinative, the findings of the CPT—a preventive and non-judicial agency—have both evidential and jurisprudential impact upon the determination of legal claims, and this is as likely to be as true at the domestic level as it is at the international level. Indeed, what might be called the “jurisprudential pull” of the preventive standards articulated by the CPT and other such bodies has recently received a potential boost from the European Court itself. In the recent case of Selmouni v. France, the European Court has said: “Certain acts which were classified in the past as “inhuman and degrading” as opposed to “torture” could be classified differently in future. [T]he increasingly high standard being required in the area of the protection of human rights and fundamental liberties correspondingly and inevitably requires greater firmness in assessing breaches of the fundamental values of democratic societies.”40

It seems clear that preventive mechanisms will have a role to play in educating the Courts on the forms of ill-treatment that may amount to a violation. It follows that preventive mechanisms may prompt more, and novel, claims to be brought before both international and domestic courts.

Prevention and Civil Redress There is also a more subtle, but perhaps more significant, contribution. The CPT has always understood that prevention must be balanced by redress for victims of torture. Though it is not in a position to provide redress, the CPT certainly considers it within its preventive mandate to ensure that there are adequate systems of accountability. The CPT calls for effective grievance and inspection procedures for all detainees, including prisons and police custody.41 It seeks information on the relevant mechanisms in the countries visited, recommending that the establishment of such systems be explored where they do not exist and examining the guarantees of their objectivity and independence 38 Peers v. Greece, App. No. 28524/95, Comm. Dec. 21 May 1998 (unpublished). This is the admissibility decision, while the citation in n.39, infra, is to the report on the merits. 39 Peers v. Greece, App. No. 28524/95, Comm. Rep. 4 June 1999. The Report noted that the Delegates’ findings varied from those of the CPT in part, but concluded that “The Commission takes into account the fact that the Delegates investigated the applicant’s complaints in depth having given special attention, during their in loco inspection, to the conditions in the place where the particular applicant had been detained. In these circumstances, the Commission considers that the findings of its own Delegates should be relied on.” (para. 92). 40 Selmouni v. France, European Court of Human Rights, Judgment, 28 July 1999, para. 101. 41 Council of Europe, 2nd General Report on the CPT’s Activities, CPT/Inf (92) 3, (Strasbourg, Council of Europe, 1992) paras 41 and 54.

Torture: Prevention Versus Punishment 147 where they do. Information is also regularly requested on the use made of complaints mechanisms and questions are raised when there is a stark mismatch between the number of complaints investigated and the resulting number of disciplinary sanctions taken.42 Moreover, it urges that such domestic agencies produce annual reports. Were state agencies to act on this recommendation, their reports might emulate the practice of some inspectorates the CPT has praised for being open, detailed and revealing.43 Indeed, such reports are sometimes more detailed, based on more systematic examination and thus even more penetrating than those of the CPT itself.44 They also have the potential to be more frequent: the CPT is unable to conduct periodic visits to a country more than once every four or five years and, although it may well find it necessary to conduct shorter ad hoc or follow-up visits in the interim, it is certainly true that, for most state parties, a CPT visit and report will be a relatively rare event. Encouraging the establishment and effective functioning of such mechanisms at the domestic level is an effective way of alleviating this problem, provided that the bodies in question are able to acquire a similar reputation for independence. This process has the general effect of challenging the culture of impunity that can so easily grow up with respect to ill-treatment in detention. It encourages and fosters an expectation that proceedings will be taken—either as an administrative matter of internal discipline or before the courts. Although the CPT is not directly concerned with the extent to which these proceedings take the form of civil actions—and it may be imagined that it will principally be concerned to ensure that appropriate criminal action is taken against those responsible,45 the overall result will doubtless be of general assistance to those seeking to do so.46 The more that the courts become accustomed to hearing and commenting on examples of ill-treatment, the more receptive they are likely to become, and the more likely it is that resistance offered by the authorities to such actions will 42

For details see Evans and Morgan (1998), supra n. 5, at 291–3 and 313–14. See Report to the United Kingdom Government on the visit to the United Kingdom carried out by the CPT from 29 July 1990 to 10 August 1990, supra n. 33, para. 203. 44 Ibid. 45 Cf. Convention against Torture, article 4, which provides that “(1) Each State Party shall ensure that all acts of torture are offences under its criminal law . . . (2) Each State Party shall make these offences punishable by appropriate penalties which take into account their grave nature.” (emphasis added) But cf. article 14(1), which provides that “Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation.” This could be read as referring to the possibility of civil actions, although it stops short of stating this clearly. 46 For example, during their visit to Cyprus in 1992, the CPT heard several allegations of, and found evidence supporting, the use of torture by the Limassol police. The Committee called for an independent inquiry to be established. The government of Cyprus established such an inquiry, which upheld the CPT’s findings. Prosecutions of police officers followed. When the CPT next visited Cyprus, in 1996, they received no allegations of torture: Council of Europe, Reports to the Government of Cyprus on the visits to Cyprus carried out by the CPT from 2 to 9 November 1992 and 12 to 21 May 1996, CPT/Inf (97) 5, (Strasbourg, Council of Europe, 1997) paras 13–22 (1992 visit) and paras 9–10 (1996 visit). 43

148 M Evans and R Morgan decrease. It should increasingly be recognised that the overall interests of the law enforcement agencies are being served through the successful prosecution of those who exceed the bounds of legitimacy, both in terms of physical illtreatment and in terms of responsibility for unacceptable custodial conditions and practices. In a sense, the preventive function, on the one hand, and prosecutions and civil litigation, on the other hand, become supportive of each other. The lack of, or inability to take, appropriate action before the courts can be seen as evidence of a lack of appropriate, or appropriately functioning, grievance procedures. The very fact of actions being taken before the domestic courts stands as clear evidence of the need for the fuller implementation of preventive safeguards, of the need to re-erect the cordon sanitaire at a place where it provides greater protection from the sources of violation. As previously argued, the jurisprudential pull of such safeguards, no matter where they are erected, is likely to produce a shift in the understanding of what comprises ill-treatment and thus the process will continue to evolve. Prevention and domestic proceedings, whether criminal or civil in nature, should combine to promote an evertightening system for the eradication of ill-treatment.

5 PREVENTIVE MECHANISMS AND TERTIARY PROTECTION : THE NEGATIVE ASPECTS OF THE RELATIONSHIP

The preceding is one version of the story. However, another, less documented version may be as compelling. An increased risk of legal action might result in increased secrecy and obstruction and thus impede what we have called primary prevention.

A Decrease of Transparency as a Response to Potential Liability Abroad We have noted that the hallmark of the CPT is its ability to intrude into what its first President, Antonio Cassese, describes as the sancta sanctorum of a state, its security apparatus.47 The trade-off for granting such unparalleled rights of access to an international human rights body lies in the confidentiality of the reports it produces. But, since so many states have authorised the publication of CPT visit reports, publication has become an expectation. Indeed, the positive aspects of the relationship between the differing approaches to torture prevention outlined in the previous section presuppose publication. Whether this expectation is legitimate may be questioned, however. Many of the states that have most recently joined the convention system were “forced participants”, in that their membership in the Council of Europe was conditional on their accep47 A Cassese, Inhuman States: Imprisonment, Detention and Torture in Europe Today (Cambridge, Polity, 1996) at 1.

Torture: Prevention Versus Punishment 149 tance of the ECPT and a raft of other human rights instruments.48 It is still too early to say, but it would be foolish to assume that all will agree to publish CPT reports as a matter of course. Turkey—ironically the first state to ratify the Convention—has been the most reluctant to publish and others—such as Spain and Cyprus—have been very slow to do so.49 If the newer states do indeed display greater reluctance to publish, then it is possible that others that have previously fallen in line with the general trend of speedy publication may choose not to do so and a counter-momentum may be established. Were this to happen, it would be a major blow to the effective operation of the Convention system. As we have said, the CPT is faced with an enormous task 48 When a state applies for membership of the Council of Europe, its human rights record is subject to close scrutiny. It is now the standard practice of the Parliamentary Assembly of the Council of Europe to include in its recommendation to the Committee of Ministers on an application for membership a detailed list of undertakings it considers necessary, and for the Committee of Ministers to reflect this in its Resolution extending membership to the applicant state. The details may vary somewhat, but it is now standard practice to require that the ECHR be signed at the time of accession and ratified within one year, and that the ECPT be signed and ratified within one year of membership. For example, the most recent state to join the Council was Georgia. Parliamentary Assembly Opinion 209 (adopted 27 January 1999) provided that

“10. The Parliamentary Assembly expects Georgia to undertake: i. with regard to conventions a. to sign the European Convention on Human Rights (ECHR), as amended by its Protocols Nos. 2 and 11, at the time of accession; b. to ratify the ECHR and Protocols Nos. 1, 4, 6 and 7 thereto within a year after its accession; c. to sign and ratify, within a year after its accession, the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment and its protocols; d. to sign and ratify, within a year after its accession, the Framework Convention for the Protection of National Minorities and the European Charter for Regional or Minority Languages; e. to sign and ratify, within three years after its accession, the European Charter of Local SelfGovernment, the European Outline Convention on Transfrontier Co-operation and its additional protocols and the Council of Europe conventions on extradition, on mutual assistance in criminal matters, and on laundering, search, seizure and confiscation of the proceeds from crime, and in the meantime to apply the fundamental principles of these instruments; f. to sign and ratify the Council of Europe’s Social Charter within three years after its accession and, in the meantime, to endeavour to implement a policy in accordance with the principles it contains; g. to sign and ratify the General Agreement on Privileges and Immunities of the Council of Europe and the protocols thereto within a year after its accession; h. to sign and ratify the Geneva Convention relating to the Status of Refugees and the 1967 Protocol thereto within two years after its accession . . .” The Committee of Ministers Resolution No 99(4) of 27 March 1999 which extended the formal invitation to Georgia was made “In the light of the commitments entered into by Georgia, as set out in Opinion No. 209 (1999) by the Parliamentary Assembly and the assurances for their fulfillment given by the Georgian Government in the letter by the Minister for Foreign Affairs of 18 March 1999, in reply to the letter by the Chairman of the Committee of Ministers of the Council of Europe of 16 March 1999, with a view to membership of the Council of Europe”. Georgia duly became a member of the Council of Europe on 27 April 1999. At the time of writing it had not yet signed or ratified the ECPT. 49 Spain did not authorise publication of the report arising from the CPT visit in 1991 until 1996 and the report on the 1992 visit to Cyprus did not appear until 1997.

150 M Evans and R Morgan spanning forty countries (including the overseas territories of France and the Netherlands) and including a prison population fast approaching, if not already exceeding, two million persons. It cannot aspire to visit more than a handful of custodial sites during each of its infrequent visits to most countries. Moreover, the visit is only a means to an end, this being the prevention of torture and illtreatment. The intermediate step between the visit and achieving that end is the dialogue ideally following each visit. The CPT has admitted that it has problems in this regard, for want of resources, and that “the absence of a sustained postvisit ongoing dialogue” would fritter away “the momentum for change”.50 In such circumstances, the baton must pass to others. In the course of our research, it has become apparent that domestic bodies—NGOs, political parties, academics, the media and others—play an important role, and in some countries a vital role, in pressing the case for the adoption of CPT recommendations. The content of a CPT report can have considerable impact when it is fed into an ongoing stream of policy discussion or is used as a platform on which to build a case for change. This is particularly true in those states in the process of reconfiguring their criminal justice systems, particularly those in the former Soviet empire. Should such states decide in future not to publish CPT Reports, or to seriously delay publication, this would have a serious impact on the effectiveness of the preventive mechanism as a whole.51 Delay, for example, tends to mean that findings of ill-treatment become a spent force: the circumstances are said no longer to apply, thereby making current claims more difficult to substantiate. It is, then, important that at this critical point in the evolution of the Convention system that nothing should be done to tip the balance against the ascending practice of reasonably prompt disclosure. The problem posed by having ever greater recourse to domestic courts, criminal or civil, for claims in relation to torture or other forms of ill-treatment is that it could encourage just such a negative effect: governments battening down the informational hatches.

Barriers to Entry: Disincentives for States to Join the Preventive System Another, and potentially even more serious, danger flowing from an increase in transnational civil litigation and criminal prosecutions is the effect that it might 50 Council of Europe, 5th General Report on the CPT’s activities, CPT/Inf 95 (10), (Strasbourg, Council of Europe, 1995) para. 10. For an example of what is suggested was an opportunity for reform, missed because the CPT failed to follow up its report with regard to the Netherlands with correspondence pressing for further information, see P Van Reenen, “Inspection and Quality Control: The CPT in the Netherlands” in Morgan and Evans (1999), supra n. 4, chapter 10. 51 It might be objected that the basic standards the CPT demands are now well known, and that it is therefore unnecessary for them to be published in the context of each individual country report. This is increasingly true. However, it appears to us that, although the existence and relevance of the CPT’s standards are increasingly known, their impact is fully felt only in the context of a particular state report. Standards extrapolating across the range of materials generated in relation to other states, or formulated in the abstract, do not have the same appeal to domestic forces, governmental or otherwise.

Torture: Prevention Versus Punishment 151 have on the prospects for the adoption and successful operation of similar preventive mechanisms in other regions or at the global level. In this regard we need to distinguish among the different types of situations in which torture and other forms of ill-treatment occur. In the majority of countries currently visited by the CPT, the increased likelihood of court actions is not likely to be a particularly important factor. With a few exceptions, most states within the European system, for example, do not employ torture or deliberately subject detainees to inhuman treatment. This is not to say that torture does not happen. It does. But the evidence suggests that in most countries it is a rogue practice not encouraged by state authorities, either explicitly or implicitly by turning a blind eye. Thus, when unequivocal evidence brings the practice to light, whether the source is the CPT or any other, the authorities are generally willing to remedy what they regard as an embarrassing blemish, through the courts or by other means. In such circumstances, there is little likelihood of, or need for, transnational tortious litigation to address the issue since there is no problem of domestic impunity. Prevention and action through the domestic civil or criminal courts co-exist and support each other, as in the model outlined in the previous section. At the other end of the spectrum lie a very different group of states, those that systematically employ torture and deliberately subject detainees to inhuman treatment in defiance of their international obligations. It is in these countries that the victims of torture are least likely to receive redress, and see their torturers punished, before the domestic courts. There is little likelihood of such states either accepting or co-operating with preventive mechanisms such as the CPT. It is this unhappy situation which prompts arguments in favour of a universalised criminal jurisdiction over torturers, as provided for in the UN Convention against Torture, and the search for appropriate routes through which the victims of torture can seek redress through civil claims for damages. This is the true “target group” whose interests inspire the claim for transnational tort litigation with respect to torture, as exemplified by the landmark case of Filártiga v. Peña Irala and the numerous cases that have followed in its wake, mentioned elsewhere in this volume.52 However, not all states are so easy to classify and many—arguably most—fall into a grey zone located between these poles. These are states in which torture does occur and in which, although not officially sanctioned, the practice is tacitly condoned or not actively prosecuted. Many states will join the chorus of condemnation but be slow to respond to allegations, fail to take adequate measures to punish those responsible for torture and not put in place those procedures that would reduce its incidence. Such states represent the key target group for preventive mechanisms, such as the CPT, since there is both the scope and the opportunity for making a major contribution to the prevention of torture. But it is necessary to win the support of these states in order that progress be made and it could well be that an increased likelihood of civil actions, 52

Filártiga v. Peña-Irala, 630 F 2d 876 (1980), 77 ILR 169.

152 M Evans and R Morgan particularly transnational actions, may be a powerful disincentive for such states to participate in such a system. The problem is simple. If such countries were to give an international body free access to places of detention, and to permit open dissemination of the resulting documentation, these records might be used in legal proceedings (civil or criminal, domestic or transnational) against those responsible for detention policies and procedures.

6 CONCLUSION

In those countries where torture and ill-treatment happen infrequently and are not authorised—officially or unofficially—as a matter of state policy, the ability to bring domestic actions to secure compensation when this is not already forthcoming from the state, for this will often be the case, is a valuable addition to the range of responses. It is also a valuable failsafe in cases where the evidence does not meet the burden of proof necessary in criminal cases (and where this is a requirement to trigger state-funded compensation) but may be adequate for civil proceedings. Such proceedings are likely to complement, and certainly not complicate, the work of preventive mechanisms such as the CPT. However, such cases are unlikely to be the subject of transnational tortious actions, for the very reason that they can successfully and fairly be brought before the most appropriate forum, the courts of the country in question. The question of transnational tortious litigation really only arises in the context of those states in which torture and ill-treatment are undertaken at the behest of the state. These are the states where it is most difficult for the preventive agencies to operate, but at the same time it is in precisely these states that it is most crucial that preventive agencies should begin to operate. It is for this reason that the principle of confidentiality surrounding CPT reports is important. It offers states the guarantee that these findings will not be used against them. Yet, as we argued above, unless the findings are placed in the public domain, the full potential of the preventive mechanism is unlikely to be realized. It may be objected that these concerns are misplaced, since the true international pariah states are unlikely to ratify a convention that allows for intrusive visits and that the exercise of international criminal, or transnational civil, jurisdiction is the only alternative. There is certainly truth in this. Even were such states to participate in preventive mechanisms, their co-operation would likely be more symbolic than real. That is scarcely the point, however. States cannot easily be grouped into categories of “good” or “bad” regarding human rights practice. There is a broad spectrum, and perceptions of where a state lies along that spectrum vary over time: just because the political elite in a country is generally regarded as acceptable today does not necessarily mean that it will be considered to be so tomorrow, and vice versa. There are many countries that know that they occupy the borderlands of international acceptability, either in general or as regards particular practices. It is in these states that most might be

Torture: Prevention Versus Punishment 153 achieved by the sensitive operation of visit-based prevention mechanisms, but whose leaders might be unwilling to gamble on opening their domains up to international inspection if they believed there to be a serious risk that the material flowing from the exercise might come back to haunt them personally at an unspecified time and place in the future. This, then, is the dilemma: is it desirable to encourage transnational litigation in order to to pursue justice for victims of systematic human rights violations, but, in doing so, take the risk of discouraging states from proper participation in preventive mechanisms? This need not be an either/or choice: a suitably nuanced approach may maximise the potential of both means of addressing the evil of torture. But it would be foolish to deny that there is a potential tension between these different forms of action at the international level, and that decisions as to the manner in which these conflicts are to be resolved raise difficult and delicate issues.

6

Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality ANNE C. McCONVILLE 1

1 INTRODUCTION H I L E T H E I N T E R N A T I O N A L legal community condemns egregious violations of basic human rights, such as torture and slavery, it provides limited effective remedies for individual victims of such conduct. War crimes prosecutions are one avenue of redress for cases with the correct factual and procedural components. Another avenue, relatively unexplored in Canada, is an action in tort for a civil remedy. Unlike a criminal prosecution, a civil action based on a human rights tort makes it possible for a victim to receive compensation from his or her abuser. Such actions, therefore, are a means simultaneously to achieve inter-party justice and vindicate international human rights. This chapter examines whether or not an Ontario court has jurisdiction to entertain an action for damages against perpetrators of systemic human rights violations in other countries.2 It considers, first, the jurisdictional framework under private and public international law and, second, the rules of service and basic jurisdiction in Ontario, including the doctrine of forum non conveniens, and it marshals arguments to overcome jurisdictional obstacles to transnational

W

1

Anne C. McConville is an associate of Fasken Martineau DuMoulin LLP, Toronto, Canada. In Canada’s federal system, private law and the rules of civil procedure are generally matters of provincial jurisdiction. Leaving aside the complication of the jurisdiction of the Federal Court of Canada over some fields of federal law, the federally-appointed courts of each province apply and interpret the law(s) of that province as well as federal law within that province’s territory. Each province has a court of appeal at the pinnacle of its judicial structure and judgments of the various courts of appeal may be appealed to the Supreme Court of Canada (SCC) with the latter’s leave. 2

158 A C McConville human rights tort litigation. At present, common law and statutory rules of jurisdiction provide sufficient scope for plaintiffs who are resident in Ontario to bring such cases in the Ontario courts, and international human rights norms support the appropriateness of Ontario providing a forum for transnational tort litigation. The approach of the Supreme Court of Canada (SCC) to conflict of laws, however, is centred on the territorial limits to the jurisdiction of the provincial courts and, concomitantly, on international comity. In light of this emphasis in the SCC’s recent decisions, the Ontario courts may be reluctant to exercise jurisdiction extraterritorially through private law actions derived from, or bolstered by, public international law norms. This is particularly so given that these public international law norms interact with the private international law principle of comity and together they reinforce the principle that there are territorial limits to the competence of a court to hear cases.

2 FACT SITUATION

This investigation of the Ontario courts’ jurisdiction to hear transnational human rights torts cases has a specific factual context. The focus of the study is a prospective suit in Canadian courts against two cabinet ministers of the government of Sudan. Reports of the UN Special Rapporteur and other human rights monitors provide evidence that the government regime in the Sudan is responsible for serious human rights abuses, including torture and slavery. In March 1997, the UN Special Rapporteur on Torture reported that torture is an extensive problem in the Sudan. Government security forces regularly torture, beat, harass, arbitrarily arrest and detain opponents or suspected opponents of the government and are responsible for extrajudicial killings and disappearances.3 The UN Special Rapporteur also reports that, although Sudanese law prohibits forced or compulsory labour, slavery persists in the Sudan.4 Government security forces are responsible for forced labour, slavery and forced conscription of children. The evidence also suggests that the government has failed to halt slave-trading practices in the country. Slaves are taken from war zones in the south and exported to central and northern Sudan. There are also reports of women and children being sold and sent to the north or abroad to work as domestic servants, agricultural labourers or concubines.5 The Report of the Canadian Assessment Mission to the Sudan, which was released in January 2000, confirms the persistence of slavery, or analogous prac3 UN Special Rapporteur on Torture, cited in US Department of State, Sudan Country Report on Human Rights Practices for 1997, Bureau of Democracy, Human Rights, and Labour, 30 January 1998, at 2 [hereinafter “Sudan Report”]. See also US Department of State, Sudan Country Report on Human Rights Practices for 1998, Bureau of Democracy, Human Rights, and Labour, 26 February 1999 at 4 [hereinafter “Sudan Report 1998”]. 4 Sudan Report, ibid. at 14; Sudan Report 1998, ibid. at 18. 5 Sudan Report 1998, ibid., supra n. 3.

Taking Jurisdiction 159 tices, in the Sudan.6 The Harker Report concluded that the allegation of slavery is not a sensational claim but a matter of fact that is exacerbated by such factors as the ongoing civil war and the internal displacement of people, activities in which the Sudanese government is directly or indirectly implicated.7 The Harker Report estimates that as many as fifteen thousand women and children have been abducted from communities in the northern region of Bahr El Ghazal alone and are probably being exploited as slaves.8 Based on this evidence, the state of Sudan is in breach of some of the most basic international human rights obligations and has thereby incurred direct responsibility for torture as well as, at minimum, indirect responsibility for failing to prevent slave-trading by nonstate actors.9 Several victims of such abuses attributable to the Sudan now reside in Ontario and, for purposes of this chapter, will be considered potential plaintiffs in a transnational human rights tort suit against certain members of the Sudanese governmental regime: Sudan’s Minister of the Presidency, who directs part of the security apparatus that engages in the practice of torture, and the Attorney General, who oversees a law enforcement apparatus that fails to call torturers and slave-traders to account. According to the best available information, both men are Canadian citizens, having acquired that status during their period as graduate students in Canada. However, it shall be assumed that both have had very limited contact with Canada since returning to their country of origin to take Cabinet posts in the Turabi regime. This chapter will address whether an Ontario court can appropriately take jurisdiction in a civil tort action against the ministers in their personal capacities. While the Sudanese situation provides a concrete context in which to consider the jurisdictional issues that transnational tort litigation raises, this chapter attempts to reach conclusions that apply more broadly to other similar types of transnational human rights litigation, including cases in which prospective defendants have no citizenship link to Canada.

6 J Harker, Human Security in Sudan: The Report of a Canadian Assessment Mission (Ottawa, Department of Foreign Affairs and International Trade, 2000) [hereinafter Harker Report]. 7 Ibid. at 29, 32, 45–9. 8 Ibid. at 25. 9 Sudan has signed and ratified the following international conventions which proscribe slavery and slave-trading: Slavery Convention, 25 September 1926, 212 U.N.T.S. 17; International Labour Organization Convention (No. 29) Concerning Forced or Compulsory Labour, 28 June 1930, 39 U.N.T.S. 55; Supplementary Convention on the Abolition of Slavery and the Slave Trade and Institutions and Practices Similar to Slavery, 7 September 1957, 266 U.N.T.S. 3; International Labour Organization Convention (No. 105) Concerning the Abolition of Forced Labour, 25 June 1957, 320 UNTS 291. Sudan has signed but not ratified the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027. [hereinafter “CAT”]. Sudan has also signed and ratified the International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976) [hereinafter “ICCPR”], which prohibits both torture and slavery.

160 A C McConville

3 JURISDICTION : FRAMEWORK CONSIDERATIONS

Territoriality Norm of Jurisdiction According to the principles of international law, a territorial nexus is generally required for appropriate exercises of judicial jurisdiction in civil cases. Jurisdiction to adjudicate should accordingly conform to the twin presumptions that jurisdiction is territorial and that respect for state sovereignty involves nonintervention within the domestic or territorial jurisdiction of other states.10 From the perspective of international law, the power and authority of a court to hear and determine an issue emanates from the sovereignty of the state that created the court. Since civil jurisdiction is an aspect of state sovereignty, its limits are also tied to the territorial boundaries of the state.11 When civil matters involve extraterritorial elements, states can rely on the principle of comity and private international law rules, in addition to domestic law, to deal with them.12 Public international law may assist or constrain states in this process. The presumption of territoriality is evident in the English common law system where civil jurisdiction is closely related to the notion of territorial sovereignty. Since the state has authority over persons who are present in its territory, the English courts have jurisdiction “as of right” over defendants who are served with a writ of process within the territory. Dicey and Morris describe the importance of service in this way: “[W]henever a defendant can be legally served with a writ, then the court, on service being effected, has jurisdiction to entertain an action against him. Hence, in an action in personam, the rules as to legal service of a writ define the limits of the court’s jurisdiction.”13

10 Charter of the United Nations, 26 June 1945, CanTS 1945 No. 7, 59 Stat. 1031, 145 UKFS 805, article 2; I Brownlie, Principles of Public International Law, 3rd ed. (Oxford, Clarendon Press, 1979) at 309–10 [hereinafter “Brownlie”]; Restatement of the Law Third, Restatement of the Law, The Foreign Relations Law of the United States (St. Paul, Minn., American Law Institute Publishers, 1987), §421 (jurisdiction to adjudicate). 11 J G Castel, Canadian Conflict of Laws, 4th ed. (Toronto and Vancouver, Butterworths, 1997) at 19 [hereinafter “Castel”]; C MacLachlan and P Nygh, Transnational Tort Litigation: Jurisdictional Principles (Oxford, Clarendon Press, 1996) at 10 [hereinafter “MacLachlan and Nygh”]; H M Kindred, International Law Chiefly as Interpreted and Applied in Canada, 4th ed. (Toronto, Emond Montgomery Publications Limited, 1987) at 457 [hereinafter “Kindred”]. 12 K C Randall, Federal Courts and the International Human Rights Paradigm (Durham and London, Duke University Press, 1990) at 156 [hereinafter “Randall”]. See text infra in this section for a discussion of additional public international law principles of jurisdiction that govern international cases in the criminal law context. It should also be noted that there are exceptions to states’ freedom in civil matters, including rules regarding state and diplomatic immunities and treaties on private international law. 13 L Collins (ed.), Dicey and Morris on The Conflict of Laws, 12th ed. (London, Stevens, 1993) at 270–1.

Taking Jurisdiction 161 This “jurisdiction as of right” extends to defendants who are not residents but are physically present in England, even if on a transient basis.14 Once a defendant enters the territory of another state, however, he or she is subject to the authority of that state and is beyond the English court’s adjudicative jurisdiction as of right. In this traditional approach, service of an English writ in a foreign state constitutes an invasion of that state’s sovereignty.15 While it is possible to effect service outside of England under Order 11 of the Supreme Court rules, this can only be done with leave of the court where the plaintiff discharges a heavy burden to show that the claim falls within an enumerated category and that England is clearly the appropriate forum for the action.16 Even in these cases of assumed jurisdiction, there must be a territorial nexus linking the cause of action, the parties and England. This narrow rule for service ex juris reinforces the primacy of the territorial principle in the common law and the idea that departures from the rule are exorbitant.17 Territoriality is also the norm that governs civil jurisdiction in common law Canada.18 Following the English tradition, service of process on a defendant in the territory of the court gives the court jurisdiction as of right. Most of the Canadian provinces also provide broad statutory rules for service ex juris in cases where there is territorial nexus. In Canada, though, the territorial principle operates at two levels. First, on the international level, it relates to the public international law principles of state sovereignty and extraterritoriality. This international aspect is evident in the importance the Canadian courts attach to comity in multi-jurisdictional cases. Second, on the federal level, territorial limits to the provincial courts’ jurisdiction are required as a constitutional imperative.19 For a court to properly exercise its authority, a real and substantial connection must exist between its province and the action.20 The international 14 Maharanee of Baroda v. Wildenstein [1972] 2 QB 283 (CA); Adams v. Cape Industries plc [1990] Ch. 433 (CA); M Baer, Private International Law in Common Law Canada (Toronto, E. Montgomery, 1997) at 184 [hereinafter “Baer”]; G Saumier, “Judicial Jurisdiction in International Cases: The Supreme Court’s Unfinished Business” (1995) 18 Dal L J 447 at 450 [hereinafter “Saumier”]. 15 Saumier, supra n. 14, at 450. 16 Spiliada Maritime Corp. v. Cansulex [1987] 1 AC 460, [1986] 3 WLR 972, [1986] 3 All ER 843 (HL) [hereinafter Spiliada]. Service ex juris was introduced in England in 1852 with the Common Law Procedure Act, now Order 11 of the Supreme Court. Order 11 enumerates the categories of claims where leave to serve ex juris may be granted. For tort, leave to serve may be granted if “the claim is founded on a tort and the damage was sustained, or resulted from an act committed, within the jurisdiction”. For a discussion of appropriateness in relation to civil jurisdiction see text infra and Section 4. 17 Amin Rasheed Shipping Corp. v. Kuwait Insurance Co., “The Al-Wahab” [1983] 2 All ER 884 (HL). 18 Baer, supra n. 14, at 184. 19 De Savoye v. Morguard Investments Ltd. [1990] 3 SCR 1077, 76 DLR (4th) 256 [hereinafter Morguard cited to SCR]; Hunt v. T&N plc. [1993] 4 SCR 289, 109 DLR (4th) 16 [hereinafter Hunt cited to SCR]. 20 Moran v. Pyle National (Canada) Ltd. [1975] 1 SCR 393, 43 DLR (3d) 239; Tolofson v. Jensen; Lucas v. Gagnon [1994] 3 SCR 1022, 120 DLR (4th) 289 [hereinafter Tolofson cited to SCR]; Morguard, supra n. 19; Hunt, supra n. 19. Note that La Forest J. wrote the Court’s judgment in each

162 A C McConville and constitutional aspects of jurisdiction to adjudicate in Canada are discussed below. Comity and Extraterritoriality In De Savoye v. Morguard Investments Ltd., a case about the recognition and enforcement in British Columbia of a default judgment granted in Alberta, the SCC discussed the principle of comity as it relates to territoriality in conflict of laws cases. As a starting point, La Forest J. observed that the principle of territoriality reflects the basic tenet of international law that each sovereign state has jurisdiction in its own territory and, as a concomitant, states are hesitant to exercise jurisdiction over matters that take place in the territory of other states.21 Consequently, the courts of one state will tend to defer to the actions of another state legitimately taken within its own territory. In Morguard, La Forest J. adopted the definition of comity from the United States Supreme Court in Hilton v. Guyot: “ ‘Comity’ in the legal sense, is neither a matter of absolute obligation, on the one hand, nor of mere courtesy and good will, upon the other. But it is the recognition which one nation allows within its territory to the legislative, executive or judicial acts of another nation, having regard both to international duty and convenience, and to the rights of its own citizens.”22

The centrality of comity as the informing principle in private international law is connected to the principle of extraterritoriality in public international law. Respect for international comity is one way to avoid impermissibly extraterritorial extensions of a state’s jurisdiction to adjudicate. Respect for the territorial sovereignty of states is, however, only one thrust of the SCC’s interpretation of comity. The SCC also emphasizes that both order and fairness underlie the principle.23 In the Court’s view, international comity is necessary for an orderly, workable system of private international law. If territorial limits to jurisdiction were not respected in the interdependent world where legal relationships often transcend state boundaries, competing exercises of state power in respect of the same activities would be the confusing result.24 At the same time, the territorial principle should not be so rigid as to allow persons to escape their legal obligations by simply exiting a jurisdiction or to force plaintiffs to sue in the jurisdiction where the defendant resides. In turn, these plaintiff-sensitive concerns must be weighed against the unfairness to defendants that would result if they could be sued in any jurisdiction, even those with no connections to the action, and be judged by foreign laws which the of the last three cases. For this reason, commentary in the text will refer interchangeably to the reasons of the SCC and of La Forest J. 21 Morguard, supra n. 19, at 1095. 22 Hilton v. Guyot, 159 U.S. 113 (1895) as quoted in Morguard, ibid. at 1096. 23 Morguard, supra n. 19; Hunt, supra n. 19. 24 Tolofson, supra n. 20 at 1047–8.

Taking Jurisdiction 163 defendants could not reasonably have been expected to observe. Fairness to the litigants thus requires that jurisdiction is properly restrained by real and substantial connections to the territory.25 The balance between the ideas of order and fairness is inextricably linked to the territorial principle. The question is what nature of connections is sufficient for an appropriate exercise of civil jurisdiction over litigation with extraterritorial elements. Connections to Territoriality Not all exercises of jurisdiction over conduct that occurs extraterritorially are impermissible at international law. They are only lawful, however, if there is a substantial connection between the subject matter and the source of the jurisdiction and if the principle of non-intervention in the domestic or territorial jurisdiction of other states is observed.26 The interacting public international law norms concerning extraterritoriality and the private international law principle of comity reinforce the common law idea that there are territorial limits to the competence of a court to hear cases. In Canada, these limits have an added dimension: the territorial limit of the provincial courts’ jurisdiction is a constitutional imperative.27 For a provincial court’s exercise of jurisdiction to be constitutionally valid, there must be a real and substantial connection between the forum court and either the subject matter of an action or the parties.28 (i) General jurisdiction based on the defendant’s presence in the territory Where the defendant is physically present in, or consents to, the jurisdiction of the forum court, the traditional common law rule determines whether there is a sufficient territorial connection for appropriate jurisdiction. When a defendant is resident in the forum, the issue of appropriateness rarely arises, as the defendant can reasonably expect to be subject to the courts of his or her place of residence. However, the sufficiency of either the transient physical presence or the consent of foreign defendants alone is less clear. In Morguard, the SCC affirmed the legitimacy of presence and consent-based jurisdiction, independent of any additional subject matter connection to the forum.29 Subsequently, however, La Forest J. suggested that the traditional rules were only starting points that may need to be considered in light of the Morguard principles.30 As a result, transient presence alone (and perhaps even transient presence in combination with some 25

Morguard, supra n. 19 at 1102–3. Brownlie, supra n. 10 at 309. 27 Hunt, supra n. 19 at 324. 28 Ibid.; Morguard, supra n. 19. 29 Morguard, supra n. 19; C Walsh, “Conflict of Laws, Enforcement of Extra-Provincial Judgements and In Personam Jurisdiction of Canadian Courts: Hunt v. T&N, plc.”, (1994) 73 CanBarRev 394 at 410 [hereinafter “Walsh”]. 30 Hunt, supra n. 19, at 325. 26

164 A C McConville added real and substantial connection) may no longer be sufficient to ground appropriate jurisdiction. (ii) Specific jurisdiction based on subject matter connections The meaning of the phrase “real and substantial connection” becomes crucial where there is no territorial jurisdiction over the defendant. In most of the provinces, the statutory rules of civil procedure provide for service ex juris and list acceptable connecting factors, such as the plaintiff’s residence or the subject matter of the action. Canadian case law since Morguard and Hunt suggests, however, that these statutory indicators will not be sufficient for constitutionally valid exercises of jurisdiction unless they also meet the test of a real and substantial connection.31 Although it is not clear precisely what connections are needed in addition to the plaintiff’s residence, both the doctrinal origins of the test and pragmatic considerations suggest that some minimum contacts are required.32 In Canada, the real and substantial connection test developed in the context of interprovincial conflict of laws cases. This context suggests that the connection must be one that would give the forum province a constitutionally sufficient interest in regulating the conduct underlying the action.33 As Catherine Walsh explains, where a court does not have presence-based jurisdiction over the defendant, the considerations that would make it constitutional for a court to take jurisdiction based on subject matter connections must be those that would give the forum a constitutionally permissible interest in legislatively regulating the outcome. On a practical level, this strict approach minimises the risks of a misapplication of foreign law and an improper application of forum law because the contacts engage the regulatory interest of the province.34 However, the interprovincial context in which a real and substantial connection became a constitutional requirement provides cause to question whether the rule applies to international cases. Lower court decisions have interpreted the Morguard and Hunt principles as indeed being the constitutional limits of the provincial courts’ jurisdiction in international cases.35 On the one hand, it 31 MacDonald v. Lasnier (1994), 21 OR (3d) 177 (Gen Div) [hereinafter MacDonald]; Canadian International Marketing Distributing Ltd. v. Nitsuko Ltd. (1990), 68 DLR (4th) 318 (BCCA); Ell v. Con-Pro Industries Ltd (1992), 11 BCAC 174; Wilson v. Moyes (1993), 13 OR (3d) 302 (Gen Div); Webb v. Hooper (1994), 19 Alta LR (3d) 269; First City Trust Co. v. Inuvik Automotive Wholesale, [1993] NWTR 273 (SC); P Finkle and C Lebrecque, “Low Cost Legal Remedies and Market Efficiency: Looking Beyond Morguard” (1993) CanBusLJ 58 at 67 [hereinafter “Finkle and Lebrecque”]; Walsh, supra n. 29, at 410. 32 Walsh, supra n. 29, at 413. The minimal contacts test is one aspect of the approach taken by the US Supreme Court in International Shoe Co. v. Washington, 326 US 310 (1945), to which La Forest J. referred approvingly in Morguard. 33 Walsh, supra n. 29, at 414. 34 I say “strict approach” because adjudicative jurisdiction and prescriptive jurisdiction do not tend to invade foreign interests to the same extent. See infra n. 37. 35 E.g., MacDonald, supra n. 31 and Frymer v. Brettschneider (1994), 19 OR (3d) 60 (CA) [hereinafter Frymer].

Taking Jurisdiction 165 seems consistent to apply these limits in an international context and restrain the courts from exercising jurisdiction over cases which have more substantial links to another state. The SCC inclined toward this view in its review of Canadian conflict of laws principles in Tolofson v. Jensen; Lucas v. Gagnon, the leading case on choice of law in tort.36 Writing for the court, La Forest J. observed that the purpose of the real and substantial connection rule is to prevent overreaching in the exercise of jurisdiction over extraterritorial and transnational transactions: “It seems to me self evident, for example, that State A has no business in defining the legal rights and liabilities of citizens of State B in respect of acts in their own country, or for that matter the actions in State B of citizens of State C, and it would lead to unfair and unjust results if it did.”37 On the other hand, the Morguard principles are most suited for the context of Canadian federalism. The SCC’s holding that the provincial courts’ jurisdiction properly extends to cases with which they have a real and substantial connection came in conjunction with a direction to the provincial courts to recognise and enforce each other’s judgments based on this same test of “real and substantial connection”. When this takes place in an integrated federation, territorial limits are a rational way to allocate judicial resources. Also, within Canada, the courts of one province can be confident that a plaintiff will have due process and fairness in the courts of another.38 These features are not necessarily present in international cases: there is neither the need to protect the idea of an integrated federation nor the assurance that plaintiffs will receive substantially and procedurally comparable justice elsewhere. Consequently, the courts need not show the same degree of restraint with international cases. Indeed, following Hilton v. Guyot, one could argue that where rights are defined by international law, fairness permits, even requires as a matter of international duty, that courts exercise jurisdiction over extraterritorial conduct.

Implications for Transnational Torts Litigation Transnational human rights torts litigation of the US variety does not easily fit into the framework of territorial jurisdiction. Rather, it permits a court to assume jurisdiction over non-resident defendants who are sued for violations of international human rights which occurred in another state. According to the principles of territoriality and state sovereignty, judicial authority to hear and 36

Tolofson, supra n. 20. See also Libman v. R. [1985] 2 SCR 178, 21 CCC (3d) 206. Tolofson, supra n. 20 at 1052. Here, the expression “defining the legal rights and liabilities” nicely evokes the way in which considerations regarding prescriptive jurisdiction (the applicability of a state’s legal norms) merge with those regarding adjudicative jurisdiction. Even when a court takes jurisdiction but decides to apply foreign law, it is still applying that law in all its indeterminacy and thus having to engage in judicial activity that is as much “interpretation” as it is “application”. Interpretation having a law-creation component, the court is, to varying degrees, defining the law of the other state for that state. 38 Morguard, supra n. 19, at 1103. 37

166 A C McConville determine such cases seems to properly fall to the foreign state’s judicial jurisdiction. First, it is unfair to haul the defendant before a foreign court to answer to the laws of a foreign jurisdiction, which is detached from the dispute, even if those laws present themselves as embodiments of international law. Second, it offends comity for a court to pass judgment on conduct that occurred within the territorial limits of another state and that, prima facie, is within that state’s jurisdiction; put another way, such overreaching may be impermissibly extraterritorial.39 Yet, territoriality is only one aspect of the principle of comity. In Hunt v. T&N plc, a case that dealt with the constitutional validity of blocking legislation in interprovincial litigation, La Forest J. reiterated that, in addition to the territorial principle, the principles of order and fairness help determine appropriate exercises of jurisdiction. La Forest J. stressed that the order achieved by territorial allocations of judicial authority was not to be rigid, but tempered by fairness. He lamented old common law rules the were “rooted in an outmoded conception of the world that emphasised sovereignty and independence often at the cost of fairness”.40 This view that fairness to the litigants should not be sacrificed to the territorial principle provides an opening to justify a departure from the territorial principle in transnational human rights cases. If the courts do not exercise jurisdiction in such cases, fairness to the plaintiffs may be sacrificed entirely. The SCC’s position that comity is not “mere courtesy” also opens the way for arguments that a departure from the territorial norm is justified in cases of serious human rights violations. If comity is the deference due to the legitimate acts of foreign states, then conduct that violates international human rights law norms, and is therefore illegitimate, does not attract comity. While it may offend comity for one state to define the rights and duties of the citizens of another state, it should not offend comity to enforce individuals’ human rights that are defined by international law. Rather than focus on sovereignty and territoriality, a better approach would be to consider fairness to both parties—the plaintiff as well as the defendant. Alternative principles of jurisdiction, developed in the context of criminal law, may bolster such a departure from the norm of territorial jurisdiction in civil cases.

Other Principles of Jurisdiction At public international law, jurisdiction to adjudicate criminal cases may be based on several recognised principles of jurisdiction, only one of which is the 39 For a discussion of similar issues in the federal context see R Wisner, “Uniformity, Diversity and Provincial Extraterritoriality: Hunt v. T&N plc.” (1995) 40 McGill LJ 759. 40 Hunt, supra n. 19 at 321–2. This language is at some variance with La Forest J.’s emphasis on order before justice in the tort choice of law case, Tolofson. See the discussion in J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11 of this volume.

Taking Jurisdiction 167 territorial principle.41 In general, these principles recognise that when a case engages either its territory, citizenry or security, a state’s interests are sufficiently involved to give that state authority to exercise jurisdiction.42 While these principles traditionally only applied to criminal law cases, there is little reason not to apply them in civil cases as well, especially when international human rights are at issue. The same international jurisdictional norms should govern human rights whether they arise in a civil or criminal context.43 Litigants in transnational human rights torts cases may be able to borrow from the principles in the sections below to justify a provincial court entertaining cases with extraterritorial elements and to show that, as a matter of fairness, transnational human rights litigation appropriately fits within the court’s civil jurisdiction. (i) Nationality jurisdiction Nationality is generally recognized as a basis for jurisdiction over extraterritorial acts.44 According to the nationality principle, a state has authority to prosecute its nationals for crimes committed anywhere in the world. The principle also applies where an accused has long residence in the state exercising jurisdiction. The rationale for this jurisdiction is that nationality and long residence are marks of allegiance to the state and, in the case of nationality, an aspect of sovereignty.45 It is also based on the idea of a state’s interest in not having its nationals cast the state in a bad light in its international relations. The Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (the “Convention against Torture”) provides an example of nationality grounding extraterritorial jurisdiction. Article 5(1)(b) obliges state parties to “take such measures as may be necessary to establish its jurisdiction over [acts of torture] . . . when the alleged offender is a national of that State”. The Convention against Torture establishes a paradigm whereby state parties ensure that acts of torture are offences under their own criminal law and then take the necessary measures to ensure that the acts can be prosecuted.46 Although this example falls in the criminal law realm, it illustrates that the nationality of the offender can be a basis for exercising jurisdiction, regardless of where the offences occur and that the nationality principle already operates to combat acts of torture. 41 In the criminal context, the territorial principle gives a court authority to prosecute offences that are committed within the territory of the forum state. 42 Randall, supra n. 12, at 153. 43 Ibid. at 157. 44 The Steamship Lotus Case (France v. Turkey), (1927) PCIJ Rep, Ser A, No.1 [hereinafter Lotus]; Brownlie, supra n. 10, at 303; M Akehurst, Modern Introduction to International Law, 6th ed. (London, Allen and Unwin, 1987) at 105 [hereinafter “Akehurst”]. 45 Brownlie, supra n. 10 at 303; Akehurst, ibid. at 105. See also R v. Cook [1998] 2 SCR 597, 164 DLR (4th) 1 [hereinafter Cook cited to SCR]. 46 CAT, supra n. 9, at articles 4 and 5.

168 A C McConville Significantly, the SCC recently invoked the nationality principle to justify an extraterritorial application of human rights protections found in the Charter of Rights and Freedoms to law enforcement activities of Canadian police officers in R v. Cook.47 The Canadian officers travelled to the United States to question the accused, who had been arrested and detained by local authorities, in connection with a murder that occurred in Canada. The trial judge allowed the accused’s statements into evidence despite the officers’ failure to properly inform him of his right to counsel before the interrogation took place, contrary to section 10(b) of the Charter.48 The majority of the court held that, even though the accused was not a Canadian citizen and the impugned statements were made on foreign territory, the Charter applied in the circumstances on the jurisdictional basis of the nationality of the law enforcement authorities who engaged in the governmental action in breach of a Charter right. Jurisdictional competence to apply the Charter to the actions of the Canadian officers gathering the evidence abroad was not based on the principle of territoriality, but of nationality. In Cook, Cory and Iacobucci JJ., writing for the majority, recognised that reference to territoriality is not the only way to resolve questions of jurisdiction: “In our view, although territory is clearly a critical element in determining the scope of a state’s jurisdiction, territory alone is not determinative of jurisdictional competence under international law. There are some circumstances where the application of Canadian law to an undertaking by Canadian law enforcement authorities on foreign territory can be grounded on other jurisdictional principles, and will not result in an objectionable interference with the exercise of foreign jurisdiction.”49

It is important to note that, while Cook signals the SCC’s recognition of acceptable alternatives to the territoriality principle, Cory and Iacobucci JJ. cautioned that the extraterritorial application of Canadian law grounded on the nationality principle was exceptional and was made on the basis of the very particular facts of the case.50 Traditionally, the common law has not recognized nationality as a source of civil jurisdiction. This may be attributable to the qualitative difference between a criminal prosecution and a civil suit.51 In a criminal prosecution, the enforcement power of the state is brought to bear against the individual and, as a result, it is appropriate for an accused to be subject to the authority of his or her own national state. This element of state power is absent in a civil suit between private litigants. Nevertheless, the nationality principle could reasonably be 47 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c.11 [hereinafter “Charter”]; Cook, supra n. 45. 48 Section 10(b) of the Charter, provides: “Everyone has the right on arrest or detention . . . (b) to retain and instruct counsel without delay and to be informed of that right;”. 49 Cook, supra n. 45 at 618. 50 Ibid. at 629. 51 Although note that, even here, Canadian criminal law does not generally apply nationality as a jurisdictional basis as do many civil law countries. Territoriality remains the norm, and nationality the explicit exception, in Canada’s Criminal Code.

Taking Jurisdiction 169 applied in the civil context. No manifest unfairness would result from a provincial court in Canada exercising jurisdiction over a Canadian national in a civil claim, even if that claim arose from extraterritorial conduct. The defendant would not be unfairly surprised by being judged against the private law rules of the state to which he or she owes allegiance. Nationality or prior long residence (even if now discontinued) are also significant personal links to the territory of the forum court. If the Canadians who are government ministers in the Sudan were charged with crimes, Canada could rightfully prosecute them in accordance with the nationality principle.52 Applying the same principle to the civil context, Canadian nationals who perpetrate human rights violations should be open to suit in the courts of their national state regardless of where the conduct took place.53 (ii) Passive personality jurisdiction Under the passive personality principle, a state may prosecute individuals for extraterritorial acts that harm nationals of the forum state. The availability of this principle as a general basis for extraterritorial jurisdiction is not widely accepted.54 In order to justify the exercise of civil jurisdiction in a suit against alleged human rights offenders, this principle would have to be extended to include individuals who were not nationals at the time the impugned conduct occurred, but subsequently became nationals or residents of the forum territory where they commenced the action.55 However, given that the passive personality principle is the least accepted of the principles of jurisdiction, such an extension of this principle is unlikely to persuade a court of its competence at international law to hear an extraterritorial civil matter. (iii) Objective territoriality A more persuasive head of jurisdiction is the objective territoriality principle. It suggests that a state has grounds for taking jurisdiction when sufficient harms or costs associated with extraterritorial acts are felt within its territory.56 Borrowing from this principle, a good case can be made in support of litigating 52 Criminal Code [of Canada] RSC 1985, c. C-46, s. 269.1, 7(3.7), 7(3.76–3.77) [hereinafter “Criminal Code”]. 53 It may be helpful to take as a hypothetical the situation in which the family of Somalian Shidane Arone, a man tortured by Canadian Armed Forces personnel, wished to sue through Canadian tort law members of the Armed Forces living abroad. See Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. The Attorney General of Canada (unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J.). And see the discussion of the problems with the judge’s dismissal of the case in J Terry, “Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad” in chapter 4 of this volume. 54 Brownlie, supra n. 10, at 303. 55 For instance, they may have fled as refugees. 56 Brownlie, supra n. 10, at 303; Lotus (majority), supra n. 44.

170 A C McConville a transnational human rights tort claim in Ontario’s civil courts. First, jurisdiction is based on the existence of harm within the territory. This territorial nexus means that the principle is close to the traditional basis of civil jurisdiction. Second, when individuals who were tortured or enslaved in a foreign jurisdiction come to Canada, they continue to suffer psychological and emotional harm as a result of the experience and may require treatment in Canada. These are harms and costs felt in a Canadian jurisdiction and which thereby trigger a state interest in curbing the conduct that led to the harm and which are similarly grounds for Canadian courts to exercise jurisdiction in actions for redress between the parties.57 Third, objective territoriality is commensurate with the “real and substantial connection” test established in Morguard to gauge the appropriateness of civil jurisdiction.58 Finkle and Lebrecque argue that under a purposive, policyoriented interpretation of the real and substantial connection rule, courts should be permitted to assert jurisdiction over non-resident defendants where, prima facie, a loss to a resident plaintiff is attributable to a non-resident. A province has a real and substantial interest in any harm or costs suffered by one of its residents regardless of where the initial cause of the loss occurs.59 The effects-based territoriality principle also satisfies Catherine Walsh’s thesis that the subject matter which engages the regulatory interest of the province meets the test for constitutionally valid jurisdiction. Since the types of harms and costs felt in the province by victims of torture and slavery, health care and social services most notably, engage the regulatory interest of the province, they are sufficiently “real and substantial” to justify a proper assumption of civil jurisdiction.60 Indeed, the rules for service ex juris in Ontario currently do provide that a coincidence of the plaintiff’s residence and loss in the province is sufficient for the province to assert jurisdiction over non-resident defendants. The broad approach in the rules results in part from recognition that a province in which a party resides is

57 But see the separate reasons of Bastarche J. in Cook, supra n. 45, at 669–70. Writing for himself and Gonthier J. and concurring with the majority in the result, Bastarche J. stated: “While the objective principle is unimpeachable as a matter of principle, it is recognised that when the interest of the forum state is less significant than the interest of the state in which the events take place, then the objective principle no longer authorises asserting jurisdiction. ” As a general statement of the relativity of the exercise of jurisdiction, this statement would equally apply to all other bases of jurisdiction than pure territoriality. Thus, even if passive personality were accepted as prima facie generally available, it would arguably represent a “less significant” interest than that of a state which could offer nationality or objective territoriality as a jurisdictional basis. 58 See the discussion in Orange, supra n. 40, of the origins of the “real and substantial connection” test in Canadian law in two cases, Moran v. Pyle [1975] 1 SCR 393 at 397 on jurisdiction over interprovincial product liability torts and Libman v. R. [1985] 2 SCR 178 on the scope of territoriality under Canada’s Criminal Code. 59 P Finkle and C Lebrecque give medical care and funeral expenses incurred in the province as examples of such costs—see Finkle and Lebrecque, supra n. 31. 60 Ontario Rules of Civil Procedure, RRO 1990, Reg 194, rule 17.02 [hereinafter Rules]; Jaffe v. Dearing (1988), 65 OR (2d) 133 (HCJ) [hereinafter Jaffe]. See also infra text at Section 4.

Taking Jurisdiction 171 expected to take a protective role towards its residents when they suffer loss by at least affording them a forum in which to vindicate their rights.61 (iv) Protective jurisdiction Protective jurisdiction allows a state to assume jurisdiction over foreigners for acts that threaten the forum state’s security interests, even when the acts are committed abroad.62 The types of acts that will give rise to valid exercises of jurisdiction under this principle include plots to overthrow the government, spying, forging currency and illegal immigration. It can be argued that Canada’s national security interests are engaged when its nationals are responsible for systemic human rights abuses in another state. To the extent that governmentsanctioned torture and slave-trading are among the reasons that resistance movements have taken up arms in the Sudan, and that Canadian nationals are part of the Sudanese regime, Canada is implicated in the situation. Canada’s legitimate interests in protecting international peace and security, and in fostering the human rights conditions conducive to the realisation of this state of affairs, also possibly support a civil action that impugns the conduct of its nationals. However, these arguments do not seem apt for the context. While they may well bolster a nationality argument, they are not likely to operate on their own in this context. (v) Universality principle The universality principle provides every state with jurisdiction over a limited category of offences that are generally recognised within the international community as being of universal concern.63 Under this principle, the jurisdiction of the domestic court does not depend on where the offence occurred, nor on the nationality of the offender. Rather, it assumes that due to the nature of the particular offences, every state has an interest in exercising jurisdiction to combat them. There is little controversy that, currently, offences that attract universal jurisdiction include at least piracy, slave-trading, war crimes (at least those that constitute grave breaches of the laws of war), crimes against humanity (at least those that are part of systematic conduct), genocide and torture.64 These features of the universality principle bolster an argument for a Canadian court to exercise discretion to entertain a civil case derived from international wrongs, including torture and slave-trading. 61 Vindication of rights depends on the plaintiff’s willingness and capacity to bring an action in court. It is better to let the court take jurisdiction and, if necessary, decline to exercise jurisdiction on forum non conveniens grounds if a serious injustice to the defendant will result. Finkle and Lebrecque, supra n. 31. Forum non conveniens is discussed infra, text at Section 4. 62 Akehurst, supra n. 44, at 105; Brownlie, supra n. 10, at 303. 63 Brownlie, supra n. 10 at 304–5; Randall, supra n. 12 at 163. 64 Randall, ibid. at 163–4.

172 A C McConville It is important, however, to distinguish between the exercise of universal jurisdiction over international crimes and the applicability of the universality principle to a private, or civil, law case. For crimes under international law, general international law has been able to fasten onto and crystallise around textual sources of the international community’s view of certain conduct as being sufficiently reprehensible to constitute crimes subject to universal jurisdiction, whether these textual sources take the form of “soft law”, such as the General Assembly’s adoption of the Nuremberg Principles, or of multilateral treaties like the Geneva Conventions on the laws of war. As international criminal wrongs, any state that obtains custody of persons suspected of the offences may prosecute them.65 In contrast, it is nearly impossible to find textual indicators that international law itself makes torture or slavery civil wrongs.66 As such, arguments that civil liability exists under international law for grave human rights violations must rely on arguments constructed on the tortuous normative terrain of customary international law and general principles of law. This criminal/civil distinction complicates an attempt to borrow the universality principle from the context of international crimes in order to apply it in the context of a civil claim. Nevertheless, the policy considerations that underlie universal jurisdiction over international crimes are analogous to concerns about the need for redress for torture and slavery and arguably justify jurisdiction to adjudicate. In R v. Finta, Cory J., writing for the majority of the SCC, observed that, while the universality principle permits a state to exercise jurisdiction over non-nationals in respect of crimes committed against non-nationals wherever they occur, this jurisdiction is based on the nature of the crime which must threaten the international order as a whole so as to justify repression of the conduct as a matter of international public policy.67 In his dissenting opinion (but not on this point), La Forest J. stated that these cases are exceptions to the general concerns about extraterritoriality, comity and the sovereignty of independent states, because the international community reviles the conduct and because it is difficult to bring the offenders to justice in the state where the acts are committed, since that state will normally be implicated in the crime.68 These arguments apply equally to cases of state-sanctioned torture and slavery: the conduct is reviled by the international community and the offenders are not likely to be brought to justice due to complicit executive or judicial state organs. 65 Geneva Conventions For the Protection of War Victims, 12 August 1949, 75 UNTS 31; R v. Finta [1994] 1 SCR 701 [hereinafter Finta]. 66 Article 14 of the CAT, supra n. 9, does not, for instance, expressly recognise universal jurisdiction to secure civil redress for torture. Article 14 reads: “1. Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation. 2. Nothing in this article shall affect any right of the victim or other persons to compensation which may exist under national law.” 67 Finta, supra n. 65, at 811. 68 Ibid. La Forest J. (dissent), at 733.

Taking Jurisdiction 173 The fact that international law itself does not make torture and slavery civil wrongs need not make universal jurisdiction inapplicable to transnational human rights torts cases. The prohibitions against torture and slavery qualify as jus cogens and erga omnes norms in international law and, therefore, they are qualitatively distinct from common tortious conduct. Erga omnes obligations refer to certain obligations that flow to all states from general principles of international law and international instruments of universal or quasi-universal character. They are the concern of all states and so all states have an interest in their protection.69 Jus cogens norms are peremptory norms accepted and recognised by the international community as a whole. No derogation from jus cogens norms is permitted.70 They can derive from basic considerations of human dignity and, thus, include the most fundamental human rights such as protection from torture and slavery.71 Since erga omnes obligations flow to all states and jus cogens norms peremptorily compel all states, it is not necessary that international law specifically make torture and slavery civil wrongs for them to be the subject of universal jurisdiction. There is ample evidence at international law that the prevention of slavery and torture qualify as the kinds of higher order obligations that attract the universality principle. The prevention of torture and slavery is the specific purpose of multilateral treaties such as the Convention against Torture, which obliges state parties to take all measures to prevent acts of torture in article 2. The proscription of slave-trading is codified in several conventions.72 In addition, slave-trading is subject to jurisdiction based on nationality and territory, informed by universality in purpose, under Article 99 of the 1982 Convention on the Law of the Sea.73 These instruments recognise that the right to be free from torture and slavery inheres in human dignity. The universality principle holds that jurisdiction is not limited by territorial boundaries where such basic human rights are infringed; the nature of the offence confers jurisdiction on all states regardless of territory. The status of torture and slavery as erga omnes and jus cogens norms, therefore, assists the jurisdictional analysis, since jurisdiction to adjudicate in these cases should not be confined to territorial boundaries. Providing a forum where victims of torture or slavery may vindicate their international legal rights is consistent with the higher obligations that flow to states, including Canada, to protect these rights.74 69 Case Concerning Barcelona Traction, Light and Power Company. Limited (Belgium v. Spain) ICJ Rep (1970) 3. 70 Article 64 of the Vienna Convention on the Law of Treaties, 23 May 1969, (entry into force 27 January, 1980) provides: “If a new peremptory norm of general international law emerges, any existing treaty which is in conflict with that norm becomes void and terminates.” All jus cogens norms are also erga omnes, but erga omnes norms may include more than jus cogens norms. 71 Kindred, supra n. 11, at 165–6. 72 See list supra, n. 9. 73 “Every State shall take effective measures to prevent and punish the transport of slaves in ships authorized to fly its flag to prevent the unlawful use of its ships authorized to fly its flag for that purpose. Any slave taking refuge on board any ship, whatever its flag, shall ipso facto be free.” Convention on the Law of the Sea, 10 December 1982, 21 ILM 1261, article 99. 74 Randall, supra n. 12, at 170.

174 A C McConville All that is being argued here is that the prohibitions against torture and slavery are the kinds of norms which permit all states to exercise jurisdiction over those persons who violate the prohibitions, whether by criminal law, civil liability, or other regulatory routes. In other words, it is sufficient for purposes of adjudicative jurisdiction for judges to know that concerns about extraterritoriality would be misplaced as a reason not to assume jurisdiction (for instance, by refusing leave to a plaintiff for service ex juris). It may even be that, at least with respect to torture, states not only may but must permit civil claims for redress to be heard in their courts. Much could depend on whether a consensus interpretation of the duty to provide for civil redress in article 14 of the Convention against Torture evolves to include an implicit universal responsibility.75 Under article 14(1) of the Convention against Torture, state parties must ensure that victims of torture have an enforceable right to obtain “fair and adequate compensation” in its legal system. While this obligation is not expressly stated to apply universally—for instance, to allow a victim in State X to obtain redress through the legal system of State Y—it can be interpreted to do so when read in the context of the rest of the Convention against Torture. Even if a universal obligation to facilitate compensation claims is still in the realm of interpretive debate, however, the fact that the proposition is arguable speaks to the permissibility of providing such redress. Thus, judicial measures are one lawful way for Canada to discharge its erga omnes obligations to redress torture and slavery and contribute to the vindication of international human rights.

The Significance of Other Jurisdictional Avenues Once international human rights law is used to bolster the argument for a court in Canada to take jurisdiction in a derivative civil action, additional jurisdictional hurdles arise. Normally, international obligations are implemented either through the legal system of the wrongdoer state or, if the state is a party to international human rights instruments, by some process of international adjudication or interstate diplomatic espousal.76 Even if the nationality, objective territoriality or universality principles can be adopted to justify an extraterritorial exercise of civil jurisdiction in transnational human rights tort cases, this jurisdiction may still conflict with the functions of international supervisory bodies and with the local remedies rule. These conflicts complicate the attempt to translate state obligations into obligations enforceable between private parties. 75 Kenneth Randall, for instance, argues that the universality principle should not be restricted in this way. In his view, the principle not only enables domestic courts to exercise civil jurisdiction and to provide private law remedies for breaches of international human rights, but international instruments also go so far as to obligate domestic courts to assume such authority: Randall, supra n. 12 at 7. 76 H Fox, “State Responsibility and Tort Proceedings Against a Foreign State in Municipal Courts”, (1989) 20 NethYB Int’l L 3.

Taking Jurisdiction 175 (i) International human rights treaties International human rights instruments make states publicly accountable, at the international level, for their treatment of individuals. Where these instruments provide a supervisory body, a domestic court in Canada may decline to exercise civil jurisdiction over a case that falls within that body’s purview. For example, the Convention against Torture establishes a process which enables state parties to submit written communications to the Committee against Torture if it considers that another party is not giving effect to its obligations under the treaty.77 The Committee is then responsible for dealing with the matter. Thus, as regards redress, the Convention against Torture envisions a two-part system: states are each responsible for providing remedies for their own acts of torture, but states impugn each other’s conduct through the supervision of the Committee and not through their domestic courts. A third component may also enter into play. Individuals may bring complaints against a state party, if that state has declared its acceptance of the application of article 22, which empowers the Committee against Torture to receive and evaluate written communications from victims alleging a violation of the Convention against Torture. The Human Rights Committee’s supervision of the ICCPR operates on the same premise that judging the wrongfulness of a state’s conduct is most appropriately left to an international body. Under the ICCPR’s first Optional Protocol, individual victims of ICCPR violations may apply to the Human Rights Committee for a “view” that the ICCPR has been violated and a recommendation that the (torturing) state provide an appropriate remedy including monetary compensation.78 The Convention against Torture and ICCPR paradigm may suggest to some that, where international adjudicative processes exist to enforce international human rights, it is inappropriate for the domestic courts of one state to assess the conduct of another through transnational litigation. It could be argued that, for states that have accepted this procedure, it is especially inappropriate to permit foreign domestic courts to operate parallel to the Committee against Torture. Others might argue for the same preclusionary effect, even when the state has not accepted this individual petition procedure, because evaluation by a foreign court circumvents a choice that the Convention against Torture itself makes optional. However, such views are based, explicitly or implicitly, on the presumption that human rights treaties are self-contained regimes which exclude recourse to any remedies outside those provided by the treaties themselves, a presumption

77

CAT, supra n. 9, article 21. Optional Protocol to the 1966 International Covenant on Civil and Political Rights, 16 December 1966, 6 ILM 383, article 2 [hereinafter “Optional Protocol”]. This avenue is only available to individuals in states that have signed the Optional Protocol and that are thereby subject to the Committee’s jurisdiction to receive and evaluate communications from individuals. The Sudan has not signed the Optional Protocol. 78

176 A C McConville which may not be well-founded.79 Unless a human rights treaty clearly stipulates that alternative remedies are incompatible with the treaty regime, there is little reason to conclude that other ordinary legal consequences cannot flow from its breach. In addition, the remedies provided in the UN human rights treaties are often weak, because they lack both formal binding force and followup enforcement possibilities, and are seldom invoked by torture victims even in those states that have accepted individual petition procedures under the Convention against Torture or the ICCPR. The effectiveness of the treaties for protecting human rights may depend on the availability of residual procedures and remedies, provided that such remedies accord with the general rules of international law and complement, rather than impede, the work of the human rights treaty bodies under the relevant UN treaties.80 Thus, where tortious conduct is also a breach of a human rights treaty, a civil suit is one way to give meaning and effect to human rights protection, particularly where a state has signed a treaty but has failed to accept its internal enforcement obligations.81 Furthermore, as a general matter of principle, public law accountability does not necessarily preclude a private aspect to international human rights protection. As Harold Koh argues, a statist view of international relations does not accurately reflect the variety of international legal relationships that exist today. In what is perhaps the leading scholarly article seeking to provide a conceptual framework for “transnational public law litigation”, Koh makes the familiar, but no less valid, point that the Tokyo and Nuremberg war crimes trials “pierced the veil of state sovereignty and dispelled the myth that international law is for states only, re-declaring that individuals are subjects, not just objects of international law”.82 Individuals and non-state actors not only hold legal rights but, as Filártiga and its progeny show, also have legal responsibilities under international human rights law that reflect the ways in which they, like states, are capable of violating the interests protected by international human rights law. Hence, following Koh, individuals are proper plaintiffs and defendants in transnational actions. (ii) Local remedies When international human rights law is used as a legal fact to assist in the jurisdictional analysis, it raises the additional hurdle of the doctrine of local remedies. According to the local remedies rule, when an international claim is brought against a state for injuries suffered by individuals, that state has the right to resist the action if the individuals have not first exhausted all the reme79 T Meron, Human Rights and Humanitarian Norms as Customary Law (Oxford, Clarenden Press, 1989) at 230 [hereinafter “Meron”]; B Simma, “Self-Contained Regimes”, 16 NethYB Int’l L (1985) 111 at 133 [hereinafter “Simma”]. 80 Meron, supra n. 79 at 230. 81 Ibid.; Simma, supra n. 79 at 129. See article 2(3) of the ICCPR, supra n. 9, and article 2(1) of the CAT, supra n. 9, for general duties to give effect to treaty rights in the domestic legal order. 82 H. Koh, “Transnational Public Law Litigation”, (1991) 100 Yale LJ 2347.

Taking Jurisdiction 177 dies available to them under its domestic law.83 The rule is intended to give the state the opportunity to provide a remedy before the matter is taken to the international level. In this way, it upholds the basic principle of respect for state sovereignty and serves a similar function in public international law as that of comity in private international law. The exhaustion of local remedies is a threshold requirement before international bodies will hear a human rights complaint from individuals. For example, under the Optional Protocol to the ICCPR, the Human Rights Committee will only accept applications from individuals who have exhausted all available domestic remedies.84 The procedure by which states parties to the ICCPR may complain of other states’ non-compliance is also subject to bilateral attempts at adjustment and other forms of dispute settlement. Also, as a matter of the general international law on state responsibility for protection of foreign nationals, the government of Canada will not normally espouse a claim of a Canadian against a foreign state until all the local remedies have been exhausted without satisfaction; this is in accordance with accepted customary law doctrine that governs diplomatic espousal of claims.85 Since the purpose of the doctrine is to allow a state to remedy situations that might otherwise attract international responsibility before an international action is brought, it would seem, as a matter of consistency of principle between public and private international law, to limit the ability of a court in one country to prematurely hear a case involving human rights abuses in another state before that state’s own courts have had the opportunity to hear it. Significantly, this idea of an available alternative forum conceptually links the local remedies doctrine of international human rights law and forum non conveniens of (common law) private international law.86 A closer examination reveals, however, that the structure of a transnational tort action between private, individual litigants does not actually raise the types of concerns that the local remedies rule addresses and thus the doctrine should not impede transnational civil litigation, even if it does have indirect residual relevance as a way of thinking about comity amongst courts under the forum non conveniens doctrine. First, take the customary process of espousal, outside the modern human rights treaty structure. The local remedies rule applies where aliens have a claim against a foreign state. When local remedies are available, the alien is not entitled to have his or her national government espouse the claim against that state on the international plane until the local remedies have been exhausted. In theory, an unredressed injury to an alien 83 Ambatielos Arbitration (Greece v. United Kingdom) (1956), 12 RIAA 83; 23 ILR 306; Draft Articles on the Origin of State Responsibility, 2 YBILC [1980], article 22; A A Cançado Trinidade, The Application of the Rule of Exhaustion of Local Remedies in International Law (Cambridge, Cambridge University Press, 1983) at 9, 11–12, 57–8. 84 Optional Protocol, supra n. 78, article 2; Kindred, supra n. 11, at 573. 85 Kindred, ibid. at 620. 86 This connection will be clearer when the forum non conveniens doctrine is discussed infra section 4.

178 A C McConville constitutes an injury to the alien’s national state and gives rise to international responsibility.87 Transnational tort claims are not structured in this way for several reasons. The plaintiffs are not often aliens suing a foreign state; rather, they are usually nationals, or former nationals, of the state in which they were tortured or enslaved who seek compensation from the individuals responsible for their suffering.88 Furthermore, the plaintiffs are not asking Canada to espouse their claims. The issue is simply one of access to the Canadian courts system to bring a private action against non-resident defendants in respect of extraterritorial conduct.89 This chapter is premised on transnational human rights suits against individuals not against states themselves and, so, the above arguments suffice for present purposes. However, for the sake of completeness, it might be noted that there are good reasons to think that any civil claim against a foreign state to which sovereign immunity does not apply cannot, by definition, engage the sovereignty interests that underpin the local remedies rule. Second, in transnational tort litigation, the plaintiffs are not prematurely bringing an action on the international plane either through the interposition of a state espousing the claim or through a complaint to an international body. The litigant is proceeding at the level of the domestic court through private law. In this context, the role of international human rights law is not to ground state responsibility for international wrongs. Third, even if the local remedies rule applies in some fashion, it only arises when effective remedies are available “as a matter of reasonable possibility”.90 Exhausting local remedies may be problematic in transnational human rights torts litigation. It requires relying on the judicial fora of a state that engages in systemic rights violations to provide a remedy for the conduct. Pressing a claim in these conditions risks leaving a victim open to further abuse. Evidence from the Sudan suggests that the Sudanese courts are not reasonably available to redress the effects of state-sanctioned torture and slavery as the judiciary is largely subservient to the governmental regime and special security courts exact severe sentences and punishments for individuals who oppose the regime.91 In December 1997, for example, thirty-five women were arrested and charged with 87

Hackworth, Digest of International Law (1943), vol. 5 at 462 in Kindred, supra n. 11, at 564–5. It is legally possible, however unlikely, for a state such as Canada to espouse a human rights claim of a non-national, such as a recently-arrived refugee against a foreign state, using the erga omnes doctrine to give it standing. Even if the doctrine of local remedies still applies to a stateto-state process, it is a fortiori no more relevant to individual-to-individual claims here than where the espousal is on a traditional nationality-of-claims basis. See M T Kamminga, Inter-State Accountability for Violations of Human Rights (Philadephia, University of Pennsylvania Press, 1992) for a full argument on why states are entitled under international law to exercise full diplomatic protection on behalf of human rights victims, regardless of nationality. 89 In any event, individuals (whatever their link to a potential espousing state) must now be viewed as holding human rights in their own right, with espousal being a procedural avenue to vindicate those rights and not reflective of any underlying substantive right of the state, other than its own interest in seeing human rights respected. 90 Brownlie, supra n. 10 at 497; Kindred supra n. 11, at 565. 91 Sudan Report 1998, supra n. 3, at 5–7. 88

Taking Jurisdiction 179 security offences when they tried to deliver a petition to United Nations offices. The women were tried and convicted on the same night as their arrest. One of the women and four of their lawyers were sentenced to lashes.92 That being said, where a state’s courts are not part of a systemically-repressive state structure, like that in the Sudan, the jurisprudence dealing with when a remedy is a “reasonable possibility” could inform that part of the forum non conveniens analysis which asks the threshold question of whether there is another alternative forum.93 A likely context in which the courts of a state that has tortured on a widespread basis will be able to provide an alternative is a situation in which a repressive regime has been replaced by a non-repressive regime and the courts in the new system are demonstrably capable of being independent and fair in their judgment of claims of torture under the previous regime. (iii) Dédoublement fonctionnel The prospect of using international human rights law and the universality principle to justify civil jurisdiction over transnational torts raises the broader issue of whether it is appropriate to adjudicate public international law obligations in a domestic court through a private action. Appropriateness will be discussed in the context of the later discussion of forum non conveniens, but it is first useful to refute the idea that such litigation is generally inappropriate. One way to reconcile the public and private aspects of transnational human rights litigation is to view domestic courts as performing a double function as agents of both the domestic and international legal systems.94 Under this conception, substantive international rights that inure to individuals should penetrate domestic law to the greatest extent possible within the constitutional structure of a given state.95 Thus, where a domestic court is in a position to give effect to international human rights values and where no other court is better placed to do justice between the parties, the court should accept jurisdiction and thereby promote inter-party justice and international justice at the same time.96 Performing this role requires that the courts treat individuals as possessing all relevant incidents of international legal personality and accept that human rights norms are amongst the norms that govern the relationships between individuals. Canadian courts are hesitant, however, to treat public international law obligations in this way. In R v. Finta, for instance, the majority of the SCC gave a narrow reading to the court’s jurisdiction to hear cases of war crimes and crimes against humanity, which attract universal jurisdiction under the Geneva 92

Sudan Report 1998, ibid. See discussion infra Section 4. 94 Randall, supra n. 12, at 152. Randall borrows from Georges Scelle’s concept of dédoublement fonctionnel. See G Scelle, “Règles Générales de la Paix”, (1933) 46 RCADI 1 at 358–9, 421–7. 95 W R Lederman, “The Private International Law System: Some Thoughts on Objectives, Methods and Relations to Public International Law” in W R Lederman et al. (eds) Canadian Perspectives on International Law and Organization (Toronto, University of Toronto Press, 1974). 96 Randall, supra n. 12. 93

180 A C McConville Conventions, other treaties and customary law. Section 7(3.71) of Canada’s Criminal Code enables Canadian courts to prosecute individuals for war crimes or crimes against humanity. Rather than viewing this provision as conferring broad jurisdiction, the majority of the court held that the section gives a Canadian court authority to try individuals living in Canada for crimes allegedly committed abroad only where certain requirements are made out. The most important elements were those that convert an ordinary crime into a war crime or a crime against humanity.97 The majority decision suggests that, even where an international convention confers universal jurisdiction and a domestic statute authorises criminal proceedings, Canadian courts may read additional elements into the international norms in order to create a high threshold for assuming extraterritorial jurisdiction.98 If the tenor of the majority’s reasoning in Finta is any indication, a Canadian court may not be willing to take civil jurisdiction over extraterritorial cases of torture and slavery in the absence of domestic statutory authority.99 The SCC has also held that the public law obligations set out in the Charter do not directly apply between private litigants; they only apply to government conduct.100 Translating international human rights obligations from the public international plane into the private law realm is an even larger step, especially in light of the prevailing rule that treaties do not have direct force of law in Canada and that one reason for this is to prevent federal encroachment on provincial legislative powers over private law.101 Transnational human rights tort litigation challenges the norm of territorial civil jurisdiction. The principle of international comity and the constitutional 97

Finta, supra n. 65. With respect to crimes against humanity, the SCC read into the textual definition the additional element that the inhumane acts must be based on discrimination against, or the persecution of, an identifiable group of people. The mental element required to be proven for either of these additional elements is that the accused was aware of, or wilfully blind to, facts that made his or her actions a crime against humanity. Ibid. at 813 and 820. Similarly, in Suresh v. Canada (Ministry of Citizenship and Immigration) [2000] Doc. A-415-99 (FCA), the Federal Court of Appeal held that the duty not to deport an individual from Canada to a country where the individual would be exposed to a risk of torture is not absolute. While deportation in such circumstances is a breach of the guarantee to life, liberty and security of the person in section 7 of the Charter (reading in the rights from the CAT), the FCA reasoned that it can be justified under section 1 as being necessary for national safety and security where the individual in question has been found to be a member of a terrorist organization. See also Ahani v. Canada (Ministry of Citizenship and Immigration) [2000] Doc. A. 413-99 (FCA). 99 Whether motivated by some antipathy toward, or ignorance of, international law or whether related to an implicit liberal perspective that interprets criminal law narrowly in favour of the accused, the result is the same. 100 Retail, Wholesale and Department Store Union, Local 580 v. Dolphin Delivery Ltd. [1986] 2 SCR 573, 33 DLR (4th) 174. Government action was also important in the SCC’s majority decision in Cook, in which case an extraterritorial application of the right to counsel in section 10(b) of the Charter right was permissible. The Charter was applicable because Canadian officers, as agents of the state, were engaged in governmental action at the time that the right was breached. See Cook, supra n. 45 at 624–5, 629. 101 AG Canada v. AG Ontario (Labour Conventions) [1937] AC 326, [1936] 1 DLR 673 (PC). Section 92 of the Constitution Act, 1867 (UK), 30 & 31 Vict., c.3 gives the provinces legislative authority over private law. 98

Taking Jurisdiction 181 real and substantial connection test reinforce the centrality of the territorial principle in Canadian conflict of laws. Several arguments have been canvassed that may be marshalled to support a departure from the norm in transnational human rights tort cases in the interests of fairness and justice, and in ways consistent with the nationality, objective territoriality and universal principles from public international law. Despite the weight of the equities, however, arguments based on public law principles may not persuade a domestic court to exercise civil jurisdiction between private litigants. This is particularly so given that transnational human rights litigation involves extraterritorial conduct and relies on public international human rights law norms, neither of which elements fit comfortably with the disposition of our non-internationalist judiciary. The structural norms of territoriality, comity and fairness to parties are not only essential framework considerations, but they also pervade the basic rules of service and jurisdiction in Ontario. These procedural rules provide the positive law basis for an Ontario court to determine when it can appropriately take jurisdiction over an action and, like the public and private international law norms, they tend to reinforce the common law idea that jurisdiction is presumptively territorial. The issue of appropriateness also links the doctrines that operate at the international level, such as the local remedies rule, with those at the disposal of the provincial civil courts, especially forum non conveniens. At either level, this testing of the plaintiff’s choice of forum against reasonably available alternatives is a way to protect against impermissibly extraterritorial assertions of jurisdiction. It also affects the prospects for transnational tort litigation in Ontario and in the other Canadian provinces that have broadly similar rules to those in effect in Ontario. 4 SERVICE AND BASIC JURISDICTION

Assessing the likelihood of litigating a transnational torts case in the courts of Ontario requires a two-part inquiry. First, does an Ontario court have jurisdiction to hear such a case? Second, if it does have jurisdiction, should the court nevertheless decline to exercise its jurisdiction on forum non conveniens grounds? Transnational tort litigation challenges the traditional basis of appropriate jurisdiction, that is service in the territory or, at a minimum, a substantial connection to the territory of the forum court. Yet, if an Ontario court can be persuaded to answer the first question in the affirmative, the absence of a practically-available alternative forum will make a forum non conveniens argument difficult for a defendant to establish. Service In Juris The starting point for the jurisdiction of the Ontario courts is territory. The Superior Court of Justice has jurisdiction over defendants who are properly

182 A C McConville served in Ontario.102 Section 11(2) of the Courts of Justice Act provides that: “The Superior Court of Justice has all the jurisdiction, power and authority historically exercised by the courts of common law and equity in England and Ontario.”103 Where a defendant is served in Ontario, therefore, the court’s jurisdiction is largely unlimited in civil matters.104 The Ontario courts recognise the plaintiff’s right to bring his or her action in the forum of his or her choice. A corollary of this right is that a court cannot refuse to hear a case simply because the relevant facts are geographically complex. Even where there is a legitimate basis for the exercise of a court’s jurisdiction, however, it must be balanced against the court’s inherent right to prevent an abuse of its process by unfair choices of forum by the plaintiff and its discretion to decline to exercise jurisdiction in appropriate cases.105 In some cases, a plaintiff who wants to bring an action in Ontario finds that the defendant is not amenable to service in Ontario. This is likely to be the case where a plaintiff sues for serious abuses of human rights that were inflicted in another country as the perpetrator will not ordinarily be present in the province. There are two possible ways to initiate proceedings in this situation: by exercising transient jurisdiction and by service ex juris. Transient jurisdiction refers to an exercise of jurisdiction by a court over a non-resident defendant who is served while temporarily present in the territory. The transient defendant may be in the jurisdiction on a short-term visit or even on a stopover en route to another destination. The foregoing assumes that civil jurisdiction by way of service on a transient defendant is justifiable under the territorial sovereignty model of jurisdiction which allows courts jurisdiction to adjudicate over all persons who are physically present within its territorial boundaries simply by virtue of them being subject to that state’s power.106 Since the SCC decision in Hunt, however, the legitimacy of jurisdiction based solely on the transient presence of a defendant in the territory of the forum court is in doubt. In Hunt, La Forest J. stated that some of the traditional rules of jurisdiction require rethinking in light of the Morguard principle that a provincial court’s jurisdiction must be properly limited to actions with which it has a real and substantial connection. Moreover, to the extent that the fortuitous presence of the defendant may result in the defendant being compelled to defend his or her conduct in a court that has no other connection to the action or parties, the transient presence rule does not comport with either the order aspect or the fairness aspect of the principle of comity.107 Given the doubtful legitimacy of transient jurisdiction and the fact that it requires knowing if and when a potential 102

Proper service usually means personal service. Rules, supra n. 60, rule 16. Courts of Justice Act, R.S.O. 1990, Chap. C-43, section 11(2). 104 Cook v. Ip (1985), 52 OR (2d) 289 (CA); leave to appeal to SCC refused (1986), 55 OR (2d) 288 (note) (SCC). 105 Moreno v. Norwich Union Fire Insurance Society Ltd (1971), 1 OR 625 (HCJ); Courts of Justice Act, section 106; Frymer, supra n. 35, at 83–4. 106 Maharanee of Baroda v. Wildenstein, supra n. 14; MacLachlan and Nygh, supra n. 11. 107 Walsh, supra n. 29 at 407. 103

Taking Jurisdiction 183 defendant will be in the province, the more likely way to commence a transnational tort action is through service ex juris.

Service Ex Juris The second way to bring a foreign defendant under the jurisdiction of the Ontario courts is to serve the individual with an originating process outside of the jurisdiction. The Ontario Rules of Civil Procedure provide broad rules for service ex juris. Service outside of Ontario may be made without the leave of the court in cases falling within the list enumerated in rule 17.02. The enumerations describe claims that have significant connections to the province. For example, service ex juris without leave is available where, inter alia, the claim is in respect of property in Ontario (17.02(a) ), a contract made in Ontario (17.02(f) ), a tort committed in Ontario (17.02(g) ), damage sustained in Ontario (17.02(h) ) or against a person ordinarily resident or carrying on business in Ontario (17.02(p) ). In addition to the conditions for automatic service ex juris, rule 17.03 gives the court a broad discretion to grant leave ex parte to serve an originating process outside Ontario in any case not coming within the enumerations in rule 17.02. Here, in the ordinary course, the court will consider forum non conveniens arguments as part of deciding whether to allow service. That being said, if the court does allow service, the served defendant may still apply to set aside the service based on the fuller forum non conveniens arguments that would not have been made in an ex parte leave proceeding. Where service is effected without leave, the burden falls on the intended defendant to apply to have the service set aside for failure to come within the rule 17.02 list or, failing that, on the basis that there is a more appropriate foreign court that should hear the case. (i) Rule 17.02(h) for damages in Ontario On its face, rule 17.02 provides potential openings for service ex juris in transnational human rights tort cases. In particular, rule 17.02(h) provides: “A party to a proceeding may, without court order, be served outside Ontario with an originating process or notice of a reference where the proceeding against the party consists of a claim or claims, ... (h) in respect of damage sustained in Ontario arising from a tort, breach of contract, breach of fiduciary duty or breach of confidence; . . .”108

This provision leaves room for a plaintiff who is currently residing in Ontario and suffering physical, mental or emotional distress in Ontario as a result of being subject to torture, slave-trading, or other serious abuses of human rights 108

Rules, supra n. 60, rule 17.02(h).

184 A C McConville in another country to serve the perpetrators of the conduct ex juris without leave.109 Jurisdiction is even clearer where new manifestations of harm occur in Ontario, such as psychological terror that results from suppressed memories coming to the surface. Once service is properly effected, the Ontario court has jurisdiction as of right over the defendant. An expansive interpretation of rule 17.02(h) is consistent with the policy underlying the rule. Before 1975, the procedural rules for Ontario required leave of the court any time a plaintiff sought to serve defendants out of jurisdiction. In 1975, the rules were changed so that leave for service ex juris was no longer required where the claims were consistent with those set out in current rule 17.02. This change was intended to make the Ontario courts more accessible to plaintiffs in Ontario.110 It also reflects the underlying principle that a plaintiff has a prima-facie right to choose the forum for his or her suit.111 The courts have construed “damage” in rule 17.02(h) to include both physical and mental suffering arising from torts in which the harm-causing conduct occurred in another province or in another country.112 In Vile v. Von Wendt; Zurich Insurance Co., for example, the plaintiff, an Ontario resident, was injured at a ski resort in Quebec when a concrete statue fell on him. He sued a German resident who was the owner of the resort. Linden J. of the Ontario High Court of Justice held that “damage” in then rule 25(1)(h) included all heads of damage. It was irrelevant that the initial injury occurred in Quebec; other damages, including pain and suffering, lost wages and profits, occurred in Ontario. The defendants in Jaffe v. Dearing challenged the validity of service ex juris of a statement of claim.113 In that case, the plaintiff had been abducted in Ontario and brought to Florida by an American bounty hunter. He was handed over to US prosecutors and held on criminal charges. When he was subsequently released on bail, he absconded. He returned to Ontario where he commenced an action against a bank and an attorney, both resident in Florida, in connection with the bank’s seizure of the plaintiff’s security in respect of a forfeited appearance bond. Even though the asset was seized in Florida, the plaintiff chose to litigate in Ontario for fear that he would be incarcerated in connection with the criminal charges against him if he returned to that state. The defendants brought a motion to set aside service ex juris or have the action stayed. Sutherland J. summarised the preceding case law as establishing that rule 17.02(h) is met when a plaintiff is resident in Ontario and the harm or damage 109 The term “damage” in rule 17.02(h), ibid., is presumptively broader than the wording of section 6(a) of the State Immunity Act, RSC 1985, Chap.S-18, which provides that a foreign state is not immune from the jurisdiction of a court in proceedings that relate to “any death or personal injury” that “occurred in Canada”. 110 Vile v. Von Wendt; Zurich Insurance Co. (1979), 103 DLR (3d) 356 (Ont HCJ) [hereinafter Vile]; S. Block, “Service Ex Juris”, (1983) 4 Advocates’ Q 332 [hereinafter “Block”]. 111 Block, ibid. 112 Vile, supra n. 110; MacDonald, supra n. 31; B.(J.) v. D.(E) (1993), 16 CPC 242 (Ont Gen Div) [hereinafter B v. D]; de Vlas v. Bruce (1994), 18 OR (3d) 493 (Gen Div) [hereinafter de Vlas]; Wilson v. Moyers (1993), 13 OR (3d) 202 (Gen Div); Power v. Probert (1987), 19 CPC (2d) 142 (Dist Ct). 113 Jaffe, supra n. 60.

Taking Jurisdiction 185 was felt, hit home or was endured in Ontario.114 In this case, the facts that the plaintiffs were resident in Ontario and that the loss of the asset hit home in the province were sufficient to bring service ex juris within the rule. Macpherson J. of the Ontario Court General Division took a similarly expansive approach in de Vlas v. Bruce.115 The plaintiffs were Ontario residents who brought an action in Ontario under the Family Law Act against an Alberta resident for damages resulting from the death of a family member in a car accident in Alberta. When the defendant moved to have ex juris service set aside, the court dismissed the motion. Macpherson J. held that the Ontario court had jurisdiction based on service ex juris under rule 17.02(h) because, inter alia, the plaintiffs sustained damage in Ontario in the form of funeral expenses, travel costs, loss of income and loss of the guidance, care and companionship of the deceased relative. It is very significant, however, that some Ontario judges may be retreating from the earlier expansive approach. In MacDonald v. Lasnier, a decision which came after the SCC decisions in Morguard and Hunt,116 the court declined to exercise jurisdiction despite finding that, prima facie, the case fitted into rule 17.02(h). Cunningham J. held that, without some other connecting factors, “damage” in Ontario was not enough to make Ontario an appropriate forum for the trial of the action. Rather, assumptions of jurisdiction under rule 17.02(h) must be considered in light of Morguard. In contrast to the “damage” connection to Ontario, the litigation had numerous contacts with Quebec including the location of the tort, the residence of the defendants, location of medical records and many potential witnesses. Cunningham J. found that, on the facts of the case, sustaining damage in Ontario did not constitute a real and substantial connection with Ontario and hence the Ontario court did not have jurisdiction. The MacDonald case has important implications for prospective transnational tort litigation in Ontario. Cunningham J.’s comparative approach to the real and substantial connection test is a stringent interpretive application of the Morguard rule.117 It requires that there be more real and substantial connections between the action and Ontario than exist between the action and any other forum before the court can properly exercise jurisdiction. This approach could effectively preclude transnational tort actions as there will almost always be more contacts with the place of the delict to outweigh “damage sustained in Ontario”. Morguard, itself, however, does not impose a comparative test according to which the place with the “more real” and substantial connections becomes the only legitimate site of adjudication. Instead, the presence of a real 114 Ibid. at 120. See also Davidson Partners v. Citibank NA (1990), 72 OR (2d) 450 (Master) at 457–8. For a discussion of Sutherland J.’s findings on forum non conveniens, see infra text following n. 152. 115 de Vlas, supra n. 112. 116 MacDonald, supra n. 31. 117 “Interpretive” in the sense that the Morguard and Hunt test is read into rule 17.02(h) as opposed to being treated as ultra vires by virtue of overreaching provincial jurisdiction.

186 A C McConville and substantial connection is sufficient for constitutionally permissible jurisdiction. Under Morguard, minimal sufficient contacts are all that is needed. Not all the post-Morguard decisions from the Ontario courts are as restrictive as MacDonald. Some judges continue to accept the validity of service ex juris provided that the case involves, prima facie, damage sustained in Ontario. When these cases involve personal injury, the “damage” typically includes medical expenses, lost income, pain and suffering that were sustained in Ontario as a result of an injury suffered in a foreign jurisdiction.118 It should be noted, however, that there has been almost invariably a number of significant factors connecting the action to Ontario in these cases such that “damage sustained in Ontario” has not in reality been the only basis for jurisdiction.119 (ii) Rule 17.03 discretion Rule 17.03 gives the court discretion to grant leave for service ex juris in cases that do not fall into rule 17.02. In contrast to automatic service, which gives the court jurisdiction as of right, on a rule 17.02 motion for leave to serve a defendant ex juris, the plaintiff has the burden to establish that Ontario is an appropriate forum for the action.120 Under 17.03, the common law determines whether leave will be granted on a general balance of conveniences to all parties.121 However, this residual discretion does not appear to greatly expand the ambit of the Ontario court’s jurisdiction. In National Bank v. Clifford Chance, for example, Ground J. of the Ontario Court General Division dismissed a motion to set aside service ex juris that was granted under rule 17.03 in an action for breach of fiduciary duty.122 While discretionary service was upheld, the decision turned on the fact that the claim was closely related to a recognised head in rule 17.02 for contract claims.123 Argument by close analogy to a rule 17.02 head 118 E.g.: Dino v. Albertson’s Inc. (1994), 28 CPC (3rd) 15 (Ont Gen Div) [hereinafter Dino]; Trepanier v. Kloster Cruise Limited (1995), 23 OR (3rd) 398 (Gen Div); Dunlop v. Connecticut College (1996), 50 CPC (3rd) 109 (Gen Div) [hereinafter Dunlop]. 119 Ibid. See also Acronym (Cayman) Inc. v. Ontario Lottery Corp. (1997), 12 CPC (4th) 331 (Ont Gen Div); Holodeck Adventurer Ltd. v. Orbotron Inc. (1996), 8 CPC (4th) 376 (Gen Div). It should also be noted that in at least two recent decisions, Ecolab Ltd. v. Greenspace Service Ltd. (1998), 38 OR (3d) 145 (Div Ct) and Ontario New Home Warranty Program v. General Electric Company (1998), 36 OR (3d) 787 (Gen Div), the court held that when a defendant moves to set aside service ex juris on the ground that it does not fall within the ambit of rule 17.02, the court must decide whether the plaintiff can demonstrate that it has a good arguable case for trial. This suggests that even where it is clear that damages were sustained in Ontario, the plaintiff’s pleadings must disclose a good arguable case in tort giving rise to the damages. As things currently stand, the cause of action presumably must be sound in tort and not be solely derived from international human rights law. 120 Rules, supra n. 60, rule 17.03. 121 K Feldman and S Vella, “The Evolution of Forum Conveniens: Its Application to Stays of Proceedings and Service Ex Juris”, (1989) Advocates’ Q 161 [hereinafter “Feldman and Vella”]. 122 National Bank of Canada v. Clifford Chance (1996), 30 OR (3d) 746 (Gen Div) [hereinafter Clifford Chance]. At that time, “breach of fiduciary duty” was not specifically listed in rule 17.02. 123 Clifford Chance, supra n. 122. Note that, although the motion to set aside service ex juris was dismissed in Clifford Chance, the action was stayed on forum non conveniens grounds. See also Upper Lakes Shipping Ltd. v. Foster Yeoman Ltd. (1992), 12 CPC 31 (Ont Gen Div).

Taking Jurisdiction 187 of service would appear to be the primary method for persuading a court to apply the rule 17.03 discretion. Also, the real and substantial connection test necessarily limits the scope of rule 17.03 no less than rule 17.02. Nevertheless, the option of seeking leave of the court to serve a defendant ex juris is a window for transnational tort litigants, particularly if the “damage sustained in Ontario” head for automatic service is unavailable or too ephemeral to ground jurisdiction on its own. On a rule 17.03 motion, the plaintiff has an opportunity to draw on public international law jurisdictional principles to give content to the court’s discretion.124 Following the cue from Clifford Chance, a way to potentially extend the ambit of the residual discretion and bring the action within the court’s jurisdiction is to use these principles to show that a transnational human rights tort case is analogous to one or more of the enumerated heads in rule 17.02. For example, rule 17.02(p) provides for automatic service ex juris on a defendant who is ordinarily resident in Ontario. It could be argued that nationality jurisdiction is analogous to ordinary residence. Like ordinary residence, nationality is evidence of a defendant’s allegiance to the state and concomitant submission to its legal system. A national can reasonably expect to account for his or her tortious conduct before the courts of the national state. By analogy, nationality is as substantial a connecting factor as ordinary residence for the purpose of service ex juris or, at least, it may be in a given context. Furthermore, this approach does not require a large departure from the jurisdictional norms. Common law courts already have authority over resident defendants who are served abroad under the service ex juris rules.125 Courts in civil law jurisdictions are also often empowered to use a defendant’s nationality as a basis for private law jurisdiction. In the case of Sudan, the prospective defendants are Canadian nationals and at least one is a past resident of Ontario. This correlation between nationality jurisdiction and the “ordinarily resident” head could justify leave to commence proceedings in a transnational tort action via rule 17.03. As an additional connection to the province, this factor supports the appropriateness of the Ontario court’s jurisdiction. Similarly, a rule 17.03 motion for leave to serve ex juris is an opportunity to argue that universal jurisdiction over jus cogens and erga omnes norms against torture and slavery is reason for the court to exercise its discretion and take jurisdiction. Linking rule 17.03 discretion to universal jurisdiction would permit the “truly” transnational claim to go forward.126 On a motion for leave to serve ex juris, a plaintiff must establish that Ontario is an appropriate forum for a transnational tort action as part of establishing a sufficient jurisdictional nexus. As outlined above, public international law principles show that such an extension of Ontario’s jurisdiction to adjudicate the legality of conduct of this 124

See supra text, at subsection (ii), for a discussion of these principles. Walsh, supra n. 29, at 409. 126 True transnationality is meant in this sense: that neither plaintiff nor defendant has any connection to Canada at the time of the acts of torture or slavery. 125

188 A C McConville nature is not impermissibly extraterritorial. In other words, rule 17.03 should be interpreted as the statutory basis for an Ontario court to take jurisdiction wherever international law permits. The courts’ traditional role in private common law claims also means that it should not be necessary to wait for legislative authorisation.127 Even though the tortious conduct occurs abroad, in most cases the plaintiff’s current residence will be Ontario, where he or she will have moved, often as a refugee, after being tortured or enslaved. This provides a link with the province, a kind of attenuated passive personality link.128 Where there is no other clearly more appropriate forum in which a victim of a human rights tort may enforce his or her rights under international law and receive a remedy, the Ontario courts would be justified, in order to prevent serious injustice, to base jurisdiction on universal jurisdiction over the most serious human rights violations in addition to the plaintiff’s presence in the jurisdiction. The presence of a discretionary power in rule 17.03 permits the court to hear claims that do not fit the usual private law paradigms and, therefore, were not contemplated in rule 17.02.129

Forum Non Conveniens Analysis Even if service of an originating process is effectively served, an Ontario court may decline to exercise its jurisdiction on the grounds of forum non conveniens, a doctrine mentioned in passing several times already in this chapter. Forum non conveniens is a doctrine whereby a court has a duty to decline jurisdiction because it is not an appropriate forum for the action. Regardless of the form of service, a defendant may invoke the doctrine to challenge the jurisdiction of the court. Under rule 17.06 of the Rules of Civil Procedure, a defendant who is served ex juris can bring this challenge in two ways: the defendant may bring a motion to set aside the service under rule 17.06(1)(a) or the defendant may move for a stay of proceedings under rule 17.06(1)(b). The court will grant either order if it is satisfied that “Ontario is not a convenient forum for the hearing of

127 The federal division of powers may come into play and prove important if the provincial courts were to take jurisdiction in this way. 128 Attenuated because, whatever the plaintiff’s current status in Canada, he or she will not have been a Canadian national at the time of the tort. 129 In Suleiman Al-Adsani v. Kuwait et al. (1996), 107 ILR 536, however, the English Court of Appeal refused to read down the English State Immunity Act of 1978 in light of the jus cogens norm against torture. This precedent may suggest difficulty in reading the Rules broadly enough to capture transnational human rights litigation, but there is an important difference. In Al-Adsani, the plaintiff asked the court (a) to take jurisdiction against a state (b) in the face of a very clear statutory provision that, on its face, precluded taking jurisdiction. Under rule 17.03, we are contemplating an individual defendant and we are faced with a broad permissive provision. Note as well a preliminary judgment in Al-Adsani granted leave to serve ex juris, one basis being the assumed likelihood that the plaintiff would be able to show harm from torture in Kuwait manifesting itself in the United Kingdom.

Taking Jurisdiction 189 the proceeding.”130 A defendant who is served within the jurisdiction can also challenge the court’s jurisdiction on forum non conveniens grounds under s. 106 of the Courts of Justice Act.131 To establish that a court is a forum non conveniens, there must exist some other forum that is more convenient and appropriate for the pursuit of the action and for securing the ends of justice.132 Canadian developments of the doctrine culminated with the SCC’s decision in Amchem Products Inc. v. British Columbia (Workers’ Compensation Board). In that case, Sopinka J. reviewed the doctrine in the course of ruling on a British Columbia court’s authority to order an anti-suit injunction against a party to legal proceedings in Texas, who was also defendant in British Columbia proceedings. Following the approach in Spiliada, Sopinka J. held that the essential question in a forum non conveniens analysis is the “appropriateness” of the local forum compared with alternatives elsewhere.133 Most commonly, Canadian courts approach this question by assessing the impact of specific facts on the parties’ respective interests and on the administration of justice. The court will consider all factors pertinent to this determination.134 These factual circumstances include: the difficulty and expense each party will have in litigating the case before one or another forum, the location from which witnesses and documentary evidence must come, the applicable law, and juridical advantage to the plaintiff in the local forum.135 As has been noted already in discussing the local remedies rule, it is crucial to note that the starting point of a forum non conveniens argument must be that there is, in fact, some alternative to the local forum.136 In a transnational torts suit, the natural alternative forum is the courts of the state in which the abuse 130 Rules, supra n. 60, rule 17.06(2)(c). The court may also grant such an order if it is satisfied that service outside Ontario is not authorised by the rules (17.06(2)(a) ) or a rule 17.03 order granting leave to serve outside Ontario should be set aside (17.06(2)(b) ). 131 Courts of Justice Act, supra n. 103, section 106 provides: “A court, on its own initiative or on a motion by any person, whether or not a party, may stay any proceeding in the court on such terms as are considered just.” 132 Antares Shipping Corp. v. The “Capricorn” [1977] 2 SCR 422, 65 DLR (3d) 105 [hereinafter Antares]; Amchem Products Inc. v. British Columbia (Workers’ Compensation Board) [1993] 1 SCR 897, 102 DLR (4th) 96 [hereinafter Amchem cited to SCR]. 133 Amchem, ibid. at 916–21. 134 Spiliada, supra n. 16; Amchem, supra n. 132; Frymer, supra n. 35. 135 In England, if the defendant shows that there is a more appropriate forum, the court will grant a stay unless the plaintiff can show that the interests of justice require that it not be granted. A substantial loss of juridical advantage is considered in the second step. Spiliada, supra n. 16. In Amchem, supra n. 132, Sopinka J. departed from the English approach and held that juridical advantages are factors to be weighed with all the other relevant factors to identify the appropriate forum. Presumably, where substantial injustice would result from loss of a juridical advantage associated with suing in Ontario, this would be given significant weight in Sopinka J.’s one-step operation. To consider simple juridical differences between the two candidate fora would, however, be inconsistent with other statements Sopinka J. made, notably that a juridicial advantage to the plaintiff is equally balanced by a corresponding disadvantage to the defendant. 136 Antares, supra n. 132; Recherches Internationales Quebec v. Cambior Inc. [1998] QJ No. 2554 (Que SC) [hereinafter Cambior]; J Blom, “Rules for Declining Jurisdiction in Civil and Commercial Matters in Common Law Canada: Forum Non Conveniens, Lis Alibi Pendens”, Contemporary Law 1994 (Montreal, McGill University, Institute of Comparative Law, 1994).

190 A C McConville actually occurred.137 On the concrete fact situation being used in this chapter, the courts of the Sudan would be the prima-facie alternative. It has already been explained why Sudan’s courts do not provide a “reasonable possibility” of remedies (the local remedies test).138 As a direct corollary, the threshold question of forum non conveniens analysis that one must ask before assessing comparative appropriateness—is there a reasonable alternative in the first place?—must be answered in the negative. However, on the counterfactual assumption that the Sudan’s courts could and would be willing to provide a remedy, the appropriateness balancing exercise should be carried out. Each of the potentially relevant factors will be considered in turn, although it is recognised that this may seem like a strained exercise given the fact that in many, if not most, human rights cases the foreign courts will be unreasonable alternatives such that one would not even need to engage in this comparative analysis. For plaintiffs resident in Ontario, litigating in the local courts will generally be more convenient and less costly than litigating abroad. There may also be some documentary evidence, such as medical reports, and possibly some witnesses located in Ontario. The majority of evidence and witnesses, however, will likely be located in the place where the torture and slavery took place. Bringing this evidence to an Ontario court will be costly for the parties, particularly for the defendants who are also currently resident in the Sudan. On the basis of convenience and expense, the Sudan would seem to be a more appropriate forum. However, a pro-plaintiff presumption operates within the Canadian forum non conveniens doctrine. In fact, in Ontario, the basic assumption is that the plaintiff’s choice of forum should not be displaced unless the defendant clearly establishes the existence of a more appropriate forum.139 This applies to foreign plaintiffs and recently-arrived non-citizens no less than to permanent residents or citizens. Where there are significant connections to more than one forum, the Ontario courts will not grant a stay of proceedings on forum non conveniens grounds unless it is established that justice can be done in the alternative forum at substantially less inconvenience and expense. In B v. D, Corbett J. dismissed a motion to set aside service ex juris and stay the proceedings in an action for damages arising from alleged sexual assaults perpetrated by the defendants in the Netherlands and in Ontario.140 At the time that the action was commenced, 137 Other possible alternative fora are international adjudicative bodies, such as the Human Rights Committee. In civil suits that avoid the operation of the state immunity doctrine, the state could argue that the international forum can more appropriately determine whether international law was violated, if the state where the tortious conduct occurred is a party to an applicable convention and its supervisory bodies. See supra text, at Section 3, for a discussion of this issue. 138 Supra, n. 93. 139 Amchem, supra n. 132; Clifford Chance, supra n. 122; Dy 4 Systems, Inc. v. Diamond Point International, Inc. (1997), 14 CPC (4th) 99 (Ont Gen Div); Frymer, supra n. 35; B v. D, supra n. 112; Dunlop, supra n. 118; Davidson Tisdale Ltd. v. Pendrick (1997), 10 CPC (4th) 336 (Ont Gen Div); Ontario New Home Warranty Program v. General Electric Co. (1998), 36 OR (3d) 787 (Gen Div); Castel, supra n. 11. 140 B v. D, supra n. 112.

Taking Jurisdiction 191 three defendants resided in the Netherlands. The fourth defendant was resident and served in Ontario where the plaintiffs were also resident. Given that the alleged assaults occurred both in Ontario and the Netherlands, it was expected that witnesses and documentary evidence were located in both jurisdictions. Corbett J. found that, although there were significant connections to the Netherlands, on balance the Netherlands was not a substantially more convenient forum for the trial for the action than Ontario.141 Another factor favouring the Sudan is the applicable law. In Canada, the law of the place of the delict is the presumed default rule for the choice of law in tort cases and, under lex loci delicti, the law of an unfamiliar legal system must be proved and applied to dispose of the matter. This presumption may discourage an Ontario court from assuming jurisdiction in a transnational tort case. In Clifford Chance, for example, the fact English law would apply to the case was one reason that Ground J. ultimately declined to exercise jurisdiction even though English common law is hardly unfamiliar to an Ontario judge.142 One way to meet this choice of law concern is to invoke the fairness exception from Tolofson, the case which established lex loci delicti as the presumptive tort choice of law rule. In his reasons, La Forest J. left open the possibility of exceptions to the rule in international litigation and where a departure is necessary to prevent an injustice.143 The absence of fairness in the legal regime of the forum of the delict is a feature of transnational human rights tort cases. Under the exception, the court could, in lieu of applying the lex loci delicti, apply either Ontario-specific or general common law tort principles to redress torture and slavery.144 If the Ontario courts were persuaded that international law applies in some fashion as the governing law of the tort, the choice of law factor would likely favour Ontario. The ideas of fairness and justice explain the importance that the notion of a legitimate juridical advantage traditionally held in forum non conveniens analyses. In earlier formulations of the doctrine, a court could keep jurisdiction, notwithstanding the existence of a more appropriate forum, where a plaintiff established that not litigating in the local forum would result in a loss of a legitimate personal or juridical advantage.145 When the SCC considered the doctrine in Amchem, Sopinka J. found that treating this factor as a distinct branch of the test was simply a product of the way the doctrine developed in England; there was no reason in principle to maintain the separate treatment. Instead, the juridical advantage is one of the factors relevant to the appropriateness of the forum. Nevertheless, in practice, juridical advantages can sometimes carry 141 In addition, the plaintiff would be deprived of the right to compel answers from the defendants on discovery if they proceeded in the Netherlands. 142 Clifford Chance, supra n. 122. See also Holodeck Adventures Ltd. v. Orbotron Inc., supra n. 119. 143 Tolofson, supra n. 20, at 1054–6. 144 This shows how intertwined the jurisdiction analysis will be with the choice of law analysis discussed by various other contributors to this volume. 145 Bonaventure Systems Inc. v. Royal Bank of Canada (1986), 57 OR (2d) 270 (Div Ct).

192 A C McConville overriding weight. They are particularly persuasive when the plaintiff’s case has a real and substantial connection with the forum and the plaintiff has a legitimate claim to the advantages that the forum provides.146 The value-laden nature of this inquiry is obvious and the fact that the tort cause of action is simultaneously a human rights violation will help to establish the legitimacy of juridical advantage.147 The Ontario courts weigh advantages according to the nature of the advantage and the context of the case. The availability of oral discovery was not a sufficient benefit to tip the balance in favour of Ontario in the Clifford Chance case in an action for negligence against solicitors who prepared an opinion on a corporate financing.148 In contrast, the court in B v. D found that, in a case arising out of sexual assaults, the ability to compel answers on discovery in Ontario was a significant advantage to the plaintiff that would not be available in the alternative forum, the Netherlands.149 A benefit like access to legal aid being available to the plaintiff may not be enough to persuade an Ontario court to keep jurisdiction,150 but a statutory right of action available to the plaintiff in Ontario has been held to be a legitimate juridical advantage.151 Where the plaintiff risks being jailed in the alternative forum, the advantage to him or her of litigating in the province is also sufficiently significant.152 In Jaffe, the court considered whether a fear or likelihood of criminal prosecution or incarceration before trial in the natural forum can amount to a legitimate personal and juridical advantage. Recall that the plaintiff, who was the subject of outstanding criminal charges in Florida, commenced a civil suit in Ontario against two defendants resident in Florida. That Florida was the natural forum was not in dispute: the proper law of the contract in question was Florida law, the contract was formed in Florida and the alleged breach occurred in Florida. Sutherland J. observed that there was no doubt that the likelihood of the plain146 Amchem, supra n. 132. In the absence of a connection, the plaintiff’s choice of a forum offering several personal advantages is tantamount to forum shopping. Forum shopping is an activity that the forum non conveniens test is intended to minimise. 147 The language of “legitimate” advantage would seem to belong to pre-Amchem case law. PostAmchem doctrine must speak in terms of whether or not a substantial injustice would occur if a plaintiff were to lose an advantage offered by the law of her or his chosen forum. When subsequent reference is made to pre-Amchem case law, the preceding rough translation of concepts should be borne in mind. 148 Clifford Chance, supra n. 122. 149 B v. D, supra n. 112. Note, however, that In re Union Carbide Corporation Gas Plant Disaster at Bhopal, India in December, 1984, 634 F Supp 842 (SDNY 1985) the US court granted a stay finding that India was the more appropriate forum for the action than the US, but required that discovery be available. However, this condition was struck from the order on appeal: In re Union Carbide Corp. Gas Plant Disaster at Bhopal, 809 F.2d 195 (2d Cir. N.Y. 1987). 150 MacDonald, supra n. 31. But see the breakthrough House of Lords case of Connelly v. RTZ [1997], 3 WLR 373 (HL), in which loss of the availability of both legal aid and contingency fees were held to be reasons not to stay proceedings where another country was the more appropriate forum, at least to the extent that the court was persuaded that the plaintiff would not be able to carry on the case abroad at all due to absence of financial resources. 151 Dino, supra n. 118. 152 Jaffe, supra n. 60.

Taking Jurisdiction 193 tiff’s incarceration in Florida deterred the plaintiff from pursuing his action there whereas bringing the suit in Ontario was advantageous to him. The issue was whether it was a legitimate advantage. Sutherland J. found that as the bail bond issue arose out of the fact that the plaintiff was in custody as a result of an abduction, any further imprisonment on that ground that he jumped bail would be tainted with the illegality of his original detention and that this factor added legitimacy to the advantage gains of suing in Ontario.153 The court also assuaged the concern that permitting the suit would assist the plaintiff to evade the operation of Florida’s criminal law. On that issue, Sutherland J. stated: “Accordingly, in interpreting the word ‘legitimate’ in the phrase ‘legitimate personal advantage,’ the court is concerned with the legitimacy in the sense of the proper operation of the civil process. On that basis, Jaffe’s liberty and his consequent ability to participate in the civil litigation constitute a legitimate advantage and if one accepts, as I do, that having to bring a suit in Florida would be an effective deterrent to pursing the action at all, then the legitimacy of the advantage is seen as central and not collateral to the litigation.”154

The court was accordingly satisfied that the plaintiffs showed a legitimate personal advantage and should not be deprived of the forum of their choice. Sutherland J.’s opinion supports transnational tort litigation for two reasons. First, it suggests that a plaintiff’s fear or the likelihood of incarceration or another reprisal is a deterrent to bringing suit in the natural forum and creates a corresponding legitimate advantage to litigation in Ontario. Forcing a plaintiff to sue government ministers in the courts of a regime which is known to arrest, detain and torture its opponents puts the plaintiff’s liberty and physical security in jeopardy and effectively deters litigation. The corresponding benefit of fair process in Ontario is the type of juridical advantage that carries overriding importance. Second, the case is authority for the proposition that the illegality of an act, which was a cause in fact of the civil suit, adds legitimacy to the advantage. It follows that this legitimacy would attach to the advantage to a transnational tort plaintiff of bringing a civil suit in Ontario as the cause of action would be derived from torture and slavery which are illegal at international law. In this way, this Ontario judge’s context-driven analysis, if adopted and built upon in the future, favours plaintiffs in transnational human rights tort actions. As noted earlier, where the courts of the place of the delict are completely inimical to a plaintiff’s claim, the plaintiff can argue that no practical alternative for trial of the matter exists at all. This is the strongest way to meet a forum non conveniens challenge. Without the threshold requirement of an alternative forum, no duty on a court to stay the proceedings can arise. The English Court of Appeal took this approach in Mohammed v. Bank of Kuwait and the Middle 153 154

Jaffe, supra n. 60 at 123. Ibid. at 124.

194 A C McConville East.155 In that case, the plaintiff was an Iraqi national who was prevented by Iraqi and Kuwaiti law from returning to Kuwait. He commenced proceedings in London against the bank for money due to him on an employment contract. The Court held that, although Kuwait was the natural forum for the claim, it was not an available forum unless it was “available in practice to this plaintiff to have the dispute resolved”.156 If no more appropriate forum exists, the court may thus properly exercise jurisdiction provided that it has reasonable grounds for doing so, on the basis of both forum non conveniens law and background public law principles relevant to judicial jurisdiction.157 Moreover, even if the Sudanese courts are available, they are not likely to be a more appropriate forum for securing the ends of justice. This aspect of the forum non conveniens test should not be lost in factor-counting exercise. In Mohammed, the English court said that it was appropriate to consider whether the plaintiff will achieve substantial justice or have a fair resolution of the dispute to determine whether a foreign court is the forum conveniens. Applying this approach to the Sudan example suggests that Ontario should not decline jurisdiction. If it does so, the plaintiffs would have to sue in courts controlled by the regime that condones the impugned human rights abuses and where an exacerbating circumstance is that one of the defendants is also the minister in charge of the judicial system. While it is appropriate to decline jurisdiction in deference to a more closely connected forum where the rules and procedures are different but fair, the duty does not arise in a case derived from systematic human rights atrocities and where fundamental concern exists about the fairness available in courts that are an instrument of an abusive governmental regime. Securing the ends of justice must include ensuring that the plaintiff will receive substantial justice or a fair resolution of the dispute.158 Assessing the quality of justice available in a foreign legal system, however, is a task few courts are eager to do. Consider, for example, the Quebec Superior Court’s recent decision in Récherches Internationales Québec v. Cambior Inc.159 In that case, the plaintiff applied to launch a class action in Quebec on behalf of the victims of a pollution spill in Guyana. The spill occurred when a tailings dam, which was constructed by Omai Gold Mines Limited, ruptured and billions of litres of liquid containing cyanide and other pollutants spilled into two rivers. The plaintiffs chose to bring the application in Quebec because the Quebec civil code provides for class action suits and because Cambior, the major shareholder of Omai and the target of the class action, is headquartered in Quebec. Maughan J. considered whether the court should decline to exercise 155 Mohammed v. Bank of Kuwait and the Middle East [1996]1 WLR 1483 (CA) [hereinafter Mohammed]. 156 Ibid. at 1490; the House of Lords judgment in Connelly v. RTZ Corp. plc., supra n. 150, stands for a similar proposition. 157 A real and substantial connection constitutes reasonable grounds. Morguard, supra n. 19; Hunt, supra n. 19. 158 Mohammed, supra n. 155. 159 Cambior Inc., supra n. 136

Taking Jurisdiction 195 jurisdiction on forum non conveniens grounds.160 As one of its arguments, the plaintiff claimed that the victims of the disaster would be denied justice if the case was heard in Guyana. It called expert evidence in support of the view that the legal system was simply an appendage of a repressive legal system.161 Cambior called a number of Guyanese and Caribbean jurists who testified to the efficiency and independence of the legal system in Guyana. Faced with this contradictory evidence, Maughan J. stated: “The Court recognizes that it is difficult, if not invidious, to make comparisons between two different systems of justice. At the same time, it acknowledges that, for the purpose of a forum non conveniens inquiry, this exercise is necessary to determine whether the remedy sought by the plaintiffs is available in a foreign jurisdiction. In the present case, RIQ had failed to bring forward any conclusive and objective evidence to substantiate its belief that Guyana is an inadequate forum due to the many deficiencies which plague its system of justice.”162

If the Court had been presented with such conclusive evidence, it may have been prepared to give greater weight to considerations of fairness than to a number of factors connecting the action to Guyana and so allow the application to proceed in Quebec.163 Absent clear evidence that litigating in Guyana would result in unfairness to the victims, the Court found that Guyana was the appropriate forum. More factors linked the action to Guyana, including the location of the victims, most of the witnesses, most of the elements of proof and the appropriate law, than connected the matter to Quebec. The reasons for decision in Cambior are indicative of the courts’ discomfort with the task of passing judgment on another court system as part of a forum non conveniens analysis. The evolution of the forum non conveniens doctrine shows that the courts are moving away from paternalistic comparisons of the quality of justice available in foreign courts towards a greater willingness to find that another forum is suitable for trying a case in the interests of the parties and the ends of justice.164 As Adrian Briggs observes, when questions of whether a foreign court will treat a plaintiff fairly arise, courts are careful not to make disparaging or politically incorrect assessments of a foreign court or its procedure.165 If the court is unwilling to find that there is no alternative forum at the threshold stage, a party that has to persuade a court to find that a trial before another court will be unfair, as part of the appropriateness inquiry, will have a difficult task.

160 Although Quebec is largely a civil law jurisdiction and forum non conveniens a common law doctrine, the doctrine was codified in the most recent revision of the private international law sections of the Quebec Civil Code (Arts. 3076–168 C.C.Q.). 161 Cambior, supra n. 136 at para 83. 162 Cambior, supra n. 136 at para 98. 163 Ibid. at para. 87. 164 The Abidin Daver, [1984] AC 398; B v. D, supra n. 112; Frymer, supra n. 35; Feldman and Vella, supra n. 121. 165 A Briggs, “Decisions of British Courts During 1996”, 67 BYIL (1996) 577.

196 A C McConville

5 CONCLUSIONS : IS AN ONTARIO COURT AN APPROPRIATE FORUM ?

Several arguments support transnational human rights tort litigation in Ontario. Although this type of litigation is distinct in form and subject matter from the types of international litigation that normally attract conflict of laws rules, transnational human rights tort litigation can be accommodated within the concept of comity by invoking the principles of fairness and justice. Additionally, the presence of the plaintiff in Ontario and the province’s regulatory interest in the welfare of its residents constitute real and substantial connections and meet the constitutional requirements of appropriate civil jurisdiction. The public international law jurisdictional principles of nationality, objective territoriality and universality also support the appropriateness of an Ontario forum for transnational litigation in the interests of fairness and justice. Similar arguments help to defend service of process on non-resident defendants. Once an Ontario court takes jurisdiction, the absence of a practically-available alternative forum in states like the Sudan will make Ontario difficult to displace as a forum conveniens. These conclusions depend, however, on the court finding that an action derived from international human rights norms properly belongs in the private law domain. Once public international law values are required to justify a court’s jurisdiction, the private nature of the dispute may be difficult to sustain. Moreover, while international obligations to prevent torture and slavery attract state responsibility, they are primarily vertical obligations owed by states to individuals that may not easily translate into duties between private litigants. This uneasy relationship between the private and public aspects is a formidable obstacle to transnational human rights tort litigation in Ontario. In light of the emphasis that the Supreme Court of Canada has placed on the territorial limits to the jurisdiction of the provincial courts and on international comity, initial signs are that Ontario courts will be reluctant to exercise extraterritorial jurisdiction over torts directly derived from public international law norms. Rather than depart from the territorial norm, they may characterize transnational human rights torts as public international law matters of state responsibility only, which matters fall outside the scope of their civil jurisdiction over private law cases. While common law and statutory rules of jurisdiction provide sufficient scope for transnational human rights tort action, the courts may not be prepared to go the next step and accept the role of agents of the international legal system if they have trouble getting their minds around the existence of the substantive cause of action. I leave inquiry into that issue to my colleagues addressing the applicable law.

7

Geographies of Injustice: Human Rights at the Altar of Convenience UPENDRA BAXI 1

1 INTRODUCTION O R F A R T O O long, the body of law known as “private” international law— known as well as conflict of laws, or, in shorthand, simply as conflicts—has sustained its enigmatic character as a craft developed by a charmed circle of highly talented jurists. Its status as an autonomous legal field arises out of the transborder movement of legal elements, causes and events. This movement or traffic raises, in the main, three large questions commonly discussed under the following rubrics: jurisdiction (whether a court may be seized of the matter); choice of law (what rule of decision, forum or foreign, may or must govern the dispute or separate elements of it); and the recognition and enforcement of foreign judgments (that is, of judgments rendered by a jurisdictionally-appropriate first court and where a foreign court is asked to give some form of legal effect to that judgment). Theories of private international law reflect a never-ending quest for the articulation of grounds on which authoritative decisions on these questions may be reached. The dominant Euro-American discourse (there is hardly any Southern voice shaping the doctrine of conflict of laws) is marked by a certain kind of epistemic social contract which contains several fundamental stipulations. Being private law, private international law must serve the overwhelming need to pursue uniformity, certainty, and predictability (or “decisional harmony”). Being private international law, it is inappropriate to bring any other consideration than that of comity from the sphere of public international law, notably concerns about human rights or global justice. Being law, it is not amenable to the higher reaches of the discourse of jurisprudence (namely, grappling with the conditions for the production of “legitimate law”), political theory (notably, issues of consent and obligation) or ethics (whether deontological or consequentialist). Only very recently have these aspects begun to be addressed in terms of ethics and political theory, notably in the work of Lea Brilmayer and in that of Friedrich

F

1 Professor of Law, University of Warwick; Honorary Professor of Law, The National Law School, University of India, Bangalore.

198 U Baxi Juenger both of whom have raised some acute questions concerning the social mission of private international law.2 But, as a general proposition, it can be said that the reception, on both sides of the Atlantic, has so far been one of benign indifference. The question that needs to be addressed in some detail is whether the dominant discourse recurrently marked by crisis and renewal (such renewals usually taking the form of those celebrated “revolutions” and “counter-revolutions” in choice of law) holds any potential for a redemptive transformation. By this latter contentious phrase, I mean no more (for the present) than the potential for redirection of the conflicts tradition to the issues of human, and human rights, violation. Some progress is visible as, for example, in conflicts principles in the family law sphere where it is no longer considered just or appropriate to follow archaic maxims like “the domicile of the wife follows that of her husband” that for a long time enabled the flourishing of an extraordinary regime of “limping marriages” and “quickie” divorces. The indeterminate category of “public policy” has indeed often been pressed to terminate long-held patterns of conflictssustained legality. But public policy contains no scope for the “trumping” feature that rights have, especially where judges take only half-seriously the general doctrinal understanding which says that “public policy” should be invoked as a way to avoid application of foreign law or recognition of a foreign judgment only where the foreign system has produced juridical content which is repugnant to, and not simply different from, the fundamental legal values of the forum. The much-vaunted distinction between “repugnance” and “difference” cannot serve regimes of principles enunciating absolute multinational enterprise liability (urged by India in the Bhopal Case) because such Southern “difference” will almost always generate the Northern “repugnance”. Further, less formalistic and more policy-oriented (and occasionally human-rights-friendly in terms of instant outcomes) performances have not led to any profound mutation in the dominant tradition. The mainstream still holds that protection, promotion and preservation of human rights is not compatible with what after all is private international law, a discipline already heavily burdened by the doctrinal twists and turns entailed in pursuing those values which are generally regarded as “properly” the concern of private international lawyers—the values of uniformity, certainty, and predictability (or “decisional harmony”). Such pursuit has to date served rather well the needs of industrial and colonial capitalism as it does now the interests of the foreign investor in these halcyon days of globalisation. However, we have to recognise that the processes of globalisation of law embody, in complex and contradictory ways, not just the power of global capital, but also the power of resistance to it. It is the latter which makes many aspects of the dominant conflicts tradition problematic. In fashioning regimes of liability for human wrongs committed in the course of 2 L Brilmayer, Conflict of Laws, 2nd ed. (1995); F K Juenger, “Mass Disasters and the Conflict of Laws”, (1989) U Ill L Rev 105.

Geographies of Injustice 199 international trade and business, the dominant tradition has not been troubled by the discourse of human rights and by people’s (and peoples’) transnational movements against human, and human rights, violation. There now exists, however, a growing social action criticism of conflicts’ performance legitimacy especially in relation to outcomes in the area of mass torts that are simultaneously social disasters. Yet, this social criticism continues to remain external to conflict of laws doctrine, and is liable to summary dismissal from an internal standpoint.3 Human rights and social activists have, at least since the Bhopal catastrophe, begun to perceive that the inner dynamic of conflict of laws constitutes an obstacle to the promotion, protection and preservation of human rights.4 But the mystery and mystique of private international law still remains beyond their praxiological grasp. Epistemic constructs such as forum non conveniens, comity, jurisdiction in personam and in rem, professio juris stipulations, lex fori, lex loci delicti, and even the seemingly flexible “public policy” are coated in a historical and dogmatic opacity that remains, as yet, impermeable to an activist gaze. It is small consolation for activist communities that these entities also mystify conflicts practitioners who light many a candle at these shrines. One thing remains clear: cursing the heart of conflict of laws darkness by way of human rights lamentation does not quite seem to help its eminent practitioners or advance causes dear to human rights and social activist constituencies— at least, not yet. On the other hand, some hope for transformation arises from the very nature of the field’s epistemic insularity. That hope arises from the fact that the conflicts practitioners have created their discipline’s knowledge (and their own selfknowledge) by constituting a community of faith. Since the Middle Ages, their mission has been to create and nurture a province of law that is capable of providing a series of pragmatic answers to specific problems posed by transborder movement of capital, and by human movement that occurs largely as some kind of byproduct of the structures of capital flow and resulting wealth allocation. Since these structures and flows cause “mass torts” affecting the life chances and 3 I address this at some length in my 1999 Hague Academy Lectures, “Mass Torts, Multinational Enterprise Liability and Private International Law”, 276 Recueil des Cours 305–427 (The Hague: Martinus Nijhoff, 2000) (hereafter referred to simply as Baxi, Mass Torts.) At points I reproduce part of its text by way of “fair use” under copyright law. I remain grateful to the Hague Academy of International Law for their generosity. 4 See my “Introduction” in U Baxi (ed.), Inconvenient Forum and Convenient Catastrophe: The Bhopal Case (Delhi and Bombay: The Indian Law Institute and N. M. Tripathi, 1985) [hereinafter IFCC]; and U Baxi and A Dhanda (eds), Valiant Victims and Lethal Litigation: The Bhopal Case (Delhi and Bombay: The Indian Law Institute and N. M. Tripathi). Eminent Euro-American scholars writing on mass torts rarely refer to Bhopal-related publications from India: third world lex loci delicti produce engaged knowledges for which the North epistemic communities have little use! See also, for an early example of scholarship linking the Bhopal events to the problematic relationship between received conflict of laws doctrine and human rights, R Kapur, “From Human Tragedy to Human Rights: Multinational Corporate Accountability for Human Rights Violations”, (1990) 10 Boston College Third World Law Jl 1.

200 U Baxi the quality of life of the First World denizens, conflicts pragmatism entails incremental doctrinal innovation, both at the level of lex fori application and of the international treaty regimes, especially under the auspices of the Hague Conference initiatives to codify private international law. One must, indeed, remain grateful for these small conflicts mercies. Pragmatism emerges as an ethical attitude of this craft legality, insisting on workable solutions in a forbiddingly diverse range of situations.5 Transformations in technology notoriously complicate the inherited menus of craftknowledges; as such, all the usual problems of conflict theory and practice become all the more acutely aggravated by the transborder mass torts.6 In this respect, conflicts lawyers and judges deserve an audience when they say that their problems are complex and complicated enough without the burdens of importing additional indeterminate vocabularies of human rights and global justice to their performative acts. In so far as their labours achieve treaty regimes governing issues of jurisdiction, choice of law and recognition and enforcement of foreign judgments, these may even be said to serve, howsoever indirectly, the cause of protection and promotion of universal human rights. To a large extent, however, the relative absence of NGO monitoring of, and participation in, these developments promotes the underlying epistemic insularity. In any case, such a sympathetic hearing need not, and must not, become an excuse for status quo worship, rationalisations posing as reasoning, and failures of judicial will. The question then presents itself as one of communicability across the divide thus configured by private international law’s self-understandings.7 Here it is crucial to observe that the resistance to human rights theory by conflicts’ craftknowledges has an inner logic (or paralogic) that communities engaged with human rights need to understand and fully address. By the same token, the masters of the dominant conflicts discourse need to equally fully address the cruelty of epistemic practices that, at least in the domain of social-disaster mass torts, produce not wholly unintended, and occasionally sinister, human, and human rights, violation. Taught law is, as Roscoe Pound used to remind us all, tough law. Private international law is continually reproduced through various forms of univer5 Such as status issues (validity of marriage, divorce, inheritance, and adoption), transaction issues (party autonomy in contracts with multi-state elements), liability issues (situations of civil wrong or tort liability invested with multi-state elements), comity issues (deference to forum jurisdiction, choice of law, recognition and enforcement of foreign judgments) and public policy issues (defining the standards of legitimation for conflicts adjudicatory performances, even feats). 6 For example, aviation and environmental disasters that render inchoate received dogma concerning the law governing the place where they occurred, simply because that “place” can be a physical space or the normative one (that is, a place determined by the site of actual, latent and/or manifest injury or the place where the harmful management decision-making occurs—or both, in some place-defying space). 7 For an elaborated discussion of a series of divides that operate as barriers between “public law” human rights principles and private law torts within a conflict of laws dynamic, see C Scott, “Translating Torture into Transnational Tort: Conceptual Divides in the Debate on Corporate Accountability for Human Rights Harms”, chapter 2 in this volume.

Geographies of Injustice 201 sity-based and then professional education. In this sense, we have a common cause: the transformation of pedagogy that produces (in a Nietzschean phrase) the Eternal Return of the Same. Put another way, the tasks confronting us are invention of new ways of understanding conflict of laws, which at the same time change its habitus.8 I essay here a few, hopefully new, beginnings against the background premise that the primary defendants in transnational humanrights-related litigation stemming from mass industrial and environmental disasters will be corporations and, in the main, multinational corporations.9

2 THE

“ SPATIAL

FIX ”: THE ORIGINAL INTENT

Private international law is inherently and overwhelmingly spatial, conditioned by space and in turn constitutive of it. Its default setting is to return to “spatial fixes” in order to reinforce its conceptual universe.10 Conflicts hermeneutics remain tethered to separate but equal spheres of sovereignty. This already determines the limits of (im)possibility of conflicts justice, consistent with “an overaccumulation of capital within a particular area” and “the uneven insertion of different territories and social formations into the capitalist world market”.11 Conflicts theory and practice provides a good example of how, to borrow the striking words of David Harvey, “places get erected as permanencies within the flux and flow of capital circulation”.12 Thus arise the geographies of injustice peculiar to conflicts adjudication. The protection of the interests of global capital, howsoever internally conflicted, requires that tort liability be localised at the place of the commission of injury and be governed by the law of that place (the lex loci delicti). In real life, this almost always means that hazardous industry, manufacture or process—usually led by entrepreneurs from the world’s affluent regions, and increasingly located in the world’s low- and middle-income regions—receive a conflict of laws incentive. The liability for harm caused by these activities is to be judged by judicial institutions of “host” countries and by their legal standards, usually underdeveloped for a whole variety of reasons which cannot be explored here except to 8 Here, I borrow from Pierre Bourdieu’s notion of habitus, “[t]he conditionings associated with a particular class of conditions of existence” that operate at some sub-conscious level on the behaviour of those imbued with such habitus: see P. Bourdieu, The Logic of Practice (Cambridge: Polity Press, 1990) at 53. For another linkage between human rights, legal knowledge and habitus, see C Scott, “Bodies of Knowledge: A Diversity Promotion Role for the UN High Commissioner for Human Rights” in P Alston and J Crawford (eds), The Future of UN Human Rights Monitoring (Cambridge: Cambridge University Press, 2000) 403 at 407. See further U Baxi, Marx, Law and Justice: Indian Perspectives (Bombay: N. M. Tripathi, 1993). 9 This, then, is the background context, but much of the larger critique will be relevant to conflicts litigation for human rights purposes in general. 10 I borrow here this illuminating notion of David Harvey in his The Limits to Capital (Oxford: Oxford University Press, 1982). 11 D Harvey, Spaces of Hope (Edinburgh: Edinburgh University Press, 2000) 23. 12 Ibid. at 295.

202 U Baxi say the following. Two facts are not a by-product or incidental effect of a fairminded or decent conflicts theory or practice: the fact that such adjudication may be inefficient or vulnerable to the greater forensic ability and staying power of global corporations; and the fact that applicable laws may produce an unconscionable lesser order of damage awards to the victims, as compared to the damages awarded to similarly-situated victims in the developed countries. Rather, these facts are a part of the original intent of conflicts, constantly enacted in conflicts adjudication. This original intent is, of course, affected by patterns of growth of new technologies. Conflicts adjudication confronted, for example, by air crashes and disasters generated a whole new order of thought and practice concerning choice of law and venue for adjudication, given that it was impossible to maintain a dual let alone multiple conflicts regime of tort liability at high altitudes.13 So, a new public international law regime emerged, partially obviating the conflicts anarchy. Similarly, the Space Objects Convention now determines the liability arising out of falling satellites or other objects launched in outer space, almost wholly removing the issues arising mainly out of the agendum of private international theory and practice.14 But these are exceptional happenings. The more routine are the problems associated with consciously planned corporate decisions that eventuate in the kinds of social disasters and mass torts that the thalidomide cases represented for the 1970s and Bhopal has since the 1980s. I say “consciously planned” because corporate decision-making needs to be anchored in cost-efficient investment practices, which are fully sensitive to the burdens that the insurance industry may “legitimately” bear in relation to risk for damages. In relation to both these aspects, the extant regime of conflictual tort liability presents and provides an inestimable subsidy for doing hazardous business abroad (that is, in the South). Both the task and the mission of conflict of laws thrive amidst the realms of “real” and “imagined” geographies.15 The “real” space of mass disasters, the constitutive geographies of injustice, is at once local and global. It is local in terms of the violation of actually existing human beings, and events and environments that shape people’s suffering, and it is global in terms of the space of production of structures of suffering.16 In contrast, the imagined—the abstract—conceptual geography creates a distinctive space of its own, through 13

W Reese, “ The Law Governing Airplane Accidents”, (1982) 39 Wash. & Lee Law Rev 1302. Convention on International Liability for Damage Caused by Space Objects, 1972, 961 UNTS 1888. See for a fuller discussion, Baxi, Mass Torts, supra n. 3 at 409–12. 15 See N K Blomley, Law, Space and Geographies of Power (New York, Guilford Press, 1994) at 36–58; R R M Verchick, “Critical Space Theory: Keeping Local Geography in American and European Environmental Law”, 73 Tul L Rev 739 (1999); E Soja, Thirdspace: Journeys to Los Angeles and Other Real and Imagined Places (Cambridge, MA:, 1996); B de Sousa Santos, Toward a New Common Sense: Law, Science and Politics in the Paradigmatic Transition (London, Routledge, 1995) 456–73. 16 D Harvey, Justice, Nature and the Geography of Difference (London, Blackwell, 1996.) 329–438. 14

Geographies of Injustice 203 the invocation of what I have called the three C’s (competence, comity, convenience) which bear little or no relation to the empirical global and local spaces of mass disasters.17 From the standpoint(s) of the violated, the subaltern standpoint, the question is not simply about the production of different, conflicting and conflicted, spaces. Rather, the question is all about using, by adapting, the techniques of private international law to secure “just production of just geographical differences”.18 For this to be possible, we must thus first of all understand the tenacity of conflicts apartheid, whether of the colonial or of the neo-colonial variety, and then grasp how extant doctrinal ideas and tools ignore real lived time-places by valorizing constructions of abstract, imagined conflicts spaces.

3 THE TROUBLESOME

“ MOTHS ”

Much has changed in the wide world of the law, its theory and practice, since the middle of the last Christian century. But, by and large, conflict of laws thought and practice remains cocooned in some sort of “time-warp”.19 Conflicts adjudicatory practice, at least in relation to mass torts, has not quite been able to emerge from its colonial episteme in a world radically decolonized, at least radically decolonised at the formal level. Nor, amidst what Ulrich Beck names as a “global risk society”, is contemporary conflicts discourse able to grasp the reality that hazardous industries pose risks that do not respect national boundaries or ideological frontiers, and ensure no specific immunities for the overdeveloped societies, as the trinity of disasters that occurred within the space of eighteen months (Bhopal, Chernobyl and the Sandoz/Basle) bring fully to view.20 That the risks thus posed by global industries menace human rights everywhere has not been a major concern of the dominant discourse, as the story of successful invocations by multinational corporations of the doctrine of forum non conveniens, even in the supposedly human-rights-oriented field of the United States Alien Torts Act proceedings reveals.21 The colonial episteme, exemplified in its pure state by A V Dicey’s argumentation in 1886,22 resonates as late as 1982 in a British court’s oft-quoted 17

See Baxi, Mass Torts, supra n. 3 at 336–64. Harvey, Spaces of Hope, supra n. 11 at 5 (emphasis in original). 19 Juenger, supra n. 2. 20 U Beck, The Risk Society (London, Sage, 1992). 21 See, C Scott, “Multinational Enterprises and Emergent Jurisprudence on Violations of Economic, Social and Cultural Rights” in A Eide, C Krause, and A Rosas (eds), Economic, Social and Cultural Rights: A Textbook, 2nd ed (The Hague: Martinus Nijhoff, forthcoming 2001) [hereinafter Scott, “MNEs and ESCRs”], chapter 32; S Zia-Zarifi, “Suing Multinational Corporations in the U.S. for Violating International Law,” (1999) 4 UCLA J International L & Foreign Affairs 81. See also, generally, A Anghie, “Time Present and Past: Globalisation, International Financial Institutions and the Third World,” (2000) 32 NYU J. Int’l L & Pol 243; F J Garcia, “The Global Market and Human Rights: Trading Away the Human Rights Principle,” 25 Brook J Int’l L 51. 22 A V Dicey, A Digest of the Law of England, with reference to the Conflict of Laws (London, Stevens, 1896); see Baxi, Mass Torts, supra n. 3, Lecture One. 18

204 U Baxi description of mass torts foreign plaintiffs suing in the United States Courts: “As a moth is drawn to a light, so is a litigant drawn to the United States.”23 Alien victims suffering from grievous corporate negligence and harm stand assimilated to “moths”. They are without agency or dignity, being genetically programmed first to victimhood by predatory global capital, and next to denial of justice by Euro-American courts; the “light” to which they flock, and which causes them to perish with a certain order of infinity, is the radiance of conflict of laws orthodoxies. Of course, when the victims of multinational aggression are co-nationals, that light suddenly provides an illuminating arc of justice, as in the Agent Orange case where the laudable desire to accomplish justice for the American Vietnam war veterans leads Judge Weinstein to even deny the existence of independent legal systems of sovereign states.24 The dominant discourse thus accomplishes a paradigmatic “cultural lag” where the adaptive technology of conflicts law lags behind human violation caused by technological development.25 But it would be an error of high magnitude to regard this simply as a problem of the law’s vaunted cultural lag. We ought to recognise that this lag may also be a part of the original intent.

4 THE CRAFT

The problematic ways in which the dominant discourse is fashioned need to be clearly grasped. The primary concerns here stand encapsulated by the three C’s—competence, comity and convenience. Each “C” is heavily overloaded with imagined geographies. Competence, or jurisdiction, is the first trope. Jurisdiction is, by definition, legislative in the sense that national legislations in the first place, and the regimes of multilateral treaties in the second place, order response to the question: Should the courts chosen by foreign plaintiffs take jurisdiction over mass torts that occur in another country, whose own legal system may have “just” title to adjudicate the event—and, if so, when? But legislation, whether national or 23

Smith Kline & French Laboratories v. Bloch [1982] 1 WLR 730 at 733 (Court of Appeal, 1982). In re: Agent Orange Product Liability Litigation 580 F. Supp. 690. The veterans there could have been governed through the usual canon governing choice of law by legal orders of South Vietnam, North Vietnam or Cambodia. But the learned Judge holds that South Vietnam no longer exists, and Cambodia, which was then at “war” with the United States, cannot be said to have a relevant legal order. He finds it “ludicrous” even to think of the application of the North Vietnamese law: ibid. at 707. See, for a critical analysis, Baxi, Mass Torts, supra note 3, Lecture One. In the UK context, the breakthrough in the corporate-defendant forum non conveniens cases, Connelly v. Rio Tinto Zinc [1997] 4 All ER 335 (House of Lords) may not be unrelated to the fact that the sole plaintiff was a British citizen (Scottish national), who was injured abroad in the services of a UK-based global corporation. 25 Charles Ogburn developed this hypothesis to explain the phenomenon that, even as industrial/workplace accidents grew, the English law of torts (the “adaptive culture of the common law”) remained unchanged (allowing an extraordinary range of defences to the employer). See the rich discussion in L M Friedman and S Macaulay, Law and the Behavioral Sciences, 2nd ed (New York, Bobbs Merrill, 1969) 629–52. 24

Geographies of Injustice 205 under a treaty regime, leaves plenty of scope for judicial determination of it. The customary law of judicial interpretation, defined in terms of ratione personae or ratione materiae, stands heavily marked by contingency, complexity and contradiction in the construction of that which either turns out to be or is commended normatively as “effective” jurisdiction. The latter principle seems clear worldwide: courts may not take jurisdiction for the sole purpose of symbolic gesturing. There is scope for merging remedial relevance and symbolism in the porosity provided by customary adjudicative law which is crucial since statutes or supranational treaties are rarely comprehensive and which can never wholly eradicate the need for interpretation.26 But, in practice, these normative pores have not permitted much breathing. This interpretive domain is rarely generous to the foreign social-disaster mass torts plaintiff, the moth that perishes. The second trope—“comity”—is a highly pliant, problematic and fuzzy one.27 Its fictional character was acutely highlighted in the Bhopal case, where a US federal judge in New York, Judge Keenan, presided over the first-ever civil mass torts suit by a sovereign plaintiff (the Government of India acting parens patriae for the Bhopal victims, empowered to do so by its passage of the Bhopal Act). He dismissed (though conditionally) India’s claim that India’s own adjudicative system did not carry the potential for justice. Judge Keenan held that to assume jurisdiction, even in the face of this extraordinary assertion, would constitute “yet another example of imperialism”.28 Accordingly, he argued that comity here required providing the Indian judiciary with an “opportunity to stand tall before the world and pass judgment on behalf of the Indian people”.29 Bhopal thus signifies that, confronted with mass torts situations, the judicial reluctance to assume jurisdiction goes so far as to deny a parens patriae foreign sovereign state a day in court on the specious ground that comity is owed not to the sovereign state but to that state’s adjudicative power! In Bhopal, piercing the sovereign veil to conceptualise comity as a judiciary-to-judiciary relationship was in effect a foreign-state-defying expression of solidarity by a US judge towards his foreign brethren who were, he seems to have thought, being embarrassed by their own state’s willingness to swallow its pride and stand before the world to tell all and sundry that its judicial system was not up to the task of such a case, at least not at that time. Convenience, the third form of adjudicative practice shaping the response to just claims of alien mass tort plaintiffs, is furnished by the Anglo-American doctrine called forum non conveniens, a doctrine which threatens to travel to European shores, given the contemporary modes of globalisation of the law.30 26

See Baxi, Mass Torts, supra n. 3 at 336–64. In re Union Carbide Corp. Gas Plant Disaster at Bhopal in India, 809 F 2d. 195 (Southern District of New York, 2dCir, 1984), reproduced in IFCC, supra n. 4 at 35–69. 28 IFCC, supra n. 4 at 69. 29 Ibid. See also the critique of the decision in my “Introduction” to that volume. 30 See, e.g., M Reimann, Conflicts of Law in Western Europe: A Guide through the Jungle (Irvington, NY, Transnational Publishers Inc., 1995); J J Fawcett, “General Report” in Declining Jurisdiction in Private International Law (Oxford, Clarendon Press, 1995) 2–70. 27

206 U Baxi In the result, forum non conveniens is the jurisprudence of judicial expediency, neither grounded, nor groundable, in any principle—or at least, to the extent some of its principles are worthy of a degree of recognition, not capable of application in any non-arbitrary way. That which is expedient for judicial decisionmakers almost always turns out to be convenient. The only good reasons for declining jurisdiction on the ground of forum non conveniens are those that judges happen to provide from case to case. These reasons may be elevated, though not without severe difficulty, to the level of the distinction between that which is justified but still not appropriate.31 In other words, in myriad ways, the invocation of the “doctrine” insists that there is a sufficient jurisdictional nexus for the court to be justified in taking jurisdiction, even while it may not be appropriate for it to do so. This distinction is mystifying, not least in the context of transnational mass torts and in the eye of social suffering. It does not help to say that the only time when it is unequivocally unjustified to decline jurisdiction is when it is manifest that “there is no clearly appropriate forum”,32 because such a test is really only capable of being met when a foreign plaintiff is able to demonstrate a total breakdown of the legal order or its scandalous incapacity to approximate even the threshold standards of minimal justice. In the absence of such a showing, there is always at hand a “more appropriate forum”. But from the internal conflicts standpoint the issue is clearly important. And that issue concerns the professional and public credibility of Euro-American judicial fora, which may not find it desirable to assume, and justify, jurisdiction in most cases when mass torts victims stand advised to sue in foreign courts but wish to maintain the possibility of keeping jurisdiction in “deserving” cases.33 A via media thus stands worked out in theory: assume jurisdiction justifiably when the “real and substantial connection” of a “connecting factor” such as corporate residence is present, but consider it appropriate to decline to exercise that jurisdiction in those cases where the “convenience of witnesses, the law governing the issue, and the places where parties reside or carry on business” (are deemed to) so require—and indeed are often presented as even dictating a stay of proceedings.34 I say in theory because, in the everyday stuff of judicial decisions, denial of jurisdiction is the norm. To quote myself somewhat immodestly: 31 See J Habermas, Justification and Application (MIT Press, 1993); K Gunther, The Sense of Appropriateness (1993) trans. J. Farrel. Application discourses are said not to relate the validity of norms to the appropriateness of their application in a fact/law complex. One way to understand the weak gravitational force of the maxim judex tentur impertiri judicium suum stands furnished by this distinction. Enwombed as it is within the discourse of contemporary public law, the distinction is unmindful of the perplexities of denying/declining jurisdiction. 32 Spiliada Maritime Corp. v. Cansulex Ltd. [1987] AC 460 at 476–8 (House of Lords, per Lord Goff). 33 Recall, for example, Agent Orange, supra n. 24. 34 Spiliada, supra n. 32. As compared to the UK’s Spiliada approach, the US approach tends much more toward naked acknowledgement that forum self-interest is the fulcrum around which forum non conveniens assessments revolve. The United States federal doctrine of forum non conveniens catalogues an assortment of “public” and” private” interest factors. The latter broadly

Geographies of Injustice 207 “Thus, the question always is how to determine the place of harm or injury in mass torts. For example, is a mighty multinational author of mass injury located three blocks away from a Texas Court an appropriate place/forum as compared with a distant Costa Rica court?35 Or is a New York federal court, situated only a half an hour cab ride across Manhattan, less appropriate a forum than a distant Indian place called Bhopal, ravaged by wicked corporate practices of the Union Carbide Corporation? Or whether the asbestos affected victims of Namibia and South Africa have any significant connection with Cape Mining Ltd. for purposes of the case being heard in London, where both the company and the House of Lords stand situated?36 The conflicts discourse thus often severs the space of production of mass disasters from the place where rights and responsibilities of global capital may be determined.”37

The question, from an internal conflicts standpoint, is then not justice but convenience—whatever may be the specific legitimating language employed by a given common law country’s courts, including, as in the UK, the language of appropriateness without strong invocation of US-style reference to forum interests. Whatever the surface text of judicial decisions in this area, forum non conveniens doctrine is configured in terms of that peculiar combination of balancing “public” and “private” interest aggregations, a mode of analysis that facilitates denial of jurisdiction in part by totally avoiding an assessment of comparative societal responsibility for ensuring civil transnational justice.38 I have explored on several occasions the foregoing interest/responsibility bifurcation with reference to the paradigmatic situation of the Bhopal catastrophe in terms of the ways in which the appropriate forum becomes constructed.39 comprise: sources of proof, access to witnesses, and possibility of view (the “ease of arranging for a view of the premises around which litigation centers”). The former centre upon the difficulties of administration of justice in heavily congested forum courts, the applicable law and related governmental/policy interests as construed ad hoc in each mass tort situation. All of this is then filtered through denial of a presumption in favour of the plaintiff’s selection of forum where the plaintiffs are non-American. In ways scarcely decisive for the mass disaster violation viewed in global perspective, some jurisdictions, like Australia, disallow recourse to the cluster of “public interest” factors. Of greater significance for the foreign-plaintiff-friendliness of Australia may be its retention of an older common law standard according to which it must be vexatious or oppressive for a plaintiff to sue the defendant in the otherwise-justifiable forum. See P Prince, “Bhopal, Bougainville and Ok Tedi: Why Australia’s Forum Non Conveniens Approach is Better”, (1998) 47 Int and Comp Law Qu 573 esp. at 595, for the argument that the mass-disaster tort stemming from the toxic spill at the Ok Tedi mine in Papua New Guinea was settled in part because the defendant Australian company, Broken Hill Proprietary, saw no hope of successful invocation of the Australian version of the doctrine once an Australian judge had ruled that the basic taking of jurisdiction was justifiable. See, for further details, Baxi, Mass Torts, supra n. 3, Lecture Two. 35 Dow Chemical Company and Shell Oil Company v. Domingo Castro Afarao, et al., 786 S. W. 2d. (Texas, 1990), cert. denied, 11 S. Ct.67 (1991) at 680–9 (per Justice Doggett). 36 Citing to Connelly v. RTZ, supra n. 24; Lubbe et al. v. Cape Plc, Supreme Court of Judicature, QBENI 98/0192/1, 30 July 1998 (unreported). 37 Citing to Baxi, Mass Torts, supra n. 3 at 337. 38 Again, Australia may currently be an exception. See supra n. 34. On the bifurcation between interest and responsibility, see Scott, “MNEs and ESCRs”, supra n. 21. 39 See, the trilogy of books on the Bhopal Case published by the Indian Law Institute: U Baxi and T Paul (eds), Mass Disasters and Multinational Liability (New Delhi and Bombay, Indian Law Institute and Tripathi, 1985); IFCC, supra n. 4; Valiant Victims and Lethal Litigation, supra n. 4 .

208 U Baxi While it would not serve much purpose to re-narrate the story here, Bhopal’s (radio)active conflicts residue warrants some reiteration in the present context, a discussion to which I turn in the following section.

5 INCONVENIENT FORUM AND CONVENIENT CATASTROPHE

The Bhopal catastrophe produced its own unique forms of juridicalisation. Responding to American contingency fee lawyers filing as many as one hundred and forty-four suits in various federal courts and their aggregation (by the Judicial Panel on Multidistrict Litigation) to the Southern District of New York, presided over by Judge John F. Keenan, the Union of India assumed, under the Bhopal Act, parens patriae responsibility for legal redress of Bhopal victims. Given the fact that the victims were incapable of litigating their individual claims against a “monolithic multinational corporation”, India brought action in Judge Keenan’s court for damages, present and future, against UCC for an amount of $3 billion, for personal injury, economic and environmental loss and punitive damages.40 India enunciated a daring principle of absolute multinational enterprise liability,41 a principle of responsibility commensurate with the global risk society and the attendant features of multinational corporate organisation, including what Beck terms “organised irresponsibility” and “organised impunity”.42 The principle, in effect, enunciated “a primary, absolute and non-delegable duty to any person or country in which it has in any manner caused to be undertaken any ultrahazardous or inherently injurious activity”. The scope of the duty included solicitude for the “highest standards of safety and to provide all necessary information and warnings regarding the activity involved”.43 Failure on both grounded the assertion by the Union of India and her peoples of absolute multinational enterprise liability. Regardless of a series of unjust and messy outcomes (at least from the standpoint of the Bhopal victims),44 the assertion of this inexorable standard of multinational enterprise liability needs to be kept alive. The ways in which conflicts adjudication proceeded was successful in imposing a regime of silence on the enunciation of this vital principle. Assume for a moment that the Bhopal litigation had run its full course (and not been aborted by an unconscionable settlement) and that the Keenanmandated Indian forum had rendered a decision in accordance with the principle of strict multinational enterprise liability—a principle not then but now an 40

See the full text of India’s statement of complaint in Mass Disasters, ibid. at 1–17 (1986). Ibid. at 4–5. 42 Beck, supra n. 20. 43 Mass Disasters, supra n. 39 at 5. 44 Outcomes that, in particular, resulted in judicial orders of settlement for $470 million, without any hearing of the victim groups who were appropriate parties, and the grant of complete civil and criminal immunity to Union Carbide, although the latter were rescinded on a review of the judgment. 41

Geographies of Injustice 209 integral aspect of Indian law45—would such a judgment have been recognised and enforced in the United States? Here assume for the moment that, even in the most traditional sense of lex loci delicti, such a decision would ordinarily compel recognition and enforcement. But the recognising forum law super-adds to the margin of appreciation for non-recognition of foreign judgments on the ground of contravention of “public policy” the additional ground that the foreign judgment should comport with the requirements of what is called “due process” in the United States. These requirements are normally understood as entailing procedural, rather than substantive, requirements, the wisdom, policy or purpose of applicable substantive law usually being beyond question. Assume further that these procedural requirements were impeccably complied with by the Indian courts rendering the decision. Would it then have been recognised and enforced by American courts? There is room for thinking that ways would have been found to deny recognition and enforcement because the more appropriate forum here would have ended up producing an unsustainable norm or standard of liability.46 It is unsustainable because the global insurance industry (always an unnamed, but potent, third party which never formally appears—that is, in its own name—before the courts) simply cannot, or does not, accept underwriting such risks in conformity with the Indian principle. Accordingly, “due process” (now in its substantive sense) will assume the same role as “public policy” in denying recognition and enforcement, but, in so doing, will lower the bar for non-recognition by invoking simple policy differences between the content of foreign law and the content of the local law which the judgment-receiving court would otherwise be bound by if the case did not have a geographically complex dimension. Where, hypothetically (but plausibly), a US court declined to enforce a damage award issued by an Indian court because it has held the multinational enterprise (MNE) liability principle to be substantively flawed, the result becomes a situation of a “choice” between impunity and imperialism: victims cannot sue a US-based MNE in its home courts but also cannot expect to collect damages against that MNE after suing in the “appropriate” forum, unless, that is, the law applied by the “appropriate” forum is in content very close to, if not the same as, American corporate liability law that prevails during the time in question.

45

See, generally, Mass Torts, supra n. 3 at 390–402. I doubt whether the outcome would have been any different under the extant treaty regimes of the Brussels and Lugano Conventions. See the reasons stated in Mass Torts, supra n. 3 at 390–1. It is unsurprising that multinational enterprises favour and promote settlements because they scrupulously avoid all precedent-creating effects while at the same time bestowing upon them, through devices of structured ambiguity, enormous public relations and mass media pay-offs due to representations of them as more caring and just than the Southern host (more accurately hostage) states! 46

210 U Baxi

6 DISSENTING VOICES OF THE TRADITION AS OPENINGS TO THE FUTURE

Some alternative possibilities exist as well, ones that remain loyal to the extant conflicts traditions and yet seek to accomplish conflicts justice. I have in mind here the different approaches of Judge Weinstein in Agent Orange (despite its neo-colonial overtones) and of Justice Doggett in Dow Chemical.47 The Weinstein approach invents a federal conflicts common law approach to accomplish justice for Vietnam War veterans. This approach is, however, limited, to doing justice to co-nationals and therefore is of limited relevance to foreign plaintiffs in social-disaster mass tort situations, it being extremely doubtful that same benign fashioning of judicial remedy would extend to non-nationals: the Vietnamese, Cambodian and Laotian war veterans.48 Judge Doggett, in contrast, does assume jurisdiction, thus avoiding the hazards entailed in our hypothetical scenario of a US court having to react to an incoming foreign judgment claim for recognition and enforcement. The way he accomplishes this feat is worthy of note in some detail. Unlike Judge Keenan, who finds no trace of a redeeming public interest in assuming jurisdiction in the Bhopal Case,49 Judge Doggett insists that the “so-called doctrine of forum non conveniens” is a “fancy name to shield alleged [corporate] wrongdoers”, a “largely impenetrable shield against meaningful lawsuits” that enables some multinational corporations to “continue to operate without adequate regard for human and environmental costs of their action”.50 Judge Doggett remains aware of what Ulrich Beck calls the boomerang impact of risk production, an image which signifies that the judicial, and the overall juridical, production of double standards for American corporations doing business at home and abroad is not conducive to American well-being. This approach at least results in jurisdiction being retained over an offending global corporation for causing mass injuries. Foreign plaintiffs alleging liability for mass torts have at least the assurance of their day in court. This shift assures all the benefits of procedural law, especially discovery processes, which enable fair access to the facts of corporate decision-making. This approach also helps redress the uneven accessibility to the “expert” scientific evidence which affects victims’ presentation of their case in many a home country (the more appropriate forum) in the South. And of course, if the case proceeds to a judgment, no significant issues or anxieties concerning enforcement against corporate assets persist. The only difficulty with the foregoing is presented by the inclination still to adopt as a determining law or rule the standards of liability available at the place of injury in accordance with the standard choice of law rule, or strong 47 48 49 50

Both cases cited supra nn. 24 and 35, respectively. See Mass Torts, supra n. 3 at 314–15. For a detailed analysis, see ibid. at 336–64. See Dow Chemical, supra n. 35 at 680, 683, 689.

Geographies of Injustice 211 presumption, that the lex loci delicti governs a tort claim. But, here already, it is crucial to realise that, in some contexts, that site is also constituted, at least under the rubric of connecting or significant factors, by the place where harmful corporate decisions were made.51 Such decision-making is the place of the tort no less than is a place where the harm transpires or later manifests itself. Moreover, compelling arguments have been made for a special regime of principles, under the uneasy rubric of “better law”, to situations of mass disasters. In the main, these approaches stress basically pro-plaintiff prescriptions that enable determination of liability by, inter alia, reference to the locus decisionis, the place of tortfeasors’ decision-making conduct.52 The prescriptions mark a shift away from parochial lex forism (apply the forum law, the summum bonum of the dominant conflicts discourse). It must be noted that these approaches are not explicitly concerned to develop a critical normative theory about conflicts adjudication, based either upon notions of conflicts justice or human rights. But human rights activism should all the same be aware of these valuable existing tendencies in a body of law which nonetheless remains generally inhospitable to human rights logic and language. In contrast to Judge Weinstein’s “compatriot patriotism” which favoured a just assumption of jurisdiction in Agent Orange, Judge Doggett adopts in Dow Chemical what might be named “cosmopolitan patriotism”.53 Cosmopolitan patriotism in the conflict of laws, of course, also stands grounded in concern for the well-being of co-nationals. But its conception of well-being extends further, to an effort to redress the plight of foreign victims of mass disasters by insistence upon human-rights-oriented corporate governance practices. This has the potential to move the dominant conflicts discourse away from its present anchorage, which treats acts of injustice merely as situations of bare misfortune.54 The human, as distinct from the merely technocratic future of law and jurisprudence, including conflicts jurisprudence, seems to me to lie in performative acts (judicial, juridical and popular) that enable for the most part an authentic recognition of situations of misfortune as simultaneously acts of injustice. For, as Judith Shklar has taught us, if it is not wholly given to us to erase the boundary between misfortune and injustice, it is also wholly unjust to maintain it in ways that reduce all acts of injustice to situations of misfortune. Such 51 See, for example, Th M de Boer, Beyond Lex Loci Delicti: Conflicts Methodology and Multistate Torts in American Case Law (1987); F Juenger, Choice of Law and Multistate Justice (1993); and Brilmayer, supra n. 2. 52 Developed notably by David Cavers, Robert Leflar, Friedrich Juenger and Willis Reese: see discussion in Mass Torts, supra n. 3, at 379–83. Although their analyses remain confined to the context of multi-state American conflicts law and do not directly address the situation of foreign mass tort plaintiffs, there is no reason in principle why this “better law” approach may not be further extended. 53 I borrow the complexities of this notion from Charles Jones’ insightful essay, “Patriotism, Morality, and Global Justice” in I Shapiro and L Brilmayer (eds), Nomos XLI: Global Justice 126 (New York, New York University Press, 1999). 54 See J Shklar, The Faces of Injustice (New Haven, Yale University Press, 1991).

212 U Baxi performative acts shrink the domains of social action in ways that naturalise predation, sapping in the process any impulse to right the inhuman wrongs. For far too long conflicts jurisprudence has shown an undue order of modesty concerning its social mission. And in the process this reticence has contributed to a great deal of human violation. Now, perhaps, is the time for a new beginning, a human-rights-sustaining future for theory and practice in the conflict of laws. We could do worse than to start with the dissenting voice in the conflicts tradition itself.

8

The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legalism ROBERT WAI 1

1 INTRODUCTION H E C O M M E R C I A L activity exception to sovereign immunity is securely established in Western common law jurisdictions such as the US, UK and Canada. The elaboration of the exception in common law courts and subsequently in statutory measures is central to a progressive narrative of the development from an absolute to a restrictive theory of sovereignty.2 This narrative of the commercial activity exception smoothes over a number of historical and continuing controversies.3 Indeed, the progressive narrative concerning the commercial activity exception sits uneasily with a number of features of contemporary doctrine on sovereign immunity. In particular, the sharp distinction between commercial activity and non-commercial activity and the dramatically different rules applicable to each category of activity signal the policy

T

1 Assistant Professor, Osgoode Hall Law School, York University. My thanks to Kerry Rittich for her helpful comments on earlier drafts of this chapter. 2 See e.g. G A Badr, State Immunity: An Analytical and Prognostic View (The Hague, Martinus Nijhoff, 1984); L Henkin, R Pugh, O Schachter and H Smit (eds), International Law: Cases and Materials, 3rd ed., (St. Paul, Minn., West Publishing, 1993), chapter 13 especially at 1126–9 [hereinafter “HPSS”]; L Collins (ed.), Dicey and Morris on the Conflict of Laws, 12th ed., (London, Sweet & Maxwell, 1993) at 241–3 [hereinafter “Dicey & Morris”]; S A Williams and A L C de Mestral, An Introduction to International Law: Chiefly as Interpreted and Applied in Canada, 2nd ed., (Toronto, Butterworths, 1987), chapter 8. C Schreuer, State Immunity: Some Recent Developments (Cambridge, Grotius Publications, 1988) at 8 observes that the commercial activity exception “is at the core of the trend away from absolute immunity and at the centre of the traditional concept of acta jure gestionis”. 3 For a critical account of narratives of progressive evolution in international law, see D Kennedy, International Legal Structures (Baden-Baden, Nomos Verl.-Ges., 1987). The commercial activity exception has been disputed as a principle of international law, especially by socialist and communist states: see e.g. J R Brierly, The Law of Nations: An Introduction to the International Law of Peace, 6th ed. by H Waldock, (Oxford, Clarendon Press, 1963) at 250.

214 R Wai controversies underlying the development of the commercial activity exception and the contemporary doctrine of sovereign immunity more generally. This chapter attempts to destabilize the perceived normalcy of the sharply distinct treatment accorded to commercial and non-commercial activity in immunity doctrine through an exploration of the public policy justifications provided for the distinction. The contrasting treatment in sovereign immunity doctrine of commercial activity and human rights abuses offers an opportunity to explore how a currently dominant set of ideas about international society, politics and economy play out in the context of a major field of public and private international law. Doctrinal challenges are first mapped that indicate the problems faced in making workable the distinction between state action that is or is not covered by the commercial activity exception. It will be argued that policy justifications offered for the distinction do not overcome these problems, and indeed rely on a set of ideas of consent and international political and economic cooperation that are erroneously presented as determinative and noncontroversial. The aims of the chapter are twofold: first to identify the operative assumptions and show how they are associated with controversial ideological and parochial concerns; and then to reconceive these international policy dimensions as structures for argumentation, analysis and contestation rather than determinate conclusions.

2 DOCTRINE

The Narrative of Evolution of the Commercial Activity Exception The doctrine of sovereign immunity provides sovereign authorities with a jurisdictional immunity. Sovereign immunity also operates as a bar to execution against, and attachment of, sovereign assets.4 If a jurisdictional immunity is found, a court will not further consider the substantive merits of a plaintiff’s claim, nor will it fully examine the procedural impediments that a plaintiff might face in seeking an effective remedy in the courts of the foreign sovereign. The practical importance of such a barrier to private law claims is obvious, and hence the classification of the contexts in which the immunity applies is of great significance to a variety of forms of litigation involving state conduct. The commercial activity exception is typically situated as central to the movement away from the absolute theory of sovereign immunity by Western jurisdictions through the twentieth century. National courts moved to restrict state immunity by distinguishing acta jure gestionis from acta jure imperii, only the 4 This chapter focuses on the issue of jurisdiction. There may be additional critiques related to the particularities of execution and attachment, but many of the general principles and the problems concerning the commercial activity are similar: see I Brownlie, Principles of Public International Law, 5th ed., (Oxford, Oxford University Press, 1998) at 346–7.

The Commercial Activity Exception 215 latter of which were covered by sovereign immunity.5 These efforts were confirmed and extended by decisions of political bodies, visible early on in the change in US State Department policy articulated in the Tate Letter.6 The movement towards restrictions on sovereign immunity was also supported by leading academic writers.7 While the status of the restrictive theory varied from jurisdiction to jurisdiction, these debates became somewhat academic in most Western common law jurisdictions as statutes were enacted during the 1970s and 1980s which codified the restrictive theory of sovereign immunity and the commercial activity exception.8 Sovereign immunity statutes in the US, the UK and Canada, for example, include a commercial activity exception, although with some variations in wording and definitions.9 As a result, the main issue for common law courts has become the proper interpretation of the sovereign immunity legislation with respect to the classification of state activity as either subject to the immunity or falling within an exception to the immunity. The Classification Problem All immunity statutes provide for differential treatment of non-commercial activity through specific provisions or implicit exclusions from the definitions of the commercial activity exception. However, the commercial activity exception to sovereign immunity is rife with classification problems that were present at 5 See Brownlie, ibid. at 330–1 and his n. 25 for a brief survey of the current status of the restrictive theory in the courts of various jurisdictions. 6 Letter to Department of Justice from Jack B. Tate, Acting Legal Adviser, Department of State, 19 May 1952, (1952) 26 Dep’t State Bulletin 984. 7 See e.g. H Lauterpacht, “The Problem of Jurisdictional Immunities of Foreign States”, (1951) 28 BYIL 220; Badr, supra n. 2. 8 Australia: Foreign States Immunities Act 1985, reproduced in (1986) 25 ILM 715; Canada: State Immunity Act 1982, RSC 1985, c. S-18, reproduced in (1982) 21 ILM 798 [hereinafter “Canada SIA”]; UK: State Immunity Act 1978, c.33, reproduced in (1978) 17 ILM 1123 [hereinafter “UK SIA”]; US: Foreign Sovereign Immunities Act, 28 USC s 1330, reproduced in (1976) 15 ILM 1388 [hereinafter “US FSIA”]. For a survey of national materials, see Materials on Jurisdictional Immunities of States and Their Property, ST.LEG/SER.B/20 (New York, UN Publications, 1982). The restrictive theory has been adopted by judicial decision in many European civil law jurisdictions including Belgium, Italy and France. In addition, see the European Convention on State Immunity and Additional Protocol, Basle, 16 May 1972, (1972) 11 ILM 470. This chapter focuses on the US, UK and Canada, but I believe the policy controversies identified are common to the justification and interpretation of the commercial activity exception in many jurisdictions. 9 Section 1605(a)(2) of the US FSIA provides that no immunity operates with respect to actions based “upon a commercial activity carried on in the US by the foreign state; or upon an act performed in the United States in connection with a commercial activity of the foreign State elsewhere; or upon an act outside the territory of the United States in connection with a commercial activity of the foreign state elsewhere and that act causes a direct effect in the United States”. The UK SIA provides that a State is not immune with respect to proceedings relating to “a commercial transaction entered into by a State” [section 3(1)(a)] or contractual obligations which are “to be performed wholly or partly in the United Kingdom” [section 3(1)(b)]. Section 5 of the Canadian SIA provides that “a foreign State is not immune from the jurisdiction of a court in any proceedings that relate to any commercial activity of the foreign State.”

216 R Wai common law and are not relieved by the immunity legislation. What is most troublesome is that significant consequences rest on a distinction that has proven very difficult to apply. For those activities that fall outside the commercial activity exception, the consequences are stark: sovereign immunity will often act as a bar to further consideration of the suit. Other exceptions are also provided in most immunity statutes, but these have added complications. For example, exceptions are provided for non-commercial torts such as personal injury torts, including any nominate tort of torture that might emerge. However, the territorial connections to the forum state are more onerous than with respect to commercial activity or to contracts such that the mainstream understanding is, for the most part, that state immunity remains intact for foreign torts.10 The difficulty of classification has been evident to many commentators on public international law.11 The rules on the commercial activity exception are obvious sources of the gaps, contradictions and ambiguities that have been used as examples for general critical analyses of legal rules in international law.12 The idea of sovereignty itself provides limited guidance, above all because most interesting cases involve conflicting claims of different sovereigns or societies and, indeed, conflicting conceptions of the concept itself. Nor do appeals to a number of underlying doctrines—such as jus cogens, sovereign equality, or international community—resolve the classification problems: they are equally contingent on contested ideas of autonomy and consent.13 Notwithstanding these difficulties, courts and commentators have endeavoured to apply the distinction set out in the statutes. The following sections explore two approaches adopted by the statutes and the courts. It will be argued that both the nature test and use of contracts in classification demonstrate how the efforts to make workable the distinction between commercial and noncommercial activity fail to escape serious indeterminacies. 10 See Schreuer, supra n. 2, at 51–4. US FSIA, section 1605(a)(5) provides that a state shall not be immune with respect to “personal injury or death, or damage to or loss of property, occurring in the United States”. UK SIA, section 5 provides:

“A State is not immune as respects proceedings in respect of (a) death or personal injury; or (b) damage to or loss of tangible property, caused by an act or omission in the United Kingdom.” Canada’s SIA, section 6 provides: “A foreign state is not immune from the jurisdiction of a court in any proceedings that relate to (a) any death or personal injury, or (b) any damage to or loss of property that occurs in Canada.” 11 See e.g. Brownlie, supra n. 4, at 335–9; J Crawford, “International Law and Foreign Sovereigns: Distinguishing Immune Transactions”, (1983) 54 BYIL 75. 12 See e.g. Kennedy, supra n. 3, at 151–65; M Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (Helsinki, Finnish Lawyers’ Publishing Company, 1989) at 394, 433–5, 450–1; E Morgan, International Law and the Canadian Courts (Toronto, Carswell, 1990), chapter 1. 13 See ibid.

The Commercial Activity Exception 217 (a) The nature or form test In the US, the principal test developed both by statute and by the courts to determine whether sovereign activity is commercial or not focuses on the nature or form of state conduct. Under this test, whether state action was directed to a commercial or non-commercial purpose is irrelevant; the test looks to “the outward form of the conduct that the foreign state performs or agrees to perform”,14 such as whether it was a market transaction which could be conducted by a private party.15 The US statute mandates that whether government conduct is commercial is “to be determined by the course of conduct or particular transaction or act, rather than by reference to its purpose”.16 The US courts have applied this test by examining whether the particular actions of the state are “the type of actions by which a private party engages in ‘trade and traffic or commerce’ ”.17 The United States Supreme Court has found that the statute prohibits recourse to consideration of objectives, no matter how “uniquely sovereign” they might be.18 The UK statute in contrast does not mandate the use of the nature test but instead lists specific examples of activities. That being said, the specified categories do seem distinguished largely by their form rather than purpose. The act includes exceptions for commercial transactions, but further defines “commercial transaction” to include “any contract for the supply of goods and services”, “any loan or other transaction for the provision of finance and guarantee or indemnity in respect of any such transaction” and “any other transaction . . . into which a State enters or in which it engages otherwise than in the exercise of sovereign authority”.19 In addition, the UK statute provides an exception for state obligations “which by virtue of a contract (whether a commercial transaction or not) falls to be performed wholly or partly in the United Kingdom”.20 The Canadian statute is more open-ended. The State Immunity Act defines “commercial activity” as “any particular transaction, act, or conduct or any regular course of conduct that by reason of its nature is of a commercial character”.21 No further specification of commercial activities is given as in the UK statute, nor is there an express prohibition on consideration of the purpose of the activity as in the US statute. The difficulty of applying the nature test was recognized by the Supreme Court of Canada in Re Canada Labour Code.22 The majority found that it was unhelpful to rule out use of the purpose test to 14 15 16 17 18 19 20 21 22

Republic of Argentina v. Weltover, Inc., 504 US 607 (1992) at 617. See e.g. Dicey & Morris, supra n. 2, at 252. US FSIA, section 1603(d). Republic of Argentina v. Weltover, Inc., supra n. 14, at 614. Ibid. at 614. UK SIA Section 3(3). UK SIA Section 3(1). Section 2. [1992] 2 SCR 50.

218 R Wai supplement the nature test.23 The majority noted that the case before them exemplified a common situation of state activity that possesses a “hybrid nature—one public, the other private”24 and a “double aspect” that was “at once sovereign and commercial”.25 The favoured approach was a pragmatic, “contextual” approach, which did not exclude consideration of purpose, and instead tried to assess the particularities of the individual facts of the activity in question. This approach should be identified for what it is: a recognition by a sophisticated court that the distinction between commercial and noncommercial activity is so complex that no simple test can be articulated. (b) Contracts and classification The presence of a contract between the state party and the private party also plays an important role in determination of whether sovereign immunity applies. The role of contracts in the classification exercise is of particular interest to those interested in private actions for human rights claims because of the comparatively favourable treatment under immunity doctrine of contractual claims as opposed to tort claims, which form the private law basis for most human rights violations such as torture.26 Contracts are relevant to determination of sovereign immunity in several respects. First, the existence of a contract indicates that a state actor is acting in a commercial capacity such as to bring it within the commercial activity exception. The common law test for the commercial exception—namely, the distinction between acts jure imperii and acts jure gestionis—was linked to the manner in which the transaction occurred “on the basis of a private law relationship, such as a contract”.27 The UK statute expressly provides that contracts of various kinds will bring state activity within an exception to sovereign immunity.28 23

Ibid. at 70 per La Forest J:

“Nature and purpose are interrelated, and it is impossible to determine the former without considering the latter. I do not accept that the definition of ‘commercial activity’ in the Act precludes consideration of its purpose. That definition, in circuitous fashion, defines ‘commercial activity’ as conduct that ‘by reason of its nature is of a commercial character’. In many cases, it may be unnecessary to delve into metaphysical distinctions between ontology and teleology of the activity in question. However, if consideration of purpose is helpful in determining the nature of an activity, then such considerations should be allowed under the Act. Further, when an activity is multifaceted in nature (as in the instant case) consideration of its purpose will assist in determining which facets are ‘related’ to the proceedings in issue.” 24

Ibid. at 72. Ibid. at 80. 26 Contracts are also treated differently under the other rules on jurisdiction and choice of law in common law jurisdictions. For example, the choice of law rule for tort tends to put much more importance on the law of the place of the tort, whereas the choice of law rule for contract emphasizes party choice as prima inter pares among a number of factors. 27 Brownlie, supra n. 4, at 335. 28 Relevant exceptions include contractual obligations (whether commercial transactions or not) which are to be wholly or partly performed in the United Kingdom [section 3(1)(b)], and which are not governed by the administrative law of a foreign state [section 3(2)]. Commercial transaction 25

The Commercial Activity Exception 219 Although no specific provision of the US statute mentions contracts, US courts have often found that the contractual relationship between sovereigns and litigants is sufficient for purposes of the act.29 Second, the existence of a contract may support a finding of waiver of sovereign immunity on the part of a foreign sovereign. Contracts with state actors can expressly provide for submission to jurisdiction and waiver of sovereign immunity.30 Furthermore, other kinds of contractual provisions, such as arbitration clauses, are sometimes found to constitute an implied waiver of normal sovereign immunity.31 In the waiver context, either an express provision or other kinds of contractual provisions may be interpreted to indicate that both parties are voluntarily entering into a transaction in which each understands that the other party expects to have enforceable legal rights and, as a necessary corollary, waives legal barriers, such as sovereign immunity, which are inconsistent with that expectation. The elements of foresight, negotiation and planning that support the waiver exception usefully suggest what it is about the presence of a contract that makes some kinds of state activity more likely to be interpreted as a commercial activity. The Canadian immunity statute does not specifically mention contracts, and the Canadian courts have been more hesitant than US courts to assume that the mere existence of the contract indicates that there is a commercial exception to sovereign immunity.32 The Supreme Court of Canada decision in Re Canada Labour Code is interesting because the court uses the presence of a contract, a lease of land to the US government as a military base, not as evidence of commercial activity, but rather as evidence of the positive assent on the part of the Canadian government to the granting of sovereign immunity to the foreign sovereign. The existence of the lease was given as much weight as the employment contracts of the employees on the base. As with the commercial activity exception, the Canadian statute and judicial interpretation suggest a more flexible, contextual approach to the use of contracts in determination of sovereign immunity. While this approach can be criticised as unpredictable and manipulable,33 it seems to more accurately reflect the complexity of classification of sovereign activity. under section 3((1)(a) are defined to include contracts for the supply of goods or services [section 3(3)(a)], and the existence of contracts is clearly relevant to the financial transactions listed in section 3(3)(b) as well as the other transactions entered into by a State otherwise than in the exercise of sovereign authority [section 3(3)(c)]. See Brownlie, supra n. 4, at 341. 29 K McGinty, “Opening the Courts to Protect Interests Abroad: The Effect of the Foreign Sovereign Immunities Act on Litigation with Developing Countries”, (1990) 10 Boston College Third World LJ 63 at 75, 82. 30 E.g. Canada SIA, section 4; US FSIA section 1605(a)(1); UK SIA, section 2(2). On the last, see Brownlie, supra n. 4, at 343–4. 31 See e.g. G B Sullivan, “Implicit Waiver of Sovereign Immunity by Consent to Arbitration: Territorial Scope and Procedural Limits”, (1983) 18 Texas Int’l LJ 329; HPSS, supra n. 2, at 1161. 32 See e.g. Re Royal Bank and Corriveau (1983) 117 DLR (3d) 199 (Ont. H.C.). 33 Morgan, supra n. 12 at 34–5.

220 R Wai The Indeterminacy of Classification The commercial activity exception has proven difficult to apply. The tests for the commercial activity exception oversimplify the complex nature of contemporary sovereignty.34 Because many fact situations straddle categories, neither the form tests nor tests based on the existence of contracts give coherence to the strong distinction between the commercial and the non-commercial. (a) Double aspect situations The difficulty of distinguishing commercial from non-commercial activities is most pronounced in situations where the interactions between governmental and business actors are multiple and intricate. Governments may be involved in commercial projects through varied policies such as public ownership, tax incentives, export financing and insurance, or granting of licences. Some highly publicised cases involving oil extraction have evidenced the multiple ways in which state action, commercial interests and non-commercial harms can be intertwined.35 In the Sudan case, an oil-extraction joint venture, the Greater Nile Oil Project, involving the Sudanese, Chinese and Malaysian state-owned oil companies and Talisman Energy Inc., a Canadian corporation, is occurring in the context of the Sudanese civil war in which the government is accused of serious human rights abuses. In addition to the complex ownership structure, commercial interests overlap with governmental conduct in that it is alleged, among other connections, that oil revenues are used by the government to finance military operations, that military aircraft have used civil airstrips operated by the joint venture, and that local populations have been attacked or displaced from the vicinity of the oil fields.36 It is also feared that government participation in commercial projects has extended to the use of state security to assist the smooth operation of business activities. In many of these situations, the sharp separation of the commercial and non-commercial is simply a fiction. 34 I Sinclair, “The Law of Sovereign Immunity: Recent Developments”, (1980) 167 Rec. des Cours 113 at 209; Lauterpacht, supra n. 7, at 222–5. 35 The role of Unocal as a joint participant with the controlling government of Burma in the Yadana Natural Gas Project has led to charges of human rights abuses and claims in the US courts under the Alien Tort Claims Act against Unocal; see National Coalition Government of the Union of Burma v. Unocal, Inc., 176 FRD 329 (C.D.Cal. 5 November 1997). The involvement of Talisman Energy in the Sudan has led to substantial governmental and non-governmental pressure, media attention and a fact-finding investigation by the Canadian government; see Human Security in Sudan: The Report of a Canadian Assessment Mission, prepared by Mr John Harker for the Minister of Foreign Affairs (Ottawa, January 2000) [hereinafter “Harker Report”]; “U.S. Rebukes Ottawa on Sudan”, The Globe & Mail (Toronto), 15 February 2000, A1. 36 See Harker Report, supra n. 35 at 10–11, 15, 16, 47–50, 64, 66–7; “Sudan’s tactics in oil fields prompt protest to Khartoum”, The Globe & Mail (Toronto), 6 January 2000, A4; “Oil drilling makes refugees of Sudanese tribespeople”, The Globe & Mail (Toronto), 8 December 1999, A13.

The Commercial Activity Exception 221 A broad reading of the commercial activity exception might potentially be a tool for implicating a wide range of governmental conduct. It may be possible for some individuals to claim in tort, for example, where the conduct arose in what were, at least initially, contractual or commercial relations between the parties. This kind of “piggyback” approach might be used by a common law court seeking to get around the absence of any international human rights exception to the sovereign immunity doctrine. Two judgments of the United States Supreme Court, however, demonstrate some of the limits on that hope under current understandings of the immunity exceptions. In Saudi Arabia v. Nelson, a US employee of a Saudi government-owned hospital brought a tort action in a US court against the hospital, its US purchasing agent and the Saudi Arabian government.37 Nelson had been employed as a monitoring systems engineer and had reported safety defects at the hospital. He was taken into custody and allegedly tortured. Nelson argued that, through the recruitment, employment contract and subsequent employment of Nelson, the hospital and the Saudi government were engaged in commercial activity and therefore that the injuries to Nelson arose “in connection” with commercial activity. The majority of the Court found that the alleged torts in this case involved wrongful arrest, imprisonment and torture, all conduct that related to exercise of the power of the police. The majority noted that Nelson did not allege breach of contract because, the majority presumed, the employment contract provided Saudi courts with exclusive jurisdiction over claims for breach of contract.38 The state exercise of the power of its police, however, “has long been understood for purposes of the restrictive theory as peculiarly sovereign in nature”.39 The connection of the arrest and torture to commercial activity was not sufficiently strong to be considered to be “based upon commercial activity” to qualify for the commercial activity exception. There are several difficulties in the reasoning in Nelson. The “nature of the activity” test was found not to permit recovery because the confinement, questioning and torture of employees is not normally considered part of commercial practice. The purpose of the confinement, however, did relate to the government’s perceived commercial interests. Indeed, similar measures are often taken by private actors. As White J observed in his partial dissent, “retaliation for whistle-blowing is not a practice foreign to the marketplace.”40 The court minimized these connections and overstated the distinction between the commercial and the police activities of the government. The majority also exaggerated the separation of different parts of running a business. The discontinuity of the employment relationship with the broader context failed to acknowledge that the disciplining of employees could effectively be achieved by imprisonment and torture. This approach artifically isolated state functions from market activities 37 38 39 40

507 US 349 (1993). Ibid. at 354. Ibid. at 361. Saudi Arabia v. Nelson, supra n. 37 at 365–6.

222 R Wai and the contract, connections that otherwise would have brought the case within the commercial activity exception.41 The US Supreme Court decision in Republic of Argentina v. Weltover, Inc. highlights the problems of the classification exercise and the differential treatment of commercial matters.42 The court ruled that the unilateral refinancing by the Argentinian government of government bonds issued as part of a currency restabilisation plan was sufficiently connected to commercial activity to come within the commercial exception. The court emphasised that the Argentinian government was acting “not as a regulator of a market, but in the manner of a private player within it”.43 The fact that the transaction was a debt contract was determinative. The purposes of the bonds, currency stabilisation, and of the refinancing itself—a regulatory response to a shortage of foreign currency—were ignored by the court. In addition, this treatment of public debt contracts ignored the degree to which investors or lenders have sufficient power, resources and expert advice to anticipate and assess risk of default and take steps to appropriately protect themselves, including through insisting on the inclusion of waiver of immunity clauses in the debt contract. (b) Increasing complexities It can be expected that neither the nature test nor the contracts/tort distinction will become any easier. In the wake of privatisation and the downsizing of government, an increasing number of traditional state functions are being devolved to the commercial realm. Examples include prisons, health care, education and dispute resolution. In this context, it becomes much more difficult to see where the boundaries of “sovereign behaviour” are to be determined. The nationality of both corporate and state actors is also becoming much less clear. As Robert Reich asked in the context of US industrial and trade policy, “who is us?” is a question that national legislatures and courts increasingly face as the corporations become less closely identified with a single jurisdiction.44 The Sudanese oil project evidences how complex the mix of public and private interests simultaneously with national and international interests, are becoming in the contemporary world system. Finally, the tests will become even more puzzling as the complexity of international society increases. In addition to transnational corporations, there is a proliferation of transnational networks, including governmental networks such as networks of banking regulators,45 and non-governmental networks such as transnational labour unions, environmental NGOs or diasporic ethnic 41 For an analysis of similar manipulation of time and sequence in the area of contract law, see S Fish, “The Law Wishes to Have a Formal Existence” in A Sarat and T R Kearns (eds), The Fate of Law, (Ann Arbor, University of Michigan Press, 1991) 153 at 188–93. 42 Supra n. 14. 43 Ibid. at 614. 44 R Reich, “Who is US?”, (Jan–Feb 1990) 90 Harvard Business Review 53. 45 See e.g. A-M Slaughter, “The Real New World Order”, (1997) 76 Foreign Affairs 183.

The Commercial Activity Exception 223 groups.46 Each of these networks has their own set of processes and norms which interact with state actors and regulations in increasingly complex ways.47

3 POLICY

The doctrine developed around the commercial activity exception has many contradictions and ambiguities. This is especially troubling given that much of consequence turns on the results of the classification. This part turns to the policy rationales for the immunity and the exception to determine whether policy distinctions might explain why the commercial activity exception exists and how the classification tests might be coherently interpreted and applied. It will be argued that analysis of the policy rationales for a change in the immunity doctrine away from absolute immunity does not support the distinct treatment of commercial and non-commercial activities. Policy argumentation therefore does not resolve doctrinal difficulties with respect to classification, but rather identifies the policy conflicts which underlie the doctrinal difficulties. In 1991, Harold Koh’s influential article on the developing field of transnational public interest litigation queried: “If contracts, why not torture?”48 Koh observed that doctrines such as sovereign immunity operated so that state activity involving contractual and commercial transactions was treated as more amenable to private litigation than more grievous state conduct involving, for example, human rights abuses. Koh argued that the distinction was neither justified in policy nor descriptive of contemporary judicial practice in the United States. Since international commercial litigation between private parties and state parties was expanding, Koh believed that expansion should also occur with respect to international litigation concerning other kinds of state misconduct, such as tort recovery for torture and human rights abuses. Nonetheless, the strong distinction between commercial and non-commercial activity has continued in immunity doctrine.49 This section explores what policy concerns justify this continued strong distinction, and seeks to provide additional explanations for its persistence. In particular, it will be argued that 46 See e.g. R Keohane & J Nye (eds), Transnational Relations and World Politics (Cambridge, Mass., Harvard University Press, 1974). 47 See e.g. G Teubner (ed.), Global Law Without A State (Aldershot, Dartmouth, 1997). 48 H Koh, “Transnational Public Law Litigation”, (1991) 100 Yale LJ 2347 at 2365. 49 One notable reform is the US Antiterrorism and Effective Death Penalty Act of 1996, section 221, which added an exception in section 1605(a)(7) to the US FSIA for claims brought against foreign states for personal injury or killing caused by various acts including torture, extrajudicial killing and hostage taking; see (1997) 36 ILM 759. There are significant restrictions under section 1605(a)(7) including that the exception does not apply unless the foreign state has been designated by the US executive as a “state sponsor of terrorism”; section 1605(a)(7)(A). States designated as state sponsors of terrorism as of 15 February 2000 included Cuba, Iran, Iraq, Libya, North Korea, Sudan and Syria: 22 Code of Federal Regulations (CFR) section 126.1(d). This is obviously a limited exception in comparison to the commercial activity exception; furthermore, creating an exception for immunity for non-commercial torts based on “terrorist state” classification raises obvious policy concerns in its own right.

224 R Wai concerns about international comity are discounted with respect to commercial activities because of influential policy ideas about the presence of consent and/or international cooperative benefit in economic matters.

The Immunity Doctrine The basic policy rationales for the immunity doctrine are usefully considered in two broad categories.50 First, concerns of “comity” require appropriate noninterference in the legal or political affairs of foreign states. Second, courts are concerned with issues of the separation of powers between the judicial and other branches of government and the relative expertise of the judicial as opposed to the political branch in international affairs. This chapter will focus on the “comity” policy rationale. Separation of powers and expertise concerns are important in many jurisdictions, but I will argue that the differential treatment of commercial activity and non-commercial activity is primarily related to differences in comity concerns. Moreover, this chapter is intended to investigate why the political and legislative levels, as well as the courts, support the differential treatment involved in the commercial activity exception. Of the various aspects of the comity rationale, three kinds of concerns are most relevant to the distinction made between commercial and non-commercial activity in immunity doctrine. (a) Impartiality One of the strongest policy rationales for the immunity doctrine is that permitting foreign courts to adjudicate disputes between foreign state actors and private parties will violate the basic objective in international affairs of impartiality among states. Unlike international institutions which at least have multinational composition, national courts are argued to lack the impartiality to adjudicate disputes between private parties and foreign states, especially in disputes involving the interests of the forum state or its nationals.51 National court judges can be internationalists and cultural sophisticates, but they can also be parochial and prejudiced. Moreover, perceptions of bias are always present when a national court adjudicates the government policy of another jurisdiction.52

50 51

This follows the useful categorisation of policy reasons adopted by Harold Koh, supra n. 48. L Henkin, “International Law: Politics, Values and Functions”, (1989) 216 Rec. des Cours 9 at

102. 52 See Henkin supra n. 51 and T Franck, The Structure of Impartiality: Examining the Riddle of One Law in a Fragmented World (New York, Macmillan, 1968).

The Commercial Activity Exception 225 (b) Restricting parochial self-interest Comity concerns also protect against the danger of the self-interested actions of legal actors. Recent studies of commercial arbitration raise concerns that a species of “regulatory competition” is occurring in which jurisdictions are changing substantive and procedural laws in order to attract the business of international dispute resolution to their cities, their lawyers and the attendant services.53 Moreover, while the pecuniary advantages of such litigation may be limited, there are other kinds of benefits that may motivate private litigants and service providers. In particular, transnational litigation may feed into the social capital and paternalistic impulses of centres of legal elites.54 (c) Self-determination Intervention by foreign courts is also criticized on the basis that it fails to respect the moral and practical concern that societies be left to develop their own mix of laws and institutions at their own pace. This argument has a long pedigree. John Stuart Mill, for example, argued in his essay on non-intervention that intervention by foreign powers, like paternalism by individuals, prevents societies from developing. Mill argued against foreign intervention to remove oppressive governments because, if the people “have not enough sufficient love of liberty to be able to wrest it from merely domestic oppressors, the liberty which is bestowed on them by other hand than their own, will have nothing real, nothing permanent”.55 More recently, Michael Walzer has revived this line of argument, arguing that non-intervention is required to permit processes of internal critique, self-choice and social struggle to occur.56 The non-intervention principle, a form of cruelty for kindness’ sake, is qualified in the arguments of both Mill and Walzer by a number of exceptions, including situations of secession struggles, massive human rights abuses and prior foreign interventions. The essential self-determination concern about foreign courts supporting private party claims is that they are pre-empting, in a deep sense, the “exhaustion of local remedies”. The stability of a domestic resolution often requires that societies go through difficult processes of confrontation and reconciliation. The value of a tort action occurring in foreign courts seems minimal in comparison 53 M Reisman, Systems of Control in International Adjudication and Arbitration: Breakdown and Repair (Durham, Duke University Press, 1992); Y Dezalay and B Garth, Dealing in Virtue: International Commercial Arbitration and the Construction of a Transnational Legal Order (Chicago, University of Chicago Press, 1996). 54 For an interesting account of the complex manner in which material, ideational and legitimation interests and functions operate in international legal fields, see D Trubek, Y Dezalay, R Buchanan and J Davis, “Global Restructuring and the Law: Studies of the Internationalization of Legal Fields and the Creation of Transnational Arenas”, (1994) 44 Case Western L Rev 407. 55 J S Mill, “A Few Words on Non-Intervention”, in J M Robson (ed.), Collected Works of John Stuart Mill, vol. XXI, (Toronto, University of Toronto Press, 1984) 110 at 122. 56 M Walzer, Just and Unjust Wars (New York, Basic Books, 1977), chapter 6.

226 R Wai to what is achieved through sustained political efforts such as the demonstrations of the Madres de la Plaza de Mayo in Argentina or the operation of the Truth and Reconciliation process in South Africa. Foreign legal actions can be consistent with such political efforts, but they can also distract from and contradict other forms of political action.

The Commercial Activity Exception An absolute theory of sovereign immunity has the virtue of a high degree of certainty. However, it is easily criticized as causing injustice that overwhelms its basic policy justifications. A jurisdictional immunity is a blunt instrument; it prevents a court from considering either the merits of a plaintiff’s claim or many of the procedural impediments to a plaintiff’s ability to seek an effective remedy in the courts of a sovereign defendant.57 In comparison, the potential for interventionism of a decision to find an exception to sovereign immunity is not as great because a foreign court may still limit its interference with the foreign state in two ways. First, limiting sovereign immunity does not automatically confer jurisdiction on a British or Canadian court. A plaintiff must meet normal requirements for jurisdiction.58 A court may also decline jurisdiction for reasons of forum non conveniens.59 Second, a court that assumes jurisdiction may very well apply a choice of law rule that results in the application of the substance of the impleaded foreign state’s own law. A choice of law rule for tort that is in whole or in part based on the lex loci delicti will result in the application of the foreign law of the jurisdiction.60 Many have questioned whether comity concerns support an absolute immunity.61 In affirming the restrictive theory, the House of Lords identified two policy foundations: first, justice to private individuals transacting with foreign states, and second, the lack of any threat to the dignity of the foreign state or interference with its sovereign functions.62 The puzzle remains, however, why foreign legislatures and courts have set the boundaries for their interference where they have: in commercial matters, but not in other realms. Lauterpacht, for example, based his critique of sovereign immunity on the vindication of individual remedies against the state under domestic principles of the rule of law.63 57

See text following n. 4, supra. For example, the UK requirements under Order 11 for service out of jurisdiction; see Dicey & Morris, supra n. 2, at 243. 59 UK: Spiliada Maritime Corp. v. Cansulex Ltd. [1987] AC 460; Canada: Amchem Products Inc. v. British Columbia (Workers’ Compensation Board) [1993] 1 SCR 897; US: Gulf Oil Corp. v. Gilbert, 330 US 501 (1947); Piper Aircraft Co. v. Reyno, 454 US 235 (1981). 60 UK: Private International Law (Miscellaneous Provisions) Act 1995, Part III; Canada: Tolofson v. Jensen; Lucas (Litigation Guardian of) v. Gagnon, [1994] 3 SCR 1022. 61 E.g. Lauterpacht, supra n. 7 at 230–2. 62 I Congresso del Partido [1983] 1 AC 244 at 262 per Lord Wilberforce. 63 Lauterpacht, supra n. 7, at 220, 235. 58

The Commercial Activity Exception 227 These policy arguments based on securing individual remedies against the state under the rule of law led Lauterpacht to a prescription for the abolition of the sovereign immunity doctrine which did not end at the boundaries of commercial activity.64 The goals of protecting individual rights or doing justice to private individuals would seem to be at least as pressing with respect to claims of non-commercial harm, such as torture. To make sense of the continued use of the distinction, I will argue that restricting the exceptions to commercial activities and contracts rests on a set of under-articulated ideas about the nature of the international economy and the role of states in it.

Increasing State Involvement in Commerce The standard policy justification for the development of the commercial exception is the claim that the commercial functions of modern governments have expanded.65 It is claimed that state governments or state entities have increasingly operated in fields that go beyond functions like policing or national defence into areas of commercial enterprise. The “sovereign” nature of the behaviour of states was argued to be losing much of its distinctively public character. The proper response in immunity doctrine, it was reasoned, was that the private system of claims that existed should be extended to sovereign entities acting in these contexts. Both the claim that states now play a larger role in the economy and the notion that the proper response was to limit sovereign immunity are questionable. Historically, there are many examples of extensive involvement of government in the market, from the granting of preferential charters to companies such as the British East India Company to the tariff protection, government contracts and subsidies provided to private industry in nineteenth century economic development in the US and Germany.66 Nor does it follow that, even if there is greater involvement of states in the commercial economy, the public interest or purpose involved is any less than with respect to traditional functions of governments. Changing social definitions of what constitute the legitimate functions of governments are precisely the kind of important social decisions that would seem deserving of respect for comity reasons. Moreover, state participation in commercial activities differs from that of private parties in a variety of ways, including the public purpose of most state commercial activities and the fact that the control of property and profits from these activities rests, at least nominally, with the state and hence the 64

Ibid. at 226, 237. See e.g. Trendtex Trading Corp. Ltd. v. Central Bank of Nigeria, [1977] QB 529 at 555–6 (CA) per Lord Denning; Badr, supra n. 2 at 1; Tate Letter, supra n. 6; Williams and de Mestral, supra n. 2, at 142–4. 66 See e.g. N. Chomsky, “Free Trade and Free Market: Pretense and Practice”, in F Jameson and M Miyoshi (eds), The Cultures of Globalization (Durham, Duke University Press, 1998) 356. 65

228 R Wai general population. It is therefore not clear why state involvement in commercial activities should lead to any lesser respect for comity concerns than state action in more traditional areas. The Importance of Consent A more sophisticated way to overcome comity concerns with respect to commercial activity is to claim that, in contrast to most other kinds of state action, the state actor in most commercial contexts has effectively consented to an exception to the normal sovereign immunity it would expect from foreign courts. Consent is argued to be present in the commercial context, and in particular where contracts are present, in a way that is not true of other areas of state action such as security or defence. The connection between doctrines of consent and international law has been explored extensively in writings on public international law.67 Consent doctrine is crucial at numerous levels in the efforts of international law to mediate disputes between actors that are, both in political reality and at the conceptual level, sovereign equals in an anarchic international structure with no overarching authority. Liberal international legalism attempts to address the anarchic nature of the international system through concepts developed in domestic liberal political theory, including theories of social contract and consent.68 The authority of rules of international law is argued to flow from state consent to both customary international law and treaty-based conventional international law.69 In the particular context of private party litigation in foreign courts, consent is an even more crucial policy rationale because the assumption of jurisdiction is being made by a foreign court rather than an international entity. Actual consent is a means for a foreign party and a foreign court to overcome legitimacy concerns about both judicial review of state actors and judicial review of foreign actors.70 The importance of consent for sovereign immunity exceptions is best exemplified by the significance accorded to the existence of a contract in the classification of facts that trigger an exception to the immunity doctrine. (a) Express contractual provisions The argument that consent is what generally distinguishes legal facts involving a contract from those involving a tort in international contexts parallels similar 67 See e.g. H Lauterpacht, Private Law Sources and Analogies of International Law, with Special Reference to International Arbitration (1927, republished Hamden, Archon Books, 1970); Kennedy, supra n. 3 at 159–65. 68 See Koskenniemi, supra n. 12, at 52–73. 69 Koskenniemi labels this structure the “ascending pattern of justification”; ibid. at 41. 70 For a similar observation with respect to choice of law, see L Brilmayer, Conflict of Laws, 2nd ed., (Boston, Little, Brown, 1995) at 234–5.

The Commercial Activity Exception 229 policy arguments for the contract/tort distinction made in domestic private law. Contracts are argued to be distinctive because they are based on the voluntary agreement of fully informed parties.71 Consequently, contracts are thought to create mutual, actual and reasonable expectations and potential reliance on the part of both parties.72 In the immunity context, it is argued that the opportunity to negotiate and agree on specific terms means that enforcement of contractual terms would accord with the expectations of both sovereign actors and the private parties who would deal with them in an international commercial context. Consent arguments for enforcement of contractual provisions are most convincing where there are not dramatic differences in bargaining power or problems with imperfect information. Sometimes, sovereign actors may suffer from duress or unequal bargaining power in relation to more powerful parties such as multinational corporations, financial institutions or international organisations.73 Generally, however, sovereign actors have their own forms of negotiating power, including legislative jurisdiction, military powers and legal advisers, which mean that they are not as powerless or uninformed as the typical parties whose claims of duress or unequal bargaining power are most convincing at contract law. In the international context, contractual provisions among sophisticated parties often specifically address issues of dispute resolution, including forum selection, arbitration, jurisdiction, choice of law and waiver of sovereign immunity.74 When the state expressly agrees to a waiver of sovereign immunity, the expectations and reliance interests of the other party are such that the state cannot opportunistically invoke its sovereign status as a trump for asserting the jurisdictional immunity it had previously promised not to assert. This is the basic, and strong, policy justification for the waiver exception to sovereign immunity.

71 M Trebilcock, The Limits of Freedom of Contract (Cambridge, Mass., Harvard University Press, 1993) at 7, citing M. Friedman, Capitalism and Freedom (Chicago, University of Chicago Press, 1962) at 13:

“The possibility of co-ordination through voluntary co-operation rests on the elementary—yet frequently denied—proposition that both parties to an economic transaction benefit from it, provided the transaction is bi-laterally voluntary and informed.” (emphasis in original) 72 For the classic articulation of the expectation, reliance and restitution purposes in contract law, see L Fuller and W R Purdue, “The Reliance Interest in Contract Damages”, (1936) 46 Yale LJ 52 at 53–7. 73 See e.g. W Mansell, “Legal Aspects of International Debt”, (1991) 18 Journal of Law and Society 381 at 391–5. 74 Practice guides and articles advise parties to expressly attend to these issues. See e.g. J G Castel, W Graham, S Hainsworth, A de Mestral and M Warner (eds), The Canadian Law and Practice of International Trade with Particular Emphasis on Export and Import of Goods and Services (Toronto, Emond Montgomery Publications, 1997) at 178; E Kwaw, “Sovereign Immunity and Jurisdiction Clauses in International Lending: The Canadian Perspective”, (1995) 11 Banking & Finance L Rev 27.

230 R Wai (b) Implied consent and penumbral cases Once removed from the context of express and specific consent to waive immunity, the strong distinction between contract and tort law based on the criteria of consent is much more problematic. Often contracts do not expressly include a waiver-of-immunity clause. In this context, can it be concluded that the state actor has agreed that there is no immunity or that the private party has formed legitimate expectations of no immunity? Is it not at least as plausible that the private party has accepted the risk and factored that risk into the decision to proceed with the transaction based on the express terms of the contract against a baseline, or presumption, of sovereign immunity? At a descriptive level, the consensual agreement model has been criticised as an inaccurate account of both the historical development and the current doctrine of contract law with respect to contexts beyond a narrow ideal form of executory contract.75 The basic conditions for the mutual agreement model often do not obtain because of power imbalances, information failures or monopolistic market structures. The fact that contract law exists within a matrix of other laws (constitutional, statutory, administrative and common law) the existence of special rules for particular kinds of contract (such as employment and consumer contracts), the myriad doctrinal exceptions (such as promissory estoppel, duress, mistake, frustration and implied terms), and the limits at the level of remedies and contractual damages (remoteness constraints, reliance and restitution in addition to or instead of expectation) belie the general model of contract contained in the mutual agreement model. At a normative level, moreover, contract law is intended to serve a number of different, sometimes cross-cutting, policy objectives that go beyond mere enforcement of agreements and that include social utility, efficiency, paternalism, distributive justice and enforcement of morals.76 Furthermore, outside of a narrow area of contracts, the idea that consent and voluntarism are what distinguish contract law from tort law has been attacked. In areas such as the intentional torts, negligent misrepresentation, remoteness of damage, voluntary assumption of risk and informed consent in medical torts, tort is shown to be as much concerned as contract with the presence or absence of consent. The absence of a contract does not necessarily indicate a lack of common understandings, shared expectations or reasonable reliance. Tortious harms may occur between parties who could have but chose not to negotiate a contractual term covering immunity. Once the immunity exception moves from the situation of clearly expressed waivers, the consensual elements of many contract situations are equalled by many tort situations. 75 See e.g. P S Atiyah, The Rise and Fall of Freedom of Contract (Oxford, Clarendon Press, 1979); G Gilmore, The Death of Contract (Columbus, Ohio State University Press, 1974). 76 See e.g. D Kennedy, “Distributive and Paternalist Motives in Contract and Tort Law”, (1982) 41 Maryland L Rev 563; A Kronman, “Contract Law and Distributive Justice”, (1980) 89 Yale LJ 472; H Collins, Contract Law, 2nd ed. (London, Butterworths, 1993).

The Commercial Activity Exception 231 What rules should apply in this context with respect to immunity? One basic response would be that the search for consent is misguided and that the policy basis for tort liability at the international level should be founded on different reasons from that of contracts. In this connection, accepted categories of prohibited behaviour, for example those covered in the Convention against Torture or by crimes against humanity would be related not to consent of states, but rather to imputed standards of justice or specific human rights. These could be treated as jus cogens norms to which all states are bound as a result of, or even as a condition of, being recognised as sovereign states in the international legal system. Interpretation of the existence and content of implied terms takes place against a backdrop of normative baselines that are either chronologically or lexically prior to the contract being construed.77 It could be argued that state immunity remains the operative baseline, such that failure to include it in a contract must be given dispositive significance. But, then, the emergent law of core norms of jus cogens itself is a counter-baseline, one whose relationship to state immunity is in need of interpretation in order to locate a kind of baseline of baselines against which state contracts can be construed. Given the strength of the consent idea in international affairs in overcoming comity concerns, however, an alternative approach may be more persuasive. This approach would emphasise that given the use of implied consent ideas with respect to commercial activities, an equivalently broad sense of implied consent is justified with respect to other kinds of state activity. First, some government activity currently covered by immunity could be understood to involve a violation of a social contract. Important work has revived the “contractarian” tradition of analysis of morality and theories of justice. This work relies not on actual historical social contract, but rather on ideas of contract in carefully defined conditions of human equality and identity.78 An elaboration of this connection might demonstrate how the consent that should be considered in addition to sovereign consent is the consent of citizens to the authority of the state itself. A sovereign or an official acting in a sovereign capacity who violates the rights of the citizen effectively loses the political legitimacy that justifies its “vertical” claim of juridical authority over that individual.79 While it is arguable that individuals might agree to a government that was not fully democratic, no government could argue that actions such as torture or slavery would ever be agreed to by an individual agent.80 Consent analysis by a foreign court could link the legitimacy of its jurisdiction to a sovereign’s violation of the fundamental 77 See Collins, Contract Law, supra n. 76 and more specifically in the context of international law, F Krachtowil, “The Limits of Contract”, (1994) 5 EJIL 465. 78 D Gauthier, Morals by Agreement (Oxford, Clarendon Press, 1986) and Moral Dealing: Contracts, Reason and Ethics (Ithaca, NY, Cornell University Press, 1990). J Rawls, A Theory of Justice (Cambridge, Mass., Harvard University Press, 1971); Political Liberalism (New York, Columbia University Press, 1993); The Law of Peoples (Cambridge, Mass., Harvard University Press, 1999). 79 L Brilmayer, Justifying International Acts (Ithaca, Cornell University Press, 1989) at 2–3. 80 See e.g. Rawls, The Law of Peoples, supra n. 78, at 65–6.

232 R Wai expectations and assumptions of its citizens. In many societies, these assumptions are reflected in constitutional documents and, in this respect, overriding public law values can join up with the idea of social contract to shape private law. As a direct parallel to the constitutional nexus, courts could imply consent when sovereign actors violate international treaties. Where states have signed international conventions promising not to violate certain human rights, for example, it may be argued they have manifested an actual, or at least reasonably imputed, consent to claims based on such violations. This kind of implied consent would not seem to be any more interventionist than implying consent from contracts where there is no specific waiver clause.

The Ideal of Cooperative Benefits What has not been fully assessed in the literature on sovereign immunity is the degree to which implied consent reasoning with respect to the commercial activity exception is now reinforced by a set of beliefs about the international cooperative benefits of the commercial activity exception. Yet, these arguments which assume non-controversial cooperative benefits are currently the most significant policy justification for the differential treatment of commercial and non-commercial activities, and accordingly warrant deeper analysis. These policy objectives inform the structure and provisions of the statutes as well as the court judgments applying the statutory provisions in difficult cases. (a) The problem of cooperation in an anarchic structure The international law tradition relies heavily on the idea of cooperative benefit to justify its own existence. International law is seen to partially address a number of situations where states could mutually benefit from an international regime: security conflicts, diplomatic processes, international public goods and externalities, and international economic cooperation. International anarchy, the absence of overarching authority, is seen to harm all national parties. Longrun mutual benefits are impeded by the lack of trust and the absence of a coordinating authority which could ensure that other parties do not cheat.81 In the immunity context, it could be argued that the broad use of sovereign immunity would lead to situations where there is a cooperative loss to private and state actors with different national identities. What is needed is for states as well as private actors to make credible commitments which will ensure that 81 In the language of game theory, international anarchy is a situation of Prisoner’s Dilemma, the Stag Hunt or the Game of Chicken. These simple games model situations where short-term selfinterest will lead participants to outcomes that are less favourable to each than if they cooperated. For a survey of the use of game theory in international law, see K Abbott, “Modern International Relations Theory: A Prospectus for International Lawyers”, (1989) 14 Yale J Int’l L 335.

The Commercial Activity Exception 233 beneficial cooperation can proceed.82 Claims for immunity by states will prevent some mutually beneficial international transactions occurring. One way of ensuring that such cooperative commitment can exist is for other legal regimes, such as international arbitration, to provide private parties legal redress in conflicts with foreign states. In the absence of international institutions, it is argued that at least the foreign courts of jurisdictions other than the state party could act as the kind of venue that could address such concerns. While this falls short of the ideal of a multinational and impartial panel with binding authority of decision, it would provide a credible commitment that is not under the direct control of the foreign state actor.83 Central to the treatment of commercial activity as distinct from noncommercial activity is a policy claim that facilitating international commercial cooperation is a situation of incontrovertible cooperative benefit, while facilitating the litigation of suits for non-commercial activity is best characterized as fraught with genuine conflicts in interests and values. Reasonable differences are thought to exist with respect to non-commercial issues such as police action in a way that are argued not to exist with respect to commercial matters. In the economic realm, a further set of arguments are used to demonstrate the cooperative benefits that can be obtained through enforcement in foreign courts. (b) Cooperative economic benefits In the commercial realm, non-controversial cooperative benefits are claimed to arise because it is in the long-run interests of state actors to waive or exclude immunity in order to encourage contracting by private parties for purposes such as international lending to states. The most fundamental economic rationale for the commercial activity exception is that there is a cooperative, Pareto-optimal welfare gain that can be made in a move from an absolute doctrine of sovereign immunity to a restrictive doctrine based on a commercial activity exception. Pareto optimality is the least contested of the basic welfare criteria in welfare economics. It provides that there is a welfare gain when at least one party benefits from a change in policy, while no party is made worse off.84 The efficiency gain supposedly arises from the ability of private parties to rely on judicial enforcement of transactional agreements. The argument extends to benefits to sovereign states. The argument is that sovereign actors will be harmed by a doctrine of immunity because private actors may refuse to deal with them because of the risk of default and immunity by the other party. More commonly, the risk of default and immunity 82 For a basic analysis of the function of credible promises in a debt contract context, see D Baird, R Gertner and R Picker, Game Theory and the Law (Cambridge, Mass., Harvard University Press, 1994), chapter 2. 83 An international level solution is, of course, preferable to the national court solution for reasons of impartiality; see supra nn. 51 and 52; McGinty, supra n. 29 at 88. 84 For a brief review of welfare economics criteria, including Pareto optimality, see Trebilcock, supra n. 71 at 7,17.

234 R Wai by the state actor will lead private actors to price into the transaction a sovereign risk premium, which will be passed on to the sovereign actor.85 Similar arguments are made with respect to the value of stable, enforceable regimes of private law. A number of economists have emphasised the importance of strong regimes of property rights and effective enforcement of contractual obligations to the development and the expansion of complex market economies based on specialisation of production and trade.86 While national markets are complex, coordination and compliance in international markets is even more complex because of problems of distance, differing traditions of exchange and diverse expectations. Predictable enforcement of rules allows state and non-state actors to make credible commitments to each other to control for future opportunism. In the absence of such regimes, which would be the case where a state immunity doctrine applied, a situation of international anarchy would obtain in which beneficial economic transactions would not be undertaken or would be more costly than they otherwise would be under a regime of enforceable law. A similar claim in tort law would be that parties will take excessive steps to avoid transacting or getting into disputes with state actors. Private actors would have to engage in costly adjustment schemes, whether they consist in taking out insurance policies or in simply avoiding operations that expose the private party to injury from sovereign actors. Both sides lose potential economic benefits. Generally, however, it seems that legislatures and courts have been unable or unwilling to extend the economic reasoning to most tort situations. Cases such as Saudi Arabia v. Nelson show how the courts are much more restrictive with tort claims even when they occur in a commercial or contractual context.87 It seems that transnational disputes about human rights or police misconduct are perceived as distinct from commercial disputes in raising fundamental conflicts in values. Foreign legislatures and courts seem to believe that such conflicts are zero-sum situations in which there can be no long-run benefit to both sides in having police powers, on the one hand, or human rights accountability, on the other hand, upheld by way of foreign court actions.

85 E.g. Schreuer, supra n. 2, at 9; G Delaume, “The Foreign Sovereign Immunities Act and Public Debt: Some Fifteen Years Later”, (1994) 88 AJIL 257. For a similar argument on the economic costs of unpredictability concerning the Act of State doctrine, see M Ramsey, “Acts of State and Foreign Sovereign Obligations”, (1998) 39 Harvard Int’l LJ 1 at 3. The same argument suggests that sophisticated commercial actors may not require protection against political risks which they were aware of when they contracted or otherwise factored into their economic calculations. Talisman Energy, for example, presumably factored in the political risks involved in its purchase in 1998 of its interests in the Greater Nile Oil projects, and allowed for this in the purchase price it was willing to pay: see supra n. 35. 86 D North, Institutions, Institutional Change, and Economic Performance (Cambridge, Cambridge University Press, 1990); M. Olson, “Big Bills Left on the Sidewalk: Why Some Nations are Rich, and Others Poor”, (1996) 10 Journal of Economic Perspectives 3 at 22. 87 See discussion of this case in text at n. 37, supra.

The Commercial Activity Exception 235 The Limits of Cooperation (a) The problematic idea of the economy as the realm of cooperation The logic of economic claims about the costs of immunity is more contestable and debatable than is often understood. It is sometimes true that sovereign risk will either prevent transactions or cause an increase in the risk premium priced into the transactional price. But on the other hand, there is no guarantee of Pareto-optimal gains from the removal of an immunity in commercial contexts. It is not possible to generalise that Pareto gains will occur in all situations. Much of the welfare gain depends on whether the existing situation is one of relatively equal bargaining power, full information and fully competitive markets.88 Some states, for example, may, because of lesser economic power or imperfect information, be unable to negotiate an immunity even where their ex ante preferences would be for such an immunity.89 The same concern motivates mandatory or default rules in, for example, consumer contexts.90 An immunity in this context would operate to insulate weak state governments from having to correct for power imbalances in negotiating loans from, for example, powerful private lenders. Cooperative game theory approaches seem to recognise that no international solution may arise that is non-controversial without taking into account the distributive consequences among states. This kind of theorising has the appearance of both recognising conflict of interests and positing the possibility of cooperation. However, in doing so, theorists of cooperation use initial welfare assumptions which are questionable given the nature of contemporary international society. The potential gains from cooperation are arbitrarily fixed through initial assumptions for the game such that the cooperative solution is better for both sides.91 Setting different payoffs/benefits may mean that cooperative benefits are no longer as obvious. Prisoner’s Dilemma, in particular, has a single point focus. The initial structure of the game is such that there is a “uniquely compelling point of mutual cooperation”;92 both parties can gain by moving to that point. However, where there are several potential points to which parties could move, 88

See Kennedy and Kronman, supra n. 76. See Mansell, supra n. 73. 90 See e.g. Kronman, supra n. 76; R Craswell, “Contract Law, Default Rules and the Philosophy of Promising”, (1989) Michigan L Rev 489. 91 A Sen, “Gender and Cooperative Conflicts”, in I Tinker (ed.), Persistent Inequalities: Women and World Development 123 (New York, Oxford University Press, 1990); A Sen, Collective Choice and Social Welfare (San Francisco, Holden-Dey, 1970). In a Prisoner’s Dilemma game, for example, the payoffs are structured such that each party has a strictly dominant strategy (confession) which is preferable regardless of what the other party chooses: see Baird, Gertner and Picker, supra n. 82, at 33. 92 D Snidal, “Political Economy and International Institutions”, (1996) 16 Int’l Review of Law & Economics 121 at 124. 89

236 R Wai a genuine conflict remains. As welfare economics has long known, the criteria of Pareto optimality typically does not lead to a single point solution.93 Moreover, given the possibility of multiple welfare equilibria, there is the possibility of lack of stability with respect to such points. It is precisely this type of conflict over different points “along the Pareto frontier” which several commentators now argue best characterises much that is interesting in international relations.94 This is as true in the economic realm (such as in trade relations) as in other areas such as military conflict or conflicts in human rights values. These criticisms identify a potential limit on the usefulness of the idea of noncontroversial cooperative benefit with respect to reform of immunity rules. As Joseph Singer and others have emphasised, the nature of conflict in private international litigation is such that “real conflicts” are endemic.95 The sense of zerosum conflicts in commercial contexts increases still more if, in addition to policies identifiable in particular state action, one considers the implied policy choices involved in decisions by jurisdictions not to act.96 For example, limiting tort liability is as much a policy choice, made by many jurisdictions, as extending tort liability, and should be recognised as such. Indeed, one indication of the lack of non-controversial cooperative benefits with respect to sovereign immunity may be the lack of any international convention or settled customary international law on immunity.97 Perhaps national courts are gradually evolving a substitute cooperative regime through internationalist policies such as selective application of the notion of comity with respect to commercial activity. Arguably, however, the courts in pursuing such reforms are effectively ignoring the lack of actual mutual cooperative gains because of a false conviction in the existence of such gains in the economic realm and the absence of such gains in the non-economic realm. (b) The overlap of economic and non-economic interests Most importantly, the arguments provided for the cooperative benefit of restrictive immunity in commercial contexts would also seem to justify restricting immunity in other contexts. The problems of indeterminacy in application of the doctrinal test for commercial activity is replicated in the indeterminacy in application of the policy distinction between economic and political conflicts. 93

E.g. Snidal, supra n. 92. E.g. S Krasner, “Global Communications and National Power: Life on the Pareto Frontier”, (1991) 43 World Politics 336. Snidal, supra n. 92, at 125 suggests that the task should be “to incorporate the bargaining problem over distribution into the cooperation problem”. 95 J Singer, “Real Conflicts”, (1989) 69 Boston University L Rev 1. 96 Singer, ibid. at 41. Joel Trachtman usefully notes in the related context of the allocation of prescriptive jurisdiction that “we must weigh not only the positive policies of other states, but also the negative, or laissez-faire policies”; J Trachtman, “Externalities and extraterritoriality: The law and economics of prescriptive jurisdiction”, in J Bhandari and A Sykes (eds), Economic Dimensions in International Law: Comparative and Empirical Perspectives, (Cambridge, Cambridge University Press, 1997) 642 at 671–2. 97 See e.g. Brownlie, supra n. 4, at 332–3. 94

The Commercial Activity Exception 237 Yet courts and legislatures have been unable or unwilling to generate a similar set of policy arguments for non-controversial cooperative benefits of enforcement with respect to other kinds of state action. Non-economic kinds of state action do not involve the flow of commercial benefits to the state and so, it is claimed, lack a similar long-run incentive structure. If the argument about commercial activities overestimates the cooperative economic benefits flowing to both private and state actors, the immunity for state conduct such as torture or other human rights abuses underestimates the degree to which economic and commercial activity is linked to a country’s general political system and record of legal protection. It might be assumed that foreign investors, for example, would generally care little about human rights abuses. However, foreign actors might care about these as a practical necessity in a range of contexts. Political corruption and unreliable legal systems, for example, are identified as a serious impediment to doing business abroad.98 Similarly, human rights abuses that affect the free movement or expression of the agents or employees of foreign businesses can impede effective economic conduct abroad. Saudi Arabia v. Nelson is a good example of sovereign misconduct which would harm the purely economic considerations of employees and future employees of the Saudi government. The argument operates in the other direction as well: states may use their police powers in ways that lead to concrete economic benefits to commercial actors. This is the nature of the complaints, for example, about the activity of governments in countries such as Burma and the Sudan where it is alleged the state security apparatus is being used to quell local or popular discontent that would disrupt economic operations, of oil companies in particular.99 However, this counter-example may prove too much as it speaks only to a short-term horizon and fails to account for how serious public attention can be focused almost instantly on a company’s operations in a way which not only affects short-term stock values,100 but also affects corporate reputation and thus medium-to-long term prospects for investment in and loans to the company for future operations. It is sometimes argued that the same economic analysis, and economic incentives or harms, are irrelevant to most state actors when they abuse human rights. Rather, it is concerns about national security or political domination that are 98 Businesses are claimed to factor in the presence of political corruption and the strength of legal systems in considering how attractive a country will be for investment purposes; see e.g. World Bank, World Development Report 1997: The State in a Changing World (New York, Oxford University Press, 1997) at 102–3 and R Barro, Determinants of Economic Growth (Cambridge, Mass., MIT Press, 1997) at 26–7. A number of indices have been created to measure the extent of corruption and other rule of law considerations; see for example, the “credibility indicator” discussed in World Bank, ibid. at 174–5, the “rule of law” index discussed in Barro, ibid. at 26–7, and the Corruption Perception Index issued by Transparency International on its website at http://www.transparency.de/. 99 See discussion in text accompanying nn. 35 and 36, supra. 100 For example, it is claimed that Talisman Energy’s Sudanese projects have led to a discount to its share price Sudan; see “Ottawa okay gives boost to Talisman”, The Globe & Mail (Toronto), 15 February 2000, B1; “Talisman divestment campaign unpopular”, The Globe & Mail (Toronto), 7 January 2000, B1.

238 R Wai pertinent. However, this is not necessarily true. The cost or loss of economic value is a factor that at least is weighed into the decisional process of most state actors. Such a belief helps explain, for example, why economic sanctions, a wellestablished if controversial practice in international deterrence, continue to be used as a policy tool.101 Concern about encouraging economic activity seems to be one of the few constants across different kinds of political regime. (c) Whose interests count? The problems of a state-centric focus Finally, it should be noted that most economic arguments and indeed most arguments about international cooperative benefit ignore the question of the unit of analysis. It is a common observation in international trade that while the argument for free trade does suggest that all societies benefit overall from trade, there may well be losers at the sub-state level. More attention must be paid to who enjoys the economic benefits and who is harmed. Too often individuals, and especially members of marginalised groups within a society, do not benefit. The claimed benefits of having the commercial activity exception but not an exception for human rights and other kinds of state misconduct is premised on the identity of states and their societies as coherent, single interests. This ignores the fact that state governments, the subjects of immunities, are not identical with their societies, let alone identical with the various sub-state constituents of a society.102 This poses problems both for descriptive analysis and for normative evaluation. On the empirical, predictive level, such a statist focus ignores the degree to which complex layers of cross-border connection and negotiation affect international processes of cooperation and conflict.103 On the normative level, this approach falsely assumes that the state’s preference for immunity is identical to the interests of its population. This ignores the possibility of a conflict of interests among different parts of the society of that state. A particularly obvious conflict of interest that could exist would be between government authorities and vulnerable minority groups within the society. Such a configuration of conflict is the basis for objections to the connection between governmental activity and the activities of foreign energy corporations in states such as Nigeria, Burma and Sudan. Whose interests are being served by the invocation of sovereign immunity in a foreign court to block claims against the governments of Nigeria, Burma or Sudan? Viewed from a human rights perspective, this question assumes great relevance. The approach to both the commercial activity exception and the non-commercial activity immunity is based on crude statist assumptions about the cooperative benefits to state parties according to 101 For a discussion of the effectiveness of economic sanctions as a tool for foreign policy, see D Baldwin, Economic Statecraft (Princeton, Princeton University Press, 1985). 102 Snidal, supra n. 92, at 130–1. 103 See e.g. R Putnam, “Diplomacy and Domestic Politics: The Logic of Two-Level Games”, (1988) 42 International Organization 427.

The Commercial Activity Exception 239 which there is no possibility of differentiating among situations based on inequitable distribution of benefits among non-state interests. (d) Globalisation of effects The commercial activity exception to sovereign immunity doctrine seems to reflect the reality of increased transnational ties. The exception on its face recognizes that sovereign entities act in ways that, intentionally or not, have effects on parties and societies outside of their own borders. Globalisation brought by technological and ideological changes has meant that communication, transportation and economic activity span the globe.104 Doctrines like sovereign immunity obscure this reality by setting out legal rules more appropriate to a world of autarkic states. Moreover, the policy analysis of mutual benefit which justifies the commercial activity exception also highlights the parochialism involved in refusing to recognise the harms caused by non-commercial acts with effects on the populations of foreign jurisdictions.105 Indeed, normative obligations would seem to flow across borders even more easily than economic benefits and effects. As the Supreme Court of Canada has noted in the context of criminal law jurisdiction, “we are all our brother’s keeper” in a global society.106 Any national court must consider carefully why it should refuse the claims of individuals based on a problematic ethics of parochialism and bounded responsibilities. A traditional defence of sovereignty also seems inappropriate for a world system characterised by significant flows of peoples across borders and individual, family and social identities that are hybrid and shifting.107

4 IMMUNITY EXCEPTIONS , IDEOLOGY AND THE DANGERS OF BIAS

The preceding analysis suggests that, at both the doctrinal and policy levels, there are no simple, coherent justifications for the distinctions made with respect to the commercial activity exceptions. Much will depend on the particularities of a case. In turn, this may support Koh’s view that prohibitory rules such as the absolute rule of sovereign immunity should be rejected.108 This position seems most consistent with the above-elaborated critical position as well. The problem of indeterminacy in doctrine and in policy would 104

C Beitz, Political Theory and International Relations (Princeton, Princeton University Press,

1979). 105

Ibid. R v. Libman [1985] 2 SCR 178 at 214. See generally P Cheah and B Robbins (eds), Cosmopolitics: Thinking and Feeling beyond the Nation (Minneapolis, University of Minnesota Press, 1998). 108 Koh argues for “doctrinal targeting” approaches which avoids blanket rules and instead gives courts greater discretion to confront policy concerns such as comity or separation of powers in adjudicating particular cases; Koh, supra n. 48, at 2382–94. 106 107

240 R Wai seem to have little political import on its own, even if it requires sometimes strained efforts to create semi-coherent tests to make the doctrines workable. That being said, in addition to the indeterminacy critique, there are further concerns that when indeterminate rules are set down in a particular ideological context, there may be a potential bias that operates in their application. Indeterminacy can lead to sub rosa determinations by legislators, judges and other legal workers that are based on dominant ideological assumptions. In this respect, the central concern is that international legalism and the growth of international law and institutions have become affiliated with promoting one specific, if increasingly dominant, set of ideas about the proper role of the state. Understood in this way, a critical perspective cannot simply assume that it is justified to achieve consistency in the treatment of the commercial and the non-commercial through eliminating sovereign immunity for both. One must look deeper at how state immunity doctrine should fit into a perspective in which sovereignty is defended as a useful corrective against contestable but increasingly dominant forms of an ideology of neoliberalism. Legalism is of course itself a distinct form of ideology.109 Legalism involves the commitment to the superiority for social outcomes of court-based adjudication over other political forms of decision-making, whether those be dictatorship or democracy. In recent times, legalism has become a much more common response to international problems. Prominent examples include the proposed International Criminal Court110 and the more efficacious international trade dispute resolution system of rules, institutions and processes of the World Trade Organisation.111 This turn to international legalism occurs at a time when there have been changes in ideological debate within the international system that track a decline in the political, economic and ideological power of the non-Western world. The actual history of the immunity doctrine belies the idea that an international consensus supports the underlying policy justifications for a commercial activity exception. Even one generation ago, for example, dissenters from the view that there were exceptions to the absolute theory of sovereignty included most of the Communist states and many of the newly-independent states of the Third World.112 To some commentators, this controversy was an unfortunate byproduct of the Cold War and the aftermath of imperialism, an era which has since passed.113 This account ignores the degree to which the 109 J Shklar, Legalism: Law, Morals, and Political Trials (Cambridge, Mass., Harvard University Press, 1964). 110 Rome Statute of the International Criminal Court, (1998) 37 ILM 999; for a useful symposium discussion of the proposed court, see (1998) 10 EJIL 93–191. 111 G R Shell, “Trade Legalism and International Relations Theory: An Analysis of the World Trade Organization”, (1995) 44 Duke LJ 829. 112 This is central to why the restrictive theory could not be claimed to be a rule of international law; see Brierly, supra n. 3. 113 G M Badr, “Recent Developments in the Dynamics of Sovereign Immunity”, (1982) 30 Am J Comp L 678 at 682.

The Commercial Activity Exception 241 exceptions to immunity are still linked to controversial policy justifications. The possible result is a new form of “neoimperialism”, a Lochnerism without frontiers,114 in which national courts of the West support the narrowing of national policy alternatives out of a claimed respect for “fairness” to commercial parties, in support of “certainty and predictability of expectations”, and on the basis of “non-controversial efficiency gains”.

International Legalism in the Service of International Neoliberalism The connection between neoliberalism and internationalism has been the subject of increasingly sophisticated critiques.115 The concern with respect to immunity doctrine and the commercial activity exception is that they might work in the service of a form of neoliberal economic and political ideology that is currently triumphant—and triumphalist.116 Neoliberal political theory emphasises the importance of the private market and international trade, and a reduction, through deregulation, privatisation and decreased public expenditure, of government in the ordering of social relations, particularly in the realm of the economy.117 Neoliberal concerns arise because sovereign immunity doctrine and the commercial activity exception are fraught with issues of the proper role of the state in the economy and society. Sovereign immunity partially constructs the relationship between the public and the private in the guise of ordering the relationship between the national and the international.118 Sovereign immunity helps to insulate a state from international private law process; in this sense, it insulates the public from the private. Consequently, there could be several impacts of ideology on the contours of the sovereign immunity doctrine and the commercial activity exception. First, the commercial/non-commercial distinction seems to rely on a public/private distinction which privileges private—particularly private economic—interests over the public interest. The problems with this distinction have been extensively canvassed in legal and political theory.119 The critique is also found in some critical writings on sovereign immunity.120 The application 114 Lochner v. New York, 198 US 45 (1905). The majority in Lochner held that state labour laws setting maximum work hours to be a constitutional violation of the rights of employers and employees to enter into contracts. Lochner has become a symbol of judicial activism on behalf of a laissezfaire view of the economy. 115 See e.g. S Marks, “The End of History? Reflections on Some International Legal Theses”, (1997) 8 EJIL 449. 116 Most commonly, F Fukuyama, “The End of History”, (Summer 1989) 16 The National Interest 3 and F Fukuyama, The End of History and the Last Man (New York, Free Press, 1992). 117 See e.g. R Unger, What Should Legal Analysis Become? (London, Verso, 1996) at 9–10; Trubek et al., supra n. 54, at 408–10. 118 D Kennedy, “Receiving the International”, (1994) 10 Connecticut J Int’l L 1 at 6. 119 E.g. M Horwitz, “The Public/Private Distinction”, (1982) 130 University of Pennsylvania L Rev 1289. 120 See e.g. Morgan, supra n. 12, at 308–9, 317.

242 R Wai of the commercial exception evidences an unsophisticated understanding of the nature of contemporary sovereignty. Often, commercial activities may be as important to a national polity as non-commercial activities. The nature test, in particular, seems to deny that commercial and economic activity frequently have important public purposes.121 In addition, neoliberal ideas of the proper functions of the state seem to lie behind the juxtaposition of the close scrutiny of commercial activity with the extreme deference to police and security actions of the foreign state even in situations where such actions occur in contexts in which commercial interests are being served. Second, the idea that commercial activity is somehow less public than other kinds of state activity can be criticised as presuming the validity of a laissez-faire view of the proper role of the government in economic markets.122 In earlier decades where strong non-capitalist states existed in the world community, international law writers were hesitant to find the commercial exception that was making its way into national law had an international law status given these controversial laissez-faire assumptions.123 As Louis Henkin noted in his Hague lectures delivered at the end of the Cold War, there is a danger that the commercial activity exception through tests such as the private person test may “build in prejudices favouring the liberal state and against the welfare state” because a government will lose immunity when it does “what private persons could do” and therefore “gives such activities an aura of illegitimacy”.124 Henkin’s concern seems justified in that the commercial activity exception has expanded to cover more and more state economic functions. As noted above, for example, public debt contracts are now usually covered by the commercial activity exception, either because of statutory provisions, as in the UK,125 or because of judicial interpretation of the classification tests, as in the US.126 Third, the idea of a commercial exception creates a preferentialism in the way that it facilitates to the litigation concerns of commercial plaintiffs. In comparison, plaintiffs suffering from non-commercial harms such as human rights abuses will be defeated by sovereign immunity. Related concerns have been expressed about the tendency for the definition of national interests in the application of US extraterritorial regulation to include commercial interests but not human rights interests.127 Jose Alvarez has noted a similar disparity between the 121

See Schreuer, supra n. 2, at 15–16. For a striking statement of the connection between international law and the “largest possible ‘depolitisation’ of the economic sphere” that would restrict interference by national governments in the economy, see W Röpke, “Economic Order and International Law”, (1954) 86 Rec. des Cours 203 at 224. 123 See Brierly, supra n. 3, at 250; O Lissityzn, “Sovereign Immunity as a Norm of International Law” in W Friedmann, L Henkin and O Lissityzn (eds), Transnational Law in a Changing Society (New York, Columbia University Press, 1972) 188. 124 Henkin, supra n. 51, at 327–8. 125 UK SIA Section (3)(3)(b). 126 Republic of Argentina v. Weltover, Inc., supra n. 14; Delaume, supra n. 85. 127 Note, “Constructing the State Extraterritorially: Jurisdictional Discourse, The National Interest, and Transnational Norms”, (1990) 103 Harvard L Rev 1273 at 1304–5. 122

The Commercial Activity Exception 243 willingness of governments to support private party complaints against state governments in the investment context and the reluctance to support such claims in human rights contexts; from this perspective, investment treaties such as Chapter Eleven of the North American Free Trade Agreement constitute “a human rights treaty for a special-interest group” which is not accompanied by more general provisions for protections of human rights.128 Similarly, Canadian courts seem more willing to find immunities where various forms of domestic public regulation of foreign states, such as labour regulation, are at stake, rather than private claims.129

International Legalism in the Service of Neoimperialism An alternative account of the commercial activity exception would point neither to a change in the global economy nor to strong functional reasoning, but rather to the parochial interests and ideological predilections of Western states. In this account, neoimperial domination continues in a more disguised form through international policies such as the commercial activity exception.130 First, the development and shaping of the commercial activity exception by Western legislatures and courts can be plausibly criticised as the parochial protection of Western economic interests. Given the historical distribution of the flow of capital, goods and services in the world economy, the primary beneficiaries of a restrictive theory of sovereign immunity in commercial affairs would be private parties from the West who are contracting with sovereign authorities of various jurisdictions.131 It is Western corporations, for example, who are likely to contract to provide cement or shipping services to foreign governments,132 or who are the bondholders on foreign debts.133 The impugned government is usually a developing state where there is more extensive state involvement in the economy, or perhaps unusual forms of state involvement.134 128 J E Alvarez, “North American Free Trade Agreement’s Chapter Eleven”, (1996–7) 28 University of Miami Inter-American L Rev 303 at 308. 129 Re Canada Labour Code, supra n. 22. 130 See e.g. McGinty, supra n. 29; J T Gathii, “International Law and Eurocentricity”, (1998) 9 EJIL 184; T Lothian, “The Democratized Market Economy in Latin America (and Elsewhere): An Exercise in Institutional Thinking within Law and Political Economy”, (1995) 28 Cornell Int’l LJ 169; O Okafor, “The Global Process of Legitimation and the Legitimacy of Global Governance”, (1997) 14 Arizona Journal of International and Comparative Law Journal 116. Current scholars in this tradition follow distinguished predecessors in anti-colonial approaches to international law; see e.g. M Bedjaoui, Towards A New International Economic Order (Paris, UNESCO, 1979) and M. Sornarajah, “The Myth of International Contract Law”, (1981) 15 Journal of World Trade Law 187. 131 For a similar critique of the lack of “symmetry of direct benefits” among investors of the three NAFTA parties under the provisions of the NAFTA investment chapter, see Alvarez, supra n. 128, at 304–5. 132 Trendtex Trading v. Central Bank of Nigeria, supra n. 65; UK SIA section 3(3)(a). 133 Republic of Argentina v. Weltover, supra n. 14; UK SIA section 3(3)(b). 134 D Tarullo, “Beyond Normalcy in the Regulation of International Trade”, (1987) 100 Harvard L Rev 546 at 569–70 argues that an ideology of the normal role of government in the economy

244 R Wai The recognition of this distribution of parties might influence the contours of Western legislation and court decisions concerning sovereign immunity. The parochialism at issue is signalled in that it is only in the commercial context that Western states have created exceptions in their immunity laws. There is no similar concern for human rights abuses abroad because of the lack of an obvious, parochial pecuniary interest at stake. Indeed, Western commercial interests other than those of the plaintiffs sometimes benefit from public order actions taken by foreign states.135 Second, the claim of neoimperialism is made because of its linkages to the sources and beneficiaries of globalised neoliberalism. The commercial activity exception plays an important part of the construction of an international economy in which neoliberal ideals are advanced and which reproduces Western economic power. In particular, the objectives of strong property rights, enforcement of contractual obligations, and rule of law are advanced as paramount to the economy.136 An important part of that programme is that state actors as much as private actors should be subject to these objectives and associated “disciplines”. This is true at the domestic level and forms the basis, for example, for prescriptions for structural reforms mandated by international financial institutions such as the IMF and World Bank.137 It also characterises the international economy where it is claimed that advancing the basic promotion of contract and property rights will facilitate more extensive commerce and hence broader benefits. Critics argue that these norms are contestable and that the commercial activity exception is portrayed and understood as necessary principally because of the greater economic, political and ideological power of Western capitalist states and their elites.138 Similarly, the agreement of many developing countries to bilateral treaties providing a commercial activity or waiver exception to sovereign immunity would not, from this critical perspective, amount to acceptance heavily influences national legislatures and courts in determinations of “subsidies” in the trade law context. 135 See discussion in text accompanying nn. 35 and 40, supra. 136 Unger, supra n. 117, at 9–10 The neoliberal programme is closely identified in the international context with the “Washington Consensus”, the set of policy values promoted by institutions such as the IMF and World Bank in their prescriptions for countries facing structural adjustment challenges; see e.g. J Williamson, “What Washington Means by Policy Reform” in J Williamson (ed.), Latin American Adjustment: How Much Has Happened? (Washington DC, Institute for International Economics, 1990). 137 See e.g. T Biersteker, “Reducing the Role of the State in the Economy: A Conceptual Exploration of IMF and World Bank Prescriptions”, (1990) 34 International Studies Quarterly 477; Mansell, supra n. 73; K Rittich, “Recharacterising Restructuring: Gender and Distribution in the Legal Structure of Market Reform”, SJD Dissertation, Harvard Law School, 1998. 138 As was recognised by the US Supreme Court in the related context of the Act of State doctrine, the legitimacy of foreign court adjudication for issues of state liability for expropriations, “rests upon the sanguine presupposition that the decisions of the courts of the world’s major capital exporting country and principal exponent of the free enterprise system would be accepted as disinterested expressions of sound legal principle by those adhering to widely different ideologies”. Banco Nacional de Cuba v. Sabbatino, 376 US 398 (1964).

The Commercial Activity Exception 245 of the correctness of the exception, but rather would be an acknowledgment of North/South power differentials and the disciplinary power of international capital markets.139

5 CONCLUSION

The analysis in this chapter has suggested that the strong distinction in immunity doctrine between commercial and non-commercial activities is not justified at the policy level. The chapter also begins the task of describing how the commercial activity exception figures in transnational public interest litigation in foreign courts for state involvement in human rights abuses and torture. Legislatures should consider removing the strong distinctions in immunity statutes between commercial and non-commercial activity, especially in the context of human rights. Similarly, judges, in interpreting immunity legislation and in applying the troubling doctrinal tests for the commercial activity exception in particular cases, should attend to the international policy dimensions— concerns such as comity, consent and cooperation—without assuming that they only operate in one direction or that they are the only policies or values in play. However, this does not mean, necessarily, that the jurisdictional treatment of the “non-commercial” should be assimilated to the “commercial” with the scope of state immunity further reduced as a result. For those acting as human rights advocates, it is important to understand the contours of current assumptions that underlie the doctrinal tests of sovereign immunity. These assumptions are as often unexamined beliefs as they are grounded policy reasons, and may need to be brought to the surface and confronted. At a minimum, these policy assumptions should be recognised as part of the strategic terrain on which decision-making currently occurs. From a strategic or instrumental perspective, both those who resist further declines in sovereignty on the commercial front and those who push for advances against sovereignty on the human rights front should be aware of how to identify and then appeal to these policy points. More particularly, those who wish to try to pursue these two agendas in tandem have a challenging task. Nonetheless, it is arguable that transnational public interest litigants do have some obligation to be cautious about the linkage of their public interest concerns to the project of international legalism, lest they find themselves allied to controversial assumptions such as those of neoliberalism which may lead to the unjustified use of foreign courts for the litigation of public policy issues that might best be left to the domestic processes of different state-communities around the globe.

139

See e.g. Alvarez, supra n. 128, at 312.

9

In Search of a Defence of the Transnational Human Rights Paradigm: May Jus Cogens Norms be Invoked to Create Implied Exceptions in Domestic State Immunity Statutes? WENDY ADAMS 1

1 INTRODUCTION

“[T]he torturer has become like the pirate and the slave trader before him hostis humani generis, an enemy of all mankind.”2 U D G E K A U F M A N ’ S reference to a torturer as “hostis humani generis” in Filártiga v. Peña-Irala expresses the uncontroversial assessment that international human rights law contains a universal prohibition against torture.3 While the precise content of international human rights may be uncertain in many areas, Judge Kaufman had no difficulty in locating within both customary and conventional sources of international law the necessary evidence to conclude that the practice of torture has been universally condemned. Why then is the enforcement of an ostensibly uncontroversial norm of international human rights law proving to be such an onerous task? The difficulty with the prohibition against torture, as with many international human rights, is that the guarantee is institutionally protected for the most part in an aspirational way. In other words, the site of controversy is not the content of the norm, but the nature of the enforcement mechanism. In the absence of an effective enforcement mechanism at the international level, victims of acts of torture are seeking redress in US courts by virtue of the jurisdiction granted to them by the Alien Tort Claims Act (ATCA) and the Torture Victim Protection Act (TVPA). They may also wish to pursue similar actions in the domestic civil courts of other

J

1 2 3

Assistant Professor, Faculty of Law, University of Western Ontario. Filártiga v. Peña-Irala 630 F.2d 876 at 890 (2d Cir. 1980). Ibid.

248 W Adams states. For courts considering such claims, at issue is whether the extraterritorial assertion of jurisdiction over foreign defendants and any concomitant denial of state immunity are legal or political exercises. In this chapter, following a brief introduction to the relevant substantive law, the author outlines the necessary doctrinal process to be followed by domestic courts as they adjudicate claims brought by victims of torture against foreign states or state officials. The central issue to be resolved, at least in jurisdictions with Westminster parliamentary traditions of government, is one of institutional competence. The question is whether domestic courts are able to rely on international sources of law in adjudicating the issues raised by these novel claims. After determining whether such institutional competence exists, and to some extent as part of the process, domestic courts must also consider the interpretive issues presented by the interpenetration of international and domestic law governing torture and civil enforcement. In order to address these issues, two major questions will dominate the analysis in the present chapter. First of all, does international law both permit a domestic court to assert extraterritorial jurisdiction over the subject matter and permit the court to deny a state or state official’s claims of state immunity? Secondly, assuming that international law does grant the necessary authority to domestic courts, how are courts to reconcile such an interpretation of international law with legislative supremacy? The problem arises from that fact that, although international law would permit the assumption of jurisdiction, the broad scope of domestic state immunity statutes are likely to be resistant to such an interpretation. In setting out the doctrinal process to be undertaken by domestic courts when presented with these types of civil actions, it will soon become apparent that resolution of the institutional and interpretive issues cannot be accomplished without factoring considerations of legitimacy into the exercise. Long-standing principles of international law create a strong presumption that a domestic court may not generally sit in judgment on the validity of acts undertaken in the territory of foreign states, nor may it generally adjudicate cases in which a foreign state is a party without the consent of that state. Where the case in question not only impleads a foreign state, but impleads it with respect to its conduct in its own territory, the sovereignty-protecting principles of international law are engaged to their greatest extent, such that the justification for assuming jurisdiction must be particularly clear and compelling under international law. Without sufficient justification pursuant to international law, an assertion of extraterritorial jurisdiction in these circumstances would be more in character with political than judicial authority. At the domestic level, the extent to which international law permits domestic courts to enforce individual accountability for violations of international norms is dependent not only on the recognition within international law of individual accountability, but also on the degree of interpenetration between international law and each domestic legal system’s constitution, legislation and common law.

In Search of a Defence 249 In states with Westminster parliamentary systems of government, such as the United Kingdom, Canada and other Commonwealth states, a formal separation of powers exists between the legislative and executive branches, which are partially integrated within Parliament through the Cabinet, and the judicial branch. Responsibility for developing the law on foreign policy matters is accorded to the non-judicial branches. The courts do, of course, have the role of ensuring that international treaty obligations, as entered into by the executive, are reflected in legislation before they will be given the direct force of domestic law. Even in these circumstances, formally speaking, it is statutory rules rather than treaty provisions that will be applied by the courts. Yet, there are two complications to the claim that it is not the courts’ role to apply international law in the absence of prior authorisation by the legislature. First, there is a recognised principle of interpretation, applicable particularly to statutes, according to which courts may (or must) interpret the statute so as to avoid placing the state in breach of international law where there is a reasonable interpretation consistent with compliance with the state’s international obligations.4 Second, nontreaty international law, or at least customary international law, is generally thought to be automatically included within the scope of the law of the land. As such, under normal rule of law premises, it is within the judiciary’s field of responsibility. In light of this doctrinal state of affairs, the legitimacy issue therefore requires a careful analysis of the point at which a Commonwealth domestic court’s reconciliation of international and domestic law crosses over from legitimate judicial interpretation to judicial activism in a manner which illegitimately intrudes into Parliament’s authority over foreign affairs.

2 A BRIEF REVIEW OF SUBSTANTIVE LAW

Domestic enforcement of a violation of international law, such as the use of torture, requires a reconciliation of the competing imperatives of state sovereignty and the universal nature of human rights norms. On one side is international law’s foundational premise, the sovereign equality of states and the concomitant role of state consent in the establishment of international norms. Whatever the degree to which a state can be bound to general international law without specific ex ante consent, it is not required to submit to the judgment of any other state as to whether it has violated an international norm. State immunity from the exercise of jurisdiction by another state’s judicial system arose as one reflection of the legal status of sovereign states as juridical equals. All states being equal, one state cannot, through its domestic courts, exercise authority over another.5 4 See the discussion of this principle infra Section 3, especially in relation to the recent Supreme Court of Canada case of Baker v. Canada (Minister of Citizenship and Immigration). 5 See Schooner-Exchange v. McFadden, 11 U.S. 116 (1812): “The immunity of a state from the jurisdiction of the courts of another state is an undisputed principle of customary international law.”

250 W Adams In tension with the foundational premise of state sovereignty is the development of international human rights law in parallel with international criminal law, particularly in the aftermath of the Nuremberg and Tokyo trials. While this development appears to be at odds with traditional concepts of sovereign equality, the application of international law by the Allies (or at least, international law as interpreted by the Allied judges) to punish Axis-power war criminals altered the reigning statist and positivist conceptions of international law. The result was a revival of a natural law strain in international law through a commitment to protect individual rights, a commitment which had diminished in importance during international law’s preoccupation with sovereignty. Whatever interpretive agreement exists over the content and limits of the relevant norms, human rights are now considered truly universal in the sense that they accrue to individuals by reason of their humanity, independently of the internal political order of the state in which these individuals reside. This principle was confirmed largely in aspirational terms in the United Nations Charter6 and the Universal Declaration of Human Rights,7 subsequently codified in more than thirty international agreements,8 and most recently strongly reaffirmed in the Vienna Declaration on Human Rights.9 The above-described trajectory of international human rights law, however, has been primarily normative in nature. While human rights norms have been accumulating at a rapid pace since the end of the Second World War, the development of institutional mechanisms of enforcement has not tracked these advances. Whether out of deliberate recalcitrance or legitimate, sovereigntybased concern regarding the imposition of Western-centric conceptions of human rights, the international community of states has not demonstrated the necessary political will to establish effective international enforcement procedures.10 The result is an enforcement gap which leaves many individuals in the untenable position of possessing rights without remedies. Some victims of human rights abuses have sought to address this enforcement gap by bringing civil actions in domestic courts, notably, as indicated earlier, in the United States. This turning toward domestic enforcement of international human rights norms is a development that has not been well-received by all members of the international community. Keeping in mind the above discussion of the tensions between universal human rights and state sovereignty, and between 6

Charter of the United Nations, 892 U.N.T.S. 119. Universal Declaration of Human Rights, adopted 10 Dec. 1948, G.A. Res. 217A (III), U.N. GAOR, 3d Sess. Resolutions, part 1, at 71, U.N. Doc. A/810 (1948), reprinted in (1949) 43 Am J Int’l L 127 (Supp.). 8 See among others: International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976) [hereinafter “ICCPR”]. 9 Vienna Declaration on Human Rights, 32 ILM 1661 (1993). 10 See, for example, M wa Mutua, “The Ideology of Human Rights”, (1996) 36 Va J Int’l L 589. Professor wa Mutua argues that international human rights discourse is premised on Western liberal democracy and represents an attempt to develop the liberal political tradition at the international level. 7

In Search of a Defence 251 normative articulation and ineffective enforcement, it can be observed that there is no dispute that a state which inflicts torture upon its citizens or residents for any reason is in violation of general international law and not only in violation of obligations in those treaties to which the state may be party. The process by which violator states are to be brought to justice, however, is much less clearly defined. In order to constitute a violation of international law, an act of torture must involve state conduct in some form: either by way of a perpetrator acting in an official capacity, or by way of a non-state perpetrator being permitted to torture by the consent or acquiescence of state officials.11 Given this state conduct requirement, foreign domestic enforcement of the prohibition against torture typically involves questions not only of jurisdiction, but of state immunity as well. Accordingly, the legitimacy of enforcement by foreign courts against a person or persons having allegedly committed acts of torture is dependent upon an international legal rule that not only justifies a state’s extraterritorial assertion of jurisdiction, but also rejects the availability of state immunity for defendants—whether they are natural persons who acted within the context of a state apparatus, or whether the state itself is being sued for the conduct of its human agents. The Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (CAT) addresses the issue of enforcement by imposing obligations on states to see to it that human actors are punished as criminals.12 In addition to prohibiting torture within their domestic criminal law, states are compelled to cooperate in international criminal enforcement efforts pursuant to a “prosecute or extradite” mechanism. States having within their jurisdiction a foreign state official who is accused of having participated in acts of torture must respond to the allegations, either by prosecuting the person within their own criminal justice system or by extraditing the person to a foreign state that has asserted criminal jurisdiction of its own. Consequently, the CAT establishes not only a form of universal jurisdiction (understood as a state’s power to subject a person to its law and processes of law enforcement) 11 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027 [hereinafter “CAT”], article 1(1), provides in part as follows:

“1(1) For the purposes of this Convention, the term ‘torture’ means any act by which severe pain or suffering . . . is intentionally inflicted on a person . . . when such pain and suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity.” Note also that this is only the definition found in the CAT. The CAT expressly says that its definition may not be invoked as a limit to broader definitions elsewhere: see article 1(2). It appears that the Human Rights Committee has interpreted torture under the ICCPR as including “private” sector torture in circumstances wider than may be caught by the CAT’s definition: see E M Hyland, “International Human Rights Law and the Tort of Torture: What Possibility for Canada?”, chapter 15 in this volume. 12 CAT, supra n. 11.

252 W Adams but goes further by creating an obligation to exercise jurisdiction so that a torturer present in its territory may not go free. The difficulty with the assertion of universal criminal jurisdiction as set out in this regime is that the CAT does not expressly address the manner in which such jurisdiction is to be reconciled with the competing customary international law norm of state immunity, understood as the ability of states and, in many cases, state officials, to resist the jurisdiction of foreign courts. The lack of international consensus on this issue was evident in the recent controversy concerning Spanish judge Baltasar Garzon’s request to the United Kingdom to extradite General August Pinochet, ex-dictator of Chile, to face charges of murder, torture and hostage taking.13 The arguments presented by both sides in the dispute illustrate the complex and uncertain nature not only of competing imperatives within international law, but also of the interpenetration of international and domestic law within national legal systems. Criminal law, whether applied at the international or national level, is a form of public law in which the state retains discretion over the decision to prosecute. Even when a criminal law system allows private prosecutions, a state official, whether an Attorney General or, as in civil law systems, an investigating magistrate like Judge Garzon, must inevitably decide whether the case should proceed. At this point, the state usually assumes carriage of the prosecution. For victims of torture desiring to take and retain the initiative themselves, the available alternative is to bring a civil claim grounded in tort against a former oppressor. Compensation is sought in the form of monetary damages as a remedy, rather than incarceration of the torturer. The complexities of criminal jurisdiction pale in comparison to the unresolved issues these claims present. Here, it shall be assumed that a civil action against a state or state officials in a domestic forum other than the jurisdiction where the alleged violation occurred constitutes a prima-facie assertion of unlawful extraterritorial jurisdiction. As a result, domestic courts may well refer to the CAT for the necessary grant of jurisdiction in these circumstances. However, the reference to civil jurisdiction in article 14 of the CAT is wholly ambiguous in terms of whether the article requires states to adjudicate domestic civil claims for acts of torture occurring abroad. This is in contrast to the more explicit provisions regarding universal criminal jurisdiction. Accordingly, civil actions as compared to criminal prosecutions present claimants with an additional obstacle, in that legal uncertainty exists not only in terms of the extent of immunity available, but also with respect to the threshold question of whether a domestic court can ever 13 For the most recent decision of the House of Lords concerning the extradition of General Pinochet, see: Regina v. Bartle and the Commissioner of Police for the Metropolis and Others Ex Parte Pinochet, 38 ILM 581 (1999). On 15 April 1999, UK Home Secretary Jack Straw decided to allow the extradition process to continue. On 11 January 2000, Secretary Straw decided that the process should be brought to a halt on the basis of the information he had received in a medical report from four doctors on the deteriorating state of Pinochet’s health: see “Comunicado del Ministerio británico del Interior”, (12 January 2000) El Pais [on-line version at http://www.elpais. com].

In Search of a Defence 253 properly be seized of jurisdiction to entertain a civil claim of this nature. At the least, it is legally uncertain whether the CAT can be invoked in support of such a taking of jurisdiction. Given the limited doctrinal development to date of the CAT by the Committee against Torture (the Committee), domestic courts can expect little substantive guidance from within the international system either as to the scope of civil jurisdiction pursuant to the CAT or as to the manner in which to reconcile a potential grant of jurisdiction with the relevant state immunity. The focus of this chapter, then, is to seek to untangle the doctrinal snarl produced by the interpenetration of international and domestic law on this issue, as well as to provide domestic courts with a coherent blueprint for adjudication. One potentially viable route of reconciliation may be to focus on the nature of the universal prohibition against torture as jus cogens. Article 53 of the Vienna Convention on the Law of Treaties (VCLT) defines jus cogens as a norm “accepted and recognised by the international community of states as a whole” and “from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character”.14 A rationale for reconciliation that relies on the jus cogens nature of the prohibition against torture seeks to justify both the assertion of jurisdiction and the denial of state immunity on the basis that enforcement of a violation of jus cogens does not conflict with the concept of state sovereignty within the international legal order. Reduced to its simplest terms, the argument is that as the exercise of sovereignty is limited by jus cogens, a violation of jus cogens such as the prohibition against torture, is not a sovereign act and accordingly does not attract state immunity. Given that an entitlement to immunity no longer exists, one state’s extraterritorial assertion of jurisdiction does not impinge upon another state’s sovereignty in a manner prohibited by law. However, even were such arguments to be accepted without controversy as elements of international law, it remains no simple matter for courts to give effect to these arguments when faced with textually-recalcitrant statutory regimes.

3 THE PROCESS OF DOMESTIC CIVIL ENFORCEMENT

In seeking to bring a civil claim of torture as a tort in a domestic forum, potential plaintiffs are attempting to address what may be viewed as an enforcement gap at the international level. At issue is whether the necessary authority over enforcement may be transferred from the international legal order to domestic forums. The most significant interpretive issues are whether international law specifically permits one state to assume adjudicative jurisdiction over the actions of another state or state officials in circumstances where allegations of torture are involved, and whether international law also precludes resort by the 14

Vienna Convention on the Law of Treaties, 1155 UNTS 331 [hereinafter “VCLT”].

254 W Adams foreign state or state officials to claims of state immunity. The inquiry becomes inseparable from, and to a certain degree recursive with, the question of whether, in the absence of legislative authorization, civil courts are able to access international law to the extent necessary in order to identify and incorporate such a mechanism directly within a domestic legal system.

Access by Domestic Courts to International Law The first difficulty faced by domestic courts in adjudicating claims against a foreign state or state officials in a civil action is that the potential defendant will no doubt claim immunity from jurisdiction pursuant to general international law as implemented, for the most part, within domestic legal systems in the form of state immunity legislation. The relevant legislation is generally quite broad in terms of coverage, with limitations to state immunity typically confined to certain enumerated and arguably exhaustive exceptions. The grant of immunity in Canadian legislation, set out in section 3(1) of the State Immunity Act, is fairly representative of the approach to immunity found in other states’ statutes in terms of establishing immunity as the default rule: “Except as provided by this Act, a foreign state is immune from the jurisdiction of any court in Canada.”15 The permitted exceptions generally include commercial activity, wherein the actions of the foreign state relate primarily to market transactions rather than government functions, and tortious conduct occurring within the territory of the enacting state.16 Domestic legislation appears to state the precise contours of state immunity in the clearest of terms. For purposes of concrete discussion, Canadian legislation will serve as a useful example. Given the clarity of the legislation with regard to state immunity, at issue is whether a court’s access to international law for assistance in interpreting the text is a legitimate exercise of judicial authority. Recall that the question of institutional legitimacy is most relevant in circumstances where the reception of conventional international law into the domestic legal regime is in accordance with a transformationist rather than adoptionist framework. Pursuant to a transformationist theory of reception, a domestic court’s ability to apply conventional international law is limited by the requirement that an international convention be of no direct force or effect within the national legal system unless and until its terms are transformed into domestic law in the form of legislation. In contrast, an adoptionist approach 15

Canada’s State Immunity Act 1985, c. S-18, s. 3(1). See, for example, Canada’s State Immunity Act, supra n. 15, section 5, setting out an exception for commercial activity: “A foreign state is not immune from the jurisdiction of a court in any proceedings that relate to any commercial activity of the foreign state”; and section 6, providing an exception for torts committed within Canada: “A foreign state is not immune from the jurisdiction of a court in any proceedings that relate to (a) any death or personal injury . . . that occurs in Canada.” 16

In Search of a Defence 255 would receive conventional international law directly into the national legal system without the necessity of involving the legislative branch. Assessment of the advantages and disadvantages of each approach is beyond the scope of this analysis, but suffice it to say that the benefit of an adoptionist stance is the seamless integration between international law and its application within the national legal system, while the advantage of the transformationist approach is the state’s ability to implement international law selectively, in a manner consistent with domestic social welfare objectives. Given Canada’s transformationist orientation, the issue of reception of treaty obligations arises in that the lack of provision for immunity in the CAT (or, more strongly, the implied prohibition of immunity under the CAT) may be one reason invoked for interpreting the State Immunity Act narrowly so as to exclude violations of jus cogens norms such as torture from the statute’s application. Yet, assuming for the moment the correctness of this interpretation of the CAT, a major problem is created by Parliament’s failure to amend the Act expressly to reflect this treaty obligation. Here it may also be relevant to know that the State Immunity Act is relatively unusual in that it legislates an international legal issue that had been dealt with previously by Canadian courts qua customary international law. It did not represent statutory implementation of any international treaty norms, and it remains the case that there is no international multilateral treaty governing the general question of sovereign immunity. In states with a Westminster parliamentary system of government, such as Canada, the distinction between the adoptionist and transformationist approaches assumes critical importance in terms of the challenge of complying with international obligations in a manner that does not circumvent the domestic democratic process. In Canada, for example, the negotiation and ratification of international conventions is solely the responsibility of the federal executive branch. This is in contrast to processes in states which involve both the executive and legislative branches. This is true of the United States, which requires that international conventions be ratified by two-thirds of the members of the Senate. Given, in the simplest of terms, that international obligations take effect upon ratification by the signatory state and the entry into force of an international convention, the most significant distinction in terms of the manner in which ratification occurs is whether the executive branch has the ability to incur international obligations without involving the legislative branch. Canada’s transformationist approach to the incorporation of conventional international law is intended to prevent the possibility that the federal executive branch may alter domestic law by ratifying international agreements without consultation or assent by federal and provincial legislative branches. Until recently, it was fairly well-settled law that an international convention, notwithstanding ratification by the executive branch and entry into force, is of no force or effect within Canada’s domestic legal system unless and until such time as its terms are incorporated pursuant to implementing legislation enacted by the

256 W Adams legislative branch.17 Canadian courts have generally applied a transformationist approach, maintaining compliance with international law and achieving consistency between international obligations and domestic legislation by use of the interpretive presumption that the legislature does not intend to act in violation of international law without an express indication to the contrary. Under the traditional transformationist approach, however, it would be difficult for a Canadian court to proceed from this interpretive presumption to the conclusion that, in the absence of any legislative implementation, domestic courts have the ability to access and interpret the provisions of the CAT as the direct basis for adjudicating civil claims initiated by victims of torture. However, the recent decision by the Supreme Court of Canada in Baker v. Canada (Minister of Citizenship and Immigration) may indicate that the Court is moving away from a categorical, all-or-nothing approach to the incorporation of conventional international law within the domestic legal system.18 In Baker, the Court was called upon to address the question of whether, in assessing a parent’s application for landed immigrant status pursuant to humanitarian and compassionate considerations provided for in the relevant legislation, immigration authorities were required to treat the best interests of the child as a primary consideration. The appellant had argued that immigration authorities were obligated to accord primacy to her children’s interests pursuant to the Convention on the Rights of the Child (the Convention), a multilateral convention signed and ratified by Canada, notwithstanding that the terms of the Convention had not specifically been subject to implementing legislation.19 In writing the decision for the Court, Madame Justice L’Heureux-Dubé expressed agreement with the long-standing rule that the terms of an unimplemented convention are of no direct force or effect within Canadian domestic law. She concluded, however, that the Court could refer to the underlying values reflected in the Convention as a feature of the Court’s contextual approach to statutory interpretation. In other words, while the appellant could not claim that the immigration authorities were formally bound to the terms of the Convention as a matter of domestic law, the values represented in international 17 Capital Cities Communications Inc. v. Canadian Radio-Television Commission [1978] 2 S.C.R. 141. 18 Baker v. Canada (Minister of Citizenship and Immigration), Supreme Court of Canada, File No.: 25823, 9 July 1999. The appellant, a woman with Canadian-born dependent children, faced deportation. She was refused humanitarian and compassionate consideration under section 114(2) of the Immigration Act, based on the requirement that an application for permanent residence be made from outside Canada. One relevant issue was the degree of Ministerial discretion under section 114(2), given the requirements of the International Convention on the Rights of the Child; notwithstanding that the Convention had not been the subject of implementing legislation enacted by Parliament. Madame Justice L’Heureux-Dubé, writing for the majority, found that a reasonable exercise of Ministerial discretion requires attention to the needs of children, as care for their interests is both indicated within international instruments such as the Convention and recognized as a value central to Canadian society. 19 Convention on the Rights of the Child, G.A. Res. 44/25, 44 U.N. GAOR, Supp. No. 49, U.N. Doc. A/44/49, (1989) at 166 [hereinafter “Convention”].

In Search of a Defence 257 human rights law were a legitimate reference point for the purposes of determining the content and requirements of the legislative reference to humanitarian and compassionate grounds. As a matter of interpretive presumption, legislative intent as to the scope of humanitarian and compassionate grounds may be interpreted as consistent with the values and principles contained within international law, of which the Convention was a significant indication. In the author’s opinion, however, the dissenting opinion of Iacobucci J (Cory J concurring) was correct in its assessment that the majority permitted the appellant to accomplish indirectly what she could not do directly in terms of compelling immigration authorities to observe Canada’s international obligations within domestic law, notwithstanding that the legislative branch had not necessarily determined that these obligations should bind either individuals or governmental authorities directly within Canadian domestic law. An indication of the instrumental nature of the majority reasoning in Baker is illustrated, with respect, by the fact that the extent to which a claimant may achieve this result depends on the level of congruence between the underlying values as articulated by the Court and the text of the otherwise-inaccessible provision contained in an international convention. In the circumstances in Baker, the appellant was seeking, inter alia, to bind the exercise of Ministerial discretion in accordance with the requirement set out in the Convention that states treat the “best interests of the child” as a “primary consideration” in “all actions concerning children”.20 The majority reasoning identified the underlying values and principles of the Convention as “being attentive to the rights and best interests of children when decisions are made that relate to and affect their future”.21 Although “being attentive” does not equate to according “primary consideration” to the best interests of the child, the degree of congruence between the underlying values identified by the Court and the relevant provision was sufficient for the appellant to obtain the assistance of international norms in an indirect fashion. In accordance with the majority reasoning in Baker, it may be possible for a Canadian court adjudicating allegations of torture in a civil claim to refer to Canada’s obligations pursuant to the CAT, notwithstanding the lack of an enabling statute in which the legislature has defined the nature and extent of these obligations. Adopting the Court’s contextual approach to statutory interpretation, a plaintiff could argue that the scope of permitted exceptions to state immunity contained in the State Immunity Act cannot be assessed in isolation from important Canadian values and principles that also operate at the international level, including those evidenced by ratification of international conventions. Given that an implied exception is a perfectly legitimate technique of statutory interpretation within the Canadian legal system, the overall objective being the determination of legislative intent, the search for exceptions that 20 Convention, supra n. 19, art. 3(1) provides as follows: “In all actions concerning children, whether undertaken by public or private social welfare institutions, courts of law, administrative authorities or legislative bodies, the best interests of the child shall be a primary consideration.” 21 Baker, supra n. 18 at para. 71, per Madame Justice L’Heureux-Dubé.

258 W Adams accord with legislative intent is an open-ended process. The domestic rules of statutory interpretation as extended by the ruling in Baker could be sufficient to permit a court considering a civil action against a foreign state or state official for acts of torture to refer to international obligations in its search for values indirectly embedded by the legislature in the State Immunity Act. A crisis of judicial authority would arise, however, given the distinction between reliance on provisions from the Convention in Baker and the manner in which civil actions initiated to obtain redress for acts of torture would need to rely on the CAT in order to argue for non-recognition of immunity. Even if a litigant were able to access the CAT directly, for example if Canada had implemented the terms of the CAT verbatim into a statute, no express term of the CAT authorises an extraterritorial assertion of civil jurisdiction with a corresponding exclusion of state immunity. By engaging in an interpretive process whereby the necessary terms would first be implied into the CAT and then drawn from “interpretively” to create an implied exception to immunity in the State Immunity Act, a Canadian court would be moving even farther from the strict transformationist approach than did the Supreme Court of Canada in Baker. In the Baker scenario, the express nature of the terms of the Convention made the underlying values more readily accessible. At issue is whether this potential development, albeit progressive and laudable, is nonetheless disruptive in terms of its effect on Canada’s internal democratic process. In the author’s opinion, no compelling justification exists that would permit Canadian courts to refer to the terms of the CAT to the extent and in the manner discussed above in the absence of implementing legislation—apart from the special case where constitutional concerns are in issue, which will be addressed later in this chapter. With respect, the difficulty with the majority’s approach in Baker is that the reference to the interpretive presumption as an indication of legislative intent is not accompanied by any statement by the majority that the presumption is not sufficient, nor has it ever been, to displace the transformationist approach towards conventional international law within the Canadian domestic legal system. If the application of the interpretive presumption is extended beyond the situation where the legislature has potentially enacted conflicting legislation (as in Baker wherein the interpretation of the scope of a legislated discretion was at issue), to include circumstances in which the legislature has not provided any indication of intent in the form of enacting any relevant legislative provision whatsoever, the potential exists for courts to fill this legislative void by identifying international obligations of its own accord and enforcing these obligations directly. To do so as a basis for upholding international human rights treaty obligations against Canada itself as a state would clearly be defensible, but to engage in such an outflanking of the treaty transformation rule in order to enforce obligations against individuals or against foreign states is not as easily justified. By necessity, Canada’s international obligations must be subjected to an internal democratic process. It is difficult to reconcile this requirement with a

In Search of a Defence 259 court’s ability to refer, however indirectly, to what are named in the interpretive principles of article 31 of the VCLT as the “objects and purposes” of international conventions. The only manner in which this could be accomplished as a matter of law, as indicated by Iacobucci J in Baker, is in situations where rights guaranteed within a state’s constitutional order are at stake. In these circumstances, the democratic consensus leading to the adoption of a written constitution has ex ante limited legislative supremacy to the extent that the legislature is not permitted to abrogate or derogate from certain constitutional guarantees. In countries such as Canada, where the baseline of constitutional protection for human rights is presumed to be at least equal to, if not greater than, the level of protection available at international law, reference to international norms is a legitimate judicial process in terms of defining the content and scope of domestic constitutional guarantees.22 However, even this approach requires judicial caution, as constitutional guarantees are of limited assistance if international human rights law is itself silent as to the manner and location of enforcement. At this point it must be acknowledged that a certain paradox exists in advocating an implicit reception of international human rights contained in unimplemented international conventions through the process of Charter interpretation, while nonetheless rejecting such incorporation through the process of statutory interpretation as an illegitimate exercise of political rather than legal authority. While the latter alternative raises legitimacy questions regarding the proper relationship between courts and Parliament within the context of legislative supremacy, any imbalance of power between these two branches is at most a temporary crisis. Parliament is free to respond to what it views as an exercise in excessive judicial enthusiasm by enacting explicit legislation to the contrary. The process of Charter interpretation, in contrast, results in binding norms that cannot be superseded by legislative action. Reconciliation is possible, however, when one considers more fully the implications of Iacobucci J’s reference in Baker to the “interpretive presumption” established by the Court’s decision in Slaight Communications Inc. v. Davidson.23 In that case, the Court held that administrative discretion involving Charter rights must be exercised in accordance with similar international human rights norms. First, it is important to set the stage for institutional competence by reiterating that the Charter is first and foremost the product of a deliberate democratic process. As the Court has repeatedly noted, the assignment of the power of constitutional review to the judiciary was the deliberate

22 Slaight Communications Inc. v. Davidson [1989] 1 S.C.R. 1038. The Court in this case ruled that the scope of legislation authorising administrative discretion cannot be interpreted in a manner that offends Charter guarantees, the content of such guarantees being generally presumed to provide at least as much protection as similar guarantees contained in international conventions ratified by Canada. 23 Slaight Communications, supra n. 22.

260 W Adams choice of the federal and provincial legislatures. 24 Accordingly, courts possess the necessary authority to embark on a process of “covenant constitutionalism”, whereby Charter guarantees may be interpreted in a manner consistent with international human rights norms. This authority was recognised not only by Iacobucci J in his dissent in Baker on this point, but also by Madame Justice L’Heureux-Dubé in her reasons for the majority. She referred to the “critical influence” of international human rights norms on the interpretation of the scope of Charter rights and guarantees.25 Second, the Court’s ability to design its own interpretive methodology can be constrained by a requirement that the Court’s reliance on international human rights law be limited to circumstances in which an international human rights obligation may clearly and incontrovertibly be said to exist. Provided that in determining domestic law the court does not proactively incorporate emerging international norms that are as yet ahead of the curve of international consensus, the legitimacy of Charter interpretation in this context remains unassailable.

Interpreting the Obligations of States pursuant to the Convention against Torture Before addressing whether a domestic court is able to justify reference to international law in interpreting the scope of domestic immunity legislation, the first task faced by the court is one of definition. It is necessary to set out the nature and extent of a state’s obligations and authority with regard to the recognition or creation of a civil cause of action for foreign acts of torture against states or state officials. Although questions pertaining to subject-matter jurisdiction and the appropriate cause of action are conceptually distinct and logically prior to any judicial consideration of state immunity, the issues are to a certain degree mutually dependent. The relevant inquiry is whether subject-matter jurisdiction 24 Justice Lamer (as he was then) commented on the legitimacy of constitutional review in Reference Re s. 94(2) of the Motor Vehicle Act (B.C.) [1985] 2 S.C.R. 486 at 497, as follows:

“This is an argument which was heard countless times prior to the entrenchment of the Charter but which has in truth, for better or for worse, been settled by the very coming into force of the Constitution Act, 1982. It ought not to be forgotten that the historic decision to entrench the Charter in our Constitution was taken not by the courts but by the elected representatives of the people of Canada. It was those representatives who extended the scope of constitutional adjudication and entrusted the courts with this new and onerous responsibility. Adjudication under the Charter must be approached free of any lingering doubts as to its legitimacy.” See also the discussion by Iacobucci J. in Vriend v. Alberta [1998] 1 S.C.R. 493 at paras 129–43. 25 Baker, supra n. 18 at para. 70. See also Madame Justice L’Heureux-Dubé’s reference in R v. Ewanchuk, [1999] 1 S.C.R. 330 at para. 73 to the Charter as “the primary vehicle through which international human rights achieve a domestic effect”. The notion of the Charter as a “primary vehicle” of reception into domestic law had been advanced before the Court in the factum of the intervenor Charter Committee on Poverty Issues in the Baker case. The expression “covenant constitutionalism” is borrowed from Craig Scott, “Covenant Constitutionalism and the Canada Assistance Plan”, (1995) 6 Constitutional Forum 79.

In Search of a Defence 261 and a related cause of action can be applied in concrete terms against a sovereign state. The CAT is the most promising source of international law in which to locate a transfer of enforcement authority from international to national jurisdiction, expressing as it does the most recent and wide-ranging consensus concerning the right of all persons to be free from torture. The interpretive issue facing domestic courts is whether the CAT permits such courts to assert extraterritorial civil jurisdiction under relevant circumstances, and, assuming sufficient jurisdiction is conferred, whether a domestic court may decline to recognise state immunity in relation to the alleged perpetrators. To an important extent, the interpretive process is recursive, in that a domestic court is interpreting of its own initiative an international document for guidance as to whether the very same document can be used to interpret the scope of the jurisdiction and authority of the domestic court. In the absence of a truly supranational adjudicative structure, such a paradox of auto-interpretation is an instance of institutional failure that cannot be avoided. The first point to note concerning the text of the CAT is the broad scope of coverage in terms of state actors. Under article 1, the terms of the CAT apply to any person who falls within the definition of “a public official or other person acting in an official capacity”.26 It does not appear necessary, therefore, for domestic courts to address as a matter of international law the issue of whether, in addition to bringing suit against a state directly, potential plaintiffs may also initiate an action against individuals acting in their capacity as state officials. This matter appears ripe for invocation by a domestic court as a clear, uncontroversial norm. The second interpretive issue is the necessary authority for domestic enforcement, initially in terms of the assertion of extraterritorial jurisdiction.27 Before addressing extraterritorial civil law jurisdiction, it should be noted that, pursuant to article 5(2) of the CAT, universal criminal law jurisdiction is conferred upon domestic courts for violations of the prohibition against torture as set out in the terms of the CAT. This jurisdiction is mandatory rather than permissive, given that article 7(1) imposes an affirmative obligation on states to either prosecute or extradite purported violators located within their territorial jurisdiction. Unfortunately, however, the CAT is silent as to the interaction between the grant of universal criminal jurisdiction and international legal rules regarding state immunity. Therefore, the first ambiguity that must be resolved is the manner in which universal criminal jurisdiction and state immunity are to be reconciled. 26 The key here is to note the broad coverage of state-related individuals. It is not the purpose of this chapter to offer an opinion on the extent to which torture can be committed by non-state actors who have the “consent or acquiescence” of the state, or to address whether such persons should be able to invoke state immunity on the basis of such a link of approbation between them and the state. 27 The issue of international law limiting the availability of state immunity will be dealt with in the subsection following.

262 W Adams At the international level, the Committee against Torture (the Committee), the international body constituted pursuant to the terms of the CAT to monitor states’ compliance, has recently expressed its view that the CAT does not permit a signatory state to recognise state immunity.28 The Committee’s interpretation seems entirely logical, as the CAT would lack internal coherence otherwise. Given the state conduct requirement contained in article 1(1), whereby torture must involve at least some conduct in the form of “consent or acquiescence” from “a public official or other person acting in an official capacity”, a grant of immunity to these very same actors would wholly abrogate the object and purpose of the CAT. While the Committee is not possessed of the necessary authority to issue binding interpretations of state obligations under the terms of the CAT, the Committee’s assessment nonetheless carries much persuasive authority. The interpretation is persuasive not least because this interpretation is entirely consistent with article 32 of the VCLT, which permits recourse to supplementary means of interpretation in circumstances where general rules of interpretation lead to a result that is “manifestly absurd or unreasonable”. By implication, the terms of article 32 appear to indicate that an interpretive result which leads to internal incoherence cannot be justified under international law. It might be noted, however, that article 32 perhaps need not even be invoked if one reads article 31(2) of the VCLT broadly enough. Article 31(2) indicates that when interpreting a treaty provision in its “context” per article 31(1), “context” includes the manner in which a suggested interpretation does or does not fit within the entire textual scheme of the treaty. Accepting, for the purposes of argument, that the Committee’s interpretation reflects the current consensus among states concerning an exception to state immunity in circumstances where the criminal sanctions contained in the CAT are implicated, it is not entirely self-evident that this reconciliation between universal jurisdiction and state immunity exists for an assertion of civil jurisdiction on the same terms. The only reference in the CAT to civil jurisdiction is contained in article 14, which requires each state to “ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation”.29 The provision does not indicate on its face whether the scope of civil jurisdiction is intended to be confined to territorial jurisdiction or whether article 14 instead tracks the universal criminal jurisdiction conferred elsewhere in the CAT. If the latter is the case, states are also required to make compensation available in the form of actions against foreign states or at least foreign state officials. However, for a domestic court to rely 28 See P Burns and S McBurney, “Impunity and the United Nations Convention against Torture: A Shadow Play Without An Ending?”, chapter 10 of this volume. See also the Statement of the Committee against Torture regarding immunity and General Pinochet cited therein, and also United Nations Press Release, HR/98/90, 25 November 1998, detailing a statement made by High Commissioner for Human Rights Mary Robinson regarding the extradition of General Pinochet. 29 A more detailed discussion of the scope of this provision can be found in A Byrnes, “Civil Remedies for Torture Committed Abroad: An Obligation under the Convention against Torture?”, chapter 20 in this volume.

In Search of a Defence 263 only on the scant reference to civil remedies contained in article 14 for the purposes of establishing a common law cause of action would equate to judicial legislation no less, in the author’s opinion, than the interpretive presumption set out in Baker, which permits courts to refer to the underlying values and principles of an unimplemented international convention for the purposes of statutory interpretation. Without a sufficient justificatory rationale, judicial interpretation of article 14 in this manner would amount to a political rather than legal exercise, and could not be countenanced within the context of legislative supremacy. Further, assuming arguendo that article 14 implicitly confers universal civil jurisdiction over violations of the terms of the CAT to the same extent as the express grant of universal criminal jurisdiction pursuant to article 5(2), the potential conflict between extraterritorial jurisdiction and the use of state immunity to resist this jurisdiction must still be resolved as was the case with a similar ambiguity pertaining to criminal jurisdiction. Here, it is important to concede a point that favours the proponents of a broader interpretation. It is no answer simply to invoke the CAT’s silence as to the limitations placed on state immunity by the exercise of civil jurisdiction as evidence, much less proof, that article 14 is intended to be merely territorial in scope of application. If this argument were accepted, it should apply with equal force to the CAT’s silence concerning the interaction of state immunity and universal criminal jurisdiction. Yet, quite rightly, the absence of any provisions setting out the manner in which these competing concepts are to be reconciled has not precluded the Committee from concluding, as discussed above, that the CAT does not permit a state party to recognise state immunity as a basis for a person to resist extraterritorial criminal jurisdiction. Having dispensed with any pre-emptive arguments relying on the CAT’s silence regarding the relationship between extraterritorial civil jurisdiction and the role of state immunity, it is still necessary to justify an expansive interpretation of article 14. In determining the scope of this provision, domestic courts must follow the interpretive process established in the VCLT. The general rule of interpretation set out in article 31 indicates that a convention is first and foremost to be interpreted “in good faith” and that terms are to be given their “ordinary meaning . . . in their context and in light of the object and purpose. The object and purpose of the CAT is expressed in straightforward terms in the preamble, among other places, as the desire to “make more effective the struggle against torture”. That the right to be free from torture originates from within a natural law tradition is indicated by the recognition within the preamble that the rights contained in the CAT derive from “the inherent dignity of the human person”. The mere text of article 14 itself, however, does not indicate how the signatories have reconciled international human rights imperatives as derived from natural law with positivistic accounts of the international legal order, in terms of the foundational principles of state sovereignty and consent. The failure to expressly address extraterritorial civil jurisdiction could suggest a

264 W Adams reconciliation of these imperatives in favour of a territorial limitation to the scope of article 14. This is especially the case in light of the express grant of extraterritorial criminal jurisdiction conferred upon states in article 5, the lack of a corresponding express grant of civil jurisdiction suggesting an intentional exclusion fatal to the case of potential civil plaintiffs. If the object and purpose of the CAT clashes in some respects with reading article 14 in the context of the criminal law provision, it may still be possible to fill the silence in article 14 on scope of applicability by reconstructing the signatories’ intent by reference to other norms of international law as permitted by article 31(3)(c) of the VCLT. A plaintiff could argue in support of an interpretation favouring universal civil jurisdiction and the exclusion of state immunity within article 14 of the CAT by reference to relevant norms contained in the international legal community, such norms being indicative baselines against which interstate understandings of treaty terms can be assessed. As the discussion in the next section indicates, the status of the prohibition against torture as jus cogens is thought by some to provide the necessary reconciliation between the competing imperatives of universal human rights and state sovereignty in the context of domestic civil enforcement—a different reconciliation from that produced above by reading article 14 in light of the criminal law provisions in the rest of the CAT. At this point it is important to note, at least in passing, that for the development of a common law cause of action in response to acts of torture, there is an alternative to reliance on the rather ambiguous grant of civil jurisdiction contained in article 14 of the CAT. An argument could be made that given the status of the prohibition against torture as jus cogens, the issue of the reception of the torture prohibition within domestic systems is not confined to the question of giving effect to the terms of a treaty such as the CAT, but involves the question of the direct incorporation of customary international law into Canadian law as well. In accordance with an adoptionist approach to custom, such as exists within Canada, customary international law is directly applicable as part of Canadian common law—the only interpretive issue being whether firm domestic common law precedent must give way to developments in customary law at the international level.30 On this argument, courts could rely on the presence of universal jurisdiction in the criminal context to recognise torture as a universal tort within the common law, it being assumed that such criminal acts would amount simultaneously to tortious conduct. The viability and acceptability of this CAT-bracketing approach will now be discussed.

30 Although it is not possible to conclude with absolute certainty that Canada is adoptionist with respect to the reception of customary international law, authority for this proposition generally rests with the judgment of Lord Atkin in Chung Chi Cheung v. R. [1939] A.C. 160 at 167–8.

In Search of a Defence 265 Reconciling International Obligations under the Convention against Torture with Domestic State Immunity Legislation through Recourse to the Jus Cogens Status of the Prohibition on Torture Given the silence and ambiguities that exist simultaneously at both the national and international level on the question of extraterritorial civil jurisdiction over torture by state officials, a domestic court must adopt some justificatory theory that explains why it is part of the judicial function to resolve these silences and ambiguities in the absence of any legislative pronouncement. In practical terms, the justification process would be carried out simultaneously at the level of both international and national norms as courts attempt to identify and reconcile international obligations pursuant to the CAT and the scope of permitted exceptions in state immunity legislation. In the search for an appropriate justificatory theory, it is assumed that the jus cogens nature of the prohibition is seen as a viable solution, to the extent that courts are seen as legitimately having a central role in giving effect to such “super” legal norms. This approach is predicated on an understanding of sovereignty within the international legal order as a conditional status; violation of jus cogens disentitles a state to the normal incidents of sovereignty, such as immunity from foreign jurisdiction. Reference to acts of torture as a violation of jus cogens, if accepted as a justificatory theory, must be sufficient both: (a) to engage the scope of article 31(3)(c) of the VCLT as a “relevant rule of international law applicable in the relations between the parties” for the purposes of interpreting the scope of article 14 of the CAT; and (b) to permit an implied exception within domestic immunity legislation for the purposes of permitting a civil claim to proceed against a foreign state or state official concerning allegations of acts of torture. (i) Reconciliation within state immunity legislation: jus cogens and the implied waiver exception Belsky, Merva and Roht-Arriaza have argued within the context of United States jurisprudence that when a foreign state violates a norm of international law that has the status of jus cogens, that state has impliedly waived its right to immunity from the jurisdiction of United States courts pursuant to the Foreign Sovereign Immunities Act (FSIA).31 The drafting of the FSIA is typical of domestic state immunity legislation in that a potential claimant seeking to bring a civil action against a foreign state or state official must demonstrate that the subject matter of the claim comes within the scope of the enumerated exceptions. Reflecting the fact that a sovereign state may voluntarily choose to waive its immunity, §1605(a)(1) of the FSIA provides that immunity will not attach in 31 23 U.S.C. §1330 et seq. [hereinafter FSIA]. See: A C Belsky, M Merva and N Roht-Arriaza, “Implied Waiver Under the FSIA: A Proposed Exception to Immunity for Violations of Peremptory Norms of International Law”, (1989) 77 Calif L Rev 365.

266 W Adams circumstances “in which the foreign state has waived its immunity either explicitly or by implication”. The essence of the argument of Belsky et al. is that there is no need to reconcile as inherent in an international legal order comprised of sovereign equals the universality of human rights norms, such as the prohibition against torture and the concept of state immunity. Sovereignty is not implicated because the constraints of peremptory norms do not recognize these violations as sovereign acts. While the doctrine of state immunity rests upon the foundation that the sovereign equality of states precludes a state from being bound without its consent, peremptory norms do not depend on the consent of any individual state for validity. Thus acts which violate these norms are not sovereign in nature, and a state impliedly waives its immunity for any claims which arise as a result: “Because jus cogens, by definition, is a set of rules from which states may not derogate, a state act in violation of such a rule will not be recognized as a sovereign act by the community of states, and the violating state therefore may not claim the right of sovereign immunity for its actions.”32

In demonstrating the qualified nature of sovereignty, Belsky et al. explore the independent development within international law of the doctrines of jus cogens and sovereign immunity. An historical analysis of sovereignty reveals its contingent nature as a construct which first originated as a discretionary principle of comity, subsequently developed into a categorical legal requirement, and eventually succumbed to a functional analysis, resulting in its current restrictive form. The limitations which international law places on the freedom of states are also best understood within their historical and contingent context. The earliest limitations of state liberty which were derived from natural law eventually gave way to the philosophy of positivism and the emphasis on state consent. In the aftermath of the Second World War and the judgments at Nuremberg and Tokyo, elements of natural law returned with the recognition within international law of jus cogens, norms which originate external to an individual state’s consent and are binding regardless of lack of specific consent. Sources of international law are now hierarchical in that jus cogens norms constrain state sovereignty by superseding voluntary or positivist notions of consent. If one accepts as legitimate the argument of Belsky et al., that a violation of jus cogens is not a sovereign act, and accordingly that a state cannot claim immunity for such acts, a functional analysis would indicate that domestic courts could have the necessary legal authority to interpret the scope of article 14 of the CAT to include both a grant of extraterritorial jurisdiction and the exclusion of state immunity in civil actions concerning the subject matter of the CAT. International law restrictions on the assertion of extraterritorial jurisdiction and the corresponding principle of state immunity are intended to maintain the structure of the international legal order in terms of the sovereign equality of states. If a state has acted in a manner that exceeds the limits of sovereignty 32

Belsky et al., supra n. 31 at 377.

In Search of a Defence 267 or, put differently, escapes the logic of the reasons why its sovereignty should be respected, these concerns are not implicated and international law has no justification for restricting a state’s autonomy in pursuing enforcement against a violator state.33 However, in terms of reconciling a state’s obligations pursuant to the CAT with domestic immunity legislation, the solution proposed by Belsky et al. is a legal fiction, and accordingly more appropriately characterised as a political rather than legal act if adopted by domestic courts. Reference to an implied waiver shifts the focus from legislative intent to the intention of the alleged violator state. Focusing on the intention of the potential defendant would alleviate concerns in parliamentary systems regarding the separation of powers between the judicial and legislative branches in circumstances where the legislation is silent as to the requirements of the CAT, although it is far from self-evident that the text of either the Canadian or UK legislation would permit this type of argument to proceed.34 Given the reliance on the requirement of intent, however, the reference to an “implied waiver” is somewhat disingenuous, and the term “constructive waiver” would be more appropriate. A foreign state does not so much impliedly waive its right to immunity, whether ex ante or ex poste, as much as the court determines that immunity does not attach to the activities in question. If the court must determine whether a state’s actions disentitle the state to claim immunity, as opposed to merely recognising a state’s own actions that truly amount to a volitional waiver, the issue of sovereignty is engaged such that a domestic court must squarely justify its assertion of jurisdiction over the foreign state. The implied waiver argument has been rejected in the United States, albeit with reluctance. In Siderman de Blake v. Republic of Argentina, the court did not so much focus on the legitimacy of the argument as a legal fiction as on the separation of powers between the judicial and legislative branches.35 Agreeing both that the prohibition against torture had the status of jus cogens and that international law did not recognise a violation of jus cogens as a sovereign act, the court nonetheless did not—and could not—limit its analysis of the relevance of these conclusions to international law, focusing instead on the scope of permitted exceptions pursuant to the FSIA. The court concluded that in this respect, it was bound by the decision in Argentine Republic v. Amerada Hess 33 This of course speaks to the jurisdictional power of the state whose courts’ jurisdiction is at issue, and not necessarily to any duty of the state to assume jurisdiction. 34 The text of the Canadian State Immunity Act precludes this type of argument. The conditions of voluntary waiver are drafted in the form of a definition for the purposes of identifying a foreign state’s submission to the court’s jurisdiction, and are limited to those set out in sections 4(2) and 4(4) of the Act pertaining to explicit submission to the domestic court’s jurisdiction or active participation in the proceedings. The State Immunity Act, 1978 of the United Kingdom may be more promising, given that section 2, pertaining to the circumstances in which a state is deemed to have submitted to a domestic court’s jurisdiction, does not necessarily constitute an exhaustive list. Neither Act, however, goes so far in scope as section 1605(a)(1) of the FSIA in terms of allowing for an implied waiver. 35 Siderman de Blake v. Republic of Argentina, 965 F. 2d 699 (9th Cir. 1992).

268 W Adams Shipping Corporation, wherein the United States Supreme Court held that it was Congress’s intention to extend immunity pursuant to the FSIA, even to alleged violations of international law.36 The court’s reliance on the decision in Amerada Hess may have been misplaced, given that in that case the United States Supreme Court was not considering a violation of international law having the status of jus cogens. Conceivably, the holding in Amerada Hess could have been distinguished on this basis in order to permit an implied exception to the FSIA for violations of jus cogens. That being said, the court’s real difficulty lay not so much with the binding force of precedent, but with the separation of powers between the legislative and judicial branches. Whatever the requirements of international law, the court was unwilling to consider those requirements independently of the jurisdiction expressly granted by or permitted under the FSIA, as is evident in the court’s conclusion that, “if violations of jus cogens committed outside the United States are to be exceptions to immunity, Congress must make them so.”37 (ii) Reconciliation in context: violation of jus cogens excluded ab initio In Siderman de Blake, as discussed in the previous section, the court had no difficulty in concluding that a violation of jus cogens was not a sovereign act, but was disinclined to follow the logic of this reasoning in assessing a state’s obligations pursuant to international law in the presence of legislative silence on the matter. A possible solution to problems involving the separation of powers is to examine the issue in terms of potential limitations imposed on legislative intent by international norms, particularly in parliamentary systems where conventional international law is of no force or effect unless and until implemented by the legislative branch. In legal systems having adopted the interpretive presumption that the legislature does not intend to enact legislation in violation of the state’s international obligations, refutable upon an express statement of intent to the contrary, there may be room to expand the enumerated exceptions to immunity to include an implied exception ab initio for violations of jus cogens. The use of this interpretive presumption may be easier to justify as an exercise of judicial rather than political authority than the legal fiction of implied waiver. Significantly, plaintiffs who have framed their arguments in this manner have not met with any greater degree of success than plaintiffs relying on the implied waiver exception in the FSIA. Courts have refused to entertain arguments from outside the conceptual contours of the relevant state immunity legislation, treating the text of the legislation as covering the field in terms of the scope of immunity. In Amerada Hess, the United States Supreme Court held that 36 37

Argentine Republic v. Amerada Hess Shipping Corporation et al., 488 U.S. 428 (1989). Siderman, supra n. 35 at 719.

In Search of a Defence 269 US courts could not exercise jurisdiction over a foreign sovereign for a violation of international law other than in accordance with the express exceptions set out by Congress in enacting the FSIA.38 The respondents had initiated an action against the Argentine Republic under the Alien Tort Claims Act (ATCA).39 The ATCA confers original jurisdiction on district courts over civil actions by an alien for a tort committed in violation of customary international law or a treaty to which the United States is party. The Court limited its inquiry to an interpretive analysis of the text of the FSIA alone, concluding that by making specific reference to an exclusion of immunity in circumstances where the foreign state has taken property in violation of international law, by necessary implication Congress intended to grant immunity for all other violations. As stated earlier, it is significant that the United States Supreme Court in Amerada Hess was not called upon to consider a violation of jus cogens. Nonetheless, the United Kingdom Court of Appeal relied in part on the decision in Amerada Hess in reaching the same result concerning a claim of extreme cruelty amounting to torture in violation of international law. In Al-Adsani v. Government of Kuwait and Others, the plaintiff, a British national and citizen of Kuwait, was a member of the Kuwait Air Force during the Gulf War.40 After Kuwait was invaded by Iraq, he continued as a member of the resistance movement, at which time he came into possession of certain video tapes belonging to one of the defendants, a relative of the Emir of Kuwait. Apparently the video tapes contained recordings of the defendant engaged in sexual activity. After the Iraqis were expelled from Kuwait, the defendants allegedly kidnapped the plaintiff. The plaintiff alleges he was subjected to acts of extreme cruelty at the hands of the defendants, as well as when he was subsequently taken to a state security prison and beaten by security guards. The plaintiff argued that the grant of immunity in section 1(1) of the State Immunity Act 1978, although apparently unqualified in terms of the text alone, was necessarily limited in scope by the existing constraints placed on state immunity pursuant to international law. The legislature could not grant states a greater degree of immunity than already exists under international law. Accordingly, the defendants could not claim immunity under the Act in circumstances where their activity amounted to a violation of the universal prohibition against torture. Lord Justice Stuart-Smith, in providing the reasons of the Court, treated the Act as a “comprehensive code”, and refused to entertain arguments to the effect that state immunity for acts of torture was excluded ab initio from the scope of coverage: “At common law a sovereign state could not be sued at all against its will in the courts of this country. The 1978 Act, by the exceptions therein set out, makes substantial 38

Amerada Hess, supra n. 36. 28 U.S.C. §1350. For an overview of the ATCA case law, see M Swan, “International Human Rights Torts Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines”, chapter 3 of this volume. 40 Al-Adsani v. Government of Kuwait and Others (1996), 107 I.L.R. 536 (Eng., C.A.). 39

270 W Adams inroads into this principle. It is inconceivable, it seems to me, that the draftsman, who must have been well aware of the various international agreements about torture, intended section 1 to be subject to an overriding qualification.”

In both Siderman and Al-Adsani, the courts appear disinclined to adopt an interpretive approach that would permit legislative intent to be examined in the context of a state’s international obligations, preferring to consider such intent in isolation by reference only to the text and structure of the legislation. The unarticulated concern running throughout the reasons in both of these decisions appears to be that any attempt to explore potential limitations on legislative intent at the time of drafting would displace legislative supremacy, and thus would be an exercise of political rather than legal authority. In the author’s opinion, this is a legitimate concern. Examining the external restraints imposed by international law upon legislative intent is in accordance with the presumption that the legislature does not intend to act in violation of international law. However, in the given circumstances, an inquiry into the extent of potential restraints is functionally similar to the initial institutional question discussed above, concerning whether courts are permitted to give effect to international law as part of the process of statutory interpretation in the face of legislative silence. This observation is qualified, however, to the extent that the prohibition against torture is set out not only within the terms of the CAT, but has independent existence, as discussed above, as a peremptory norm within customary international law. Accordingly, the legitimacy of judicial implementation rests not only on the principle of legislative intent with respect to statutory interpretation, but also on the accepted authority of domestic courts to adjudicate upon matters of customary international law as they have been received into the common law. In this context, given that the principle of state immunity has not yet been codified within an international convention, one could argue in support of a certain amount of flexibility in this area. In other words, domestic courts should be permitted to adapt the common law in accordance with emerging norms of customary international law, particularly in circumstances such as these where principles of state immunity are being developed in tandem with jus cogens norms. From this perspective, the relevant inquiry for the Court of Appeal in Al-Adsani should not simply have been the content of the international norm of state immunity at the time of the enactment of the State Immunity Act 1978 for the purposes of interpreting the scope of permitted exceptions, but whether any emerging consensus existed within customary international law, as received into domestic common law, concerning the interaction of state immunity and the jus cogens prohibition against torture. Assuming that courts are competent to consider the scope of legislative exemptions in the context of common law baselines, the argument that violation of jus cogens is incompatible with sovereignty might seem compelling. To insist that the courts of one sovereign state may not sit in judgment of another

In Search of a Defence 271 sovereign state’s acts is to restate a traditional rule in formal terms, but not to provide a functional analysis of the values and interests promoted by the rule.41 If the argument proceeds by way of first principles, then a persuasive case can be made that as jus cogens limits the exercise of sovereignty, an act that violates jus cogens is by definition not a sovereign act and therefore does not attract immunity. The difficulty with the argument as presented by Belsky et al. is not with the utility of the justificatory rationale, but with its implementation firmly against the grain of the text of domestic legislation. As such, it endorses the theory that courts may rewrite legislation by treating its external normative context as if it were itself legislation.

4 CONCLUSION : JUDICIAL OR POLITICAL ENFORCEMENT ?

This chapter began by noting that any domestic resolution of the issues surrounding an alleged violation of the international law prohibiting torture cannot proceed without an assessment of the legitimacy of a domestic court’s actions. The search for a defence of the transnational human rights paradigm is a search for a justificatory theory that not only permits domestic courts to access international law in resolving disputes, but also provides these courts with the means to reconcile the simultaneous and potentially conflicting demands of international and national law. The source of the necessity for justification at the international level is the well-established principle of international law that a domestic court may not sit in judgment of the sovereign acts of another state undertaken within its own territorial jurisdiction. Without sufficient justification for international law to modify or offset this principle, an assertion of extraterritorial civil jurisdiction by domestic courts, along with the concomitant denial of state immunity, amounts to an exercise of political rather than legal authority. In states with Westminster parliamentary systems of government, an additional question of legitimacy arises in circumstances where the legislative branch has not legislated international obligations into the national legal system. As to the ability of domestic courts to access the CAT directly in circumstances where the legislature has not enacted the necessary transformative legislation, the author views the example set by the recent decision of the Supreme Court of Canada in Baker as insufficient precedent to justify reference to the CAT in the process of interpreting the scope of permitted exceptions to the State Immunity Act. The danger of the ruling in Baker, wherein a court is 41 M. Reimann, “A Human Rights Exception to Sovereign Immunity: Some Thoughts on Prinz v. Federal Republic of Germany” (1995) 16 Mich J Int’l L 403 at 419 (arguing that a human rights exception to sovereign immunity is a logical subsequent development to the commercial activity exception, the ideal solution being multilateral rather than unilateral action, but that, at any rate, such an exception requires an amendment to the relevant US legislation, and cannot be accomplished by judicial fiat).

272 W Adams permitted to refer to the underlying values and principles of unimplemented international conventions, is that it permits a plaintiff to circumvent the democratic process, allowing indirect access to international norms that would otherwise be unavailable. The only justification for this manner of displacement of the democratic process would be circumstances involving a potential infringement of the plaintiff’s constitutional rights. No doubt a case can be made that the State Immunity Act should be interpreted in a manner consistent with constitutional values as they interpretively interact with international human rights law, but rendering the CAT visible to the domestic legal system is but the first step in the adjudicative process. Domestic courts would still need to locate the necessary grant of universal civil jurisdiction within international law, along with the authority to deny state immunity which would otherwise be available. As to whether a justificatory theory premised on jus cogens can provide the necessary grant of domestic authority pursuant to international law, a functional analysis might indicate that an extraterritorial assertion of civil jurisdiction, along with the parallel denial of immunity, is not necessarily incompatible with existing positivist and consent-based understandings of international law. The very concept of jus cogens has received widespread state support in the form of ratification of the VCLT in which the concept is decisively recognised and defined.42 If sovereignty must be exercised in a manner that complies with jus cogens imperatives such as the prohibition against torture by virtue of the consensus understanding of states, a violation of jus cogens need not be treated as a sovereign act even within a positivist framework. However, while the VCLT may represent an interstate consensus on the definition of jus cogens, it does not as a matter of positive law recognise its legal effects outside the particularities of treaty law with which the VCLT is concerned. There is no comparable interstate recognition, either in hard, treaty law or soft law, that jus cogens somehow serves as a basis in private international law for granting domestic, civil legal systems jurisdiction over foreign sovereigns or officials breaching jus cogens. It might be argued that such an effect of breach of jus cogens has a certain logic if understood as a lawful countermeasure (i.e., a breach of an international legal obligation by one state as a proportionate response to a prior breach of an obligation by the delinquent state).43 Taking into account the significance of the general prohibition on extraterritorial enforcement by domestic courts against foreign states for conduct within 42

VCLT, supra n. 14, at art. 53, defines jus cogens as follows:

“For the purposes of the present Convention, a peremptory norm of general international law is a norm accepted and recognized by the international community of states as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character.” 43 Here, one would note in addition that erga omnes obligations are owed by all states to all other states. While not all erga omnes obligations are jus cogens, all jus cogens obligations are also erga omnes obligations: see brief discussion in A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality”, chapter 6 in this volume.

In Search of a Defence 273 those states’ own territory, it is nonetheless a proportionate response given the seriousness of the violation of international law that torture represents. Yet, again, this may well work at the doctrinal level by drawing on natural law arguments, but the fact remains that no interstate consensus can be clearly identified as endorsing this argument qua positive international law. The mention of countermeasures brings us finally to the question of unilateralism in the enforcement of international law by one state against another, or its agents. Even if the loss of immunity did follow conceptually from jus cogens breaches, it simply does not follow that domestic enforcement is justified against the actor which, by international law, is not entitled to plead immunity as a bar to jurisdiction. The argument may be made that a violation of jus cogens is not a sovereign act, and accordingly that domestic enforcement does not implicate concerns regarding the sovereign equality of states. However, such reasoning is entirely circular in nature, in that a domestic court must first assume that which has not yet been adjudicated, namely that the state or state official has violated jus cogens. Violation of jus cogens may indeed disentitle a state from the traditional incidents of sovereignty, such as immunity from a foreign court’s jurisdiction, but unilateral assessment of whether a violation has occurred itself threatens domestic sovereignty in a manner that is not likely to be well received within the international system. If the assertion of universal criminal jurisdiction and the denial of state immunity is controversial even in the face of the express language and necessary implications contained in the CAT, as has been illustrated by the reactions engendered by the case of General Pinochet, then unilateral assessment by one state’s courts of a violation of one of the most significant norms of international law is unlikely to be perceived as a legitimate exercise of judicial authority across the community of nations. It is not surprising that a justificatory theory relying on the identification of torture as a violation of jus cogens would not necessarily do the work of grounding domestic adjudicative jurisdiction over torture as a universal tort. The challenge presented by international human rights law is that while sovereignty is no longer uncritically accepted as independent of a state’s internal political structure, the reigning statist paradigm of international law has yet to fully reconcile positivism and voluntarism with the re-emergence of natural law tendencies. Justification for universal civil jurisdiction over torture must be located within a political theory that national legislatures are willing to invoke as a basis for statutory action.44 States—in other words, their political authorities—that are willing to work to secure true universality of human rights guarantees should not be deterred by the fact that their efforts are not universally recognised, at the outset of an initiative, as legitimate at the international level. It is important to remember that the controversy over domestic enforcement, particularly in cases involving allegations of torture, does not concern the content of the international norms. The actual site of controversy is the effect of unilateral 44

Note especially the US Torture Victim Protection Act.

274 W Adams enforcement on the principle of sovereign equality, and the significance accorded voluntariness and consent. What is required is that within states the political authorities responsible should openly acknowledge that their unilateral efforts are justified by reference to natural law imperatives rather than by existing positive international law, and that auto-interpretation, including that performed by domestic judges operating pursuant to express statutory authorisation, is a necessary evil given the lack of political will in terms of establishing international enforcement mechanisms. Perhaps at some point a critical mass of legislatively-driven domestic enforcement will lead to the necessary political will to establish a truly supranational institution for adjudication of human rights abuses, including compensation in appropriate cases.

10

Impunity and the United Nations Convention against Torture: A Shadow Play without an Ending PETER BURNS 1 and SEAN McBURNEY 2

1 INTRODUCTION

asked to address the lawfulness of impunity for torturers. Specifically, is it a violation of international human rights law and especially the UN Convention against Torture (the Convention or CAT) for a state to refuse to extradite or prosecute a person accused of torture on the basis, firstly, of international immunity or, secondly, of national amnesty?3 The answer to each of these questions, upon the basis of both the Convention and general international law, is: yes, in most instances—emphatically, so far as the Convention is concerned, and more softly so far as general international law is concerned. As the eradication of torture through the twinned mechanisms of international human rights law and international criminal law has, to date, received far more treatment than the doctrinal interface of international human rights law and civil liability, we have opted to relate the problem of impunity to

W

E HAVE BEEN

1 Professor of Law, University of British Columbia, and Chairperson of the United Nations Committee against Torture. While the views expressed herein benefit from the co-author’s official role in interpreting the UN Convention against Torture, they do not necessarily represent the views of the other members of the Committee against Torture. 2 Student, Third Year, Faculty of Law, University of British Columbia. 3 Convention against Torture and Other Cruel, Inhuman or Degrading treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027 [hereinafter “CAT” or “the Convention”]. The Convention created a monitoring body, known as the Committee against Torture, to oversee its implementation pursuant to art. 17. At the time of writing, there are 117 state parties to the Convention. J Voyame and P Burns, “The Convention against Torture” in Manual on Human Rights Reporting (U.N.O., Geneva, 1977) at 367. For a full account of the origins of the Convention and how it applies, see M Lippman, “The Development and Drafting of the United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment”, (1994) 17 Boston College Intl and Comp L Rev 275; J H Burgers and H Danelius, The United Nations Convention against Torture (Dordrecht: Martinus Nijhoff, 1988); P Burns and O Okafor, “The United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment or How it is Still Better to Light a Candle than to Curse the Darkness”, (1998) 9 Otago L Rev 399.

276 P Burns and S McBurney the more developed of the two contexts. While, in answering these two questions, we are thus confining ourselves to the context of individual criminal responsibility for torture, we believe the logic of our arguments extends to the context of civil liability. We would also note that the question of impunity in the criminal context has direct implications for civil redress where a state which is seeking extradition (or a state which has a duty to prosecute if it does not extradite) has a legal system in which civil compensation is treated as an adjunct procedure to criminal law prosecution. For our present purposes the crucial parts of the CAT are contained in articles 1, 2, and 4–8.4 Article 1(1) defines torture for purposes of the Convention and is worth reproducing in full here: “For the purposes of this Convention, the term ‘torture’ means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. It does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions.”

The features of this definition which should be noted are that the pain or suffering must be severe, that it must be inflicted to accomplish one of the specific purposes outlined (or similar purposes by virtue of the use of the words “such . . . as”), and that it must be inflicted in relation to some form of “state” agency.5 A single act of torture would suffice to bring the provisions of the Convention into effect. It should finally be noted that, to the extent the definition in the Convention is narrower on its face—or comes to be interpreted as being narrower—than definitions found elsewhere in domestic or international law, for example, in article 7 of the International Covenant on Civil and Political Rights,6 the Convention itself makes very clear that it cannot be invoked to cut back on these more ample definitions.7 Article 2 of the Convention goes on to oblige states to take effective measures against torture in any territory under their jurisdiction. This article also 4

CAT, supra n. 3. We have said “in relation to some form of ‘state’ agency” due to the ambiguity of the words “with the consent or acquiescence” in article 1(1). Where consent has been given, this wording may be interpreted to include conduct in the “private” sphere by non-state actors. This need not necessarily create a legal relationship of agency between the actor and the state. More commonly and with greater significance, this interpretation may include conduct occurring in the private sphere of which “a public official or other person acting in an official capacity” has knowledge but does not make any serious attempt to prevent or stop the conduct. See also the extended meaning given by the CAT to “persons acting in official capacity” in Sadiq Shek Elmi v. Australia, Communication No. 120/1998, UN Doc. A/54/44 at 109. 6 Adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N.GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976) [hereinafter “ICCPR”]. 7 See article 1(2) of the CAT, supra n. 3. 5

Impunity and the CAT 277 excludes any defence of necessity or duress, as well as any defence of superior orders. An obligation to criminalise torture, and related offences associated with it, is found in article 4. But it is articles 5–8 that are the source of answers to the questions posed above, concerning the Convention. However, before examining these provisions we must briefly tease out the dimensions of impunity with which we are concerned. To this we now turn. 2 THE GRADUAL DEMISE OF IMPUNITY FOR INTERNATIONAL CRIMES ?

Forms of Impunity In recent years the interplay between impunity and accountability for international criminal conduct has begun to absorb academics and practitioners alike.8 Impunity can be defined as exemption from punishment or penalty.9 Understood in this very broad manner, impunity can arise out of a variety of circumstances. (i) Policy-based decisions States that have jurisdiction to investigate and prosecute international crimes may choose not to do so for reasons of policy. Such policy can range from a calculated analysis of the political costs and benefits at stake to cases of the state not having the resources to embark on such an enterprise even if it had the will to do so. (ii) Traditional immunities under the international law Historically, certain entities have been immune from prosecution in foreign courts and even before international tribunals. Immune entities included: heads of state and government, possibly with special immunity for ministers of government; foreign states themselves, including institutional appendages of those states; and members of the diplomatic corps accredited to a host state, most notably ambassadors. For our purposes, and in light of the attention which has been paid to the Pinochet extradition case in the United Kingdom, the immunity of greatest interest is that related to the conduct of current or former foreign heads of state concerning acts taking place in their own country.10 But the seeds

8 See, for example, the excellent and comprehensive examination of the topic contained in the various papers taking up a full issue of Law and Contemporary Problems, Vol. 59, No. 4 (1996), papers having been delivered at a symposium at the International Institute of Higher Studies, Syracusa, Italy, earlier that year; see also K Ambos, “Impunity and International Criminal Law: A case study on Colombia, Peru, Bolivia, Chile and Argentina”, (1997) 18 Hum Rts L Jl. 1. 9 The Shorter Oxford English Dictionary (Oxford, Clarendon, 1973) at 1038. 10 Hatch v. Baez, 7 Hun. 596 (1876); 5 Amer Intl L Cases 434 (NYCA)

278 P Burns and S McBurney of the erosion of these immunities were planted in the Treaty of Versailles after the First World War, before flowering in the London and Tokyo Charters after the Second World War.11 With the rise of the United Nations legal order as the primary source of legitimate constraint upon conduct during both conflict and peacetime, such immunities are less frequently available for crimes that also amount to the most serious kinds of human rights violations.12 (iii) Amnesties and immunities as part of the price of a transition to democracy As part of the price of a transition from a regime of brutality to more open forms of governance, new regimes have often been willing to grant immunities to even the most brutal participants in the old regime, by statute and even constitutional amendment. Immunity has often been proffered by way of blanket amnesties from criminal prosecution, sometimes also extending to civil suit. In Central and South America, this usually occurred in the form of a decree of immunity enacted by the former repressive regime in favour of its own officials prior to power being passed over to the new government, giving rise to a new term of art—the autoamnistia (self-amnesty).13 (iv) Truth and reconciliation commissions A very recent development has been the creation of truth and reconciliation commissions as part of the process of facilitating a transition from a conflictridden and brutal regime to a more open and inclusive form of governance.14 These commissions investigate grave breaches of international human rights, and tend to have the power to grant amnesties on a case-by-case basis.15 They 11 Treaty of Versailles, 1919, Arts. 228–30. See also the Treaty of Sèvres, 1920, which contained provisions to punish the Turkish perpetrators of the Armenian genocide of 1915. The Treaty of Sèvres was replaced by the Treaty of Lausanne, 1923, which deleted these provisions. Charter of the International Military Tribunal, annexed to the London Agreement, 8 August 1945; reproduced in M C Bassiouni, Crimes Against Humanity in International Criminal Law (Dordrecht, Martinus Nijhoff, 1992) at 582–6. Charter of the International Military Tribunal for the Far East, Tokyo, 19 January 1946, as attached to Proclamation by the Supreme Commander for the Allied Powers of the same date: reproduced in M C Bassiouni, ibid., at 606–11. 12 See especially the central subject of this chapter, the Convention against Torture, as well as the various instruments cited infra nn. 18 to 20. 13 These are usually referred to as amnesty decrees and can even be enacted by the subsequent regime. For detailed discussion of this phenomenon and its relevance for transnational human rights litigation, see Jennifer Llewellyn, “Just Amnesty and Private International Law”, chapter 22 in this volume. 14 See P B Haynor, “International Guidelines for the Creation and Operation of Truth Commissions: A Preliminary Proposal”, (1996) 59 Law and Contemporary Problems 173; D Cassell, “Lessons From the Americas: Guidelines for the International Response to Amnesties for Atrocities”, (1996) 59 Law and Contemporary Problems 197. J M Pasqualucci, “The Whole Truth and Nothing But the Truth: Truth Commissions, Impunity and the Inter-American Human Rights System”, (1994) 12 Boston U. Intl. Law Jnl 321. 15 P Parker, “The Politics of Indemnities, Truth Telling and Reconciliation in South Africa: Ending Apartheid Without Forgetting”, (1996) 17 Hum. Rts. L. Jnl. 1.

Impunity and the CAT 279 usually require the individuals involved to admit their guilt and demonstrate contrition in order to receive an amnesty, but may reserve the power not to grant amnesty for crimes of a particularly horrific nature.16

The Doubtful Legality of International Immunities and Domestic Amnesties as Jurisdictional Shields for Torturers (i) International immunities Public international law was once thought to be essentially concerned with state relations, to the point that in mainstream international legal circles individuals were defined and treated as “objects” rather than “subjects” of the law. However, developments since the Second World War have shown this analysis to be specious.17 The Nuremberg and Tokyo Charters placed individual responsibility upon those engaging in war crimes, committing crimes against humanity, or waging wars of aggression. This individual responsibility is also reflected in the treaties adopted shortly after the end of the Second World War and the advent of the era of the UN Charter, notably the Genocide Convention and the Geneva Conventions on the laws of war.18 More recently, the tribunals created by the Security Council to prosecute the perpetrators of war crimes, crimes against humanity and genocide in the former Yugoslavia and Rwanda specifically exercise jurisdiction over the responsible individuals at all levels.19 At some future time, the permanent International Criminal Court (ICC) will also exercise a more general jurisdiction over individuals who have committed such crimes—and may have its jurisdiction extended to include crimes against the peace.20 The significance of these tribunals includes the fact that the two ad hoc statutes, those concerning the ICTFY and ICTR, and the statute for the permanent ICC specifically exclude any defence based upon the various forms of 16

See Llewellyn, supra n. 13. See M C Bassiouni, “International Crimes: Jus Cogens and Obligatio Erga Omnes”, (1996) 59 Law and Contemporary Problems 63, especially the sources in n. 2. 18 Convention on the Prevention and Punishment of the Crime of Genocide, 1948; U.N. G.A.D.R. Res. 96 (Dec. ll, 1946); U.N.T.S. 277; Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field, 1949; 6 U.S.T. 3114, 75 U.N.T.S. 31; (1949) Geneva Convention for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea, 6 U.S.T. 3217, 75 U.N.T.S. 85; (1949) Geneva Convention Relative to the Treatment of Prisoners of War, 6 U.S.T. 3316, 75 U.N.T.S. 135; (1949) Geneva Convention Relative to the Protection of Civilian Persons in Time of War, 6 U.S.T. 3516, 75 U.NT.S. 28. 19 Statute of International Criminal Tribunal for the Former Yugoslavia, Security Council Res. 808 (1993), UN Doc. S/RES/808/1993, esp. at arts. 3, 4 and 5, [hereinafter “ICTFY”]; Statute of the International Criminal Tribunal for Rwanda, Security Council Res. 955 (1994, UN Doc. S/RES/994 esp. at arts. 2, 3 and 4 [hereinafter “ICTR”]. 20 Rome Statute of the International Criminal Court, 17 July 1998, UN Doc. A/Conf. 183/9, art. 5. This treaty establishing the ICC will enter into force when 60 states have ratified it: see art. 126. 17

280 P Burns and S McBurney sovereign and diplomatic immunity that might otherwise have been argued to be available to charged individuals.21 From this, it can be concluded that today international tribunals have the power to try offenders against international criminal law, and to exercise universal jurisdiction over the application of this body of law.22 State courts which function as the domestic agents of these regimes, notably within the pending ICC regime, can also exercise such a jurisdiction without the constraint of pleas of immunity. (ii) Domestic amnesties As for national amnesty laws, none of the recent instruments cited specifically mentions these, either negatively or positively. However, this silence—juxtaposed both with these treaties’ imperative phrasing of the obligations upon states to prosecute or extradite and with the need to be normatively consistent with the express non-availability of immunity—is strongly supportive of a conclusion that the validity of laws granting amnesty for human-rights-related crimes cannot be sustained under international law. This is especially the case when the amnesties are of a general, or blanket, nature. At a formal level, then, the range of traditional methods of impunity that are mandatory under international law (international immunities) or treated as permissible (amnesties) is diminishing. The foregoing short description of trends in international criminal law leads us back to the Convention against Torture.

3 THE CONVENTION AGAINST TORTURE AND IMPUNITY

Apart from the developments in international criminal law outlined above, another trend has developed under the sponsorship of the United Nations, but this time under the conceptual umbrella of international human rights law. Six international human rights treaties have been developed by the United Nations, each having its implementation supervised by its own monitoring body. The CAT is one of these. The Convention creates a Committee against Torture (the Committee). Its members are elected by the state parties and charged with the mandate of overseeing compliance with the obligations in the treaty. 21 ICTFY, supra n. 19 at art. 7; ICTR, supra, n. 19, art. 6; International Criminal Court, ibid., proposed art. 27. 22 For an excellent account of the legal issues leading to this conclusion, and the further conclusion that states have not merely the right to prosecute such crimes but an obligation to do so, see Bassiouni, supra n. 17, and M Scharf, “The Letter of the Law: the Scope of the International Legal Obligation to Prosecute Human Rights Crimes”, (1996) 59 Law and Contemporary Problems 41. Some states would still require domestic law to empower their courts to exercise such a universal jurisdiction. This, at least, seems to be one of the implications of the reasoning in the decision of the House of Lords in the Pinochet case: R v. Bow Street Metropolitan Stipendiary Magistrate and Others, Ex Parte Pinochet (No. 3) [1999] 2 W.L.R. 827, [1999] 2 All E.R. 97, (hereinafter Pinochet (No. 3) ). All subsequent references are to the W.L.R.

Impunity and the CAT 281 In the case of the CAT, a state is obliged by article 5(1) to assume criminal jurisdiction over acts of torture committed in its territory, including onboard a ship or aircraft registered in that state, and if the alleged offender is the state’s national. If it is only the victim who is a national, the state may assume jurisdiction if it regards such jurisdiction to be appropriate.23 Article 5(2) reads: “Each State Party shall likewise take such measures as may be necessary to establish its jurisdiction over such offences in cases where the alleged offender is present in any territory under its jurisdiction and it does not extradite him pursuant to article 8 to any of the States mentioned in paragraph 1 of this article.”

Article 5(2) provides not only for a universal jurisdictional power or liberty but also for a form of universal obligation. The foregoing statement requires some elaboration. A conservative, and perhaps even strained, reading of article 5(2) would interpret a state’s duty to “establish its jurisdiction” to arise only where that state has chosen not to extradite when faced with an actual extradition request made pursuant to article 8 by a state meeting article 5(1) jurisdictional criteria. In the absence of such an article 8 extradition request, the condition precedent for the state in whose territory the alleged offender presides would not, on this interpretation, be fulfilled. In such a case, the state where the offender is present would retain the liberty not to prosecute. A reading finding no requirement for an article 8 extradition request would be better, by far. On the plain wording of article 5(2), the state where the alleged offender is found must “establish” its jurisdiction where “it does not extradite.” It is the fact of non-extradition, not the fact of declining a request to extradite, which article 5(2) designates the condition precedent for the state’s obligation to prosecute. Of course, the obligation would also exist if a request for extradition had been made by a state party and had been rejected for some reason. Here it is crucial to emphasise that what is triggered is not simply jurisdiction, but the duty to exercise that jurisdiction: not only the power to prosecute, but the obligation. Furthermore, that obligation is universal in the fullest sense. Article 5(2) must be interpreted as applying even if the alleged torture took place in the territory of a state which is not party to the Convention, and even if neither the offender nor the victim is a national of either the state party where the alleged offender is found or of any other state party. Nothing in article 5(2) limits the duty to situations in which another state party, or parties, would also have jurisdiction based upon article 5(1) criteria. In the case that a request is made by an article 5(1) state, article 5(2) does offer extradition as an alternative to the state where the alleged offender is found. Otherwise, it imposes a duty founded in a universal jurisdiction to prosecute torture wherever it occurs. This duty applies even if there is no state party that satisfies the article 5(1) jurisdictional criteria based on the territoriality of the crime, the nationality of the 23

CAT, supra n. 3, art. 5 (including any associated inchoate offences).

282 P Burns and S McBurney perpetrator, or the nationality of the victim. It is for this reason that we speak of the obligation as universal in the fullest sense. In short, article 5(2) imposes a universal duty to prosecute in the absence of extradition, whatever the reason for non-extradition. A state incurs responsibility under the Convention if it fails to prosecute, and in doing so allows the alleged torturer to remain free to either stay in its territory or to leave. The foregoing interpretation is consistent with the purposes of the Convention.24 Although the Committee has itself not decided this issue, the tenor of the questions that it poses to state parties during the discussion of the reports they periodically submit to the Committee indicates that it considers a state’s duty to exercise universal jurisdiction to be based on simply the presence of a torturer in the state’s territory. Support for this stance is slowly emerging from within domestic legal systems, and is further reinforced by the language of article 7(1), which reads: “1. The State Party in the territory under whose jurisdiction a person alleged to have committed any offence referred to in article 4 is found shall in the cases contemplated in article 5, if it does not extradite him, submit the case to its competent authorities for the purpose of prosecution.”

According to article 31(2) of the Vienna Convention on the Law of Treaties, any given provision of a treaty must be interpreted in the context of other relevant provisions within the same treaty.

4 THE PINOCHET CASE AND IMMUNITY UNDER THE CONVENTION AGAINST TORTURE

The House of Lords and the Unavailability of Immunity for Torture The most authoritative pronouncements by a national court upon the status of torture as an international crime, upon the jurisdictional reach of the CAT, and upon the viability of sovereign immunity as a jurisdictional defence were delivered by the House of Lords in R v. Bow Street Metropolitan Stipendiary Magistrate and Others, Ex Parte Pinochet (No. 3).25 By an overwhelming 24 One of the rationales for the CAT was expressed by Lord Browne-Wilkinson in Pinochet (No. 3) as being: “The Torture Convention was agreed [by the State Parties] not in order to create an international crime which had not previously existed but to provide an international system under which the international criminal—the torturer—could find no safe haven”: Pinochet (No. 3), supra n. 22, at 842. If this is so, no reasonable policy would promote an interpretation that would restrict the obligation to assume jurisdiction. This is clearly the view of Lord Browne-Wilkinson, who also stated without qualification in Pinochet (No. 3), supra n. 22, at 842: “Under article 5(2) a state party has to take jurisdiction over any alleged offender who is found within its territory.” See also the judgment of Lord Millett, ibid., at 913. 25 Supra n. 22, noted by R Piotrowicz, “The Governments, the Lords, the Ex-President and His Victims: Limitations to the Immunity of Former Heads of State”, (1999) 73 Austr L Jnl 482. Pinochet (No. 3) was the second time the House of Lords found, by a majority, that Pinochet could not rely

Impunity and the CAT 283 majority of 6–1, the House of Lords affirmed the view of an earlier House of Lords panel that sovereign immunity does not represent a blanket defence to international crimes that are jus cogens pursuant to the opinio juris of international law, including torture.26 As part of their reasoning, the Lords examined in detail the effect of the CAT upon the correct interpretation of the scope of immunity for a former head of state under the UK’s State Immunity Act of 1978. This statute was drafted in a sufficiently ambiguous way such that Pinochet was able to argue that it conferred immunity from any criminal proceeding not only for himself as an existing head of state, but also for a former head of state. The UK’s statute addressed the state immunity of a former head of state in an unusual way relative to the formulations in other national legislation, namely by using the same language as a test for a former head of state’s immunity as is used in the Vienna Convention on Diplomatic Relations, as incorporated into UK law, with respect to former diplomats such as ambassadors. Article 39(2) of the Vienna Convention on Diplomatic Relations directly applies to former ambassadors. According to the UK statute, it applied mutatis mutandi to Pinochet. Article 39(2) reads as follows: “When the functions of a person enjoying privileges and immunities have come to an end, such privileges and immunities shall normally cease at the moment when he leaves the country, or on expiry of a reasonable period in which to do so, but shall subsist until that time, even in the case of armed conflict. However, with respect to acts on immunity as a former head of state to shield him from criminal prosecution for torture, including extradition to face prosecution in Spain: see also Pinochet (No. 1), [1998] 3 W.L.R. 1456, [1998] 4 All E.R. 897. In Pinochet (No. 3), a 6–1 majority of the House held that only those allegations that referred to conduct after 8 December 1988, the date on which the UK ratified the Convention, and on which relevant provisions were given the force of statutory law in the UK’s own legal system, could sustain any extradition because, prior to this date, the majority were not convinced that the sovereign immunity of Senator Pinochet had been abolished. But they were convinced that, once the Convention became UK law, it had been so abolished. This meant that almost all the counts alleged in the provisional warrants were struck out. Pinochet (No. 1) was vacated by the Lords in response to a motion brought by Pinochet that one of the judges in the majority had links to Amnesty International, an intervenor in the case, which created at least the apprehension of bias: see Pinochet (No. 2), [1999] 2 W.L.R. 272, [1999] 1 All E.R. 577. Following Pinochet (No. 3), the magistrate deciding the extradition issue ruled that Pinochet (No. 3) governed and ordered Pinochet’s committal for extradition. The judge accepted that new allegations from Spain of torture occurring after the cut-off date of 8 December 1988 could be considered as part of the extradition request along with the few incidents that were still left after the Lords’ ruling in Pinochet (No. 3). At the time of writing, the magistrate’s decision to order Pinochet’s extradition was being appealed at the same time as which representations were being made to the relevant UK Cabinet minister, Jack Straw, that Pinochet should be exempted from further legal process on grounds of poor health. On 11 January 2000, the relevant UK Cabinet minister, Home Secretary Jack Straw, decided that the process should be brought to a halt on the basis of the information he had received in a medical report from four doctors on the deteriorating state of Pinochet’s health: see “Comunicado del Ministerio británico del Interior”, (12 January 2000) El Pais [on-line version at http://www.elpais.com]. Senator Pinochet returned to Chile and is presently facing criminal charges in that country. 26 Pinochet (No. 3), supra n. 22 at 846 (Lord Browne-Wilkinson); at 881 where the conduct is for the pleasure or benefit of the person claiming the defence otherwise only where excluded by convention (Lord Hope); at 913–14 (Lord Millett); at 924 (Lord Phillips). Lord Hutton confined his comments to the CAT, and did not deal with the general international law in this respect.

284 P Burns and S McBurney performed by such a person in the exercise of his functions as a member of the mission, immunity shall continue to subsist.”27

Thus, in reading this provision to apply mutatis mutandi not only to former diplomats but to former heads of state, the House of Lords were faced in essence with the interpretive issue of whether Pinochet’s conduct relating to “acts performed”, incidents of torture in Chile, should be interpreted as having been “in the exercise of his functions” as a head of state. Drawing on both general international law and the CAT, incorporated into UK law by virtue of s. 134(1) of the Criminal Justice Act 1988, the Lords ruled that their interpretation of article 39(2) for this purpose must be consistent with this broader relevant law. The most significant pronouncement of the Lords was the view expressed by a majority that torture, as defined in the CAT, cannot be a legitimate state function since it is a jus cogens international crime, and therefore no defence of sovereign immunity can apply. It was found that torture cannot be considered conduct part of the “functions” of any state official, even a head of state. Pinochet was thus extraditable for crimes contrary to both UK and Spanish law. Significantly, the Committee had the opportunity to issue its own pronouncement on the matter of Pinochet’s immunity, due to a timely coincidence which saw the United Kingdom appearing to defend its record of compliance with the CAT before the Committee only days before Home Secretary Jack Straw was due to issue his response to Pinochet (No. 1). As part of the Concluding Observations adopted by the Committee on 23 November 1998, it said the following: “E. Recommendations ... (f) The Committee finally recommends that in the case of Senator Pinochet of Chile, the matter be referred to the office of the public prosecutor, with a view to examining the feasibility of and if appropriate initiating criminal proceedings in England, in the event that the decision is made not to extradite him. This would satisfy the State Party’s obligations under articles 4 to 7 of the convention and Article 27 of the Vienna Convention on the Law of Treaties 1969.”28 27 Vienna Convention on Diplomatic Relations, UN Doc. A/CONF. 39/26, article 39(2) (emphasis added). 28 UN Doc. CAT/C/SR. 360, 23 November 1998, at 5–6. In the course of the meeting with the United Kingdom delegation the Chairman of the CAT, who was the Rapporteur, made the following observations:

“. . . In the Committee’s view, . . . paragraph [2 of article 5] conferred on states parties universal jurisdiction over torturers present in their territory, whether former heads of state or not, in cases where it was unable or unwilling to extradite them. Whether they decide to prosecute would depend on the evidence available, but they must at least exercise their jurisdiction to consider the possibility . . . . . . If the High Court position [to allow the sovereign immunity defence to lie] was carried to its logical conclusion, the United Kingdom would have no jurisdiction to prosecute Adolph Hitler if he had visited England after the Second World War. The absolutist notion of sovereign immunity had been seriously undermined by state and international practice over the past 25 years . . . :” CAT/C/SR. 354, 18 November 1998, at 11–12.

Impunity and the CAT 285 Quite apart from constituting a contribution to the specific case regarding Pinochet, the Committee’s pronouncement constitutes a major precedent for all state parties to the Convention. The Committee’s view is clearly consistent with the foregoing arguments, in section 3 above, that states have a duty to prosecute based on simple territorial presence triggered by the fact of non-extradition, and that the failure of the Convention expressly to rule out immunity cannot be interpreted to mean that it is permitted. It is to this latter point that we now return to develop our earlier rather schematic arguments. It is important for it to be clear why the Convention must not be interpreted as silent on immunity, but as containing a bar on immunity by necessary implication. One reason this is important is that in future other states and judges in other states may be asked to draw normative guidance not only from the House of Lords’ interpretation of lack of immunity under the Convention, but also from the Committee’s own interpretation. Some states and judges may be tempted to treat as more persuasive the dissenting view of Lord Slynn in the first of the Pinochet cases; that silence in a treaty must mean that immunity is retained.29 The central reason Lord Slynn gave was that there are treaties, such as the ICC, that expressly remove immunity. That being the case, Lord Slynn reasoned, the drafters of any treaty that does not also do so must be interpreted as not having intended to remove an immunity defence. Despite himself having an international law background, Lord Slynn made use of an extremely legalistic technique of interpretation to arrive at his result, a technique inconsistent with the holistic and purposive way in which human rights treaties must be interpreted. The House of Lords in Pinochet (No. 3) saw beyond Lord Slynn’s approach and, in this, there is complete concordance between the Lords’ view in that case as to the significance of the lack of an express bar on immunity in the Convention and the Committee’s view, addressed to the United Kingdom delegation by the Committee only a couple of months before the Lords’ ruling. The reasoning of the majority in Pinochet (No. 3) is clearly to be preferred for a number of reasons. First of all, article 32(2) of the Vienna Convention on the Law of Treaties of 1969 obliges states and institutions to resort to the state of general international law in interpreting treaties. Thus, in interpreting treaties dealing with human rights norms, interpretive “gaps” must be filled with reference to current jus cogens obligations. Secondly, if applied to the CAT, Lord Slynn’s approach would negate its very substance. Article 1 defines torture in terms of “state conduct”, and article 2 permits no justification for torture. To render immune sovereigns, diplomats or the state itself is to render the CAT largely nugatory. These actors are usually the sources of the policies that lead to the prohibited behaviour by lower state functionaries. Finally, it is specious to read down one human rights treaty, in light of the greater protection expressly found in others. This is reflected in article 1(2) of the CAT itself which reads: 29 Supra n. 25, at 1474 et seq. See also the views of Lord Goff in dissent in Pinochet (No. 3), supra n. 22 at 856–8.

286 P Burns and S McBurney “This article is without prejudice to any international instrument or national legislation which does or may contain provisions of wider application.” The specific reference to sovereign immunity being excluded as a defence in the ICTFY, the ICTR and the ICC was probably based upon excessive caution on the part of the drafters, and was designed to remove the very type of argument raised in this respect by Lords Slynn and Goff.

5 THE CONVENTION AND IMPUNITY

Torture, as defined by the CAT, is all too common in authoritarian states, particularly in times of fragile government control, or when civil conflict has occurred. The whole premise of the CAT is that torturers are not to enjoy impunity—that they must be investigated, arrested and tried for their crimes. From Pinochet (No. 3) and the Committee’s own view of the CAT, we know that individuals cannot successfully raise a defence of sovereign immunity, nor, logically, diplomatic immunity, for acts defined as torture under the Convention, whatever the inconsistencies and complexities with the way the situation is perceived in domestic law. To what extent, then, can general or specific amnesties be regarded as a breach of a state party’s obligations under the Convention? Clearly, general amnesties would constitute such a breach by their very nature. By excluding investigation and not addressing guilt or innocence, they obviously conflict with the obligations contained in articles 12 and 13 of the CAT. Where there is no provision to compensate torture victims, the conflict with article 14 of the CAT is even more obvious. Since such general amnesties are usually proclaimed by a brutal regime on the eve of its demise, or by a successor regime that is reliant upon the goodwill of the former regime’s military and police agencies, the removal of impunity for torturers, the primary purpose of the CAT, is frustrated. Such general amnesties cannot be in conformity with the obligations of the CAT. Such, too, is the legal situation when truth and reconciliation commissions are driven, as they very often are, by policies that have no real bearing upon the guilt or innocence of an alleged perpetrator or the vindication of the interests of the victim. When this is the case, it seems impossible to say that the prohibition on torture and the duty to sanction it once it has occurred have at all been taken seriously. It is not the case that the truth and reconciliation process has fulfilled the state’s responsibilities in a different way from that contemplated in the Convention. Rather, it has derogated from its norms completely. Specific amnesties present a harder case, particularly where they are granted after a truth commission hearing. If a specific amnesty did not involve an inquiry into the substantive question and did not involve evaluating the merits of prosecution against the objectives of the specific amnesty, such amnesty would also be in breach of the terms of the Convention—for reasons similar to those whereby general amnesties are unacceptable. But what of an amnesty granted by

Impunity and the CAT 287 a truth commission for the worthwhile purpose of fostering national reconciliation, where the amnesty is conditional upon the recipient admitting his or her guilt? Even here a case can be made that the terms of the Convention are being breached. Article 7 imposes an obligation upon a state party to submit the alleged torturer’s case to its competent authorities “for the purpose of prosecution”. It obliges such authorities to “make their decision in the same manner as in the case of any ordinary offence of a serious nature under the law of that state”. This does not necessarily mean that every case will be prosecuted. Most notably, where the evidence is too weak to anticipate the probability of a conviction, prosecution would not be required. But, to remain consistent with article 7, it does mean that each case must be pursued in the same way and evaluated for the purposes of a criminal prosecution in the same manner as any ordinary criminal case would be. Truth and reconciliation commissions are not usually part of the standard criminal process. That being the case, even in instances of individual admissions of guilt before truth commissions, the granting of an amnesty for the purposes of national reconciliation may not meet the obligations contained in this provision. On the other hand, if a truth and reconciliation commission has been created with the power to order prosecution or amnesty in any given case pertaining to the conduct of state or even non-state but “authorised” actors over a period of time, it may be regarded as part of the normal investigative and adjudicative criminal procedure of that state for the period that the commission operates. If all similar allegations of criminal conduct are dealt with in the same manner and even-handedly, such amnesties as are granted may not be in breach of the CAT. This will be the case especially if provision was made to ensure compensation of torture victims despite the amnesty. In such cases, the primary purpose of the CAT, the abolition of the torturer’s impunity, would be met by applying the same, “ordinary” investigative and adjudicative processes of the commission to all instances, and by including the bottom-line requirement of admission of guilt. The torturer would be exposed to the world and subjected to all the consequences of such exposure. In addition, another main purpose of the CAT would be met—the compensation of torture victims. Where the amnesty does not estop legal pursuit of the torturer or state by the victim for compensation, part of the obligations imposed by article 14 are met. If a state is further prepared also to commit resources to rehabilitation of torture victims, such amnesties are probably in conformity with the CAT.

288 P Burns and S McBurney

6 CONCLUSION

Torture is conduct that may never be eradicated, but it must be condemned— and torturers must be pursued and punished.30 These are the fundamental goals of the CAT. Amnesties for political purposes may provide gains, however ephemeral, but the damage they do to the development of international criminal law, based as it is upon individual responsibility, will weaken its conceptual foundations. The damage they could do to the integrity of the CAT may prove to be irreparable.31 In recent years the tide has turned against international criminals, but despots and their brutal agents may be heartened by the proliferation of truth and reconciliation commissions, which may allow them to ultimately avoid criminal responsibility for their conduct. In contrast, the proposed International Criminal Court should provide a mechanism that accomplishes both the vindication of the interests of victims and the punishment of offenders. The ICC statute does not appear to have left any more room for the application of amnesties than does the CAT. Hopefully, its adoption will reinforce the concerns we have expressed about this phenomenon, which apply no less when amnesty is granted as part of a truth and reconciliation process than when enacted by way of the self-amnesties of outgoing repressive regimes. However, the ICC’s jurisdiction with regard to torture is generally limited to cases which are widespread or systematic and thus constitute the most serious instances of crimes against humanity.32 The CAT, on the other hand, covers all torture as defined in its article 1, including single instances. Its jurisdictional utility will therefore continue beyond the initiation of the work of the ICC. Given developments such as the ICC and the dilution of traditional immunities as described above, it can be seen that, at a formal level, impunity is shrinking. Whether or not impunity will shrink in reality depends upon the will of the international community. In this respect, the jury is still out.

30 Our views here accord with those of Justice Cassese in his Chorley lecture (1997); republished as A Cassese, “Reflections on International Criminal Justice”, (1998) 61 Mod L Rev 1. Of course, the assumption is that trials of human rights violators will never become the instruments of simple revenge: J M Goti, “Dignity, Vengeance, and Fostering Democracy”, (1998) 29 Univ Miami InterAm L. Rev. 417. 31 Except in the cases of individual amnesties after a truth and reconciliation hearing; which maintain the victims’ rights to compensation and rehabilitation. 32 Supra n. 20 art. 7(f), or those pertaining to armed conflict amounting to a war crime, supra n. 20 art. 8(2) (a)(ii), that do not necessarily demand the torture to be widespread or systematic.

11

Torture, Tort Choice of Law and Tolofson JENNIFER A. ORANGE 1

1 INTRODUCTION F A M I L Y L I V I N G in New York City discovered that the man who tortured their 17-year-old son to death in Paraguay was living in New York. They told the local police, and the man was arrested. Although the activities occurred in Paraguay and both the parties were Paraguayans resident in Paraguay at the time of the torture, an American court took jurisdiction and applied its interpretation of the law of nations, public international law, to find the man liable for the tort of torture. These are the simplified facts of Filártiga v. Peña-Irala.2 In Canada, the application of public international law by a domestic court to a tort claim seems strange. In the US, this action is grounded in the Alien Tort Claims Act (ATCA)3 and the Torture Victim Protection Act (TVPA).4 The ATCA, originally enacted in 1789, provides that “[t]he district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.”5 The ATCA thus recognizes an international tort counterpart to individual criminal responsibility under the law of nations.6 In Canada, no statute exists upon which to found such international tort claims. Instead, a torture victim in Filártiga’s shoes, or his family, could argue that Canadian tort law should be applied as the outcome of private international law (conflict of laws) analysis. Until 1995, Canada’s choice of law rule was that established in McLean v. Pettigrew.7 In order to invoke Canadian law, the wrong had to be both actionable in Canada and not justifiable under the law where the act occurred. A Filártiga in Canada might have been able to meet the

A

1

Jennifer Orange is a litigation lawyer of Torys, Toronto. 639 F.2d 876 (2nd Cir. 1980) [hereinafter Filártiga]. 3 28 USC §1350 (1994) [hereinafter ATCA]. The ATCA provided the statutory basis for the Filártiga adjudication, supra n. 2. 4 Pub. L. No. 102-256, 106 Stat. 73 (1992) (codified at 28 U.S.C. §1350 note (1994) ). 5 ATCA, supra n. 3. 6 Stephens, “Conceptualising Violence Under International Law: Do Tort Remedies Fit the Crime?”, (1997) 60 Albany LR 579 at 594 [hereinafter “Stephens”]. 7 [1945] SCR 62 [hereinafter McLean]. 2

292 J Orange McLean test due to the very forum-centred nature of that test. Although there is no tort of “torture” per se in Canadian private law, the underlying harms of torture are actionable in Canada as the torts of assault and battery. As for the second inquiry under the McLean test (non-justifiability under the law of the state where the tort occurred), the Supreme Court of Canada had held, in McLean itself, that the conduct need not be actionable as a civil wrong in the foreign jurisdiction as long as it is unjustifiable in a looser sense of being wrong by some branch of the foreign law, most notably its criminal law. Thus, a Filártiga would still be able to sue a foreign state official, such as a police torturer, as long as their conduct was considered criminal, even if that state’s law made such officials immune from civil suit or the practical likelihood of prosecution was slim. In 1995, the Supreme Court of Canada substantially reformed Canada’s common law choice of law doctrine in Tolofson v. Jensen,8 steering it away from its bias towards the lex fori. Tolofson was an interprovincial tort case arising from a car accident where the parties were from different provinces. Mr Justice La Forest established a new general rule for interprovincial choice of law in tort: the lex loci delicti. Further, La Forest J. held that, in interprovincial cases, the lex loci delicti shall govern without exception. This approach is new in Canada, not only for stating a single criterion basic test but also for failing to retain the possibility of justice-based exceptions that would allow the lex fori to be applied in situations where the acts were neither criminal nor tortious under the foreign law.9 If this strict rule were held to apply not just to interprovincial disputes, but also to those with an international axis, then Canadian law loses its potential to provide justice in certain situations. In a Filártiga situation, Paraguayan law would govern under a lex loci delicti choice of law test. Whether recovery in tort would be possible would depend entirely on the existence of a legally viable civil cause of action for the acts of torture, either through a cause of action called “torture” or through a cause of action called something more general like “battery” or “intentional personal injury”. If a Filártiga in Canada could sue under the letter of Paraguayan law, then he or she would be able to sue in Canada under the lex loci delicti test—assuming of course that Canadian courts take and keep jurisdiction over the case. In Filártiga, it seems the Filártiga family was prevented from taking legal action in Paraguay due to an atmosphere of government intimidation, but such prevention was an extra-legal fact that does not affect the formal status of the acts of torture as actionable in law. If, however, a police official’s acts of torture did not give rise to civil liability in Paraguay—for example due to immunity from lawsuits of state officials or for acts while 8 [1994] 3 SCR 1022 (subnom. Lucas v. Gagnon); (1994), 120 DLR (4th) 289 (SCC) [hereinafter Tolofson; future references cited to DLR]. 9 For a leading pre-Tolofson example of flexible application of the two-limbed choice of law rule in the name of justice, see the Ontario Court of Appeal’s decision in Grimes v. Cloutier (1989), 69 OR (2d) 641, and two leading cases decided by the House of Lords: Chaplin v. Boys [1969] 2 All ER 1085 (H.L.) [hereinafter Chaplin] and Red Sea Insurance Co. Ltd. v. Bouygues S.A. [1995] 1 AC 190.

Torture, Tort Choice of Law and Tolofson 293 employed as state officials—then the lex loci delicti choice of law rule would result in the victim not being able to sue in Canada. This would be the case even if the acts of torture were considered criminal in Paraguay or if a Paraguayan public law, for instance the constitution, prohibits torture. It has not been, and does not remain, uncommon for some states’ legal systems to bar civil actions against state officials, at least for acts done while on official duty. And it has been increasingly common for states to adopt amnesty laws to protect human rights abuses from civil suits even if the law had not, to that point, recognised an immunity from suit.10 Even if Paraguayan law at the time of Filártiga may not have barred suit, it would now be something of an exception—or, at least not the rule—in repressive regimes or in societies in transition from such regimes. The foregoing discussion demonstrates the danger of applying an absolute lex loci delicti rule to international torts. However, La Forest J. did not make it clear how strictly the rule applies to international situations, and left open the possibility of “rare” exceptions in international cases. In this chapter, I will discuss the extent to which Tolofson’s application is limited to interprovincial conflicts, and how far La Forest J.’s arguments in favour of the general lex loci delicti rule should extend to the international, as opposed to interprovincial, arena. Tolofson clearly stated that the old McLean rule for choice of law is insufficient, but I will argue that the Tolofson replacement suffers from its own inadequacy. The straight application if lex loci delicti to international fact situations can lead to unacceptable injustice, most notably when the wrong done can also be characterised as a human rights violation. In the Tolofson judgment, La Forest J. offered as a rationale for the lex loci delicti rule the need to support the growth of the pan-Canadian and global economies. I will argue that the rationales for the choice of law rule for tort should not be limited to those associated with the private commercial realm. Canada’s obligations in relation to global business and trade cannot be approached as if there were a normative void in the realm of non-economic globalisation of values. All of Canada’s public international obligations should be factored into a decision on choice of law. These include the fact that international human rights law obligations are being given an increasingly privileged role within the international legal order itself. Here, this issue is discussed in relation to the role of the doctrines of territoriality and comity, as invoked by La Forest J. in his reasoning regarding the contours of a tort choice of law rule suitable for the current international legal order. It is also considered in relation to the possibility of an “international human rights tort” in Canadian law. Throughout this chapter, I will use the human rights violation of torture as my focus, but much of the analysis will be applicable to torts related to other kinds of serious human rights violations such as slavery. 10 For a discussion of trends along these lines, see P Burns and S McBurney, “Impunity and the United Nations Convention against Torture: A Shadow Play Without an Ending?”, chapter 10 of this volume, and J Llewellyn, “Just Amnesty and Private International Law”, chapter 22 of this volume.

294 J Orange It is important at the outset to note that La Forest J. himself recognised that the strict lex loci delicti rule may be unsatisfactory in situations where elements of the harm occur in more than one jurisdiction, when the act and the harm occur in different jurisdictions, or when elements of the harm occur in different jurisdictions.11 Torture can fall into such categories. The initial physical injury may occur in one place, but the psychological damage can progress after the victim moves from the original jurisdiction. Breaking down where the tort occurred may be as difficult as analysing the psychological impact on the victim, or as straightforward as using the conventional foreseeability test from tort law. In a given instance of torture, was it foreseeable that after the beating the victim would suffer psychological harm, and that the harm would travel with the victim? If a tort can be reasonably connected to more than one jurisdiction, there then may be more than one “proper” law of the tort.12 Tolofson’s apparently strict rule for interprovincial tort cases and La Forest J.’s comments on the possibility of exceptions for international claims leave us wondering where the choice of law doctrine will go from here. The Supreme Court of Canada may have overruled McLean, but, in doing so, it left the law incomplete. From the subsequent case law it seems clear that the Tolofson rationales apply to international situations, but also that there must be exceptions to lex loci delicti rule. The exceptions, however, have yet to be defined. This makes it worthwhile to look at the US and UK regimes for guidance. The American Law Institute’s Second Restatement on the Conflict of Laws13 and the UK Private International Law Act14 provide us with various factors to consider when constructing exceptions to the lex loci delicti rule. I will consider the two regimes, and recommend in conclusion that Canada enact legislation that clarifies tort choice of law in international scenarios. Such enactment would benefit parties to litigation, victims and defendants alike, as well as the judges who grapple with conflict of laws decisions.

2 THE JUDICIAL HISTORY

Canadian tort choice of law rules have evolved from, and largely with, the English case law. In 1870, Willes J. established the basic rule in Phillips v. Eyre, that in order for an action to be brought in England, two conditions must be met: (1) the matter must be actionable in the forum; and (2) the act must not have been, or have since been made, justifiable in the place where it

11

Tolofson, supra n. 8 at 305. In Section 5 I will discuss some of the possibilities for choice of law when dealing with complex, multijurisdictional facts. 13 American Law Institute, 1 Restatement (Second) of Conflict of Laws (1971) [hereinafter Second Restatement]. 14 Private International Law (Miscellaneous Provisions) Act 1995 [hereinafter the PIL Act]. 12

Torture, Tort Choice of Law and Tolofson 295 occurred.15 The first condition required that the statement of claim contain a tort cause of action recognized by the forum. The second condition concerns comity and public policy. If the action was not justifiable in the place where the act occurred, then to allow the same action in the forum and to apply the forum’s law would constitute a lesser incursion into the other state’s jurisdiction. Also, this condition provides a defence if in the place of the act, such act is not considered legally offensive. A predictable problem arose out of Phillips v. Eyre, namely, defining what is “justifiable”: “What the phrase ‘not . . . justifiable’ meant was always a subject of intense debate. It has been held that an act is ‘not . . . justifiable’ if it is criminal, if it gives rise to liability for property damage, even if not for physical injury, if it exposes the defendant to any civil liability, even if the amount of damages is a tiny fraction of the liability under the lex fori, or, in perhaps the extreme case, if the defendant simply did something that the court thought was bad.”16

A questionable spin was put on Phillips v. Eyre in Machado v. Fontes.17 In that case, it was held that an act was not justifiable in the jurisdiction where the tort occurred if it was not legally innocent. Therefore, when no cause of action in tort was recognised in the place of the wrong, the existence of a criminal or regulatory offence in that jurisdiction was still enough to allow the English courts to choose English tort law to govern the action.18 Machado demoted the local law to a status well below that of the law of the forum. However, it took almost three-quarters of a century before the House of Lords would rule that this extraterritorial imposition was offensive and that the tort choice of law rule was a double-actionability rule.19 Double actionability required that the wrong be actionable both by the lex fori and by the lex loci delicti. The requirement of actionability by the lex loci delicti includes assessing the availability of substantive defences under the lex loci delicti so as to narrow the scope of how much of the lex fori can be applied. However, the House of Lords also held that this 15

(1869), LR 6 QB 1 (Ex Ch). As Mr Justice Willes stated at 28:

“As a general rule, in order to found a suit in England for a wrong alleged to have been committed abroad, two conditions must be fulfilled. First, the wrong be of such a character that it would have been actionable if committed in England; therefore, in The Halley (1), the Judicial Committee pronounced against a suit in the Admiralty founded upon a liability by the law of Belgium for collision caused by the act of a pilot whom the ship owner was compelled by that law to employ, and for whom, therefore, as not being his agent, he was not responsible by English law. Secondly, the act must not have been justifiable by the law of the place where it was done. Therefore, in Blad’s Case (2), a.d. Blad v. Bamfield (3), Lord Nottingham held that a seizure in Iceland, authorised by the Danish Government and valid by the law of the place, could not be questioned by civil action in England, although the plaintiff, an Englishman, insisted that the seizure was in violation of a treaty between this country and Denmark—a matter proper for remonstrance, not litigation.” 16 John Swan, “Federalism and the Conflict of Laws: The Curious Position of the Supreme Court of Canada:, (1995) 46 South Carolina LR, 923 at 941. 17 [1897] 2 QB 231 (CA) [hereinafter Machado]. 18 Machado, ibid. at 235. 19 Chaplin, supra n. 9.

296 J Orange general rule provided an exception when it would otherwise result in an injustice.20 In Canada, the double-actionability rule in its Chaplin v. Boys form was never established. In the already mentioned leading case, McLean v. Pettigrew, the Supreme Court of Canada had followed Machado and continued to consider whether the wrong was actionable in the forum and not justifiable—in the loose sense—in the jurisdiction where the act occurred.21 When the Supreme Court of Canada next revisited tort choice of law in Tolofson, it had the benefit of the Chaplin v. Boys analysis, and commentary on it. Tolofson nonetheless rejected the House of Lords’ approach, as we have seen and as will now be discussed in more detail.

3 THE TOLOFSON CASE

The Tolofson decision was actually a determination of two separate appeals with similar facts, Tolofson v. Jensen and Lucas v. Gagnon: they both dealt with automobile accidents that led to interprovincial negligence issues. Tolofson involved a car accident in Saskatchewan where the passengers of one car were both residents of British Columbia. Lucas v. Gagnon involved a car accident in Quebec where the passengers of one car were residents of Ontario. More specifically, the plaintiff in Tolofson, Kim Tolofson, was 12 years old when she and her father were in a car accident. They were both residents of British Columbia, where the car was registered. While the father was driving in Saskatchewan their vehicle hit that of Leroy Jensen, a resident of Saskatchewan, whose car was properly registered in that province. Eight years later, Kim Tolofson brought an action against her father and Mr Jensen in British Columbia (BC). She assumed that the action was doomed to fail if brought in Saskatchewan, as any action there would be statute-barred by a short limitation period under Saskatchewan law. In addition, Saskatchewan law did not permit a gratuitous passenger to recover unless it was proved that the driver had acted with wilful or wanton misconduct as opposed to standard negligence. She therefore sued in BC hoping to have both BC’s law of limitations and its substantive law apply, both being more favourable to her case than Saskatchewan law. The defendants brought an application seeking a determination that the BC court was a forum non conveniens or, alternatively, that Saskatchewan law applied. On appeal from the motions judge’s dismissal of the application,22 the BC Court of Appeal found that the BC courts could hear the case and that the law of the forum should apply.23 20 It is arguable that Chaplin simply went back to the original version of Phillips v. Eyre (1870) LR 6 Q.B. 1 itself, in that Willes J. prefaced his statement of the test by noting that it was a general rule. However, the explicit injection of justice-based exceptions in Chaplin v. Boys could be read as having substantially modified Phillips v. Eyre to the extent of allowing us to say the House of Lords established a new rule. 21 McLean, supra n. 7. 22 (1989), 40 BCLR (2d) 90. 23 (1992), 65 BCLR (2d) 114; 89 DLR (4th) 129.

Torture, Tort Choice of Law and Tolofson 297 In Lucas v. Gagnon, Mrs Gagnon brought an action on her own behalf and as litigation guardian of her children against her husband and the driver of the other car, Mr Lavoie, for injuries suffered in a car accident in Quebec. As Mrs Gagnon’s action was barred by Quebec’s Automobile Insurance Act,24 she brought her action in Ontario arguing that Ontario courts should take jurisdiction and that Ontario law should be applied. The Ontario Court of Appeal found that Ontario law applied to the action against Mr Gagnon because he was a resident of Ontario, but not to the claim against Mr Lavoie, since he was a resident of Quebec and the accident occurred there.25 These two cases met at the Supreme Court of Canada. It is important to highlight a few obvious points regarding the shared fact situations of these cases. First, they dealt with issues of tort law negligence and statutory liability arising out of car accidents. Second, each of the legislatures of the provinces in which the accidents occurred had enacted laws to manage the compensatory schemes of the insurance industry. As such, the sophistication of the regulation of the insurance industry brought interprovincial commerce into play with tort law. Third, as these were both interprovincial cases, the Court only had to choose between the laws of two provinces, not two countries. In his decision, La Forest J. referred to these points repeatedly. They will be useful for us to recall in later discussions of the factors that must be taken into account in choosing the applicable law for international conflicts cases not involving the routine harm caused by car accidents and not involving standardized insurance schemes.

La Forest J.’s Opinion and His Reasons In his judgment for a unanimous Court, La Forest J. discussed the history of the choice of law doctrines in Canada. It is clear from his description and analysis that he found the system unpredictable and lacking in a logical grounding: “What strikes me about the Anglo-Canadian choice of law rules as developed over the past century is that they appear to have been applied with insufficient reference to the underlying reality in which they operate and to general principles that should apply in responding to that reality. Often the rules are mechanistically applied. At other times, they seem to be based on the expectations of the parties, a somewhat fictional concept, or a sense of ‘fairness’ about the specific case, a reaction that is not subjected to analysis, but which seems to be born of a disapproval of the rule adopted by a particular jurisdiction.”26

La Forest J. called for a new rule that would be simultaneously based on the general principles of conflict of laws and responsive to the reality of the public international order. As such, he proceeded, in effect, on the basis of unifying 24 25 26

Automobile Insurance Act, LQ 1977, c. 68, ss. 3–4. (1992), 11 OR (3d) 422; 99 DLR (4th) 125. Tolofson, supra n. 8, at 302.

298 J Orange principles of international law as a whole, public and private. The first general principle he discussed was territoriality according to which “each state has jurisdiction to make and apply law within its territorial limit”.27 The second, and corresponding principle is comity, which requires that a state respect the actions of a foreign state within that state’s territory. Unless there is a breach of some “overriding norm”, the outside state should not second-guess the applicability of a foreign state’s law to events and activities on its own territory.28 As potentially significant as such an “overriding norm” exception might be, La Forest J. did not provide a definition of such norms, although, to be consistent with his focus on principles of the international legal order which structure conflicts analysis, he could easily have had in mind the peremptory norms of public international law known as jus cogens norms.29 He did, on the other hand, give us some sense of the current purpose of comity, namely “accommodat[ion of] the movement of people, wealth and skills across state lines, a byproduct of modern civilisation”.30 Out of the need for a new standard that respected his understanding of territoriality and comity, La Forest J. overruled McLean and stated that the general rule for choice of law in interprovincial tort cases shall be the law of the place where the tortious activity occurred, the lex loci delicti.31 He reasoned that people expect to be governed by the law of the place where they are.32 He rejected exceptions to the lex loci delicti rule, because any exception would lead to uncertainty in the law: “While, no doubt, as was observed in Morguard, the underlying principles of private international law are order and fairness, order comes first. Order is a precondition to justice.”33 La Forest J.’s focus on the predictability and order inherent in Canada’s interprovincial system may be warranted, since a hard-and-fast rule based on the local law would not unduly prejudice parties in actions within Canada. Although provincial laws may differ, we share the same basic legal principles. However, while clearly still wishing to cling as much as possible to the certainty of a strict rule, La Forest J. did acknowledge that an exception to the rule might be appropriate in different circumstances. Both major circumstances he mentioned—where the wrongful act or the damage occur in different jurisdictions and when the claim deals with different nation states—are very relevant to acts of torture. As often happens in today’s global political economy, the repercussions of an act which occurred in one place can be felt in several foreign jurisdictions.34 In 27

Tolofson, supra n. 8, at 303. Ibid. The importance of La Forest J.’s reference to an “overriding norm” will be discussed infra text at n. 51. 29 For some discussion on this linkage, see Llewellyn, supra n. 10. 30 Tolofson, supra n. 8. 31 Tolofson, supra n. 8 at 305. 32 Ibid. at 310. 33 Ibid. at 311. 34 For a discussion of the problems arising from international defamation claims see Craig Martin, “Tolofson and Flames in Cyberspace: The Changing Landscape of Multistate Defamation”, (1997) 31:1 UBCLR at 126. 28

Torture, Tort Choice of Law and Tolofson 299 his judgment, La Forest J. specifically recognised that this situation may require a more flexible standard.35 Furthermore, he noted that international differences between legal regimes can provide more difficult problems than interprovincial differences: “[B]ecause a rigid rule on the international level could give rise to injustice, in certain circumstances, I am not averse to retaining a discretion in the court to apply our own law to deal with such circumstances. I can, however, imagine few cases where this would be necessary.”36

These comments are obiter dicta, as they were not necessary to decide Tolofson. However, these words were supported by the separate but concurring joint opinion of Major and Sopinka JJ., who disagreed with the absolute application of the lex loci delicti rule in international situations.37 At the end of Tolofson, the rule for Canadian choice of law in tort for international cases is unclear. La Forest J.’s disdain for the McLean rule is plain. The majority’s words lead us to believe that the new standard is international choice of laws would be a general rule of lex loci delicti with some room to apply the law of the forum. However, the Court left Canadian judges with no definite rule to apply when faced with international choice of laws problems.

Post-Tolofson A brief glance at the case law since Tolofson demonstrates that the lex loci delicti rule will not govern absolutely in international torts, but that La Forest J.’s rationales from Tolofson are nevertheless holding their ground. The Supreme Court of Canada has referred to Tolofson in two recent decisions: R v. Terry38 and R v. Harr.39 Both dealt with the admissibility of criminal evidence obtained in the United States that would not be admissible in Canada under the Canadian Charter of Rights and Freedoms because of the manner in which the accused was informed of the right to counsel.40 In these cases, Tolofson’s rationales of territoriality and comity were cited, and the Court found that the accused should have expected their activities to be governed by the law of the place where they were. If the Charter were applied to evidence collected in another state by that state’s police, the Court held that there would be an unjustified extraterritorial reach of Canadian law. Although these cases dealt with the interface between constitutional and criminal law and not with tort law, they 35

Tolofson, supra n. 8, at 305. Ibid. at 307. 37 The fact that they saw it as necessary to pen their brief reasons on this point may suggest that they understood La Forest J.’s “discretion” in a “few cases” to be too limiting. 38 [1996] 2 SCR 207; (1996), 135 DLR (4th) 214. 39 [1995] 3 SCR 562; (1995), 128 DLR (4th) 98. 40 Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11, s. 24(2) [hereinafter “the Charter”]. 36

300 J Orange are significant in so far as the Court relied on Tolofson’s rationales in relation to the applicability of Canadian legal norms in an international situation. The Ontario court found an exception to the strict lex loci delicti rule in Hanlan v. Sernesky.41 The Hanlan case, involving a car accident in Minnesota between residents of Ontario, is significant because it finds an exception to the lex loci delicti rule. The Court held that the lex loci delicti rule would work an injustice. The factors that the motions judge considered in making the decision were: (1) that the parties were both resident in Ontario; (2) that the contract of insurance was issues in Ontario; (3) that there was no connection with the State of Minnesota other than that it was the place of the accident; (4) that, although the accident occurred in Minnesota, the consequences to members of the injured plaintiff’s family were directly felt in Ontario; and (5) that the uncontradicted evidence before him was that family law claims of this nature were not permitted under Minnesota law.42 The situation in Hanlan brings to mind the leading US case of Babcock v. Jackson.43 In that case, all parties involved in a car accident in Ontario were residents of New York State. In Ontario, gratuitous passengers could not be compensated by drivers for accidents. The Court found that, despite the accident occurring in Ontario, New York law should be applied because the parties were residents of New York. Similarly, Hanlan found an exception to the lex loci delicti rule based on the injustice that the parties would suffer if the lex loci delicti rule were applied. Although the Court in Hanlan claims that such exception is in accordance with Tolofson, La Forest J. discussed such arguments directly in his judgment. He rejected the notion that parties expect to be governed by their home law when travelling abroad.44 Furthermore, he rejected a public policy exception based on the notion that it would be more convenient for the parties to litigate at home: “These ‘public policy’ arguments simply mean that the court does not approve of the law that the legislature having power to enact it within its territory has chosen to adopt . . . True, it may be unfortunate for a plaintiff that he or she was the victim of a tort in one jurisdiction rather than another and so be unable to claim as much compensation as if it had occurred in another jurisdiction. But such differences are a concomitant of the territoriality principle . . . Why should be allow an exception at all where two residents of the forum fortuitously happen to meet each other head-on on the road? Should luck be on your side because you happen to crash into another Ontario resident while driving in Quebec, instead of crashing into a Quebecer?”45 41 42 43 44 45

[1997] OJ No. 3658 (Ont. Ct. (Gen. Div.) ) [hereinafter Hanlan]. Ibid. at 1. (1963), 12 N.Y. 2d 473 [hereinafter Babcock]. Tolofson, supra n. 8, at 310. Ibid. at 311.

Torture, Tort Choice of Law and Tolofson 301 It seems that the Court in Hanlan found that justice required an exception in just that situation. “Further, I do not read the Reasons of La Forest J. in Tolofson as creating a ‘without exception rule’ which would include accidents occurring in foreign jurisdictions in some circumstances. The circumstances highlighted in this case by the plaintiffs which I consider to be relevant are that the principal parties are both residents of the same province (forum) and the contract of insurance in effect was issued in Ontario. I am not unmindful that at the time this action arose, there was no question but that Ontario law applied. Taking into account a test suggested by Major J. in Tolofson, I am also of the view that in the particular circumstances before me, it can be said that the operation of the lex loci rule would work an injustice. But for the actual location of the incident, there is absolutely no connection with the State of Minnesota.”46

After Tolofson, the courts are struggling to define what rare circumstances should result in an exception to the lex loci delicti rule, with little guidance. In the sections that follow, I will look in more detail at the rationales underlying La Forest J.’s approach to choice of law and discuss how they should guide us in dealing with international issues, in particular with actions which also constitute international human rights violations or international criminal conduct. International human rights violations, I will argue, should be considered among those circumstances where the courts are permitted the discretion to consider applying forum law.

4 APPLYING TOLOFSON ’ S RATIONALES TO INTERNATIONAL HUMAN RIGHTS VIOLATIONS

As already noted, La Forest J. outlines in Tolofson the underpinnings of a new Canadian interprovincial doctrine of choice of law in tort. He refers to several rationales for this doctrine: territoriality, comity, order and the global reality. These rationales should also be considered when creating a Canadian international choice of law rule, but they are somewhat transformed by the different expectations of, and within, the global community as opposed to the interprovincial one. In this section I will discuss these rationales as they apply at the crossroads of international private and public law. I will argue that when confronted with a human rights violation such as torture, the considerations and responsibilities of the Canadian judiciary change.

The Relationship of Territoriality and Comity to Jus Cogens Norms In his judgment, La Forest J. states that territoriality and comity must be acknowledged by any choice of law rule. Territoriality and comity are basic 46

Hanlan, supra n. 41 at 5–6.

302 J Orange concepts in international law which have stood with relatively few exceptions. One of these exceptions, however, is where jus cogens norms are violated. Jus cogens norms are customary law rules which are generally accepted by the community of nations as being so fundamental to humanity that no breach is justifiable: “The major distinguishing feature of [jus cogens] rules is their relative indelibility. They are rules of customary law which cannot be set aside by treaty or acquiescence but only by the formation of a subsequent customary rule of contrary effect.”47

Jus cogens norms prohibit torture, amongst other offences.48 The state of international opinio juris is such that prohibition of torture can clearly be considered a peremptory norm as defined in Article 53 of the Vienna Convention on the Law of Treaties: “. . . [It] is a norm accepted and recognized by the international community of States as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character.”49

Since jus cogens norms reflect the deepest values of the international community, when such a norm is violated, it is in the interest of all states to rectify and repair the situation to whatever extent possible.50 It is significant that La Forest J., despite the generally territorial case of his judgment in Tolofson, seems to have deliberately signalled the relevance of peremptory norms of international law in the private international law context of choice of law analysis, if somewhat cursorily, in saying: “On the international plane, the relevant underlying reality is the territorial limits of law under the international legal order. The underlying postulate of public international law is that generally each state has jurisdiction to make and apply law within its territorial limits. Absent a breach of some overriding norm, other states as a matter of ‘comity’ will ordinarily respect such actions and are hesitant to interfere with what another state chooses to do with those limits.”51

The language of “overriding norm” is too close conceptually to the discourse of peremptory norms of international law to be treated as a mere coincidence or as an inadvertence.52 47 I Brownlie, Principles of Public International Law, 4th edn., (New York: Oxford University Press, 1990) at 513 [hereinafter “Brownlie”]. 48 J H Burgers and H Danelius, The United Nations Convention against Torture: A Handbook on the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (Dordrecht: Martinus Nijhoff Publishers, 1988) at 12. 49 (1969) 8 ILM. 50 It may be sufficient for a norm to qualify as “overriding” in a sense relevant to choice of law if it is “merely” an erga omnes norm and not also a jus cogens norm. For the distinction and the observation that jus cogens norms are also erga omnes but not vice versa, see A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality”, chapter 6 of this volume, and J Llewellyn, supra n. 10. 51 Tolofson, supra n. 8, at 1047 (emphasis added). 52 See a similar point in J Llewellyn, supra n. 10.

Torture, Tort Choice of Law and Tolofson 303 When a Canadian court is faced with an international tort issue that also can be classified as a violation of a jus cogens norm, it is important that the court consider international norms in determining which law to apply. Although looking beyond the host state’s laws may be an extraterritorial move on the part of our courts, it is more accurate to see this as a legitimate effort to internationalise choice of law. Apart from La Forest J.’s reference to overriding norms, support for this kind of argument can be found in Tolofson where he refers to the Hague Convention on Traffic Accidents.53 It may be that the Hague Convention can be drawn upon for normative support when Canada is not even a party to it, and when its content is by no stretch of the imagination reflective of jus cogens norms. If so, how much more justified is recourse to jus cogens international law in seeking international principles to determine choice of law? Indeed, while this chapter’s minimal premise is that at least peremptory international law norms should be available to help structure tort choice of law principles, we would be justified to follow La Forest J.’s lead and consult international human rights agreements that overlap with interests tort law seeks to protect. This is especially justified when such treaties are widely ratified. On this more optimal internationalist approach, it would not be necessary to ensure such a broadly ratified human right has a parallel high status as jus cogens. While jus cogens norms can justify overriding whatever law is selected by a given choice of law rule, general international human rights law should be available in helping formulate what the most desirable rule should be in the first place. After World War II, numerous human rights instruments were adopted which can be useful references for Canadian courts’ interpretations of Canadian law. The Universal Declaration of Human Rights aimed to set out the basic human rights for all people worldwide.54 The following articles are particularly relevant to an analysis of torture as a potential tort: “Article 3: Everyone has the right to life, liberty and security of person. ... Article 5: No one shall be subjected to torture or to cruel, inhuman or degrading treatment or punishment. ... Article 8: Everyone has the right to an effective remedy by the competent national tribunals for acts violating the fundamental rights granted him by the constitution or by law.”55

While the Universal Declaration is a non-binding document, its principles have taken on legal significance, as states as well as international courts and 53 Mémorandum Dutoit in Actes et documents de la onzième session, as cited in this form in Tolofson, supra n. 8, at 306. 54 Universal Declaration of Human Rights, adopted 10 Dec. 1948, G.A. Res. 217A (III), U.N. GAOR, 3rd Sess. Resolutions, part 1, at 71, U.N. Doc. A/810 (1948), reprinted in (1949) 43 Am J Int’l L 127 (Supp.) [hereinafter “the Universal Declaration”]. 55 Ibid. (emphasis added).

304 J Orange tribunals have recognised, and acted on the basis of, its moral force.56 In 1988, for example, a sub-commission of the UN Commission on Human Rights defined the right to an effective remedy to include restitution, compensation and rehabilitation.57 Canadian courts could offer obvious support to torture victims’ rights to restitution by recognising their claims in tort law. The International Covenant on Civil and Political Rights (ICCPR), to which Canada is a State Party, along with a large majority of the other states in the world today, contains a binding treaty commitment to the prohibition on torture.58 Article 7 deals specifically with torture in providing that “[n]o one shall be subjected to torture or to cruel, inhuman or degrading treatment or punishment.” Article 4 of the ICCPR provides that States Parties may take some measures derogating from their obligations under this covenant in times of public emergency. However, the fact that article 4(2) prohibits any derogation from Article 7 helps confirm its status as a jus cogens norm.59 The most comprehensive treaty on torture is the UN Convention Against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment (the CAT).60 Each state party is required to make torture a criminal offence, and to take jurisdiction over such offence when: it occurs in that state’s territory; the offender is a national of that state; the victim is a national of that state and the state considers this jurisdictional basis to be appropriate; and finally, when the alleged offender is present in that state.61 Article 7 of the CAT requires the prosecution of any alleged offender for which jurisdiction is grounded. The

56 I Brownlie (ed.), Basic Documents in International Law, 4th ed., (New York: Oxford University Press, 1995) at 255. 57 Stephens, supra n. 6 at 602. 58 (1976), 999 UNTS 171, [1976] CTS 47. 59 The above-mentioned agreements form the foundations for human rights documents which prohibit torture. If Canadian judges required more proof of the international importance of eradicating torture, then they could refer to many other regional conventions, and to conventions regarding the treatment of persons during wartime. For example, the African Charter on Human and Peoples’ Rights Text: 21 ILM (1982) 58; The Universal Islamic Declaration of Human Rights; the European Convention for the Protection of Human Rights and Fundamental Freedoms, (1955) 213 UNTS 221, Europ. T.S. No. 5 (entered into force 3 Sept. 1953) reprinted in (1994) 33 ILM 943, 960, as amended by Protocols Nos. 3 and 5, which entered into force on 21 Sept. 1970 and 20 Dec. 1971 respectively [hereinafter “the European Convention”]; European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment, adopted 26 November 1987, ETS No. 126; and the Inter-American Convention to Prevent and Punish Torture. All of the foregoing contain provisions prohibiting torture. On the laws of war, see the Declaration on the Protection of Women and Children in Emergency and Armed Conflict, the Geneva Convention Relative to the Treatment of Prisoners of War, 6 U.S.T. 3316, 75 U.N.T.S. 135 (1949); and the Geneva Convention Relative to the Protection of Civilian Persons in Time of War, 6 U.S.T. 3516, 75 U.N.T.S. 28 (1949). 60 Adopted 10 Dec. 1984, GA Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 art. 1, at 197 [hereinafter “CAT”]. Canada signed the CAT on 23 Aug. 1985 and ratified it on 24 June 1987: Outlawing an Ancient Evil: Torture; Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment; Initial Report of Canada, (Multiculturalism and Citizenship Canada: Ottawa, 1989) at 46. 61 CAT, Supra n. 60, articles 4 and 5.

Torture, Tort Choice of Law and Tolofson 305 breadth of the CAT’s criminal provisions provides us with a clear indication of the gravity with which international law treats torture. Significantly, the CAT does not limit states parties’ obligations to criminal law. In particular, article 14 states: “1. Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation. 2. Nothing in this article shall affect any right of the victim or other persons to compensation which may exist under national law.”

The means by which redress and compensation are to be pursued are not detailed. However, while compensation is sometimes provided through punitive damages in criminal law, it is more typically a civil law remedy. The requirement that the victim be compensated in an amount that permits the fullest possible rehabilitation dovetails with the basic restitutionary goals of tort law remedies. Both put the plaintiff as far as possible in the position in which (s)he would have been had the tort not occurred.62 While it may be argued that article 14(1) does not require state parties to provide tort remedies for torture victims, neither does the article prohibit such remedies due to the presence of the disclaimer in article 14(2). Arguably, article 14(2) can even be interpreted as a grant of permission to establish universal civil jurisdiction over torture.63 According to the CAT, states do not have to defer to the territorial sovereignty of states that participate in torture, with respect to a given violation. A lack of normative force for territoriality negates comity, so that La Forest J.’s Tolofson rationales are transformed. In Filártiga, the Court accepted this reasoning, as Harold Koh has explained: “Filártiga convincingly rebutted the comity, separation-of-powers, and incompetence objections to domestic judicial decision of human rights cases. . . . ‘For purposes of 62 In Canada, the domestic legal system does provide the full rehabilitation standard of the CAT. The Criminal Code stipulates a right to restitution: RSC 1985 c. C-46 ss. 725 & 737. Section 737 allows the court to apply a victim fine surcharge to an offender convicted of an offence under the Criminal Code. The surcharge is in addition to any other penalty, not just fines. The maximum victim fine surcharge is $10,000. This provision is rarely used or, when used, successful: A M Linden, Canadian Tort Law, 6th ed. (Toronto, Butterworths, 1997) at 48 [hereinafter “Linden”]. And, while compensation regimes for victims of crime have been enacted in Canada, levels of compensation do not meet the full restitution standard set out in the CAT. See the Compensation for Victims of Crime Act, RSO 1990, c. C-24. Section 5 of the Act provides compensation to the victim and his or her dependants where there has been a commission of a crime of violence against the Criminal Code. The Act goes further than the Criminal Code, as the Criminal Injuries Compensation Board may award compensation whether or not any person is prosecuted or convicted of an offence (s. 16(1) ), and the maximum award is $25,000 (s. 19). 63 For some exploration of whether article 14 can be viewed as containing a power even if not an obligation, see A Byrnes, “Civil Remedies for Torture Committed Abroad: An Obligation under the Convention against Torture?”, chapter 20 in this volume, and M Sornarajah, “Linking State Responsibility for Certain Harms Caused by Corporate Nationals Abroad to Civil Recourse in the Legal Systems of Home States”, chapter 18 in this volume.

306 J Orange civil liability,’ the court declared, ‘the torturer has become—like the pirate and slave trader before him—hostis humani generis, an enemy of all mankind.’ By so saying, the Second Circuit reaffirmed the Nuremburg ideal: that torture (like genocide) is never a legitimate instrument of state power. Thus, official torturers may not invoke comity or cloak themselves in state sovereignty to avoid individual responsibility to their victims before a court of law.”64

When confronted with torture, international law does not simply permit, but affirmatively requires that states take action to protect victims and promote justice. States’ jurisdiction is expanded where needed to allow such obligations to be met. The CAT regime is consistent with more general normative developments: in situations of international concern, jurisdiction can be expanded. Hari Osofsky has usefully summarised this normative trajectory: “With the development of the international human rights regime, state sovereignty diminished. Universal jurisdiction was no longer confined to violations of the high seas, but rather extended to behavior occurring purely within a state’s borders. International law began to recognise that some norms were so fundamental that they could not be legally violated anywhere and that all nations have jurisdiction over their violation.”65

In order to deal with international crimes, the principle of “universal jurisdiction” gives any state the power to prosecute those who have allegedly committed international offences. Universal jurisdiction is triggered by the concerns of all states regarding international crimes.66 When dealing with civil remedies, jurisdiction is also satisfied by the international community’s shared interest in holding human rights violators accountable.67 Just as for some crimes, notably torture, the exercise of states’ “permissive” jurisdiction has become not just permissive, but obligatory. It appears that international law not only permits but obligates states to provide the international avenues for the victims of human rights abuses to litigate redress as civil tort claims.68 Order and Global Reality In the Tolofson decision, La Forest J. cited order as a key rationale in establishing rules in conflict of laws.69 However this may apply among provinces of a 64 H H Koh, “Transnational Public Law Litigation”, 91991) 100 Yale LJ 2347 at 2367 (notes omitted). A separate point is that, even were international law not brought to bear on the duty to respect Paraguay’s territorial jurisdiction, lack of respect was even less at issue because Paraguay’s own law did not validate the torture. Koh has put it as follows: “The court showed no disrespect to Paraguay by trying Peña-Irala, because his acts were illegal even under Paraguayan law.” 65 H M Osofsky, “Domesticating International Criminal Law: Bringing Human Rights Violators to Justice”, (1997) 107 Yale LJ 191 at 205. 66 Stephens, supra n. 6 at 586. 67 Ibid. at 600. 68 Ibid. at 601. 69 “While, no doubt, . . . the underlying principles of private international law are order and fairness, order comes first. Order is a precondition to justice.” Tolofson, supra n. 8, at 311.

Torture, Tort Choice of Law and Tolofson 307 single country, the certainty and predictability promoted by prioritising order may not be possible, nor desirable between different countries having radically different commitments to basic human rights underlying their laws and government policies. La Forest J. relied on the reality of the global situation as supportive of the prioritising of order over case-by-case justice. He stated that the lex loci delicti rule has the virtue of being stable, predictable and respectful of the territorial sovereignty of other nations, and therefore of falling into line with the requirements and practicalities of international law.70 He warned the courts against an uncertain rule: “If other states routinely applied their laws to activities taking place elsewhere, confusion would be the result. In our modern world of easy travel and with the emergence of a global economic order, chaotic situations would often result if the principle of territorial jurisdiction were not, at least generally, respected. Stability of transactions and well grounded legal expectations must be respected. Many activities in one state necessarily have impact in another, but a multiplicity of competing exercises of state power in respect of such activities must be avoided.”71

La Forest J. was concerned to align tort civil suits with the “underlying reality of the international legal order”.72 Note, however, that the realities to which he refers are those of the global economic order and related transactions, as well as of “easy travel” of the more privileged members of global society (including those whose mobility is tied to economic globalisation). Predictability in economic and business matters is certainly helpful for people, companies and states trying to organise their affairs. Victims of human rights violations, however, live in a very different reality, one where they have little or no ability to structure the situations surrounding them. Physical abduction and beatings are poignant examples of the loss of control. A strict legal rule that does not allow an exception for fairness or justice in cases of torture runs too high a risk of making the legal system useless for the people who need its protections most. This is not to say that Canadian judges should not look at global realities and at the structure of the international legal system when addressing torts that are also human rights violations. But, what they will find is different from the picture portrayed by La Forest J. It will emerge that, as a normative matter, international law places ever greater value on protecting individual human rights, but also that, as an institutional reality, the systems set up to deal with human rights issues do not effectively address the needs of their constituents.

70 “It is to the underlying reality of the international legal order, then, that we must turn if we are to structure a rational and workable system of private international law.” Tolofson, supra n. 8 at 303. 71 Ibid. at 305 (emphasis added). 72 Supra n. 70.

308 J Orange Characterising Torture Canadian law’s compartmentalization of human rights violations into public law categories, as opposed to the private realm of tort law, should not apply to the characterisation of tort claims in international conflict of laws cases. An international case must be characterised by the forum, but with a view to the laws of all applicable jurisdictions. In a torture case, these sources consist of the forum law, the law of the place where the initial injury occurred, and any relevant international law that applies. Torture becomes problematic at this stage because it can be characterised in multiple ways—as a crime, as a human rights violation and as a tort. In a choice of law analysis, trying to characterise the facts as falling into a certain parcel of law causes a dilemma. Jurisdictions have rules for choice of law which differ depending on the issue at hand. For example, the conflicts laws for marriage are different from those for contract, even though a marriage is a type of contract. In order to characterise the action as tort, one needs to identify in advance the place of the law that will apply. For example, the Canadian delineation of the boundaries of tort law may be different from delineation of civil delict in a civil law jurisdiction such as Japan. We cannot, therefore, know if there is a legal relationship arising from a certain fact situation until we have selected an initial legal system to apply to our problem.73 But—and here is the dilemma—we cannot select the law to apply until we characterise the problem and look at the choice of law rules of the forum. In other words: “[w]e have to know the juridical nature of the real facts before our conflictual rules for choice of law will indicate the governing dispositive rules or system, and yet only dispositive rules can define the juridical nature of the real facts for the purposes of this choice.”74

To challenge the face of this paradox, we must look to all possible choices of law before we characterise a certain bundle of facts. Falconbridge suggests a “tentative characterisation” before selecting the proper law. First, we should be informed of the content of the two laws that are candidates for the governing law, and then characterise the legal question by taking both legal systems into account.75 Lederman suggests that it is methodologically important to list the possible contending characterisations by the candidate laws and then make a choice between them.76 He recognises that it is basically the lex fori that is used to classify the facts, but that the forum judge should make the characterisation with an eye to the broad categories of law which may accommodate foreign concepts.77 73 J D Falconbridge, “Conflict Rule and Characterisation of Question”, (1952) 30 No. 2 Can BR 103 at 106 [hereinafter “Falconbridge”]. 74 W R Lederman, “Classification in Private International Law”, (1951) 29 Can BR 3 at 17 [hereinafter “Lederman”]. 75 Falconbridge, supra n. 73, at 117. 76 Lederman, supra n. 74, at 23. 77 Lederman, supra n. 74, at 24.

Torture, Tort Choice of Law and Tolofson 309 Although both Falconbridge and Lederman persuasively contend that a judge should look to possible foreign laws and their characterisations before finally characterising the issue, they provide little guidance on how to make the final decision. Castel recognises that such choices are often made with a view to achieving justice in the final result: “As will be noted, characterisation is not a purely mechanical process. If more than one characterization is available for a set of facts, the choice between the characterisations may turn upon the court’s desire to achieve justice in the particular case or preference for one rule of law over another.”78

Characterising an issue with a goal of achieving fairness in the case seems to be the type of uncertain rule that La Forest J. would dislike. Indeed, it can be imagined that he might respond to these three Canadian conflicts scholars by pointing out how messy their approaches to characterisation are compared to his advocacy of the need for order over chaos in private international law. Indeed, it is true that a characterisation mixing the forum’s understanding of legal issues with its knowledge of the law of the other pertinent jurisdictions cannot be reduced to a single phrase. As Falconbridge puts it: “It is important to note, however, that the definition of the scope of conflict rules in accordance with the law of the forum does not mean that the definition should be based solely on the strictly domestic or legal concepts of the law of the forum. Conflicts rules are of course the expression of the policy of the law of the forum for the purpose of the conflict of laws and must be interpreted in such a way to carry out that purpose. They must be interpreted sub specie orbis—from a cosmopolitan or world-wide point of view—that is, in such a way as to render the conflict rules susceptible of application to foreign rules of law, which may be based on concepts that do not coincide exactly with the domestic or local concepts of the law of the forum.”79

Falconbridge’s mid-century “world-wide point of view” is far more in keeping with the evolving nature of the interaction of legal systems at the millennium than La Forest J.’s overly territorialist and order-bound approach. However, the time may have come to give priority in characterisation less to melding two legal systems’ categories and more to using relevant international law as a kind of law of “choice of category” for characterisation. Such privileging of international law is consistent with Falconbridge’s valuation of a “cosmopolitan” approach, an approach articulated well before international law had evolved to the point it now has regarding torture. When a Canadian court is faced with a claimant who, like Filártiga, wants to sue in tort for torture suffered abroad, the court must decide if this claim is capable of being characterised in Canadian law as a tort (of torture). Assuming there is no existing nominate tort of torture, but only kindred traditional torts like those of assault and battery, the instinct of a Canadian court might be to draw 78 79

J G Castel, Conflict of Laws, Cases, Notes and Materials, 6th ed. (1987), pp. 1–35. Falconbridge, supra n. 73, at 112.

310 J Orange on the legal tendency in Canada not to translate public law values into torts.80 However, a truly cosmopolitan characterization, where the international legal order lays a key role, requires a different analysis. First of all, if a Canadian court disallowed a tort claim for torture that occurred in Canada, it would likely be breaching Canada’s obligations to ensure that its legal system has an enforceable right to adequate legal compensation for torture, pursuant to Article 14 of the CAT.81 From the perspective of the CAT, then, Canada’s legal system has an obligation to have a civil category of “torture”—at least where no comprehensive statutory compensation regime is in place. Secondly, under international law, states are not prohibited from allowing civil claims characterisable both as criminal offences or human rights violations. States (and courts exercising common law duties) are free to recognise tort liability in their domestic legal systems for those who have violated international criminal law or public law human rights norms.82 Thirdly, when a Canadian judge is attempting to characterise a civil law case of international torture, the judge should look at the applicable foreign laws. However, the normative status of torture in international law as jus cogens should affect the weight to be given to the foreign law’s lack of civil liability for torture in a triangulated analysis of Canadian law, foreign law and international law. While the article 14 and general international law duty to afford civil redress may not share the jus cogens status of the basic norm of prohibition of state involvement in the actual torture, the close proximity of the duty to have civil redress for torture to the basic jus cogens prohibition of acts of torture speaks against the judge giving any weight to the foreign law in the characterisation process. As a matter of overriding international public policy, the foreign legal system’s lack of a category of torture in tort need not be given legal effect by a foreign court.83

80 Board of Governors of Seneca College v. Bhadauria (1981), 124 DLR (3d) 193 [hereinafter Bhadauria]. The SCC held in Bhadauria that the comprehensive human rights statute barred human rights torts. See the discussion of S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, chapter 14 in this volume, and E M Hyland, “International Human Rights Law and the Tort of Torture: What Possibility for Canada?”, chapter 15 in this volume. 81 Article 14 reads:

“1. Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victims as a result of an act of torture, his dependents shall be entitled to compensation. 2. Nothing in this article shall affect any right of the victim or other persons to compensation which may exist under national law.” 82 83

Stephens, supra n. 6, at 587. For a similar reference to international public policy, see M Sornarajah, supra n. 63.

Torture, Tort Choice of Law and Tolofson 311 Summary of the Legacy of Tolofson In the Tolofson decision, La Forest J. attempted to rectify the inconsistencies of past Canadian choice of law rules by establishing one precise rule which can be applied to all situations. The above discussion aimed to demonstrate that the rationales upon which La Forest J. based the lex loci delicti rule are transformed by a fuller picture of the international legal realm. Territoriality and comity are changed by public international obligations. The global reality is unstable for the victims of human rights abuses, and it is nearly impossible for a claimant to receive compensation at the international level. A straight application of the lex loci delicti rule ignores the order created by the public international law system. Similarly, Canadian judges cannot characterise the issue of a conflicts problem without looking to all relevant laws. Canadian law requires a more complex and sensitive approach to international problems that implicate public international law obligations. In the upcoming sections, I discuss what law should be applied to a tort characterised as torture, in light of Tolofson’s own signalling of caveats and in light of principles of US and UK law.

5 COMPLEX FACT SITUATIONS AFTER TOLOFSON

An issue of civil liability that also implicates international human rights law and international criminal law, such as torture, changes the rationales from those applicable to the traditional domestic or to the international commercial spheres. A situation of complex facts, either where the wrongful conduct and the harm occur in different places or where the harm occurs in several places, may also require a different choice of law analysis. As indicated in the Introduction, La Forest J. referred to this issue in Tolofson: “There are situations, of course, notably where an act occurs in one place but the consequences are directly felt elsewhere, when the issue of where the tort takes place itself raises thorny issues. In such a case, it may well be that the consequences would be held to constitute the wrong. Difficulties may also arise where the wrong directly arises out of some transnational or interprovincial activity. There territorial considerations may become muted; they may conflict and other considerations may play a determining role. But that is not this case.”84

The harm caused by torture may occur in several jurisdictions, causing debate over where the harm is felt. For example, a woman is physically beaten and threatened in one country, then flees to Canada. The consequences of the torture continue in Canada, as a physical disability and psychological damage. It may be that some psychological damage manifests itself long after the initial act occurred. It is then important to establish whether the harm occurring in 84

Tolofson, supra n. 8, at 305.

312 J Orange Canada is enough to ground jurisdiction and establish that Canadian law should govern the dispute.85 It is important to note at the outset that the issue of where the tort occurred has never been the subject of a Canadian judicial decision regarding choice of law.86 In the conflicts setting, the place of the tort has been discussed in terms of issues of jurisdiction, particularly related to service ex juris, as well as in terms of application of limitation periods.87 Now that Tolofson has established the place of the tort as the key to choosing the applicable law, we can expect that the problem will present itself. Presently, the Canadian judiciary can draw from general tort principles, cases defining the place of the tort in other conflict of law areas, and cases regarding the place of the criminal act for the establishment of jurisdiction. In the criminal law case of Libman v. The Queen,88 La Forest J. lays out the possible tests for establishing jurisdiction which could be transposed to a tort problem: “One is to assume that jurisdiction lies in the country where the act is planned or initiated. Other possibilities include the place where the impact of an offence is felt, where it is initiated, where it is completed, or again where the gravamen or essential element of the offence took place. It is also possible to maintain that any country where a substantial or any part of the chain of events constituting an offence takes place may take jurisdiction.”89

In the end, La Forest J. decided that all that is necessary for Canadian courts to take jurisdiction in a criminal case—and remain within the Criminal Code’s directive that the Code’s scope of application is territorial, unless expressly indicated to be broader—is that “a significant portion of the activities constituting the offence took place in Canada” so that there is a “ ‘real and substantial link’ between the offence and this country”.90 Of course, the relationship of criminal law to our society as a whole is very different from the relationship between the tortfeasor and his or her victim. Criminal law may treat preparatory conduct and attempted harm as criminal, whereas the purpose of tort law in general is to compensate the person injured.91 Thus, in negligence actions, the focus is on the harm done to the plaintiff. As Viscount Simonds stated in the famous Wagon Mound case: “It is, no doubt, proper when considering tortious liability for negligence to analyse its elements to say that the plaintiff must prove a duty owed to him by the defendant, a 85 Although the issue is well beyond my expertise, it would make sense that a litigant in such circumstances should investigate medical issues related to psychological harm. The litigant should also investigate whether a particular mental illness can be considered a discrete problem, as opposed to illnesses materialising at other times but connected to the same incidence of torture. If one illness can be considered a discrete harm, then the tort may be considered “born” in Canada. 86 J G Castel, Canadian Conflict of Laws, 4th ed. (Toronto, Butterworths, 1997) at 692 [hereinafter Castel]. 87 Ibid. 88 [1985] 2 SCR 178 [hereinafter Libman]. 89 Ibid. at 185. 90 Ibid. at 213. 91 J G Fleming, The Law of Torts, 8th ed. (NSW, The Law Book Company, 1992) at 1.

Torture, Tort Choice of Law and Tolofson 313 breach of that duty by the defendant and consequent damage. But there can be no liability until the damage has been done. It is not the act but the consequence on which the tortious liability is founded. Just as (as it has been said) there is no such thing as negligence in the air, so there is no such thing as liability in the air.”92

Although one may instinctively locate the tort where the wrongful act occurred, Viscount Simonds’ statement and the underlying rationale of tort law to compensate injury actually done leads to the conclusion that the locus of the tort should be defined as the place of the harm. Yet, debate among scholars on this point has not led to any single test in the conflict of laws context. As the Supreme Court of Canada noted in Moran v. Pyle: “[t]he task of determining the situs of a tort is of some difficulty. A number of tests which one might apply are to be found in the cases and in the learned journals but none has been free of criticism and I think it fair to say that no clear principle has emerged.”93

In Moran, a jurisdiction case in which the Court had to interpret the scope of an ex juris service rule that allowed such service where a tort took place “in” the jurisdiction, Dickson J. (as he then was) put forth a flexible test for establishing the place of the tort, criticising stricter rules: “Generally speaking, in determining where a tort has been committed, it is unnecessary, and unwise, to have resort to an arbitrary set of rules. The place of acting and the place of harm theories are too arbitrary and inflexible to be recognised in contemporary jurisprudence.”94

He then established a test for locating the tort of careless manufacture based on the principle that it is not “inappropriate to regard a tort as having occurred in any country substantially affected by the defendant’s activities or its consequences and the law of which is likely to have been within the contemplation of the parties”.95 La Forest J. would subsequently analogise from the Moran test in articulating the earlier-mentioned test for territoriality in the criminal law case of Libman.96 The application of this test to torture would determine that Canada is substantially affected by the consequences of a defendant’s activities, which cause mental illness in one of its residents or citizens. This is true for such 92 Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd. (The “Wagon Mound”) [1961] AC 388 at 425. 93 Moran v. Pyle National (Canada) Ltd. [1975] 1 SCR 393 at 397 [hereinafter Moran]. 94 Moran, supra n. 93 at 408. 95 Ibid. at 409. 96 Supra n. 88. Note that there is another difference between Moran and Libman, other than tort law context versus criminal law context. This is, that Moran dealt with adjudicative jurisdiction and Libman with prescriptive jurisdiction. That being said, in the criminal law context, adjudicative and prescriptive jurisdiction become partly fused, given that once a court determines that the Criminal Code applies, it acquires jurisdiction to adjudicate as long as the requisite jurisdiction over the person is also present. In the tort context, a court assumes the power to adjudicate on the basis of the Moran test, but this does not necessarily mean that the court automatically applies that jurisdiction’s tort law. It could conceivably take jurisdiction and apply another legal system’s rules, in whole or in part, or apply a hybrid set of rules and principles.

314 J Orange reasons as the general state interest in the well-being of persons with close links to the society in question and specific state expenditures related to mental and physical healthcare for the victim. Canada is naturally concerned about the well-being of its people and the social and fiscal costs of psychological injuries—health care, welfare, underemployment costs, to name a few. Establishing, per Moran, that it was also likely in the reasonable contemplation of the parties that Canadian law would apply to the action may be a more arduous task. If the initial torture occurred in Paraguay, unless the parties already had some connection to Canada, Canadian law would be far from their minds. It is, arguably, at this stage of contemplation at which La Forest J. stopped in Tolofson when he adopted as a generally true premise that people expect the law of the territory on which they are located to apply. It may be countered that it is in the reasonable contemplation of any torturer that his or her victim may leave the country (as a result of the torture) and take the psychological damage with him or her. Although the torturer could not reasonably foresee any given law applying other than that of the territory in which he tortured, he or she could see that other nations’ laws could well apply. More significantly, it is increasingly reasonable for torturers to expect that international law will govern their conduct whether or not they make a mental distinction between criminal responsibility and civil liability. This line of reasoning is consistent with Moran in two respects. Firstly, no one country alone can prescribe its laws but rather any country with a real and substantial connection to the tort can do so. This being so, turning to international law to supply principles of tort liability becomes a non-arbitrary way to choose applicable law in contrast to what results when two or more countries’ divergent tort regimes can end up being applied to the same root conduct. Secondly, international law itself can fit within the Moran framework in that the erga omnes character of the torture prohibition means all states are substantially affected by breach of the norm. The fact that the actual act of torture cannot correctly be characterised as negligence may cause additional complications.97 Any tort of torture would consist of the established intentional torts of assault and battery and, where it exists as a recognised tort, intentional infliction of mental suffering. Although the characterisation of torture as assault and battery is a rough compartmentalisation, it does give us a legal framework from which to work. As La Forest J. stated in another of his judgments with regard to the offence of incest in M. (K) v. M. (H): “Although assault and battery can only serve as a crude legal description of incest . . . there is no question . . . that incest constitutes an assault and 97 Note however that negligence principles may be relevant to the responsibility of state officials further up the chain of authority who do not do the actual torturing, but also do nothing to stop it. See the contribution of V Oosterveld and A Flah, “Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability”, chapter 16 of this volume, on the responsibility of leaders.

Torture, Tort Choice of Law and Tolofson 315 battery.”98 Crucially, intentional torts, with the exception of intentional infliction of mental suffering, differ from negligence actions in that they generally do not require proof of actual harm to the plaintiff. In this, they resemble criminal law’s attempt offences. Once the wrongful conduct is proved, the harm to the plaintiff is presumed.99 Tort law’s general focus on remedying actual harm reappears, however, in that the amount awarded in damages will reflect the gravity of harm suffered.100 A counterpoint is that, when harm is intentionally caused, the defendant will be liable for all damage caused, whether remote, unforeseeable or unintentional.101 If no injury is required to prove an intentional tort, and foreseeability of injury is not necessary, then the place of the harm may be of less importance in discerning the place of this kind of tort for choice of law. The place of the tort, in that case, may only be considered the place of the wrongful act. Although there are no Canadian cases distinguishing intentional torts from negligence in the conflict of laws realm, this concept certainly requires more investigation. Such investigation must take into account the paradox that a focus only on the act as locus helps territorialise the (arguably) more serious intentional torts while the focus on harm helps deterritorialise the (arguably) less serious tort of negligence. But a reconciliation seems, intuitively, available. Where actual harm occurs, as it will in cases of actual torture, harm should remain a relevant basis for the legal interests of foreign countries to be triggered. Furthermore, where torture is at stake, the presumption of harm, without need for proof, is eminently justifiable. It follows that mere presence of the victim in another country triggers a presumptive adjudicative interest. At the same time it is international law itself that should apply, as it has the primary role in establishing a global prohibition on the practice of torture whatever the form or degree of harm.

6 LESSONS FROM OTHER JURISDICTIONS

Both the United States and the United Kingdom have developed multilayered tort choice of law rules that allow considerations, other than those pointing to the lex loci delicti, to be taken into account. In this section I will briefly recount the main sources of such laws, analyse the factors considered, and discuss their 98

[1992] 3 SCR 3; 14 CCLT (2d) at 16. M Kerr, J Kurtz and L M Olivo, Canadian Tort Law in a Nutshell (Scarborough, Thomson, 1997) at 11. 100 Ibid. 101 Linden, supra n. 62 at 45. “If physical contact was intended, the fact that its magnitude exceeded all reasonable or intended expectations should make no difference. To hold otherwise . . . would unduly narrow recovery where one deliberately invades the bodily interests of another with the result that the totally innocent plaintiff would be deprived of full recovery for the totality of the injuries suffered as a result of the deliberate invasion of his bodily interest . . . the intentional wrongdoer should bear the responsibility for the injuries caused by his conduct and the negligence test of “foreseeability” to limit, or eliminate, liability should not be imported into the field of intentional torts”: Bittel v. Yim (1978), 20 OR (2d) 617 at 628–9, as cited in Linden, at 45. 99

316 J Orange implications for an international human rights tort in light of La Forest J.’s own reference to “overriding norms” in tort choice of law.

The United States The US Second Restatement of Conflict of Laws (the Second Restatement) sets forth a list of factors for judges to consider when analysing a choice of law tort problem.102 The Second Restatement is an unofficial and, as such, non-binding document drafted by the American Law Institute, but is generally accepted as a summary of principles of American law. It is the source most often used and accepted as persuasive by judges resolving choice of law cases in the United States.103 With regard to tort law, section 145 of the Second Restatement provides the following general principle: “(1) The rights and liabilities of the parties with respect to an issue in tort are determined by the local law of the state which, with respect to that issue, has the most significant relationship to the occurrence and the parties under the principles stated in s. 6. (2) Contacts to be taken into account in applying the principles of s. 6 to determine the law applicable to an issue include: (a) the place where the injury occurred; (b) the place where the conduct causing the injury occurred; (c) the domicile, residence, nationality, place of incorporation and place of business of the parties; and (d) the place where the relationship, if any, between the parties is centered. These contacts are to be evaluated according to their relative importance with respect to the particular issue.”104

This general rule applies to all torts and all issues in tort law.105 Section 6, to which section 145 refers, sets out the general choice of law principles: “(1) A court, subject to constitutional restrictions, will follow a statutory directive of its own state on choice of law. (2) When there is no such directive, the factors relevant to the choice of the applicable rule of law include (a) the needs of the interstate and international systems; (b) the relevant policies of the forum; (c) the relevant policies of other interested states and the relative interests of those states in the determination of the particular issue; (d) the protection of justified expectations; 102

Second Restatement, supra n. 13. L L McDougal III, “Toward the Increased Use of Interstate and International Policies in Choice-of-Law Analysis in Tort Cases under the Second Restatement and Leflar’s ChoiceInfluencing Considerations”, 91996) 70 Tulane LR 2465 at 2466 [hereinafter “McDougal”]. 104 Second Restatement, supra n. 13, s. 145. 105 Ibid., s. 145 cmt. a. 103

Torture, Tort Choice of Law and Tolofson 317 (e) the basic policies underlying the particular field of law; (f) certainty, predictability and uniformity of result; and (g) ease in the determination and application of the law to be applied.”106

Comment d to section 6 gives an explanation for the reasons the American Law Institute considers the “needs of the interstate and international systems” (section 6(2)(a) ) important considerations in a choice of law exercise.107 The Comment states that the primary function of choice of law rules is to make the “interstate and international system work well”, promoting “harmonious relations between states” and “facilitat[ing] commercial intercourse between them”.108 Courts should also consider the “needs and policies of other states and of the community of states”.109 The Comment then states, that when the needs policies of the international system are taken into account in forming choice of law rules, “the values of certainty, predictability, and uniformity of result” will be furthered.110 The words of this Comment bear a striking resemblance to La Forest J.’s rationales in Tolofson, but in American law the same rationales lead to a different rule. La Forest J. sees that order will result from a strict rule that applies only the lex loci delicti, whereas the American Law Institute finds that, by considering the needs of international systems, order will result. The American Law Institute’s approach seems to consider alignment between international legal order and national law, including private international law, as itself an element of order. Unlike La Forest J.’s approach in Tolofson, this approach does not conceive of international law as coordinating territorial state sovereigns deserving of equal respect. Nor does it isolate law as, first, the only relevant structural feature of the international legal order or, second, so dominant a feature that all the other relevant factors (set out in sections 6(a)(a)–(g )) virtually collapse into this single factor. There are two other sections of the Second Restatement relevant to this chapter. Following the general tort principle in s. 145, the next section addresses particular torts. Section 146 instructs us on choice of law for personal injuries: “In an action for a personal injury, the local law of the state where the injury occurred determines the rights and liabilities of the parties, unless, with respect to the particular issue, some other state has a more significant relationship under the principles stated in s. 6 to the occurrence and the parties, in which event the local law of the other state will be applied.”111

This section clarifies several points left ambiguous in La Forest J.’s strict rule as it is articulated in Tolofson and discussed above. First of all, this section 106

Second Restatement, supra n. 13. Bear in mind that the Second Restatement was published in 1971, before the deepening of consensus on many norms in international human rights law. 108 Second Restatement, supra n. 13. 109 Ibid. 110 Ibid. 111 Second Restatement, supra n. 13, s. 146. 107

318 J Orange applies to personal injuries which are caused by negligence and intentional torts.112 Secondly, “personal injuries” include physical harm and mental disturbances.113 Finally, the place of the injury is the significant place of the tort. The Tolofson decision leaves the law lacking direction with regard to the location of a tort’s occurrence, but section 146 recognises and solves this problem. Comment c to section 146 adds another factor to the choice of law analysis. A decision that another state’s legal system should be selected under the choice of law principles of section 6 will depend to some degree on the purpose of the relevant local law.114 Thus, if the rule is viewed as compensatory in nature, the courts may choose the law that would most suitably compensate the victims of the tort and address concerns such as the cost of health care and insurance. On the other hand, if the tort rule is viewed as meant to punish misconduct, then the place where the misconduct occurred may be most concerned about the tort. It may then follow that, if the tort is one that concerns the international community as a whole, as torture quintessentially is, then the goals of the international community should be taken into account when choosing applicable law. A purposive selection of the legal system which most advances international law principles of accountability for torture—or which indeed applies, adapting as necessary, international law itself—would be no less legitimately instrumental an approach than what section 146 already concedes is appropriate. The Second Restatement rules are useful in vindicating an international tort of torture in several ways. First, an American court could apply US law by initially defining the choice of law as the place of the injury, if the torture victim could prove that some injury had been suffered in the United States—such as psychological harm of mental illness. This would be possible even in the absence of the ATCA or TVPA. Secondly, once it had been established that an injury occurred in the US, a defendant would then have the burden of proving that another state has the “most” significant relationship to the parties and the occurrence. If one of the parties is living in the United States, then the plaintiff has one important element already established. In Filártiga, both the plaintiff and the defendant were residing in the US, having moved there from Paraguay. This improved the case for the application of the forum’s law. Finally, considerations of “the needs of the interstate and international systems”, as set out in section 6, seem broader than La Forest J.’s statements and likely include public international law institutions and conventions as they evolve. Such a statement seems to contemplate more than international commercial transactions and interstate comity, and can easily and justifiably extend to the international community’s rapidly evolving and deepening legal treatment of human rights problems. 112 113 114

Second Restatement, supra n. 13, s. 146 cmt. a. Ibid., s. 146 cmt. b. Ibid., s. 146 cmt. c.

Torture, Tort Choice of Law and Tolofson 319 The United Kingdom In the United Kingdom, the common law rules for choice of law in tort were recently replaced by the Private International Law (Miscellaneous Provisions) Act 1995 (the PIL Act).115 The new general rule for choice of law in tort is set out in section 11: “(1) The general rule is that the applicable law is the law of the country in which the events constituting the tort or delict in question occur. (2) Where elements of those events occur in different countries, the applicable law under the general rule is to be taken as being— (a) for a cause of action in respect of personal injury caused to an individual or death resulting from personal injury, the law of the country where the individual was when he sustained the injury.”116

Subsection 11(3) clarifies the meaning of “personal injury” to include any physical or mental impairment. Thus, lex loci delicti is the general rule in the UK, with an explanation that the tort occurs where the injury is sustained.117 The rationales for this general rule were, according to a Scottish Law Commission study that preceded adoption of the PIL Act, those of certainty, uniformity and prevention of forum shopping.118 However, at the same time, the UK Parliament sought a new rule to replace the double-actionability rule of Chaplin v. Boys which was considered too inflexible to achieve justice between the parties.119 Therefore, unlike the new Canadian Tolofson rule, the PIL Act provides for some exceptions to the rule. Section 12 is set out below: “(1) If it appears, in all the circumstances, from a comparison of— (a) the significance of the factors which connect a tort or delict with the country whose law would be the applicable law under the general rule; and (b) the significance of any factors connecting the tort of delict with another country, that it is substantially more appropriate for the applicable law for determining the issues arising in the case, or any of those issues, to be the law of the other country, the general rule is displaced and the applicable law for determining those issues or that issue (as the case may be) is that law of that other country. (2) The factors that may be taken into account as connecting a tort or delict with a country for the purposes of this section include, in particular, factors relating to the 115 Supra n. 14. The PIL Act does not cover torts of defamation. Also, the Act applies only to tort causes of action arising after entry into force of the Act, such that Chaplin, supra n. 9, continues to cover pre-Act causes of action. 116 PIL Act, ibid., s. 11. 117 Note the use of “the” in “the injury”. Note also that it does not say “where the individual was when he first sustained the injury”. Thus, section 12 seems to leave open locating the tort outside the place of initial human-rights-violating conduct and locating it where discrete injury develops, notably new dimensions and manifestation of psychological harm. 118 P Kincaid, “Jensen v. Tolofson and the Revolution in Tort Choice of Law”, 74 Can BR 4 at 549 [hereinafter “Kincaid”]. 119 C G J Morse, “Torts in Private International Law: A New Statutory Framework”, (1996) 45: Oct. Int’l Comp LQ 888 at 897 [hereinafter “Morse”]. One can contrast this reasoning to La Forest J. in Tolofson, supra n. 8 seeing Chaplin as too flexible.

320 J Orange parties, to any of the events which constitute the tort or delict in question or to any of the circumstances or consequences of those events.”120

Section 12 gives the courts the discretion to look at all the factors involved in the event, especially those concerning the parties and the constitution of the tort, in order to apply the most appropriate law. In effect, the PIL Act has creates a test that has the lex loci delicti as the general rule, with a “proper law” of the tort exception.121 It is relevant that the section speaks in qualitative terms of the “significance” of factors and “appropriate[ness]” of a law’s application, as this appears to signal the need for a normative evaluation of the relevance and weight of factors rather than anything resembling the stereotypical mechanistic listing and counting of contacts that is notorious in a “proper law” style of analysis. There may be concerns that exceptions may be found too easily, forgoing certainty in the law. Perhaps the words “substantially more appropriate” will act as some safeguard to the hasty application of the forum’s law. In addition to the exceptions allowed in section 12, section 14(3) prohibits the application of the law of another country that conflicts with the principles of public policy of the forum.122 Recall La Forest J.’s comments in Tolofson regarding an exception for public policy. He suggested that a public policy exception was just another way of saying that we do not approve of the laws of another jurisdiction, which conflicts with the rules of comity. In the PIL Act, however, the public policy exception is much more restricted than the Chaplin v. Boys application as it no longer applies to actions in tort which are not recognised in the UK as torts.123 Any public policy exception in a tort case would, thus, have to be consistent with the UK’s general tort principles. Finally worthy of mention is that the PIL Act addresses characterisation issues in section 9(2): “The characterisation for the purposes of private international law of issues arising in a claim as issues relating to tort or delict is a matter for the courts of the forum.”124 This statement does not mean that the issue should be characterised by the law of the forum, as this would in effect bring back the problems of the double-actionability rule: “The key to questions of characterisation, for the purposes of section 9(2), probably lies in the presence of the expression ‘for the purposes of private international law’. This expression suggests that courts should take a broad ‘international’ view of legal concepts, paying attention to the nature and content of the foreign rule so as to ascertain whether or not the liability under that rule falls within the general rubric of liability in tort, flexibly interpreted.”125 120

PIL Act, supra n. 115, s. 12 (emphasis added). Kincaid, supra n. 118, at 549. 122 PIL Act, supra n. 115, s. 14(3)(a). 123 S. 14(3) provides that: “Nothing in this part—(a) authorises the application of the law of a country outside the forum as the applicable law for determining issues in any claim insofar as to do so: (1) would conflict with principles of public policy . . .” 124 PIL Act, supra n. 115, s. 9(2). 125 Morse, supra n. 119, at 894 (notes omitted). 121

Torture, Tort Choice of Law and Tolofson 321 The above passage accords with Falconbridge’s assessment of characterisation. The domestic courts must look to a broader concept of law before characterising the issue. The PIL Act recognises this necessity. Although the US Second Restatement and the UK PIL Act use basic methods for choice of law in tort that are quite similar, the latter seems simpler and less ambiguous. Both sources recognize the importance of different elements of the tort that could assist Canadian judges in forging common law principles appropriate for claims of torture. The US and the UK also stress the importance of the values and norms of the international community. Finally, both jurisdictions found it necessary to allow exceptions to a presumptive lex loci delicti rule, risking uncertainty for the sake of justice.

Torture, Tort Choice of Law and Tolofson It is now timely to consider again the internationalist hint buried in La Forest J.’s own invocation of the idea of overriding norms as an exception to the application of the lex loci delicti in international conflicts cases. Given that invocation, it would seem that he overstated his own case regarding “public policy” collapsing into simple policy differences between jurisdictions as a reason for excluding a public policy exception in tort choice of law. He was surely correct that judges can manipulate “public policy” by calling the concept in aid at too low a threshold of offensiveness to the forum. And he was probably correct that it is a concept at odds with his conception of absolute comity in the interprovincial context. But this does not mean that it is not an important doctrinal concept for Canadian conflict of laws to retain if our courts continue to adhere to lex loci delicti as a near-absolute rule in the international context. La Forest J.’s reasoning in Tolofson more or less excludes the kinds of analysis set out in the United Kingdom’s PIL Act as a method for selecting a law other than the lex loci delicti. Displacing the lex loci delicti was, for him, a true exception, rarely to be invoked. If this remains the case, a high-threshold public policy doctrine seems to be the doctrine exactly designed to trigger such an exception. In other words, “public policy” is the existing conceptual rubric into which to slot the “overriding norms” La Forest J. invoked by way of passing mention. Whether or not such norms should be limited to jus cogens norms such as the torture prohibition, or whether they can encompass a fairly robust list of human rights norms which create erga omnes obligations for all states under customary international law or general principles of law, it is clear that the Supreme Court did leave open the door for the application of Canadian tort law—where the lex loci delicti is tangibly repugnant to overriding international human rights norms. It would further appear that it would be faithful to Tolofson’s own general reasoning and its specific data to draw on overriding norms for aid in making principled choices between more than one plausible lex loci delicti for a given

322 J Orange torture tort claim. We can expect that Canadian law would often be one plausible candidate because of significant injury occurring or worsening in Canada. In other situations, additional factors may help tip the balance toward application of Canadian law as, for instance, in the torture-death of the Somalian Shidane Arone at the hands of Canadian soldiers in Somalia in the course of their posting with a UN mission in that country.126 If either Canadian law or, say, Somalian law is more conducive to redress and compensation for tort victims, then international human rights law provides a principled basis for selecting one law over the other for application by Canadian courts. In either of these guises—as a direct public policy exception to the sole possible lex loci delicti or as a principle of choice between two or more plausible lex loci delictis—there is sufficient consensus in the international legal system over a cluster of core basic rights. These are clearly erga omnes, and either also jus cogens or candidates for elevation to that status in fairly short order. This being the case, La Forest J.’s concerns with certainty and predictability will largely be respected in the quest for justice. Furthermore, when future Canadian courts (and a future Supreme Court of Canada in the absence of the now-retired La Forest J.) come to consider a claim of torture, slavery or genocide as a tort, they would also do well to remember La Forest J. in Tolofson. There, he situated the “order versus justice” within a particular conception of the international legal order. Judges may adopt a conception of that order that is less centred on the interaction between territorially sovereign states, and more inclusive of international human rights as central pillars of modern international law. If so, perhaps the dicta in Tolofson on lex loci delicti as the tort of choice of law rule should be approached as a first crack at the problem, rather than as comments that seriously constrain the future development of the law. Indeed, some sign that Canadian courts may be open to the above kind of analysis can be seen in the recent case of Davidson Tisdale Ltd. v. Pendrick in which the Ontario Court (General Division) had the following to say: “It is difficult to see how applying the law of New York to the conduct of the defendants in bringing [this] class action could cause injustice to DTL for a like class action could have been brought in Ontario by virtue of the Class Proceedings Act. . . . This cannot be one of those ‘few cases’ La Forest J. spoke of [in Tolofson] where Ontario would have to apply Ontario law to conduct in a foreign jurisdiction to prevent injustice. We are not dealing here with conduct lawful by the laws of a state run by a despot, but unacceptable in a democratic society a matter that might require the application of Ontario law to prevent injustice.”127

126 Report of the Commission of Inquiry into the Deployment of Canadian Forces to Somalia, Dishonoured Legacy: The Lessons of the Somalia Affair, Volume 1 (Ottawa, Minister of Public Works and Government Services Canada, 1997) at 321; M L Friedland, “Military justice and the Somalia affair”, Crim LQ, Feb–May 1998, v. 40, nn. 3–4, pp. 360–99; Report of The Special Advisory Group on Military Justice and Military Police Investigation Services (Ottawa, 1999). 127 [1998] OJ No. 5308 at 7–8.

Torture, Tort Choice of Law and Tolofson 323

7 CONCLUSION

My purpose here has been to establish the shortcomings of the strict lex loci delicti rule as expressed in Tolofson, and to explore considerations that should be taken into account in the construction of any new rule in the international context. This development may occur by future courts interpreting (and distinguishing) Tolofson, or by legislative action. The example used here was extreme: a victim of torture in another country now residing in Canada. La Forest J. noted that some international situations, such as torts which occur in more than one jurisdiction, would require a more flexible test and he invoked, without explanation, the notion of “overriding norms” as relevant to tort choice of law in international contexts. A violation of jus cogens, and perhaps erga omnes, international human rights norms must be one of the situations which calls for an exception to the lex loci delicti rule, on Tolofson’s own reasoning. Ideally, in order to develop a sensitive and consistent approach to international choice of law in tort, legislative action is desirable. A Canadian version of the UK PIL Act, one which sets out a general rule with listed considerations for exceptions, a definition of where the tort occurs and how courts should characterise legal issues, would be a relief to plaintiffs, legal scholars and judges alike.128

128 It is beyond the scope of this chapter to address the particular division of powers issues in Canada as to whether the federal Parliament may adopt such a statute or whether it must be left to each province. However, at first blush, it would seem that the international dimension of the private law subject matter would accord the federal Parliament at least concurrent jurisdiction with the provinces.

12

Characterisation, Choice of Law and Human Rights GRAHAM VIRGO 1

1 THE NATURE OF THE PROBLEM

has suffered torture in one country and wishes to commence civil proceedings in another, there are two key hurdles that need to be surmounted before the substantive issues of law can be considered. First, the forum must have jurisdiction to hear the case. Secondly, the forum must decide which country’s law should be applicable to determine the substantive issues. In both cases, but especially when determining the applicable choice of law, the identification of the nature of the claim or the issue in dispute is of vital importance. The process of identifying the nature of the claim or issue is called “characterisation”. In many cases the process of characterisation causes no problems, since it is clear that there is only one relevant cause of action, and that cause of action is recognised in the forum, or it is clear that the issue in dispute can only be placed in one category of claim. But the characterisation question becomes much more important where the nature of the claim is uncertain, as will be the case, for example, where a cause of action is pleaded that is neither recognised in the forum nor in any other relevant national law. So, for example, if the claimant is tortured in Utopia and commences proceedings in the forum for the tort of torture, and neither Utopia nor the forum recognise such a tort, what can the forum do? This turns on the function and application of the doctrines of characterisation and choice of law.

W

HERE A CLAIMANT

1 Fellow of Downing College, Cambridge, and Senior University Lecturer at the Faculty of Law, University of Cambridge.

326 G. Virgo

2 CHARACTERISATION

What is Characterisation? When the doctrine of characterisation is referred to in private international law, it can relate to two different matters.2 First, and most simply, the nature of the cause of action needs to be characterised in broad terms. For example, where the claimant sues his or her employer for injury suffered in the course of employment, is this a claim in contract or in tort? Secondly, even though the nature of the cause of action may be obvious, the nature of the issue in dispute may be less clear. For example, is it a question of substance or procedure? Regardless of whether causes of action or issues are involved, the consequence of characterisation is important because it will determine the choice of law rule applicable to the dispute, and ultimately the governing law. In essence, the role of the doctrine of characterisation is to allocate the issue in dispute to a specific legal category, which will in turn indicate the relevant choice of law rule. As regards a claim arising from torture, the crucial question of characterisation is whether a claim explicitly founded on torture falls within the category of tort, some other legal category or, most importantly, no recognised category at all. It is only if the claim can be treated as falling within the category of tort that the choice of law rule for tort will be applicable.

Which Law Determines Characterisation? Where there is only one category into which the claim or the issue can be placed, there is no problem concerning characterisation. Where, however, the law of the forum and the prospective governing laws would characterise the claim or the issue differently, or would not be able to characterise it at all, it is necessary to determine how the claim or the issue should be characterised. A number of different theories have been suggested in this regard, but two are particularly relevant to the characterisation of torture as tort, especially where no relevant national law explicitly recognises a tort called “torture”.3 (i) Law of the forum The simplest theory of characterisation is that it is a matter for the law of the forum to determine. It follows that the forum will characterise the cause of action or issue by reference to its own categories. If the pleaded cause of action 2 As recognised by the English Court of Appeal in Macmillan Inc. v. Bishopsgate Investment Trust plc. (No. 3) [1996] 1 WLR 387 [hereinafter MacMillan v. Bishopsgate]. 3 See L Collins, Dicey and Morris on the Conflict of Laws 13th ed. (London, Sweet and Maxwell, 2000), Chapter 2 [hereinafter Collins, Dicey and Morris].

Characterisation, Choice of Law and Human Rights 327 does not exist, the forum will characterise it by reference to the nearest existing category.4 If there is no relevant category, it will either create a new choice of law rule5—or conclude that the case should be dismissed.6 If the judicial power to create a new choice of law rule were exercised, it might still lead to the dismissal of the case if the newly created rule pointed to the lex fori, and that lex fori did not recognise the cause of action for which the choice of law rule had been created. However, and significantly, if the new choice of law rule pointed to a legal system that did recognise the cause of action, and the court proceeded to apply that other system’s substantive rules, then we would see in operation a high degree of judicial cosmopolitanism bound up in characterisation and choice of law. This is the case because one court system is willing to enforce foreign legal rules which have no concomitants in its own private law. The operation of characterisation by the lex fori is illustrated by the following example. Some jurisdictions, such as Singapore and most Canadian provinces, have a statutory cause of action, a hybrid of tort and family law, which allows parents to sue their adult children for financial support. England does not have a similar civil right. It is nonetheless possible, if English courts have acquired jurisdiction to determine the matter, that a lawsuit before an English court could proceed based on the application of, for example, Singapore’s “parental support” statute, if the English court characterised the claim as tortious and treated Singapore as the lex loci delicti. This would be so even though such an action is not available domestically to a parent domiciled in England. Proceeding now to the subject of this book, we would note that, if the state of Utopia created a nominate tort called “torture”, perhaps because it views itself as obliged to do so as part of its international obligations with respect to torture, this tort of torture could still be the basis of an adjudication in a forum state such as England, which has no domestic common law or statutory tort of torture. From the perspective of English law, this theory of characterisation by the lex fori has received explicit judicial7 and statutory8 recognition. Characterisation by reference to the law of the forum can also be justified by reference to the logic of private international law itself. The argument proceeds as follows. It is a fundamental principle of private international law that questions of procedure are matters for the law of the forum, whereas questions of substance are matters for 4 See, e.g., Re Bonacina [1912] 2 Ch. 394, where an agreement unsupported by consideration was characterised as a contract. 5 See Collins, Dicey and Morris, supra n. 3, at 42. 6 Although this “lex fori theory” is supported by many authors, a number suggest that it should not be a straitjacket to characterisation, since regard should be had to the context in which the issue arises and the way other systems would characterise the action or issue. See, in particular, O KahnFreund, General Problems of Private International Law (Netherlands, Sijthoff, 1976) at 227 et seq. 7 Macmillan Inc. v. Bishopsgate, supra n. 2 at 407 (Auld L.J.). 8 The Private International Law (Miscellaneous Provisions) Act (UK), 1995, (PIL Act), s. 9(2) states that the characterisation of issues as issues which relate to tort is a matter for the law of the forum. The 1995 Act legislates a new choice of law rule for tort in the United Kingdom.

328 G. Virgo the governing law of the dispute. So, is the process of characterisation itself to be characterised as procedural or substantive? Surely it is a procedural matter, since characterisation is simply a mechanism of private international law that assists in the determination of the governing law, which, once identified, determines the rights of the parties.9 Consequently, the inherent logic of private international law demands that the characterisation of the cause of action or issue is to be determined by the law of the forum. (ii) Analytical jurisprudence, comparative law and international law10 An alternative theory of characterisation is that disputes are not to be characterised by reference to the national law of the forum, but instead by reference to the laws of nations.11 Essentially, this theory assumes that, by careful analysis of the laws of different nations, common norms can be identified, and that these will produce common characterisations of issues and causes of action. On a theoretical level this theory seems particularly naïve, because only a rudimentary knowledge of comparative law suggests that some of the most basic causes of action are characterised very differently by different countries. But in specific areas of the law, this theory of characterisation may be preferable—especially if it is expanded to include not simply an empirical comparison of the laws of nations but also the clear normative dictates of the law of nations, i.e. public international law as a required standard for domestic law. This will be particularly important in the context of serious human rights violations, a category that would clearly include torture, where such violations are considered to be both contrary to the law of nations and capable of creating individual liability under that body of law.12 (iii) Law of the forum as informed by the law(s) of nations It may be possible to avoid the need to choose between the two preceding approaches by amalgamating them. If a judge is given power by the law of the forum to recognise a new choice of law rule to fit a new kind of claim, the judge should be permitted to look to both comparative law and international law as informing values for the exercise and content of the judicial decision. Consequently, although characterisation generally occurs by reference to the 9 Compare the judgment of La Forest J in Tolofson v. Jensen, [1994] 3 SCR 1022 at 1070, who adopts a more restrictive interpretation of procedural matters by confining the term to “those rules which make the machinery of the forum run more smoothly”. 10 J Collier, Conflict of Laws 2nd ed. (Cambridge, Cambridge University Press, 1994) at 18–19. See also, W E Beckett, “The Question of Classification (‘Qualification’) in Private International Law”, (1934) 15 BYIL 46. 11 The “laws of nations” are distinct from the “law of nations”, since the latter is sometimes used as synonym for public international law. 12 See, for example, the decision of the House of Lords in R v. Bow Street Metropolitan Stipendiary Magistrate, ex parte Pinochet Ugarte (No. 3) [1999] 2 WLR 827 [hereinafter Pinochet No. 3].

Characterisation, Choice of Law and Human Rights 329 law of the forum, where the forum does not recognise the cause of action which has been pleaded or the issue which is in dispute, it should at least be permissible to have regard both to the laws of nations and to the law of nations to determine how to characterise the cause of action or the issue.13

3 CHOICE OF LAW

If the claim has been characterised as tortious, it is then necessary for the law of the forum to determine the choice of law rule for the applicable legal category, which will then identify a governing law. The forum will then apply that governing law to resolve the dispute. Typically, the choice of law rule for tortious claims is that the law of the place where the tort was committed is applicable.14 This is called the lex loci delicti. This is usually defined as the place where, in substance, the cause of action arose.15 With respect to torture, presumably this will be the place where the victim was tortured, rather than where any consequences of that torture developed. Consequently, if the lex loci delicti does not recognise any liability for torture, then the civil claim of the victim will fail unless the forum is able to avoid the operation of the lex loci delicti by some subsidiary rule of law.16 But if the lex loci delicti does recognise civil liability for the conduct which amounts to torture, it is of no consequence, at least as regards private international law, that that country does not recognise a tort of torture—or that it would not characterize the claim as tortious at all, but rather as some other category of private law.17 It is sufficient that the foreign law recognises that the torturer will be liable to the claimant regardless of how the foreign law would have characterised the action or the issue had the dispute been brought in that country.

4 CAN TORTURE BE A TORT ?

According to the law of the forum theory of characterisation, the victim of torture will only be able to bring a tortious claim founded on torture if the forum 13 See J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11 in this volume. See also J D Falconbridge, “Conflict Rule and Characterization of Question”, (1952) 30 No. 2 Can Bar Rev 103 at 112 [hereinafter Falconbridge] and W R Lederman, “Classification in Private International Law” (1951) 29 Can Bar Rev 3 at 34 [hereinafter Lederman] both of whom recognised the importance of characterising by the lex fori initially, and subsequently having regard to other laws to determine the appropriateness of the initial characterisation. Of course, this raises difficulties in determining what is appropriate in the circumstances of the case. 14 This is now the general rule in England after the enactment of the PIL Act, supra n. 8. It is also the general choice of law rule in Canada: Tolofson v. Jensen [1994] 3 SCR 1022. 15 Distillers Co (Bio-Chemicals) Ltd. v. Thompson [1971] AC 458; Castree v. Squibb Ltd. [1980] 1 WLR 1248; Metall und Rohstoff AG v. Donaldson Lufkin and Jenrette Inc. [1990] 1 QB 391. 16 Such subsidiary rules of choice of law are examined infra, Ssection 5. 17 Boys v. Chaplin [1971] AC 356.

330 G. Virgo is prepared to characterise torture as a tort. Clearly, if the forum recognises a nominate tort of torture there will be no obstacle to the claimant’s claim being founded specifically on the torture. But, on the assumption that the forum does not recognise a tort of torture, how will the claimant’s claim be characterised? To answer this question it is first necessary to define torture itself.

What is Torture? Torture can cover a wide variety of human rights abuses including killing, injuring, sexual assault, psychological harm and other cruel and inhuman treatment. Torture has usefully been defined by the Convention against Torture as: “. . . any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity.”18

There are two key features of this definition worth emphasising. Firstly, torture is not confined to causing physical harm, since it also encompasses mental harm. Secondly, the conduct must be committed by a state official, or be performed with the consent or acquiescence of such an official.19

Can Torture be Characterised as a Tort? Whether torture can be characterised as a tort will be affected by the nature of the law of tort in the domestic law of both the law of the forum and the lex loci delicti. (i) The law of the forum recognises a tort of torture If the law of the forum recognises a specific tort of torture, a claim founded on torture will clearly be characterised by the forum as tortious so that the tort choice of law rule will apply. As mentioned earlier, it is irrelevant that the lex 18 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 [hereinafter “CAT”]. 19 This would include a head of state: Pinochet No. 3, supra n. 12, at 843 (Lord BrowneWilkinson). The Convention contains a provision which says that its definition may not be invoked to limit wider definitions of torture found in domestic law or other treaties, a significant proviso in light of article 1’s requirement that a non-state actor can only be found to have engaged in torture if it was somehow consented to or acquiesced in by the state.

Characterisation, Choice of Law and Human Rights 331 loci delicti does not recognise such a tort, it being sufficient that the lex loci delicti recognises civil liability of some kind. Consequently, “tort” as a category will trigger the choice of law rule, but the law that is then applied need not itself be tort law. So, for example, the lex loci delicti may not recognise tortious liability for torture, but might recognise civil liability under public law instead. (ii) The lex loci delicti recognises a tort of torture Where the lex loci delicti recognises torture as a tort, but the forum does not, the forum may accept that characterisation because it is the nearest category into which the claim can be placed.20 (iii) No relevant law recognises a tort of torture: other possible torts Where neither the law of the forum nor the lex loci delicti recognise the tort of torture, the claimant may plead, as an alternative, a nominate tort recognised by the forum. Clearly there is no problem of characterisation in such circumstances, since the forum will accept that characterisation. Alternatively, the claimant may decide not to plead an available nominate tort, and instead plead that he or she has been tortured. In such circumstances, the law of the forum will only consider this to involve the pleading of a tortious cause of action if the judge is willing to consider whether the elements of a recognised category of tort are present in the facts alleged.21 But, either way, what tort will torture be?22 The most obvious nominate tort on which a torture victim can found a claim is the tort of battery, which can be defined as the “actual infliction of unlawful force on another person”.23 This tort will be committed regardless of the degree of force used.24 Battery can only be committed if the tortfeasor intends harm of some kind to the victim. In England, if the victim suffers physical harm as a result of the defendant’s intentional conduct which does not involve the infliction of force, this is a separate tort,25 as is the tort of assault where the defendant causes the claimant to apprehend immediate personal harm.26 Alternatively, the victim of torture may rely on the tort of harassment,27 if it is

20

By analogy with Re Bonacina, supra n. 4. In the exercise of their discretion, some legal systems or judges may not see it as the role of a judge to “convert” a claim in a direction not specifically pleaded by the claimant. In such cases, the judge would dismiss the case on the basis of no recognised cause of action having been alleged. 22 See M R Brazier (general ed.), Clerk and Lindsell on Torts 17th ed. (London, Sweet and Maxwell, 1995) Ch. 12. 23 Collins v. Wilcock [1984] 1 WLR 1172. See Alomang v. Freeport-McMoran Inc., 718 So. 2d 971 (La. App. 4 Cir. 1998). 24 Cole v. Turner (1704), 6 Mod. 149. 25 Wilkinson v. Downton [1897] 2 QB 57. This tort is known by the name of this case which first recognised it. 26 Collins v. Wilcock, supra n. 23. 27 This is a statutory tort in England: Protection from Harassment Act (UK), 1997, s. 3. 21

332 G. Virgo recognised by the law of the forum, or even false imprisonment, to catch at least one aspect of some torture situations. With respect to the immediate torturer, the claimant cannot found a claim on negligence, since torture by definition involves intentional conduct. However, the Convention against Torture makes clear with regard to torture that the complicity of superiors triggers individual responsibility.28 That being the case, negligent supervision of underlings or the failure to put torture-preventing procedures in place could well be sufficient to create liability. Even though torture may be characterised as a battery, it may not be possible for the claimant to bring a claim founded on that tort because the context in which the torture was committed may be such as to constitute a bar to the tortious claim. This is a matter for the governing law rather than a matter of characterisation, save to the extent that it is necessary for the forum to characterise the bar either as procedural, calling for the application of the law of the forum, or substantive, calling for the application of the governing law. Assuming the bar has been characterised as a matter of substance, then it is for the governing law to determine whether the claim is barred. So, for example, if the victim of the tort has died from the injuries suffered during the torture, his or her estate may not be able to bring a claim if the governing law states that tortious claims are barred when the victim dies. Similarly, the governing law may state that acts of state officials are justified and so cannot ground tortious claims.29 These examples show why it may be crucial for a nominate tort of torture to be the basis of liability; so that adjunct principles appropriate to the nature of torture and to the content of international law principles on torture can thus be the basis of adjudication. (iv) No relevant national law recognises a tort of torture: other non-tortious causes of action If the forum does not recognise the tort of torture and the claimant does not wish his or her claim to be characterised as a battery, he or she may instead plead an alternative cause of action. This will be particularly attractive to the claimant if this means that different law applies by virtue of a different choice of law rule more advantageous to the success of the claim. However, in the context of torture, it is difficult to conceive of any other cause of action which would be relevant other than tort, save that in some contexts it might be possible for the victim to invoke civil law redress contingent on a criminal law conviction or involving restorative justice regimes where compensation is part of a truth and reconciliation, or similar, process.30 28

CAT, supra n. 18, article 2(3). See, for example, the classic tort choice of law case of Phillips v. Eyre (1869), LR 4 QB 225, where the actions of a state official in quelling a rebellion were subsequently justified by way of a legislative indemnity—thereby negating tortious liability. 30 See e.g. J Llewellyn, “Just Amnesty and Private International Law,” chapter 22 in this volume. 29

Characterisation, Choice of Law and Human Rights 333 (v) No relevant national law recognises a tort of torture: reference to public international law Where neither the law of the forum nor the lex loci delicti recognise the tort of torture, the claimant may alternatively resort to public international law to establish a claim founded specifically on torture rather than any other tort. The question of whether an international tort of torture exists is canvassed in some of the other chapters in this volume.31 This is initially a matter for public international law. But, if it is assumed that such a tort does exist, the question then is whether the rights protected by that tort can be vindicated in national courts. This will usually be a matter for private international law. Private international law can make an important contribution toward the recognition of such an international tort for two reasons. Firstly, one method of encouraging states to recognise the introduction of a transnational tort of torture is by reference to characterisation theory. Although, traditionally, emphasis has been placed on the forum as being able to characterise the cause of action or the issue by reference to its own law, in this field straddling public and private international law, the emphasis should shift to the approach drawing on comparative law and international law.32 This approach encourages states, and judges, to widen their perspective away from the national and towards the international plane so that human rights are recognised in a way that allows them to be vindicated in the national courts. Secondly, what should the choice of law rule be for this type of liability? This will turn on how liability for torture is characterised. The nearest category would be tort, so it would appear to follow that the tortious choice of rule should apply. But, in fact, if public international law recognises civil liability for torture as a customary norm, it follows that this should be recognised in every state—even if the domestic law in the relevant states is such that there is no liability, or only limited liability. Consequently, no conflict between the laws of different states should be recognised as part of private international law analysis if those laws are required to be the same by public international law. There is thus no need to identify a choice of law rule, in the traditional sense of choosing one private law system’s rules or those of another system, since the law of the country where the torture occurred must be treated as the same as the law of the forum by normative fiat. In other words, where there is no relevant conflict between the laws of particular nations, there is no role for private international law. The foregoing analysis works at the level of recognising the basic cause of action, but there may be other subsidiary issues in dispute when the international tort of torture is litigated in national courts and, as between these 31 See especially S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, chapter 14 in this volume, and E M Hyland, “International Human Rights Law and the Tort of Torture: What Possibility for Canada?”, chapter 15 in this volume. 32 Examined here, supra Section 2.

334 G. Virgo issues, a conflict between laws might still arise. One of the most important examples of this concerns limitation periods. In a standard conflict of laws analysis, if a limitation period is characterised by the forum as procedural, it follows that the forum’s limitation period will apply. But if the limitation period is characterised as substantive, it follows that the governing law’s limitation period is applicable. The limitation periods of these two laws might be very different, in which case it would be a matter of some importance as to which limitation period applies.33 There would be a conflict of laws, and so a choice of law rule would be necessary. For example, if the forum characterises limitation periods as substantive then it would be necessary to identify a governing law, which would presumably be the lex loci delicti. Nevertheless, there is a strong argument to the effect that, where the claim is founded on an international tort, national rules such as limitation periods should not be applicable. Instead, public policy, ascertained with regard to the international interest (especially insofar as it is expressed in international legal rules and principles dealing with torture), should determine the course of the litigation. Arguably, this is just an application of the approach outlined earlier, whereby public international law norms provide the framework for the applicable law. Consequently, at a minimum, it could be said that the claimant’s case should be allowed to proceed despite delay, save where that delay is considered to be unconscionable or affects the fairness and efficacy of the trial. Where, however, the claim is founded on the commission of torture, it is unlikely that the claimant’s delay in bringing proceedings would outweigh the seriousness of the conduct alleged against the defendant. A useful analogy can be made with international criminal law categories, such as crimes against humanity and war crimes, in which torture is a kind of conduct criminalised within these broader concepts. To the extent that international criminal law does not recognise any time limit on prosecuting torture, it could reasonably be argued that it is a fortiori the case that no limit should apply to civil claims where the more serious consequence of imprisonment is not a potential legal consequence.

Does it Matter How Torture is Characterised? As far as the law of the forum is concerned, it makes no difference what kind of tort torture is characterised as being, as long as it is characterised as a tort. This is because, for the purposes of private international law, it is only necessary to place the cause of action within a broad category of legal claims to determine what the applicable choice of law rule is. As different choice of law rules are not

33 In England, all limitation periods are characterised as substantive rather than procedural so the limitation period of the governing law is applicable: Foreign Limitation Period Act (UK), 1984, s. 1(1).

Characterisation, Choice of Law and Human Rights 335 generally the case for different types of tort, it is sufficient that the claim be characterised as tortious.34 From the perspective of the victim, however, the fact that torture is characterised as a tort, such as battery, will matter—simply because characterising torture in this way does not necessarily reflect the seriousness of the conduct involved. In the context of violations of human rights, particularly by States and officials, symbolism is crucial. The relevance of symbolism has been recognised in the context of the criminal law, where notions of “fair labelling” are of growing importance in the analysis of the law.35 In that context, the offence of which defendants are convicted should reflect the seriousness of the harm caused. It is for this reason that a defendant who has caused serious injury to the victim would be charged with an offence of causing grievous bodily harm, rather than simply with battery, even though a battery would invariably also have been committed. This flexibility in the criminal law tends not to be reflected in the law of tort. Tort law tends not to have degrees of liability in the same way as the criminal law, presumably because the law of tort is concerned with compensating victims for harm suffered rather than with characterising the defendant’s conduct as especially culpable. Despite this, in the context of characterising torture as tort, it should matter what the tort is called, especially because the main reason why the victim wishes to commence civil proceedings will presumably be to ensure public awareness of the violation of fundamental human rights. The remedial consequence of successfully bringing a case is often, or even usually, only a secondary concern. In fact, bringing civil proceedings in respect of international human rights violations should properly be treated as straddling the divide between the civil and the criminal law. Perhaps the torturer cannot be prosecuted in the forum’s courts because of the more restrictive approach adopted by the criminal law as regards the acquisition of jurisdiction. Although many countries are able to prosecute torturers regardless of where in the world the torture was committed, it is still necessary in most jurisdictions (at least Westminster common law jurisdictions) to acquire jurisdiction over the defendant either by virtue of the defendant’s voluntary presence in the forum or by virtue of the defendant being extradited to that country.36 As regards the civil law, the forum may be able to acquire jurisdiction over the defendant even though he or she is not present within the forum’s territory, although some other kind of territorial link, such as submission to the jurisdiction or the exercise of legitimate extraterritorial jurisdiction, would have to be established. 34 Exceptionally, there may be different choice of law rules for different torts. For example, in England, the PIL Act, supra n. 8, preserves the common law double-actionability choice of law rule for torts of defamation. However, this is clearly of no significance to torts relating to torture. 35 See in particular A Ashworth, Principles of Criminal Law 3rd ed. (Oxford, Oxford University Press, 1999) at 90–3. 36 This is possible in England by virtue of the Criminal Justice Act 1988, s. 134(2).

336 G. Virgo This analysis suggests that it is not enough that torture is characterised as a tort; the nature—indeed the name—of the tort is a matter of importance. It is for this reason that the recognition of an international tort of torture is particularly significant, even though the forum and the governing law would accept that the claimant only had a claim against the torturer for battery. But, the importance of labelling also indicates that national courts should explicitly recognise a tort of torture in their domestic law.

5 THE SITUATION IN ENGLAND

Although there has not yet been a reported case where a claimant has successfully sued in England for torture committed abroad, it is possible in principle for such a claim to be brought in this country, although the success of such a claim may depend on how it is characterised.37 Jurisdiction For the purposes of acquiring jurisdiction in England, it is generally of no significance how the cause of action is characterised. Usually, it is sufficient to establish that the defendant is present in England or has submitted to English jurisdiction and the nature of the claim is irrelevant. Exceptionally, however, the nature of the claim may matter. This will be the case where the claimant petitions the court to exercise its discretion to serve a writ on the defendant abroad.38 This extraterritorial jurisdiction may be particularly significant where the claimant wishes to sue a torturer for torture abroad. To obtain service of the writ abroad, the claimant must show that there is a serious issue to be tried,39 that England is the most appropriate forum,40 and, crucially, that there is a jurisdictional link between the cause of action and the forum. This is where the question of characterisation is important, since, where the claimant is suing in tort, he or she must establish that the claim relates to an act which was committed in England or that damage was suffered in England.41 In 37 A claim was brought in Al-Adsani v. Government of Kuwait (1996), 107 ILR 536 but it was defeated by state immunity. 38 Schedule 1 of the Civil Procedure Rules 1998 (CPR), preserving Order 11 of the Rules of the Supreme Court (RSC). 39 Seaconsar Far East Ltd. v. Bank Markazi Jomhouri Islami Iran [1994] 1 AC 438. 40 Spiliada Maritime Corporation v. Cansulex Ltd. [1987] AC 460. 41 Ord. 11, r. 1(1)(f), supra n. 38. If the claimant wanted to sue a torturer in England and that torturer was habitually resident in a Contracting State to the Brussels and Lugano Conventions, in other words a State which is a member of the European Union or European Free Trade Agreement, then the English court could also acquire jurisdiction by virtue of the fact that the claim was founded on tort, if either the tort was committed in England or the harm was suffered in England; whether the claim relates to tort and, if so, where the tort was committed or the harm was suffered, is a matter for community law rather than English domestic law: Kalfelis v. Bankhaus Schroder, Munchmeyer, Hengst and Co. [1988] ECR 5565 (Case 189/87).

Characterisation, Choice of Law and Human Rights 337 Al-Adsani v. Government of Kuwait, the Court of Appeal held that damage can be said to have been suffered in England, even though physical injuries were suffered abroad, if the victim suffered serious mental injuries in England as a result of what had happened abroad.42 Crucially, when establishing that a tort has been committed in England, or that harm resulting from a tort has been suffered in England, it does not matter what type of tort has been committed as long as the conduct can be characterised as tortious. So it will not matter that the claimant sues for the tort of torture, because this would be treated as falling within the category of tort. The Al-Adsani case is important not only because of its interpretation of the rules on service out of the jurisdiction, but also because it is the only English case to date which involves a tortious claim relating to torture. The claimant alleged that he had been detained and tortured in Kuwait by officials of the Government and at the instigation of the royal family of Kuwait. He came to England and alleged that he received death threats there. As a result of the torture and death threats, he alleged that he had suffered serious physical and mental injury. He sought to sue the Government of Kuwait and three other defendants who were in Kuwait, so he applied for leave to serve proceedings on the defendants abroad. The Court of Appeal confirmed that leave to serve proceedings on all the defendants should be granted because the claimant had established a good arguable case that, although a tort was committed abroad, the harm was suffered in this country. Subsequently, however, the Court of Appeal rejected the claims against all the defendants by reason of the State Immunity Act 1978.43 However, this litigation does suggest that, if there is no issue relating to state immunity, then the English courts are prepared to contemplate proceedings in this country in respect of torture committed abroad, and are willing to treat such conduct as tortious.44

Choice of Law Once the English court has determined that it has jurisdiction to hear the dispute and it has characterised the claim as tortious, it will apply the choice of law rule. This application is now governed by Part III of the Private International Law (Miscellaneous Provisions) Act 1995. By this statute, the previous doubleactionability test, whereby the tort had to be both actionable as a tort by the law of the forum and civilly actionable by the lex loci delicti, has been replaced by a 42

(1994), 100 ILR 465. (1996), 107 ILR 536. For information on the challenge that has since been made to the application of immunity to a human rights claim in the Al-Adsani case under the European Convention on Human Rights, see A Clapham, “Revisiting Human Rights in the Private Sphere: Using the European Convention on Human Rights to Protect the Right of Access to the Civil Courts”, chapter 19 in this volume. 44 See especially Al-Adsani, supra n. 42 at 469 (Evans LJ). 43

338 G. Virgo general rule of lex loci delicti.45 There is a presumption that the applicable law is the law of the country where the significant elements or most of the elements constituting the tort were committed.46 So if the claimant is tortured in Arcadia, Arcadian law will apply. If Arcadia recognises that torture is a tort or, at the very least, that there is some form of civil liability for torture, the victim’s claim is likely to succeed. If, however, Arcadia does not recognise civil liability, the application of that law by the English court will mean that the claim fails. It is possible, however, to avoid the rigid application of the lex loci delicti in two different ways. First, there is a statutory exception enabling the court to apply the law of another country if the tort is more closely connected with that country, but only if it would be substantially more appropriate to apply the law of that other country.47 To determine whether, for example, the tort is more closely connected to England than the place where the tort was committed, it would be necessary to examine all the connecting factors in the case and then consider whether these factors point to England in particular and whether it is substantially more appropriate to apply English law rather than the lex loci delicti. The most usual situation for this exception to operate will be where the claimant and the defendant are not resident in the lex loci delicti, so that the commission of the tort in that country is deemed to be fortuitous.48 But this is unlikely to be the case where the tort involves the serious violation of a human right. By its very nature the tort is likely to be closely connected with the country where the torture occurred, so this exception is unlikely to be of any use. Alternatively, the English courts will not apply the lex loci delicti if it is considered to be contrary to public policy.49 The notion of public policy in the context of private international law is narrowly defined, but it has been accepted that the English courts should not recognise a foreign law if it involves “grave infringement of human rights”.50 So, by logical extension, if the foreign law does not recognise civil liability for serious violations of human rights, it should follow that the English court will, in the exercise of its discretion, not apply that foreign law. But if the lex loci delicti is rejected in favour of the application of English law, will this be of any assistance to the claimant, since English law itself has not yet expressly recognised a nominate tort of torture? English law would, of course, be prepared to characterise the torturer’s conduct as involving a battery, but, as 45

Chaplin v. Boys [1971] AC 356, interpreting the rule in Phillips v. Eyre (1870) LR 6 QB 1. If the tort relates to personal injury or death, the applicable law is the law of the country where the victim was when he or she sustained the injury: s. 11 (2)(a), PIL Act, supra n. 8. 47 Ibid., s. 12. 48 See Chaplin v. Boys, supra n. 45. 49 PIL Act, supra n. 8, s. 14(3)(a)(i). 50 Oppenheimer v. Cattermole [1976] AC 249 at 263 (Lord Hailsham of Marylebone), at 265 (Lord Hodson), at 278 (Lord Cross), and at 283 (Lord Salmon). Lord Pearson dissented. In this case a German decree of 1941, which deprived a German Jew of his nationality if he was ordinarily resident abroad, was not recognised for reasons of public policy. See also Williams and Humbert Ltd. v. W and H Trade Marks (Jersey) Ltd. [1986] AC 368 at 379 (Nourse J). 46

Characterisation, Choice of Law and Human Rights 339 has been seen, such a claim would appear to be unsuited to the claimant who wishes the conduct to be dealt with for what it is, namely the infringement of human rights by agents of a state or by persons acting with the acquiescence of the state. However, with the recent enactment of the Human Rights Act 1998, it appears that the English courts will be able to recognise a tort of torture, at least where the torture occurred in Europe. The European Dimension: The Human Rights Act 1998 The Human Rights Act 1998 incorporates the European Convention for the Protection of Human Rights and Fundamental Freedoms into English law.51 Article 3 of this Convention specifically prohibits torture as well as “inhuman or degrading treatment or punishment”. This is an absolute right not subject to any qualifications. Although the Act does not enable one private individual to sue another for a new tort of “breach of the Convention”, as regards the prohibition of torture it appears that the Act does indeed create a tort of torture.52 This is because, by virtue of sections 6 and 7 of the Act, it is possible for a person to bring proceedings directly against a public authority which has acted incompatibly with a Convention right.53 A public authority is defined to include a court 51

The Act came into force in October 2000. J Wadham and H Mountfield, Blackstone’s Guide to the Human Rights Act 1998 (London, Blackstone Press Ltd., 1999) at 26. 53 Sections 6 and 7 of the Human Rights Act (UK), 1998, are as follows: 52

Public authorities 6. Acts of public authorities (1) It is unlawful for a public authority to act in a way which is incompatible with a Convention right. (2) Subsection (1) does not apply to an act if— (a) as the result of one or more provisions of primary legislation, the authority could not have acted differently; or (b) in the case of one or more provisions of, or made under, primary legislation which cannot be read or given effect in a way which is compatible with the Convention rights, the authority was acting so as to give effect to or enforce these provisions. (3) In this section “public authority” includes— (a) a court or tribunal, and (b) any person certain of whose functions are functions of a public nature, but does not include either House of Parliament or a person exercising functions in connection with proceedings in Parliament. (4) In subsection 93) “Parliament” does not include the House of Lords in its judicial capacity. (5) In relation to a particular act, a person is not a public authority by virtue only of subsection (3)(b) if the nature of the act is private. (6) “An act” includes a failure to act but does not include a failure to— (a) introduce in, or lay before, parliament a proposal for legislation; or (b) make any primary legislation or remedial order. 7. Proceedings (1) A person who claims that a public authority has acted (or proposes to act) in a way which is made unlawful by section 6(1) may— (a) bring proceedings against the authority under this Act in the appropriate court or tribunal, or

340 G. Virgo or tribunal and “any person certain whose functions are functions of a public nature”.54 For the purposes of these provisions, a victim is defined as any person who would be treated as a victim if proceedings were brought before the European Court of Human Rights,55 namely a person who claims that his or her rights have been violated by one of the contracting parties to the European Convention for the Protection of Human Rights and Fundamental Freedoms.56 The application of this provision to the victim of torture who wishes to commence legal proceedings against the torturer is not straightforward, because the victim does not complain that his or her rights have been violated by a contracting party to the Convention, but rather by an individual or public authority. Despite this, the provision presumably means that the victim can only bring a claim if he or she has been tortured by a public authority, or by an official representing a contracting state to the Convention. There is no other territorial limitation in the Act, so it seems that, as long as the English court has civil jurisdiction to hear the case,57 it would be possible for a victim of torture to bring proceedings against a public authority of another European state party to the European Convention, where that authority was a person who tortured the victim in the course of his or her public duties, or had authorised the torture of the victim.58 This would, for example, enable a victim (b) rely on the Convention right or rights concerned in any legal proceedings, but only if he is (or would be) the victim of the unlawful act. (2) In subsection (1)(a) “appropriate court or tribunal” means such court or tribunal as may be determined in accordance with rule; and proceedings against an authority include a counterclaim or similar proceeding. (3) If the proceedings are brought on an application for judicial review, the applicant is to be taken to have a sufficient interest in relation to the unlawful act only if he is, or would be, a victim of that act. (4) [application in Scotland] (5) Proceedings under subsection (1)(a) must be brought before the end of— (a) the period of one year beginning with the date on which the act complained of took place; or (b) such longer period as the court or tribunal considers equitable having regard to all the circumstances, but that is subject to any rule imposing a stricter time limit in relation to the procedure in question. (6) In subsection (1)(b) “legal proceedings” include— (a) proceedings brought by or at the instigation of a public authority; and (b) an appeal against the decision of a court or tribunal. [the remaining subsections omitted] 54

Human Rights Act, ibid., s. 6(3)(a) and (b). Ibid., s. 7(7). 56 European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 221, Europ. T.S. No. 5 (entered into force 3 Sept. 1953, reprinted in (1994) 33 I.L.M. 943, at 960, Article 34. The contracting parties are members of the Council of Europe. This has over 40 members, including many former Communist states from Eastern Europe. 57 Under the Brussels Convention or Lugano Convention, supra n. 41, or at common law depending on where the torture is alleged to have occurred. 58 Proceedings are subject to a limitation period of one year, beginning with the date on which the act complained of took place, or a longer period if the court considers it equitable having regard to all the circumstances: s. 7(5). 55

Characterisation, Choice of Law and Human Rights 341 of torture to bring proceedings under the Act against a foreign police authority having authorised torture, or even against an individual police officer who tortured the victim in the course of his or her police duties. Consequently, as regards European public authorities and officials, the English courts could recognise a statutory tort of torture if they adopt the foregoing interpretation. The English courts may wish to interpret the Act in such a way that it is presumed not to have an extraterritorial effect, since no mention is made in the Act of its territorial application. This would, however, be inappropriate, both because the statute incorporates a European Convention and because of its subject matter. A further difficulty with the interpretation of the Act as having extraterritorial effect concerns the extent to which a claim under the Act will be qualified by the State Immunity Act.59 There is, however, a very strong argument that, with the incorporation of the European Convention into English law, the State Immunity Act must be interpreted in such a way that it is compatible with the Convention.60 This interpretation of the Human Rights Act giving it extraterritorial effect may appear somewhat radical and, as such, may not prevail in the courts. However, even if the courts do not interpret the Act in this way, it is clear that the Act will have two important implications for the recognition of a tort of torture in English law. Firstly, it is clear that proceedings could be brought under the Act against a United Kingdom state official in respect of torture committed in the UK. It follows that, at least as regards state actors, the Act does recognise a statutory tort of torture. Consequently, if a defendant is sued in this country in respect of torture he or she committed abroad in an official capacity and English law is applicable, because the lex loci delicti is excluded for reasons of public policy or because English law is significantly more appropriate, then this new English statutory tort of torture will apply. This should be the case regardless of the country where the torture was committed. Although this does involve the effective extraterritorial application of the Human Rights Act, this is only the case because the Act creates new causes of action in domestic law applicable only if English law applies to the tort. Secondly, it is likely that the Human Rights Act will have a “horizontal effect”, in that it is likely to influence the judiciary in the United Kingdom to develop existing private law claims in a domestic context, so that they are compatible with the rights recognised by the European Convention on Human

59 It was this Act which defeated the claim of the torture victim in Al-Adsani v. Government of Kuwait (1996), 107 ILR 536, supra n. 43, but this case was decided both before the decision of the House of Lords in Pinochet (No. 3), [1999] 2 WLR 827, supra n. 12, and before the European Convention on Human Rights was incorporated into English law. 60 See section 3(1) of the Human Rights Act 1998, which states that: “So far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights.”

342 G. Virgo Rights.61 It is unlikely, however, that the Convention will be interpreted in such a way as to create new causes of action against non-state actors.62 It should follow that the Human Rights Act will have a significant effect on the development of the existing torts recognised in England in their application to non-state actors. It follows that the Human Rights Act 1998 should be the springboard for the recognition of a nominate tort of torture, regardless of where the torture is committed. Thus it should soon be possible to say, without fear of contradiction, that, in England, torture is a tort.

61 See M Hunt, “The Horizontal Effect of the Human Rights Act”, (1998) Public Law 423, 442. See also G Phillipson, “The Human Rights Act, ‘Horizontal Effect’, and the Common Law: A Bang or a Whimper?”, (1999) 62 MLR 824 and N Bamforth, “The Application of the Human Rights Act 1998 to Public Authorities and Private Bodies”, (1999) 58 CLJ 159. 62 See M Hunt, supra n. 61 at 442.

13

The Emperor’s New Clothes: Defabricating the Myth of “Act of State” in Anglo-Canadian Law MARTIN BÜHLER 1

1 INTRODUCTION

24 M A R C H 1999, the House of Lords ruled that Chilean Senator Augusto Pinochet, the former head of state of Chile, could be arrested and held for extradition to Spain on criminal charges involving human rights violations, including torture.2 While the decision was rightly celebrated by many as a major step in holding former heads of state responsible for violations of human rights, the reasoning used by a number of the Law Lords in analysing the issues addressed in the case should raise concern among those hoping to achieve similar results in the future. This chapter focuses on one concern raised by the Pinochet cases: the apparent acceptance by many Law Lords that English law contains an “act of state” doctrine which operates to limit the ability of courts to inquire into the laws and acts of other states.3 That which has been referred to as the “act of state doctrine” has often been used to support the claim that the courts of one country should, or even must, refrain from inquiring into the legal validity of actions by foreign states or their agents. Problems for those seeking to implead state officials, including heads of state, for violations of human rights will arise when the supposed act of state doctrine interacts with a claim of sovereign immunity ratione materiae by the state official. Sovereign immunity ratione materiae is a public international

O

N

1 B.A.(Criminology), LL.B, barrister and solicitor with Borden Ladner Gervais LLP in Toronto, Ontario. 2 R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No. 3) [1999] 2 All E.R. 97 (hereinafter Pinochet No. 3). This decision replaced the decision of the House of Lords of 25 November 1998, in R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No. 1) [1998] 4 All E.R. 897 (Pinochet No. 1) that was set aside on 15 January 1999 ([1999] 1 All E.R. 577) (Pinochet No. 2). 3 In Pinochet No. 1, Lord Slynn of Hadley discussed the English doctrine at 919, and Lord Lloyd of Berwick equated it to a principle of non-justiciability at 933–4. In Pinochet No. 3, Lord Saville of Newdigate (at 170) and Lord Millett (at 171) both used the term, and Lord Millett actually spoke of the “anglo-american ‘Act of State’ doctrine” at 172.

344 M Bühler law rule which operates to prevent a former head of state or other eligible state official from being subject to the jurisdiction of a foreign court for his or her official conduct as an officer of the state by virtue of the “official” or “sovereign” nature of the conduct.4 By precluding an inquiry into the facts of the conduct and/or into the validity of the laws or decrees invoked to justify the actions in question, the application of an act of state doctrine can frustrate an attempt to argue that the actions of the person concerned were not official in the proper sense or otherwise should not attract immunity. In such an interaction between act of state and immunity ratione materiae, a former state official could shield himself or herself behind laws or decrees of his or her own creation. The immunity claim puts the official nature of the acts in question at issue, but the supposed act of state doctrine precludes examination of the acts and the supposed authority for their performance, solidifying the immunity claim.5 In the Pinochet judgments, a significant number of the Lords in the House of Lords were unwilling to accept the argument that conduct of state officials should be precluded from being treated as “official” for the purposes of immunity on the basis that the conduct violated jus cogens principles of international law.6 If this view of official conduct is adopted in the future by Anglo-Canadian courts, the ability to challenge a claim of official function by a former state official by challenging the legality of the function or action in question under the foreign legal system will become even more important to those seeking to implead former state officials. This possibility adds urgency to the goal of this chapter, which is to show that Anglo-Canadian law does not contain an act of state doctrine as understood by a number of the Law Lords in the Pinochet cases. The supposed act of state doctrine as referred to by certain of the Law Lords in the Pinochet cases is the product of recent court decisions which blended together distinct Anglo-Canadian common law rules with American rules and doctrines, many of which evolved at different times and have divergent premises. The resulting doctrine expressed in the Pinochet cases runs counter to Anglo-Canadian law’s traditional treatment of foreign laws and the acts of foreign states and state officials in Anglo-Canadian law, and should not be followed. To sustain this claim, this chapter will begin by examining the historical treatment of foreign laws and actions in English and Canadian courts, revealing that they were not accorded such a special status as the Pinochet cases suggest. The American act of state doctrine is then examined for comparison. Finally, the chapter attempts to reconstruct the legal path that led to the assertions in the Pinochet cases of the existence of an Anglo-American act of state doctrine and 4 Discussed in Ian Brownlie, Principles of Public International Law (5th ed.) (Oxford: Clarendon Press, 1998) at 333. 5 Perhaps the clearest example of this line of reasoning in the Pinochet cases was that displayed by Lord Millett in Pinochet No. 3 at 172–3. 6 Pinochet No. 3: Lord Browne-Wilkinson at 114 (Lord Saville of Newdigate concurred with Lord Browne-Wilkinson’s reasoning (at 168) ), Lord Hope of Craighead at 147. Pinochet No. 1: Lord Slynn of Hadley at 913, Lord Lloyd of Berwick at 928. The notable exception was Lord Nicholls of Birkenhead, Pinochet No. 1 at 939.

The Myth of “Act of State” in Anglo-Canadian Law 345 to show these assertions are inconsistent with the pre-existing Anglo-Canadian rules in this area.

2 THE ANGLO - CANADIAN LAW TO THE 1980 S

In 1981, the issue of “act of state” came before the House of Lords in Buttes Gas v. Hammer.7 Lord Wilberforce, giving the opinion of the House, examined how this phrase had been used in English jurisprudence. His Lordship described it as “a generally confused topic”, adding that “[n]ot the least of its difficulty has lain in the indiscriminating use of ‘act of state’ to cover situations which are quite distinct, and different in law”.8 He found that the phrase applied to two general categories of situations. One category concerned the applicability of the legislation of a foreign state within its own territory, and the question of whether and, if so, to what degree such legislation was examinable by English courts. It was suggested that this situation was one falling within the field of conflict of laws (private international law)—specifically, the issue of the proper law to be applied.9 The second category concerned situations where officers of the British Crown took action against non-British residents outside British territory otherwise than under colour of legal right. This category is not concerned with a conflict of laws question; its relevance will be addressed at the end of this section of the chapter. Lord Wilberforce went on to identify and explain a separate doctrine of non-justiciability that he felt to be different from either of these two “act of state” categories.10 This non-justiciability doctrine will be discussed after the examination of the act of state issue falling within Anglo-Canadian conflict of laws rules.

Conflict of Laws This section’s analysis of the Anglo-Canadian jurisprudence begins by focusing on what Lord Wilberforce identified as a traditional “act of state” argument: that Anglo-Canadian courts will not examine the validity of foreign laws applicable within the territory of foreign states. Such an assertion is incorrect, for foreign laws can be and have been challenged both on grounds of formal invalidity within the foreign legal system itself and on grounds of repugnancy to the basic values of the Anglo-Canadian legal system. In the case of formal invalidity, courts have examined foreign laws against the constitution of the foreign state to determine whether they are valid under that state’s legal system. In the case of repugnancy, foreign laws which were formally valid have been given no effect 7

[1981] 3 W.L.R. 787. Discussed in greater detail infra n. 55 and accompanying text. Ibid. at 803. 9 Ibid. at 804. 10 Buttes Gas v. Hammer, supra n. 7 at 804 ff. Discussed infra n. 55 and accompanying text. 8

346 M Bühler due to their repugnancy with Anglo-Canadian public policy, including as it incorporates public international law norms. Courts may inquire into the formal validity of a foreign law In Anglo-Canadian jurisprudence, foreign laws are matters of fact to be proved. Decrees and similar non-legislative legal instruments may be lumped in with legislation for this purpose, as such legal instruments will have their own tests for validity within the legal order of the foreign state. A head of state may have the discretion to issue certain decrees by virtue of a power assigned by legislation, or the head of state may have a recognised constitutional power to do so. Either may be proved by experts in the legal system in question. Nevertheless, the argument has been made before Anglo-Canadian courts that, at least where the facts of the case concern the applicability of the foreign law within the territory of the foreign state, Anglo-Canadian courts must stop their inquiry into a foreign law once the content of the law has been proved. That is, the courts must take a foreign law’s validity at face value, and not look behind (or above) the law to see whether it is constitutionally valid in the foreign legal system. An examination of Anglo-Canadian cases shows that, at least since the Second World War, this assertion is without merit. The annexation of Estonia by the USSR during the Second World War resulted in litigation in both Canada and England after the new Soviet government of Estonia attempted to nationalise major capital assets and business, including shipping. That litigation involved title to ships which the Soviet Estonian government had purported to nationalise. Briefly, the background facts of the cases were as follows. Estonia’s government underwent a radical change in late June, 1940, after the Soviet Union invaded Estonia. On 6 August Estonia was admitted into the Soviet Union as the Estonian Soviet Socialist Republic. On 25 August a new constitution was published. In seeking to give effect to the constitution, a number of decrees were issued which purported to nationalise various property, including shipping property.11 In England, the legal validity of the decrees was challenged. The English case involved a dispute over the right to insurance proceeds relating to the steamship Vapper.12 The Vapper had been owned by A/S Tallinna Laevauhisus and insured with Lloyd’s underwriters. The Vapper was torpedoed and lost at sea. When it came time to claim the insurance proceeds, both the plaintiff company and the new Estonian State Cargo and Passenger Steamship Line—to which all nationalised Estonian ships had been transferred by the Soviet Estonian government—claimed ownership of the proceeds. The trial judge found that the Soviet Estonian decree was formally invalid for two reasons. First, a de facto 11

See infra n. 12 at 254–7. A/S Tallina Laevauhisus v. Tallinna Shipping Co. (“The Vapper”) [1946] 79 Ll. L. Rep. 245 (K.B.). 12

The Myth of “Act of State” in Anglo-Canadian Law 347 government must abide by the constitution of the still-recognised de jure government (now in exile);13 second, even if the new constitution were valid, the trial judge accepted the evidence of the legal expert that the nationalisation decree was also ultra vires the new Soviet Estonian constitution.14 On appeal, the first reason of the trial judge was rejected.15 The Court of Appeal stated that the recognition by the British government of the Soviet Estonian government as the de facto government rendered the old constitution irrelevant.16 However, the trial judgment was affirmed on the second ground. Tucker L.J. found that the expert evidence showed the decrees to be unconstitutional,17 while Scott L.J. stated that the lack of evidence on Soviet Estonian law meant that the validity of the decrees had not been proven as a matter of evidence.18 In Canada the same decrees also fell to legal scrutiny. The S.S. Elise was an Estonian steamship owned by two Estonian citizens. The ship ran cargo between the UK and Canada and, as such, was outside Estonian territorial waters when the changes in Estonia’s government and laws occurred. In 1940, the Elise was arrested in Canada by her crew and sold to pay the crew’s wage claims and other debts. Both the Estonian citizens and the Estonian State Cargo and Passenger Steamship Line (the Estonian Line) claimed the remaining proceeds of the sale. At trial, Anglin D.J.A., like the English Court of Appeal, found that the recognition by the Canadian government of the Soviet Estonian government as the de facto government of Estonian made the old Estonian constitution irrelevant.19 In the course of arriving at his decision, Anglin D.J.A stated that the courts had not only a right but also a duty to examine the validity of a foreign legislative act.20 However, he interpreted the agreed statement of facts as having admitted the formal validity of the decrees and, on that basis, found for the Estonian Line. On appeal, the Supreme Court of Canada reversed the decision on the basis that the decree in any case should be considered as confiscatory due to the low rate of compensation to be paid for the ship under the Soviet expropriation law, which was considered repugnant to Anglo-Canadian public policy.21 The decree was therefore held to be unenforceable at least as it purported to effect property located outside the territory of Estonia; as the Elise had been in Canadian waters when the decree had been made, and had never

13

Ibid. at 258. Ibid. at 256–7. 15 [1947] 80 Ll. L. Rep. 99 (C.A.). 16 Ibid. at 112. 17 Ibid. at 114. 18 Ibid. at 110. 19 Estonian State Cargo and Passenger Steamship Line v. S.S. “Elise” (The “Elise”) [1948] 4 D.L.R. 247 (Exch. Ct.) at 259, 266–7. The action was in rem, with the Estonian citizens as intervenors. 20 Ibid. at 264. 21 Laane and Baltser v. Estonian State Cargo and Passenger Steamship Line [1949] 2 D.L.R. 641 (S.C.C.). 14

348 M Bühler entered Estonian waters afterwards, the decree would not be effected against the Elise by the Canadian courts.22 As regards the reasoning of Anglin D.J.A. regarding the right and duty of Canadian courts to examine the formal validity of a foreign statute within its own system, the Supreme Court did not comment unfavourably. In fact, comments by a number of the Supreme Court justices suggest that they supported this statement. Both Rinfret C.J.C. and Kellock J. noted that the decrees in question had been found to be unconstitutional under Soviet Estonian law and therefore unenforceable by the English Court of Appeal in The Vapper and did not comment further on this fact, suggesting that such a finding by an AngloCanadian court was within its judicial ambit.23 In the course of his reasons, Kerwin J. (with whom Estey J. concurred) expressly held that the agreed facts admitted the effectiveness of the decrees, a finding which would have been unnecessary if a foreign law had to be taken at face value once its content was proved.24 Recent case law has confirmed that Anglo-Canadian courts may examine the formal validity of foreign laws within the foreign legal system. After Sierra Leone was established as an independent nation by the United Kingdom in 1961, several residents of the former colony sought to challenge the formal validity of a British Order-in-Council which had been issued in the process of creating the constitution of Sierra Leone. The English Court of Appeal in Buck v. AttorneyGeneral characterised the Order-in-Council as a constitutional (as opposed to merely statutory) instrument.25 The Court refused to issue a declaration of invalidity, stating that where the very subject matter of an action was a claim of invalidity of a foreign constitution, English courts would not entertain the 22 With respect to this extraterritoriality dimension to the invocation of public policy in this case, some of the judges considered the leading English case of Luthor v. Sagor [1921] 3 K.B. 532, as a backdrop. In that case, the English Court of Appeal dealt with the issue of ownership of lumber that originated from a saw mill in Russia. The lumber made its way to the United Kingdom after the mill had been nationalised by the Russian Soviet government and its ownership accordingly transferred from the previous owner under Russian law. The former owner sought a decree that the lumber belonged to him, damages for conversion, and an injunction restraining the defendant, an agent of the Russian state. The Court of Appeal held that the rules governing property required that acts of expropriation by a foreign state within its territorial jurisdiction were not open to challenge on the basis of repugnancy to English law. Note, however, that Bankes LJ limited this rule to situations involving a “civilised country” (p. 546). While this statement may have reflected a European bias against non-European (or Western) legal systems, it could also be the basis for distinguishing situations involving states using their legal systems in a manner totally repugnant to the public policy of the courts before which the matter is adjudicated. This interpretation would complement the reasoning displayed in Helbert Wagg & Co. Ltd., infra n. 36. Compare the Canadian case of Juelle v. Trudeau (1968), 7 D.L.R. (3d) 82 (Que. S.C.). In that case the plaintiffs’ horses were confiscated by Cuban troops under the Castro regime. The horses were alleged to have been brought to Canada to be raced, and certain horses were allegedly sold by the defendants who argued that they had bought the horses from the Cuban government. Mr. Justice Smith ruled that the evidence showed that the confiscation was illegal under Cuban law and thereby found that the plaintiffs were still the owners of the horses (p. 84). 23 Supra n. 21 at 649 and 660. 24 Ibid. at 652. 25 [1965] Ch. 745 (C.A.).

The Myth of “Act of State” in Anglo-Canadian Law 349 action.26 However, Diplock L.J. differentiated cases where the subject matter of a case (such as Buck) was the validity of a foreign constitution from cases where the validity of a foreign law came into question incidentally as a necessary matter for decision in the course of adjudicating private rights and obligations; in the latter situation, the (un)constitutionality of the foreign law could be proved.27 In 1990, the Dubai Bank Ltd. sued for the recovery of funds it alleged had been misappropriated. The defendants sought to establish that the Dubai Bank lacked a relevant legal status under Dubai law because the legal instrument on which it relied was unconstitutional. In Dubai Bank Ltd. v. Galadari and Others (No. 5), the Dubai Bank moved to strike the defence, saying that the English court could not make such a determination.28 Morrit J., relying on Buck v. Attorney-General and The Vapper, stated that while English courts would not entertain an action the object of which was to determine whether a foreign law was constitutionally valid, an English court could determine such a question as a matter of fact where the court was required to do so in order to determine the issue before the court. Furthermore, he stated that certificates from ministers or other officials of the foreign state expressing their view on the constitutionality of the foreign law are not conclusive of the question. The distinction drawn in Buck v. Attorney-General has been adopted by the Supreme Court of Canada. In Hunt v. T & N plc, the Supreme Court had occasion to examine whether the courts of one Canadian province could rule on the constitutionality of legislation enacted by another province.29 In delivering the opinion of the Court, La Forest J. examined the ordinary power of Canadian courts to consider the constitutionality of foreign law in general: “In determining what constitutes foreign law, there seems little reason why a court cannot hear submissions and receive evidence as to the constitutional status of foreign legislation.”30 He noted that Buck v. Attorney-General stood for the proposition that one could not challenge the validity of a foreign constitution itself, but the court could enquire the constitutionality of foreign legislation where this question arose incidentally to the issue before the court. La Forest J. then wrote: “The constitution of another jurisdiction is clearly part of its law, presumably the most fundamental part. A foreign court in making a finding of fact should not be bound to assume that the mere enactment of a statute necessarily means that it is constitutional. Why should a litigant not be able to argue constitutionality in the course

26

Ibid. at 768 and 770. Ibid. at 770. Diplock L.J. used the common situation of an action upon a contract governed by foreign law as an example in which the formal validity of a foreign law might arise incidentally. 28 [1990] Times L. Rep. 490 (H.C.J. (Ch. Div.) ). 29 [1994] 1 W.W.R. 129 (S.C.C.). 30 Hunt v. T & N plc, supra n. 29, “Ordinary Power of Courts to Consider the Constitutionality of Foreign Law”, at 142 ff. 27

350 M Bühler of litigation that directly raises the issue? As a practical matter, it is not much more difficult to determine constitutionality than any other aspect of foreign law.”31

La Forest J. probably overstated his point in making this statement. Canada’s own legal history shows that determining the constitutional validity of legislation can be a far from simple task. Nevertheless, Hunt v. T & N plc should dispel any doubt as to whether there is a rule requiring Canadian courts to take foreign legal instruments at their face value. There is not. The constitutionality of foreign legal instruments is to be treated as a (legal) fact that may be argued like any other. This is compatible with a rule requiring that a foreign law be presumed valid until proven otherwise. Such a presumption, which is arguably already in existence given the respect by courts for comity between legal systems (discussed below), would place the burden of proof on a party seeking to establish the formal invalidity of a foreign law or decree. As a practical matter, cases involving former heads of state or other high state officials are likely to involve resources sufficient to support the production of documents and legal experts able to assist a court in determining the validity of a foreign law as a matter of fact. The preceding cases are critical to actions alleging human rights violations by former heads of state and high officials because they remove the shield which such an official may create for himself or herself simply by issuing decrees purporting to allow or require the actions in question. Treating foreign laws and constitutions in this way would allow courts to find, at the very least, that illegal actions taken to further state goals need not qualify as “official” actions for the purposes of claims of immunity ratione materiae. Finally, it is suggested that treating foreign laws and constitutions in this manner will have the added benefit of forcing foreign states to take their obligations under international treaties seriously, as state officials will not be able to enact municipal laws purporting to make international instruments effective and then ignore their own laws in the belief that the validity of official actions cannot be questioned. Legal ineffectiveness due to repugnancy A court may refuse to give effect to a foreign law even if it is formally valid. The law may be substantively repugnant to the domestic legal system for some reason. Of importance is that in refusing to give effect to a law in such instances, a court is not concerned with the validity of the law (at least within the foreign municipal legal system to which the law belongs); neither is the court finding that the court is prevented from determining the issue in question. The court deals with the matter in issue. However, it declines to apply or recognise within the court’s own legal order the effect of a foreign law that purports to address the matter in issue. The court is still operating within the doctrinal arena of conflicts of laws. 31

Hunt v. T & N plc, supra n. 29 at 143.

The Myth of “Act of State” in Anglo-Canadian Law 351 Anglo-Canadian common law has a few well-recognised categories of foreign laws to which it will not give effect. Of importance is the category of extraterritorial confiscatory laws. Anglo-Canadian courts will not, generally speaking,32 give effect to laws seeking to transfer title to property without giving reasonable compensation to the original owner where the property in question was located outside the territory of the enacting state at the time the law had effect. It was on this basis that the Supreme Court of Canada refused to recognise the transfer of title in the steamship Elise in Laane and Baltser.33 This rule does not affect the recognition of such laws where they operate inside their own territory.34 Courts have, however, refused to give effect to foreign laws even when those laws have purported to operate within the territory of the enacting state, where those laws have been found so repugnant to the Anglo-Canadian legal order that courts should refuse to give them any legal effect. In Laane and Baltser at least one judge would have invoked this ground had it been necessary.35 Courts have offered varying explanations as to the exact nature of the repugnancy, which will be discussed below. First, however, an examination of the cases is required. The laws of Germany established prior to and during World War II raised difficulties for the courts of other countries after the war. Though often used for anti-Semitic purposes, the laws were formally valid within the German legal system and some may even have had acceptable purposes before being used by the Nazi government to further its agenda. In 1955, Upjohn J. of the English High Court of Justice (Chancery Division) was faced with a contract which he determined to be governed by German law.36 The question was raised as to whether a certain German law purporting to regulate currency was really a confiscatory law. The law was formally valid, but Upjohn J. inquired further. He wrote: “[T]his court is entitled to be satisfied that the foreign law is a genuine foreign exchange law and is not a law passed ostensibly with that object, but in reality with some object not in accordance with the usage of nations. The title and the expressed purpose of such legislation are not conclusive upon the point.”37

Upjohn J. actually applied a test of repugnancy to English public policy, but it is clear that he felt considerations of international law to amount to the same thing in this area or, to put it differently, to inform the content of English public policy.38 The important point is that Upjohn J. felt compelled to go behind the formal validity of the law and to examine it for repugnancy. In the result, the law was found to have been a genuine foreign exchange law and was applied. The case of Oppenheimer v. Cattermole is another example of such an examination, one in which four of five Law Lords endorsed a finding that a formally 32 33 34 35 36 37 38

But see Lorentzen v. Lydden & Co., discussed infra n. 52. Supra n. 21. Recall Luthor v. Sagor, supra n. 22. Supra n. 21 at 657. Helbert Wagg & Co. Ltd. [1956] 1 Ch. 323 (H.C.J. (Ch. Div.) ). Ibid. at 352. Ibid. at 349.

352 M Bühler valid foreign law should be given no legal effect by the British courts, albeit a finding that was obiter.39 The case was an income tax case of some complexity, involving double taxation on a German pension by Germany and the United Kingdom. The key issue was whether the British courts would give effect to a Nazi law purporting to strip Jewish citizens of their German citizenship. The Court of Appeal had stated that it was for a state to decide whether a person qualified as its citizen, and that British courts would apply foreign citizenship law “however inequitable, oppressive or objectionable”.40 The House of Lords refused to follow this reasoning. Lord Cross of Chelsea followed Upjohn J. in Helbert Wagg & Co. Ltd., stating that it is the public policy of Britain to give effect to clearly established rules of international law. He then stated: “To my mind a law of this sort constitutes so grave an infringement of human rights that the courts of this country ought to refuse to recognise it as a law at all.”41

The remainder of the Law Lords in the majority gave reasons that were substantially the same, with only Lord Pearson dissenting on this point. Rationales The cases examined above establish two clear rules: 1. Courts can and should inquire into the formal validity of a foreign law within the foreign legal system and decline to give legal effect to an invalid foreign law within the inquiring court’s own legal order. 2. Even if a foreign law is formally valid within the foreign legal order, courts may refuse to give effect to a foreign law if it is repugnant to the public policy of the domestic legal system. The first rule is a logical derivative of a system in which foreign laws are treated as fact and the constitution of a foreign state is its fundamental law. Respect for foreign states requires respect for their constitutions and associated rules of recognition for establishing the validity of their laws. The explanations provided by courts for the second rule have not been a model of consistency. Nonetheless, it is important to develop an explanation for the repugnancy exception if it is to be applied and developed properly. A variety of factors have been named in these decisions to explain the decisions reached by the various courts: comity, public policy of the forum, and public international law. In the end, it is a respect for public international law which emerges as the dominant factor in the decisions, whether in terms of specific norms or by reference more generally to its fundamental values. The fact that contradictory results have been reached over time even when courts have been trying to respect public international law is to be expected. The rules contained in public international 39 40 41

[1976] A.C. 249 (H.L.). Ibid. at 276–7, quoting Buckley L.J., [1973] Ch. 264 at 273. Ibid. at 278.

The Myth of “Act of State” in Anglo-Canadian Law 353 law relating to the nature of state sovereignty have changed dramatically over the past century. As was noted by Viscount Simonds in Regazzoni v. K. C. Sethia (1944) Ltd., in the area of international relations older cases may not provide direct assistance, as the nature of international relationships and associated normative theory (political and legal) change over time.42 Comity Earlier cases tended to use the notion of comity to support their decisions. Comity is a vague concept which often seems to stand in for formal rules of international law. It also speaks to politesse between states, although over time this dimension has become secondary to the recognition that courts should follow the rules of international law. Viscount Simonds in Regazzoni refused to examine the intent of a foreign legislature in enacting a particular law, stating that to do so would be a violation of international comity, which was important to British public policy.43 (This conclusion was effectively overruled by Helbert Wagg & Co. Ltd.). It is clear, however, that Viscount Simonds understood comity to be synonymous with international law.44 In Oppenheimer v. Cattermole, Lord Cross of Chelsea stated that a court should be very slow to deny effect to a foreign law in any sphere in which the foreign state has jurisdiction according to the principles of international law.45 This can be taken as an aspect of comity, requiring courts to be cautious when deciding whether a foreign law is repugnant to international law such that courts ought not to give effect to that law. In Buck v. Attorney-General, Diplock L.J. stated that comity required that courts not exercise jurisdiction with respect to the internal affairs of another state except in accordance with the rules of public international law.46 Thus comity seems to encompass two ideas: first, that courts be cautious in exercising their power to refuse to give effect to foreign laws on the basis of repugnancy; second, that courts make judgments of repugnancy using the rules and principles of international law as the measure. The aspect of caution is consistent with the principle that foreign laws should be presumed valid until proven invalid by the party pleading their invalidity, as was suggested earlier. What must be realised is that comity itself is not a shield, and it does not supply a test for repugnancy.

42

[1957] 2 Ll. L. Rep. 289 at 298 (H.L.). Ibid. at 295–6. Ibid. at 298: “The real question is one of public policy in English law: but in considering this question we must have in mind the background of international law and international relationships often referred to as the comity of nations.” 45 Supra n. 39 at 277. 46 Supra n. 25 at 770. 43 44

354 M Bühler Interference with foreign relations Embarrassment of the government and interference with its ability to conduct foreign relations is, as will be seen, a driving force behind the American act of state doctrine. However, this consideration has never been as important a factor in Anglo-Canadian jurisprudence.47 Lord Salmon in Oppenheimer v. Cattermole stated that he could not see how refusing to give legal effect to a foreign law out of respect for the rules and principles of international law should cause embarrassment to the government of the foreign state.48 But Lord Cross of Chelsea in that case, after warning courts to be slow to hold foreign laws ineffective, did note that a refusal to recognise a foreign law may cause embarrassment to the government of the forum state and affect friendly foreign relations.49 A concern for embarrassment to the government may be seen as informing the rationale for the comity requirement to exercise caution when determining the validity of foreign laws or whether they should be given legal effect by the court. Public international law and public policy The need to respect international law has been frequently cited by courts in reaching their decisions as to whether they could inquire into the formal validity of foreign laws and/or refuse to give effect to them as repugnant. Given the transformations in the role of territoriality and state sovereignty in international law in the twentieth century, it is not surprising to see that earlier decisions placed a premium on respect for territoriality and were very hesitant to do more than take foreign laws at face value, while later decisions have been more willing to inquire into the validity and character of foreign laws.50 The concern for territoriality was reflected in the law surrounding confiscatory legislation, as demonstrated in Laane and Baltser. Formal validity notwith47 The non-justiciability doctrine discussed by Lord Wilberforce in Buttes Gas v. Hammer could be viewed as an exception to this statement, but the circumstances in which it applies are restricted to very specific circumstance as is discussed below. 48 Supra n. 39 at 283. 49 Ibid. at 278. 50 Cranstoun v. Bird (1896), 4 B.C.R. 596 (Div. Ct.), is an example of how strongly Canadian courts at the turn of the last century felt international law linked a state’s territory and the effectiveness of its legal order. Cranstoun was an American citizen in Hawaii who was deemed a security threat by the Hawaiian government. The Hawaiian government had Cranstoun arrested an placed upon a British ship, the captain of which, Bird, was a British citizen. Cranstoun sued the captain for trespass and false imprisonment. The trial court held that Bird had been an agent of the Hawaiian government, and that a court lacked the jurisdiction to discuss the validity of foreign sovereign acts. The Divisional Court commented that Bird would probably have been found to have committed no wrong in Hawaii and that, had Bird been on a Hawaiian ship, the laws of Hawaii would have been applied and no wrong committed. However, once a British ship crossed the threemile-limit of Hawaiian territorial waters, Hawaiian law was of no effect and British law governed Bird. Bird could therefore be found guilty of a tort under British law. The case shows how deeply ingrained the principle of territoriality was in the courts at the time.

The Myth of “Act of State” in Anglo-Canadian Law 355 standing, Anglo-Canadian courts found a taking of property to be repugnant to their legal system and would not give effect to such laws where the confiscated property in question was located outside the foreign state at the time of the confiscation. Conversely, if the foreign law was formally valid, the repugnancy of the foreign law to the Anglo-Canadian legal system would not prevent the courts from giving effect to the law if conflict of laws held that the subject matter was governed by foreign law exercised within the territorial jurisdiction of the foreign state. Territoriality was the critical factor. After the Second World War, a new concern for individuals qua individuals (as opposed to qua state subjects) emerged in international law. A proper application of the rules and principles of international law required that courts respect the new principles of human rights. Jurisdictional territoriality, while still an important principle of international law, was no longer dominant. In delivering their judgments in Helbert Wagg & Co. Ltd. and Oppenheimer v. Cattermole, the courts stated that they were following principles of international law by refusing to give effect to formally valid foreign laws even when these laws had been enacted and exercised within the territorial jurisdiction of the foreign state. English courts have not applied international law directly to conflict of law issues; rather, international law has been applied indirectly through public policy. Foreign laws have been held to be repugnant to public policy because courts have held public policy to require that courts give effect to the rules and principles of international law.51 It may be that judges have been unwilling to bind themselves wholly to international law for fear that a case might arise where the domestic national interest demands that recognised principles of international law not be followed in a conflict of laws situation.52 It may also be that courts have felt that direct application of the rules and principles of public international law is not possible in actions where individuals, not states, raised the international legal validity argument regarding foreign state conduct. In my opinion this indirect method of applying the principles of international law is both inappropriate and unnecessary. It is unnecessary because, in the AngloCanadian legal system, customary international law is a part of the common 51 Regazzoni v. K. C. Sethia, supra n. 42 at 298; Helbert Wagg & Co. Ltd., supra n. 36 at 349; Oppenheimer v. Cattermole, supra n. 39 at 278. 52 Such a situation arose in Lorentzen v. Lydden & Co. [1942] 2 K.B. 202. The Norwegian government, just before departing Norway for England, had decreed that all Norwegian ships outside German-occupied territory would be requisitioned for the war effort. Atkinson J. had to decide whether to give extraterritorial effect to the decree. He decided on the facts that the decree was not confiscatory. He then went on to say that the public policy of Britain demanded that extraterritorial effect be given to the decree in any case as the decree was made by an allied country in time of war and would assist the British war effort. This interpretation of public policy was criticised and not followed by Devlin J. in Bank voor Handel en Scheepvaart N.V. v. Slatford [1953] 1 Q.B. 248 at 265–6. Devlin J. ruled that a similar decree by the Dutch government-in-exile nationalising the property of Dutch residents would not be given extraterritorial effect. Devlin J. said that the public policy test as used by Atkinson J. would require courts to determine the impossible question of whether or not a foreign decree would assist in the realisation of the goals of the British government. Of course, Devlin J. had the comfort of knowing that the war was won when he issued his decision.

356 M Bühler law.53 Rules and principles of general international law that are opposable erga omnes between states, especially those principles that also have the status of jus cogens norms, are directly applicable by courts through the common law. It would be inappropriate—if not improper—for a court to make a decision on a question of conflict of laws that runs counter to principles embodied in the common law. In the end, the norms and principles of public international law have been the dominant—though not the only54—factors guiding the application of the repugnancy exception. Comity and other related concerns have served a secondary role, directing courts to be cautious when determining the validity of foreign laws and in the application of the repugnancy doctrine.

Non-justiciability: Buttes Gas and Matters where no Rules Capable of being Applied by a Municipal Court are to be Found It was noted above that Lord Wilberforce in Buttes Gas v. Hammer identified a principle of non-justiciability that applied to certain subject matters tied to state action.55 Lord Wilberforce said that this principle was distinct from the issues of conflict of laws covered earlier in this section. In Buttes Gas, one party sued the other for slander. The defence put forward would have required the English court to determine the true boundary between sovereign states in the Persian Gulf area, a boundary that at the relevant time was disputed by at least four states in the region. Lord Wilberforce, speaking for the House of Lords, held that this matter was beyond the ability of municipal courts to render decisions. Such subject matter was situated completely within the domain of international law and relations: “It is one thing to assert that effect will not be given to a foreign municipal law or executive act if it is contrary to public policy, or to international law [cf. In re Helbert Wagg & Co. Ltd’s Claim] and quite another to claim that courts may examine the validity, under international law, or some doctrine of public policy, of an act or acts operating in the area of transactions between states.”56

This was not a situation where a municipal court could look to settled international law and apply it to a fact scenario. In Buttes Gas, there was no settled law applicable by a municipal court on the issue before the court. Rules existed in international law for determining such issues, but those rules had to be applied by states themselves. The position of state borders is largely determined 53

Trendtex Trading Corp. v. Central Bank of Nigeria [1977] Q.B. 529 at 553–4. For example, the reasons of Rand J. in Laane and Baltser, supra n. 21 at 657, show that he was concerned with forwarding the principles of a free market system when he applied the repugnancy exception. 55 [1981] 2 W.L.R. 787 at 804 ff (HL). 56 Ibid. at 804. 54

The Myth of “Act of State” in Anglo-Canadian Law 357 by negotiation leading to agreement between states.57 Failing agreement, international processes of arbitration or adjudication become the relevant avenues of dispute settlement. In short, in boundary disputes the law is in the hands of states and, where states do not agree, there is nothing for a municipal court to apply to the facts. The matter was non-justiciable, not because state actions were involved but because the required state actions—determination of the boundary—had not yet occurred. The narrowness of the area of non-justiciability identified by Lord Wilberforce cannot be overemphasised. As Lord Wilberforce stated, not all sovereign acts are non-justiciable. Only a “certain class of sovereign acts”, which exist solely in the domain of international law and which auto-define their legality, fall under the Buttes Gas category of non-justiciability.58 It is worthwhile to consult the cases cited by Lord Wilberforce to understand the types of sovereign acts to which he was referring. The English cases cited by Lord Wilberforce were: Duke of Brunswick v. King of Hanover,59 Cook v. Sprigg 60 and Secretary of State in Council of India v. Kamachee Boye Sahaba.61 Duke of Hanover involved a decree issued by the German princes of sovereign states addressing the recognition of the head of another German state. In Cook v. Sprigg and Secretary of State in Council of India, parties sought to challenge the legitimacy of the acquisition of territory by the British government, by secession and seizure respectively. In each of these cases, the subject matter was a transaction between states involving recognition of statehood and state boundaries, characterised as being determined by rules of international law outside the jurisdiction of municipal courts.62 Conversely, Lord Wilberforce cited the American case of Banco Nacional de Cuba v. Sabbatino as an example of a situation that was justiciable and to which the conflict of law rules would apply.63 The sovereign action in that case was nationalisation of property within Cuban territory by the Cuban government. This involved the determination of title to property, a situation for which municipal courts have clear rules, including the conflicts of laws rules discussed earlier in this chapter. It is clear that Lord Wilberforce intended to keep the rule of non-justiciability separate from the conflict of laws principles discussed earlier in this chapter. The 57

Ian Brownlie, supra n. 4 at 214. Supra n. 55 at 806. 59 (1844) 6 Beav. 1; (1848) 2 H.L.Cas 1, ibid. at 805. 60 [1899] A.C. 572, ibid. at 806. 61 (1859) 13 Moo.P.C.C. 22, ibid. 62 Certain manners of acquisition of territory would likely be justiciable by municipal courts today in light of the development of international law since these cases were decided. See e.g. Friendly Relations Declaration of the United Nations General Assembly, UNGA Res. 2625 (XXV) of 24 October 1970, providing that states shall not use force to threaten the territorial integrity of other states. This application of international principles might be prevented by the fact that AngloCanadian courts will look to their governments for a decision on whether the government of a foreign territory is recognised, a decision which could be seen as implicitly commenting on the legitimacy of the formation of the government in question. 63 (1964) 376 US 398, supra n. 55 at 806. 58

358 M Bühler non-justiciability rule does not operate to make all foreign laws or state actions unexaminable. Neither does the non-justiciability rule deem certain laws or acts of a foreign state as valid and effective. It is not based upon a desire to avoid the embarrassment of states. It outlines an area in which the courts cannot function because there is no law available to them to apply to state action. This can occur where interstate action, agreement and/or interpretation is required before the international law on a specific matter can be said to be known, and the relevant states or international bodies have not yet acted, agreed and/or interpreted.

Application to Cases Involving Violations of Human Rights The conflict of laws and justiciability rules examined in this section should not impede an action in a Canadian court against a foreign state or state official (including a head of state) for actions involving the violation of human rights. The rule of non-justiciability in Buttes Gas would not be triggered, as international law now has clear principles and rules governing actions by states and their officials against individuals. Indeed, actions such as torture by state officials are the subject of jus cogens norms, the highest governing principles in international law from which states and their agents may not deviate. It is likely that the issue of formal validity could rarely be raised successfully by a defendant. A state is likely to have laws prohibiting actions of the nature of torture even if those laws are routinely ignored; plaintiffs could thereby turn to such laws to support their cases. It may be, however, that the defendant would rely on a decree or order authorising the actions in question. If the decree or order were characterised as a law itself, then the plaintiff would be able to argue that it was formally invalid due to the foreign state’s other laws or its constitution. If the decree or order was characterised as an interpretative action—that is, as an implicit statement by the person ordering the action that it was legal under the laws of the state in question—this would be analogous to expert evidence as to the formal validity of the decree or action, but would not be conclusive of the matter. Finally, even if a formally valid foreign law were found to authorise the action, that law would be repugnant to public international law and Canadian public policy and should therefore be given no effect by the court. The issue of amnesties poses a difficulty. Should an Anglo-Canadian court give effect to a foreign nation’s amnesty law that puports to bar all criminal charges and civil claims against persons guilty of violations of the foreign legal order? I would suggest that the amnesty law should be examined in the same manner as any other law, first for validity within the foreign legal order and then, if the law is valid, for repugnancy to the Anglo-Canadian legal order. It would likely be a matter determined by an evaluation of whether the amnesty law should be characterised as a genuine attempt to benefit the citizens of the state concerned or instead as a mechanism to further prevent a wronged popu-

The Myth of “Act of State” in Anglo-Canadian Law 359 lace from seeking compensation for the illegal actions of a group of victimisers.64

The Crown’s Prerogative Power and “Act of State” It was noted above that Lord Wilberforce had described two distinct categories of situations which had been referred to as “act of state” situations. The conflict of laws category has been examined. The other category was identified as encompassing situations where officers of the Crown took some disputed action against non-British residents outside British territory otherwise than under colour of legal right. This category cannot be summed up by a simply stated rule. Generally, it operates to limit the ability of British courts to either adjudicate or award damages for situations where the Crown has exercised some aspect of its prerogative powers outside the territory of its realm. The issue was discussed by the House of Lords in Nissan v. Attorney-General.65 That case concerned a British citizen who had operated a hotel in Cyprus. In December 1963, British troops arrived in Cyprus at the invitation of the government to help keep the peace after civil strife erupted. The troops became part of a United Nations force in March 1964. In December 1963, British soldiers occupied the plaintiff’s hotel and subsequently damaged it. The plaintiff attempted to sue the Crown for damages. The Crown claimed that the suit must fail as what had occurred had been an “act of state”. The Law Lords noted that the authorities in this area were divergent. Lord Reid, for example, stated: “I think that a good deal of the trouble [in this area] has been caused by using the loose phrase ‘act of state’ without making clear what is meant.”66 In the end, the Lords decided that the Crown could not benefit from the rule on the facts before the House. It is clear that this category of “act of state” is a product of the historical development in Britain of the nature of the prerogative power of the Crown. It is not a principle derived from international law, public or private. As such, it should not be used to assist in the analysis of the rules in Anglo-Canadian law examined earlier. Unfortunately, this category was not addressed by Lord Wilberforce as clearly as it could have been, thereby undermining his attempt to have it recognised as a distinct legal rule.67

64 See J Llewellyn, “Just Amnesty and Private International Law”, chapter 22 in this volume, for discussion of this issue. 65 [1970] A.C. 179 (H.L.). 66 Ibid. at 211. 67 Infra n. 93 and accompanying text.

360 M Bühler

3 THE AMERICAN ACT OF STATE DOCTRINE

Courts in the United States recognise the existence in American law of a doctrine called “act of state”. The American act of state doctrine touches upon many of the issues referred to by the Anglo-Canadian conflict of laws rules examined earlier. However, the rationale for the doctrine and the way in which the doctrine deals with the issues are very different from the AngloCanadian rules. Overview The Restatement of the Law (Third): The Foreign Relations Law of the United States describes the American act of state doctrine this way: “In the absence of a treaty or other unambiguous agreement regarding controlling legal principles, courts in the United States will generally refrain from examining the validity of a taking by a foreign state within its own territory, or from sitting in judgment on other acts of a governmental character done by a foreign state within its own territory and applicable there.”68

This American act of state doctrine can be described as a non-absolute conflict of laws rule with quasi-constitutional strength. According to the latest word on the subject from the United States Supreme Court, it operates to prevent an American court from treating as invalid and ineffective the official act of a foreign sovereign performed within that state’s own territory.69 A difficulty with the American act of state doctrine is that American courts have changed the rationale for the rule over time. In 1964, the United States Supreme Court stated, in Sabbatino, that while the doctrine had at one time been seen as a derivative of international law and of the very nature of sovereign authority, it was actually based on the principle of separation of powers between the executive and judicial branches of the American government—specifically, the powers concerning the conduct of foreign affairs.70 One must be cautious, therefore, when attempting to derive principles of the doctrine from the earlier cases. The Sabbatino case concerned a challenge to the legality of the nationalisation by Cuba of sugar located in Cuba. The challenge was made on the grounds of formal validity. The nationalisation decree had not been published in the Cuban Gazette. The US Supreme Court held that a trial court could not inquire into the validity of the Cuban law; rather, the decree had to be taken as valid.71 The Supreme Court referred to the division of powers between the judiciary and 68 The American Law Institute, Restatement of the Law (Third): The Foreign Relations Law of the United States (St. Paul, Minn., American Law Institute Publishers, 1987), §443(1). 69 W.S. Kirkpatrick v. Environmental Techtonics, 493 US 400 at 405, 110 S.Ct. 701 (1990). 70 Banco Nacional de Cuba v. Sabbatino, 376 US 398 at 421 ff, 84 S.Ct. 923 (1964). 71 Compare to Juelle v. Trudeau, supra n. 22.

The Myth of “Act of State” in Anglo-Canadian Law 361 the executive in American law. It held that an examination by an American court of the formal validity of a taking of property by a foreign state within that state’s territory would impede the ability of the executive to conduct foreign relations. The plaintiff had also alleged that such a taking violated customary international law. The Supreme Court stated that the doctrine prevented courts from commenting on sensitive and unsettled areas of international law, leaving such areas to be determined by the foreign relations branch of the executive.72 In keeping with this justification, the Supreme Court wrote the following key passage: “It should be apparent that the greater the degree of codification or consensus concerning a particular area of international law, the more appropriate it is for the judiciary to render decisions regarding it , since the courts can then focus on the application of an agreed principle to circumstances of fact rather than on the sensitive task of establishing a principle not inconsistent with the national interest or with international justice.”73

Significantly, in this separation-of-powers formulation, the Court noted that not all questions of international law were of equal importance to the American government, and that courts would have less justification for applying the doctrine where the question at issue would not seriously impact on American foreign relations. Furthermore, where the foreign government concerned had changed since the time of the relevant “act of state”, foreign relations might not be affected by a judicial determination of the question at issue.74 The Supreme Court had occasion to re-examine the nature of the American act of state doctrine in W.S. Kirkpatrick v. Environmental Techtonics. The Court seemed intent on limiting the scope of the doctrine. The suit was launched after the plaintiff and the defendant had competed for a contract with the Nigerian government to build an aeromedical centre at a Nigerian Air Force base. The defendant won the contract, and the plaintiff launched a suit under various American statutes on the allegation that the defendant had bribed 72 Supra n. 70. at 418, 428–30, 437. The foreign relations branch would presumably act through diplomacy or international tribunals to forward claims on behalf of individuals who could find no relief in American courts due to the operation of the American act of state doctrine: Ibid. at 423. A question unaddressed by the Supreme Court is whether the American government can act in any particular case by informing the court whether or not the matter before the court would, in the opinion of the government, prejudice the foreign relations of the government. The American government has acted in this manner; such letters are referred to as “Bernstein letters” after the twin cases of Bernstein v. Van Heyghen Freres Societe Anonyme, 163 F.2d 246 (2nd Cir. 1947) and Bernstein v. N.V. Nederlandsche-Amerikaansche Stoomvaart-Maatschappij, 210 F.2d 375 (2nd Cir. 1954). In the former case, the United States Court of Appeal (Second Circuit) had applied the act of state doctrine to refuse to hold a law of the Nazi government to be invalid. The American government wrote a letter stating that it was of the opinion that the doctrine should not be applied in the circumstances, and the Court allowed the matter to proceed in the second case. The American Supreme Court has avoided ruling on the status of such letters: Sabbatino, supra n. 70 at 420; W.S. Kirkpatrick v. Environmental Techtonics, supra n. 69 at 405. 73 Sabbatino, supra n. 70 at 428. 74 Ibid.

362 M Bühler Nigerian officials to obtain the contract.75 The Supreme Court noted that in order to prove the action, the trial court would have to find that Nigerian officials had awarded the contract on the basis of bribes, but it would not be necessary for the dispute at hand to look at the legal status of the bribery.76 The Supreme Court held that the American act of state doctrine did not apply in such circumstances.77 In its reasons the Supreme Court expressly affirmed Sabbatino;78 however, less imperative language was used to describe the basis for the doctrine. Specifically, the Court identified the following as policies supporting the application of the doctrine: (1) international comity; (2) respect for the sovereignty of foreign nations on their own territory; and (3) avoidance of embarrassment to the Executive Branch in its conduct of foreign relations.79 Additionally, the Supreme Court stated that a prerequisite for triggering the application of the doctrine was that a court be required to find the act of a foreign state within its own territory to be invalid.80 The court differentiated between the motivation for an “act of state” and the validity of an “act of state”. The American act of state doctrine was said to apply only to the latter. A court may decline to apply the doctrine when this factual prerequisite exists if the policies behind the doctrine are not present. However, even where the policies driving the doctrine are present, a court cannot apply the doctrine unless the factual prerequisite exists.81 The American act of state doctrine “merely requires that the acts of foreign sovereigns taken within their own jurisdictions shall be deemed valid” where the policy considerations demand it.82

Illegality and Official Acts The issue of the illegality—both under the foreign municipal law and under international law—of a foreign “act of state” under the American act of state doctrine is complex. On the one hand, if the doctrine applies, then the foreign “act of state” in question is deemed to be valid even if it is illegal under the foreign legal system or under international law.83 The United States Court of Appeal (Second Circuit) has stated that if an “act of state” was carried out by a state official in his or her official capacity, the American act of state doctrine can be applied even if the act in question was illegal under the foreign municipal

75 Supra n. 69.The plaintiff alleged violations of, inter alia, the Racketeer Influenced and Corrupt Organisations Act, 18 U.S.C. §1961 et seq. (RICO). 76 Supra n. 69 at 401. 77 Ibid. at 409–10. 78 Ibid. at 403, 409. 79 Ibid. at 408. 80 Ibid. at 406. 81 Ibid. at 409. 82 Ibid. 83 Sabbatino, supra n. 70 at 422.

The Myth of “Act of State” in Anglo-Canadian Law 363 law.84 Similarly, it has been stated that, where an act is within a state’s discretion, the American act of state doctrine may apply notwithstanding that the act violates international law.85 On the other hand, in other cases both foreign municipal law and international law have been used to determine whether the act in question was a true “act of state” such that it triggers the application of the doctrine. For example, the United States Court of Appeal (Second Circuit) has stated that an act by a foreign official which violates the constitution of the foreign state and which has not been ratified by the government of that state is probably not an “act of state” afforded the benefit of the American act of state doctrine.86 The United States Court of Appeal (Ninth Circuit) has stated that in determining whether an official acted in his or her official capacity, a court can look to the statutes of the foreign state to determine the scope of their official duties.87 This court has also stated that the American act of state doctrine does not exist for the purpose of providing the equivalent of a sovereign immunity defence to a former head of state.88 Given that former head of state immunity exists ratione materiae (as a function of the nature of the conduct in question), this would imply that the illegality of acts taken by state officials in performing an official function should be considered when deciding whether to apply the doctrine.

The American Act of State Doctrine and Torts Involving Violations of Human Rights Despite the lack of clarity in American law on the relationship between illegal action by foreign officials and the American act of state doctrine in general, it appears settled that, in the area of human rights violations, it will be almost impossible for a party to successfully seek the application of the doctrine. The Third Restatement of the Foreign Relations Law of the United States takes the view that, where an action is for a fundamental violation of human rights such as torture or genocide, the American act of state doctrine would probably not apply because the international law of human rights is both well established and contemplates scrutiny of such acts by foreign states.89 American courts have 84 Republic of Philippines v. Marcos, 806 F.2d 344 at 359 (2nd Cir. 1996) (action by Philippines against ex-head of state to recover funds allegedly embezzled while in office.) 85 Flatlow v. Islamic Republic of Iran, 999 F.Supp 1 at 24 (D.D.C. 1998) (action against Iran and Iranian public officials for supporting terrorist attack in Israel that killed an American citizen.) 86 Kadic v. Karadzic, 70 F.3d 232 at 250 (2nd Cir. 1995) (action against head of de facto state— Serbian Republic “Srpska”—for human rights violations.); Filártiga v. Peña-Irala, 630 F.2d 876 at 889 (2nd Cir. 1980) (action against Paraguayan state official for torture). 87 Hilao v. Marcos (In re Estate of Ferdinand E. Marcos Human Rights Litigation) (“Estate II”), 25 F.3d 1467 at 1471–2 (9th Cir. 1994), cert. denied 115 S. Ct. 934 (action for human rights violations against estate of former head of state.) 88 Republic of the Philippines v. Marcos, 862 F.2d 1355 at 1361 (9th Cir. 1988). 89 Supra n. 68, §443, comment “c”.

364 M Bühler used international law in two ways to achieve such a result. First, they have found the international law to be so well established on state accountability for human rights violations that the policy concerns for the application of the doctrine do not apply to such cases.90 Secondly, the courts have found that where human rights violations can be described as being the subject of jus cogens prohibition by international law, such acts are incapable of being official acts of a state or state official.91

Comparison to Anglo-Canadian Law The American act of state doctrine bears a superficial resemblance to the conflict of laws rules and the doctrine of non-justiciability in Anglo-Canadian law. All arise when foreign laws or state action must be considered in a domestic forum. Public policy, international law, comity and the tension between courts and executives in the formation and interpretation of international law have been considered by courts in the development of the American and AngloCanadian rules. At times they lead to the same result. However, the AngloCanadian rules differ from the American act of state doctrine in fundamental ways. First of all, the Anglo-Canadian rules allow full inquiry into the formal validity of foreign laws. Where a foreign law or act is found to be formally invalid according to the foreign legal system, it will not be applied. The American act of state doctrine looks only at content, deeming foreign laws and actions to be valid within foreign territory. It applies unless one can show that the policies which drive it—policies derived from the American constitutional order—do not apply. The effect of illegality of actions by state officials on the application of the American doctrine is unclear. Secondly, the American act of state doctrine is only applicable where the acts in question occurred in the territory of the foreign state. The Anglo-Canadian non-justiciability rule has no such limitation. And thirdly, the American act of state doctrine is grounded in American constitutional theory. The status of international law is but one policy informing the application of the doctrine, which is primarily concerned with preventing interference in the conduct of foreign relations by the executive. The doctrine can be applied even where the “acts of state” violate international law. The AngloCanadian repugnancy rule is based predominantly on the rules and principles of public international law, although it may be applied through the gateway of 90 Doe v. UNOCAL Corp., 963 F.Supp 880 at 893 (C.D.Cal. 1997) (action by Burmese citizens against Burmese government and an American oil extraction company acting in a joint venture alleging, inter alia, torture and slavery (forced labour) ). 91 Hilao v. Marcos (“Estate II”), supra n. 87 at 1471–2; Paul v. Avril, 812 F.Supp 207 at 212 (S.D. Fla 1993) (action against former head of state of Haiti for acts including torture); Xuncax v. Gramajo, 886 F.Supp. 162 at 175–6, 176 n. 10, 184 (action against former military commander and later Minister of Defence for Guatemala alleging acts including torture); Doe v. UNOCAL, supra n. 90 at 894.

The Myth of “Act of State” in Anglo-Canadian Law 365 public policy. In Anglo-Canadian law, the concern for embarrassing states is properly seen as best encompassed by the principle of comity, which in essence instructs courts to proceed carefully when examining the laws and actions of foreign states.

4 THE PATH TO PINOCHET

The Anglo-Canadian Law becomes Confused Two related decisions in the early 1980s set the stage for a drastic turn in AngloCanadian jurisprudence surrounding the “act of state” and related rules. Both cases originated on the same facts. A Cuban state corporation, Cubazucar, had had a contract with a Chilean corporation, Iansa, to deliver sugar to Chile. Delivery occurred on ships owned by Cuba. Delivery had been partially completed when the Pinochet-led coup occurred on 11 September 1973. Cuba broke off relations with the new right-wing Chilean government and Cubazucar refused to complete the contract. The first case, I Congresso del Partido, was an in rem proceeding against a sister ship of the ships which had been carrying the sugar in question, the Playa Larga and the Marble Islands.92 Iansa brought an in personam action against Cuba for the loss of the sugar. This case was concerned with the state immunity of Cuba, the defendant. In introducing the issue of state immunity, Lord Wilberforce stated:93 “The basis upon which one state is considered to be immune from the territorial jurisdiction of the courts of another state is that of ‘par in parem’ which effectively means that the sovereign or governmental acts of one state are not matters upon which the courts of other states will adjudicate.”94

This looks similar to the rationale for the American act of state doctrine. It must be realised that the issue under discussion was of the nature of state immunity ratione personae. Such immunity is rooted in the nature of the state itself; it does not depend on the nature of the facts before the court.95 While Lord Wilberforce’s statement appears to emphasise acts, it is the nature of the (state) actor that drives this rule. Unfortunately, to someone unfamiliar with this area of the law, the quotation could be taken to mean that acts by states are in a category of actions that may not be adjudicated upon, a prohibition akin to the American act of state doctrine. 92

[1981] 3 W.L.R. 328 (H.L.). Dissenting in part on the result, but speaking for the House on this point. 94 Supra n. 92 at 334. 95 This is true notwithstanding the move to a “restricted” state immunity for commercial subject matters. The exception speaks to a change in the nature of states—from sovereign actor to market participant—not to some fundamental importance about commercial subject matter. 93

366 M Bühler In the second case, Iansa sought to recover from Cubazucar on the basis of their contract. The contract went before English arbitration applying English law, and eventually appeared before the Court of Appeal.96 Cubazucar’s defence included the plea that the refusal to deliver the sugar was an “act of state”. In a stunning manoeuvre, Ackner L.J. (writing for the court) took a quote from Nissan v. Attorney-General out of context, treated the different rules of “act of state” that had been identified in Buttes Gas—namely, the rule concerning the action of the Crown in exercising its prerogative powers outside the British territory and the rule on examining the formal validity of foreign laws within their territory—as if they represented one doctrine, and then discussed the resulting doctrine as if it and the American act of state doctrine were the same thing.97 In doing so, Ackner L.J. referred back to I Congresso del Partido when comparing the nature of state immunity to the nature of this new “act of state” doctrine.98 This breathtaking recreation of the law occurred with minimal analysis. Suddenly the different rules which had unfortunately been identified by the same label—“act of state”—were transformed into Act of State, a new and powerful doctrine shielding foreign laws from judicial scrutiny. This reasoning was completely at odds with Lord Wilberforce’s careful separation of the rules in Buttes Gas, and his description of the situations which they address as “situations which are quite distinct, and different in law”.99 The unfortunate wording of the rationale for ratione personae state immunity in I Congresso del Partido and the discussion of distinct situations—the conflict of laws and justiciability rules—as if they were one coherent doctrine, similar to the American act of state doctrine, set the stage for further misconception by judges as to the manner in which Anglo-Canadian law treats foreign laws and the actions of foreign states. The misconception surfaced in the Pinochet cases.

Pinochet The speeches in the House of Lords in the Pinochet cases bore the sour fruit produced from a cross-pollination of these two lines of thought. Suddenly the laws and actions of foreign states were afforded a protection from examination they had never had under the standard Anglo-Canadian conflict of laws rules examined earlier. Much of the Law Lords’ description of the nature of sovereign immunity ratione materiae resembles the principles behind the American act of state doctrine, principles which had not before been applied in conflict of laws to state laws or actions. Lord Millett stated, 96 Empresa Exportadora de Azucar v. Industria Azucarera Nacional S.A. (The “Playa Larga” and “Marble Islands”) [1983] 2 Ll. L. Rep. 171 (C.A.). 97 Ibid. at 193. 98 Ibid. 99 Supra n. 55 at 803.

The Myth of “Act of State” in Anglo-Canadian Law 367 “Given its [immunity ratione materiae] scope and rationale, it is closely similar to and may be indistinguishable from aspects of the anglo-american ‘Act of State’ doctrine.”100

Similarly, Lord Phillips of Worth Matravers stated, “Where a state is not directly or indirectly impleaded in the litigation, so that no immunity as such arises, the English and American courts have nonetheless, as a matter of judicial restraint, held themselves not competent to entertain litigation that turns on the validity of the public acts of a foreign state, applying what has become known as the act of state doctrine.”101

His Lordship felt this conclusion to be supported by Buck v. Attorney-General. Lord Hope of Craighead felt comfortable making an equivalent statement, citing I Congresso del Partido.102 In the end, the Law Lords supporting this “act of state”-based immunity ratione materiae determined that it afforded Pinochet no defence only because the terms of the International Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment,103 as incorporated in material respects into UK law, were held to be irreconcilable with the operation of such an immunity.104 This line of reasoning is irreconcilable with the rules of Anglo-Canadian law examined earlier in this chapter. This is not to say that immunity ratione materiae cannot exist for former heads of state for such crimes; it may be that it can, but it cannot be founded on such reasoning unless all the jurisprudence examined in this chapter has been overruled implicitly by this decision. In a civil suit against Pinochet, these rules, if unchanged, would have operated as follows. The formal validity rule would allow the lawfulness of the state action to be examined to see if it was valid under Chilean law. Pinochet would have had to introduce evidence that he had the authority to perform the acts in question either through legislation or the constitutional status of his position. Failure to do so would mean that his actions could be found to have been formally invalid and of no applicability. Even if his conduct was formally valid under Chilean law, the repugnancy rule would require that his actions, found to be jus cogens violations by the House of Lords, be given no effect by the English courts, even when the conduct occurred within Chilean territory, thus depriving Pinochet of a valid defence. The Buttes Gas non-justiciability rule would not be triggered as the subject matter of the international rules is far removed—even by analogy— from the boundary and sovereign recognition issues of Buttes Gas, especially given the settled jus cogens nature of torture. Buck v. Attorney-General would not be a barrier, as the action would be for damages due to harm to the person, 100

Pinochet (No. 3), supra n. 2 at 171. Ibid. at 186. Ibid. at 146. 103 Adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), 1465 U.N.T.S. 85, reprinted in 23 I.L.M. 1027. 104 Pinochet (No. 3), supra n. 2: Lord Hope of Craighead at 152; Lord Millett at 178–9; Lord Phillips of Worth Matravers at 189–90. 101 102

368 M Bühler not a direct challenge to the validity of a Chilean law. The adjudication of the validity of Chilean law would arise only incidentally if Pinochet relied upon it in his defence. In his judgment, Lord Hutton made reference to a decision of the Ontario Court of Appeal, Jaffe v. Miller,105 as supporting the proposition that officials of a foreign state may claim immunity for actions carried out by them in ostensible exercise of their governmental functions, notwithstanding that their actions were illegal (under the foreign legal system). Such a broad statement, if taken at face value, would seriously impair attempts to implead foreign officials—or persons who were formerly foreign officials—in Anglo-Canadian Courts. While this chapter is concerned with act of state and not sovereign immunity, the issue addressed in Jaffe v. Miller raises many of the same issues addressed in this chapter. If Lord Hutton’s obiter is picked up in the future by courts, it has the potential to outflank the arguments made in this chapter by recreating the new “Act of State” doctrine within the rubric of sovereign immunity. In Jaffe v. Miller, the plaintiff brought an action in tort against officials in the Office of the Attorney General of the State of Florida for malicious prosecution and conspiracy. The plaintiff alleged that he had been kidnapped in Canada by bounty hunters hired by Florida State officials, and that he had been maliciously and falsely prosecuted in the State of Florida. The defendants who were persons working in the Office of the Attorney General brought a motion to have the plaintiff’s claim against them dismissed on the grounds of immunity from suit. These defendants submitted, successfully, that they were included in a definition of “foreign State” found in Section 2 of the State Immunity Act.106 The Ontario Court of Appeal affirmed this decision. Finlayson, J.A., for the Court, adopted an expansive definition of foreign state for the purposes of Section 2 of the State Immunity Act.107 He based his decision both on cases he felt supplied precedent for who and what fell under the definition of a foreign state, and upon the reasoning that to deny immunity to state functionaries—which he felt to include the defendants in this case—would be to improperly allow states to be impleaded indirectly, as states would presumably indemnify their officials against any judgment obtained in a law suit. The plaintiff submitted that the defendants in question had acted illegally, and that this deprived them of the benefit of a claim to state immunity. Finlayson J.A. noted that the plaintiff’s Statement of Claim did not contain an express allegation that any of the defendants were acting outside of their official capacities.108 In any case, he also stated that “the illegal and malicious nature of the Acts alleged do not of themselves move the actions outside the scope of the official duties of the responding defendants.”109 105 106 107 108 109

103 D.L.R. (4th) 315, referred to by Lord Hutton in Pinochet (No. 3) at 157 and 167. R.S.C. 1985, c. S-18. Supra n. 105 at 325 ff. Ibid at 329. Ibid.

The Myth of “Act of State” in Anglo-Canadian Law 369 The first important aspect of this decision is that it concerns the question of state immunity ratione personae. That is, the immunity being discussed was the immunity associated with a state qua state, not immunity associated with a particular class of action. As such, Jaffe v. Miller only applies to situations in which the plaintiff seeks to implead persons who are currently officials of a state. As it is based upon the theory that the state itself should not be impleaded, the reasoning in Jaffe v. Miller should not apply to situations in which the party impleaded as a defendant is an ex-official of a state. If an ex-official were to be impleaded for acts done while he or she was an official with a state, the exofficial cannot plead state immunity ratione personae. If the ex-official wished to be indemnified by the state for which he or she was employed, the ex-official should have to attempt to launch a third party action for indemnity. In such an action, the state itself could raise the issue of its own immunity ratione personae in its own defence. While immunity ratione personae would (or should) be unavailable, the ex-official could plead that his or her acts were authorised by the laws of the state for which he or she worked—a typical act of state defence. In such a case, the acts would have to be examined against the laws of the country in the manner discussed in this chapter to see whether or not they could provide a valid defence for the ex-official. If courts insist on adopting such an expansive definition of which actors may be included under the definition of state for the purposes of claims for state immunity, then they must also be vigilant in remaining true to the theoretical underpinnings for claims of state immunity ratione personae. As already indicated, in Jaffe v. Miller, the expansive definition was based on the reasoning that to allow state officials to be impleaded, plaintiffs would be achieving indirectly what they could not achieve directly—impleading states themselves. This was said to be so because states would be obliged to indemnify their officials against any judgment obtained against the officials. This reasoning assumes a situation in which the state official is acting on behalf of the state. For this to be the case, the state official must at minimum have been acting within his or her jurisdiction as a state official. This appears to have been recognised in the decision of Finlayson J.A. when he stated, “the illegal and malicious nature of the acts alleged do not of themselves move the actions outside the scope of the official duties of the responding defendants.”110 Jaffe v. Miller then, recognised that where actions of a state official fall outside the official duties of that official, sovereign immunity ratione personae does not apply. Thus Jaffe v. Miller recognised, at least implicitly, that actions of an official of a foreign state claiming the benefit of state immunity must be examined to determine whether they come within the scope of their official duties. The issue is not one of illegality per se, but rather one of jurisdiction and ultra vires actions by state officials. In the end, this requires an examination of the laws of a foreign state to see whether actions taken by a state official were within the jurisdiction granted to them by the laws 110

Ibid. (emphasis mine).

370 M Bühler of the foreign state. The analysis in this chapter must be as applicable to such an examination of the jurisdiction of a foreign official when he or she raises a sovereign immunity defence as to the examination of the legal status of acts of foreign states when raised as defences by non-state actors. Were a different analysis to prevail instead, the scope of the defence of sovereign immunity ratione personae would be expanded in a manner that is completely at odds with the principles embodied in the cases addressing the issues of conflict of laws examined in this chapter—a key principle being a respect for the principles of public international law. A court should accordingly examine the actions at issue when a foreign state official is impleaded, and determine first of all whether the acts in question were ultra vires the jurisdiction of the official. If they are within his or her jurisdiction, the court must go on to ask whether the foreign laws authorising such action were so repugnant to Anglo-Canadian legal values that they should not be given recognition by Canadian courts.

5 CONCLUSION

As Lord Wilberforce recognized in Buttes Gas, the phrase “act of state” has been used indiscriminately to refer to concepts that operate in the same field of law— conflict of laws—but that are very different in their underlying premises. Use of the phrase without care for the differing principles that have shared its name has led to incorrect generalisations about how the laws and actions of foreign states and their officials are to be treated by Anglo-Canadian courts. The Pinochet decisions are cases in point. Unfortunately, the very ground-breaking nature of arrest and detention of Pinochet may result in the use of the Pinochet decisions as the starting point for analysis of the issue of act of state in subsequent cases, when the Pinochet decisions have misstated pre-existing rules relating to conflict of laws and public policy. Despite the impression one might have from the more recent decisions such as The “Playa Larga” and “Marble Islands”, Jaffe v. Miller and the Pinochet decisions, the laws and actions of foreign states and their officials are not possessed with some potent quality that prevents Anglo-Canadian courts from examining their validity and/or refusing to give them effect. To the contrary, the cases examined in this chapter show that Anglo-Canadian courts can and should examine such foreign laws and actions for formal validity within the foreign legal system and be alert to the possibility that public policy, informed by the norms and principles of public international law, may require that no effect be given to certain foreign laws and actions. It may be that the determination of a case involving foreign laws or actions could now very well turn on whether the words “conflict of laws rules” or “act of state” are spoken first in court. The former would lead the judge to the traditional Anglo-Canadian rules examined earlier in this chapter; the latter could lead to the dramatically different principles applied by a number of Law Lords in Pinochet (No. 3). Resolution of this confusion will likely require that lawyers

The Myth of “Act of State” in Anglo-Canadian Law 371 and judges retire the phrase “act of state” for good and instead refer precisely to the principle they seek to apply in any particular case. In any event, lawyers and judges will have to work vigilently to ensure that the promise of the Pinochet decisions for an advancement in human rights is not crippled by the mistaken view of state actions contained within those same decisions.

14

Grounding a Cause of Action for Torture in Transnational Law SANDRA RAPONI 1

1 INTRODUCTION E S P I T E T H E D E V E L O P M E N T of international principles, standards, procedures, and bodies that address fundamental human rights violations such as torture, international law lacks effective enforcement mechanisms that provide protection or remedies to individuals.2 Due to the fact that international law is predominantly state-based, individuals cannot seek redress under international treaties or from international bodies unless their state accepts the authority of these instruments and bodies. Even the prohibition of torture under international criminal law has hitherto been dependent upon the ability and willingness of other states to undertake prosecution in their domestic courts or on the ability of the United Nations to establish international criminal tribunals for particular situations. Although the creation of the International Criminal Court (the ICC) will mean that torturers may be prosecuted before an international adjudicative body, the circumstances in which such individuals will be prosecuted are significantly limited.3

D

1

Ph.D. candidate in philosophy at the University of Toronto. International treaties and documents that prohibit torture include article 5 of the Universal Declaration of Human Rights, adopted 10 Dec. 1948, G.A. Res. 217A (III), U.N. GAOR, 3rd Sess. Resolutions, part 1, at 71, U.N. Doc. A/810 (1948), reprinted in (1949) 43 Am J Int’l L 127 (Supp.) [hereinafter “UDHR”]; and article 7 of the International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976) [hereinafter “ICCPR”]. Torture is prohibited in regional human rights treaties as well. For example, in article 3 of the European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 221, Europ. T.S. No. 5, entered into force 3 Sept. 1953, reprinted in (1944) 33 ILM 943, 960 [hereinafter “ECHR”]. The significance of the prohibition of torture in international law is evidenced by the fact that an international convention has been created specifically to address torture: Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 [hereinafter “CAT”]. 3 The Court can exercise a general jurisdiction in cases that are initiated by the Security Council. However, where an investigation is initiated by a State Party or by the Prosecutor, the Court cannot exercise jurisdiction unless either the state in which the act was committed or the state of the nationality of the accused is a party to the Statute or gives its consent. This principle excludes the state of 2

374 S Raponi By bringing a civil action based on international human rights law in the domestic courts of other states, individuals may be able to obtain redress for human rights violations when international mechanisms fail to provide effective protection or remedies. Not only will domestic courts be able to fill the gap between the existence of international principles and the ineffective enforcement of these principles, but they will also be able to effectively complement the work of international treaty and adjudicative bodies in enforcing international law. In addition, such actions will enable individual claimants and domestic courts to contribute to the development of the principles and norms of international human rights law. Although a traditional “conflict of laws” framework may be able to provide a civil remedy based on the domestic private law of either the foreign or the forum state, my argument is that it is conceptually more appropriate to formulate and conceptualise an action that is based on international human rights law within a framework of “transnational law.”4 Phillip Jessup defined “transnational law” as “all law which regulates actions or events that transcend national frontiers”, including both public and private international law and other rules which do not wholly fit into these standard categories.5 In the United States, foreign citizens have been able to bring a civil action against foreign defendants for acts committed in a foreign state if the act violates international law.6 Harold Hongju Koh describes the essence of this kind of litigation, which he characterises as “transnational public law litigation”, as “the coupling of a substantive notion—individual and state responsibility—with a familiar process— adjudication—and a normative goal—the promotion of universal norms of international conduct.”7 A transnational human rights action lies at the interface between private and public international law because the form of the action is that of a private law tort, but its substance is primarily derived from the rules and principles of public international law. Such derivation involves a mix of direct application and the nationality of the victim and the state which has custody of the accused. Articles 12 and 13 of the Rome Statute of the International Criminal Court, adopted 17 July 1998, U.N. Diplomatic Conference of Plenipotentiaries on the Establishment of an International Criminal Court, U.N. Doc A/CONF.183/9 (1998), reprinted in 37 ILM 999 (1998) [hereinafter “ICC Statute”]. 4 My conception of a transnational human rights tort action is based on Harold Hongju Koh’s conception of “transnational public law litigation.” “Transnational Public Law Litigation”, (1991) 100 Yale LJ 2372. 5 P C Jessup, Transnational Law (New Haven, Yale University Press, 1956) at 4–5. 6 In Filártiga v. Peña-Irala, 630 F.2d 876 (CA, 2n Cir. 1980) at 890 [hereinafter “Filártiga”], Paraguayan plaintiffs brought an action in a US federal court, suing a Paraguayan official for torture committed in Paraguay. Such actions are made possible in the United States by the Alien Tort Claims Act 28 U.S.C §1350 [hereinafter “ATCA”]: “The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.” This statutory provision was originally enacted in the late 18th century. As the wording indicates, while the plaintiff must be an “alien”, there is no parallel requirement that the defendant need also be. Nor must the alleged tortious act have occurred outside the United States, it seems. 7 Supra n. 4, at 2372.

Grounding a Cause of Action for Torture 375 extrapolation, for instance by analogy. Second, the action blends international law with domestic law by fusing international legal rights with domestic adjudication and domestic judicial remedies. Third, the action also merges international law with domestic law in a substantive manner because the legal rights claimed are not based solely on international law; they are also based on the substantive domestic laws of the states involved in the action.8 Conceptualising such actions as a hybrid between international and national law, where the interpretation and application of international human rights law is influenced by the national laws and the cultural context of all the states involved, will strengthen the legitimacy of allowing the domestic courts of one state to judge another state’s human rights violations. The argument that transnational human rights litigation is conceptually more appropriate and effective in promoting the rights and values at stake in the case of torture than the traditional conflict of laws analysis is developed in the first section of the chapter. The second section examines some of the challenges that may prevent a Canadian court from adjudicating a transnational human rights action. While the chapter focuses specifically on the Canadian judicial context, much of the analysis will be relevant, some of it directly, to other Commonwealth jurisdictions. The first set of challenges concerns the court’s conception of its own judicial competence and its concern for maintaining the separation of powers. In order to overcome these challenges, the judiciary must recognise itself as having the authority and capability of applying conventional and customary international law in order to adjudicate a transnational human rights action that draws upon international, domestic and foreign law.9 A second significant challenge concerns the reluctance of some, if not many judges, to regard a private cause of action as arising from international human rights instruments or from international customary norms.10 The second challenge constitutes a serious obstacle for Canadian courts for two reasons. First, only customary international law is available as a legal source for deriving a cause of action directly from international law. Whereas customary law is generally considered to be directly incorporated into the law of Canada, an international treaty will only 8 Koh states that “[a]ctors invoke claims of right based not solely on domestic or international law, but rather, on a body of transnational law that blends the two.” Ibid. at 2348. Koh’s conception of transnational litigation suggests that the domestic law of the forum state does not merely provide the procedural rules in which international human rights law is applied. Rather, domestic law also contributes to the substantive law that is applied. In addition to blending the substantive principles and rules derived from international law and the domestic laws of the forum state, the court may also include substantive principles and rules from other involved states (i.e. the domestic laws of the state in which the act was committed, the state of nationality of the defendant, the state of nationality of the victim, the current state of residence of the victim, and so forth). 9 Whereas rules or principles of international conventional law only apply to a state if it is a party to the treaty, rules or principles of international customary law apply to all states, irrespective of whether they are parties to any relevant treaties. Timely and persistent objection to certain customary norms may be sufficient to fend off the application of those norms to the protesting state, at least for a period that allows the state to bring its laws and practice in line with a new norm. 10 In the United States, these challenges were presented most strongly by Judge Bork in Hanoch Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, (D.C. Cir. 1984) [hereinafter “Tel-Oren”].

376 S Raponi be given direct domestic effect if it is implemented through domestic legislation.11 Second, existing Canadian jurisprudence on the relationship between private and public domestic law seems to reinforce the second obstacle. Not only have Canadian courts rejected the claim that a private cause of action may arise from statutory law, but they have also held that the existence of a statutory regime may in fact preclude the development of a private action that promotes the same norms.12 Although I will present arguments that seek to overcome these challenges, convincing a Canadian court to recognise such actions as “transnational” may be difficult given the novel and radical character of such a transnational human rights action. A gradual process will be needed to encourage courts to go beyond the existing conceptual barriers between international and national law, and between public and private law. Despite the challenges raised above, Canadian courts have increasingly given indirect effect to international human rights norms through an interpretative presumption that domestic law be construed in conformity with these international norms. These cases provide a starting point. A tort action brought within the traditional conflict of laws framework may be able to succeed if a court construes the rules determining jurisdiction and the choice of law question in conformity with international human rights norms. However, as long as the court stays within the traditional conflict of laws model by applying existing Canadian tort law as the rules of decision, the action will be limited in terms of its legitimacy and its effectiveness in promoting the interests and values at stake. At the very least, even if Canadian courts are not ready to accept an action grounded in “transnational law” as such, they should be encouraged to draw upon international human rights norms as much as possible in interpreting and developing the substantive law that is to be applied.

11 For example see Capital Cities Communications Inc v. Canadian Radio-Television Commission [1978] SCR 141 at 172–3 for the latter proposition. This principle was most recently affirmed by the Supreme Court of Canada in Baker v. Canada (Minister of Citizenship and Immigration), [1999] SCJ No.39 at para. 69 [hereinafter “Baker”]. The Court also affirmed in Baker that international treaties, at least international human rights treaties of the sort at issue in the case, have an indirect application in Canada by virtue of the principle of statutory interpretation that the legislature is presumed to have legislated in conformity with Canada’s international law obligations. The principle applies to treaties that are unincorporated no less than to those that are incorporated. As it did not arise for decision, the Court did not address whether the presumption of compliance with international law extends to all law, including the common law, such that unincorporated treaties might be used as persuasive sources of law for judges having to decide whether or not to recognise a new cause of action. For a discussion of this relationship between domestic and international law see: A Bayefsky, “International Human Rights Law in Canadian Courts”, in B Conforti and F Franconi (eds), Enforcing International Human Rights in Domestic Courts (The Hague, Martinus Nijhoff Publishers, 1997) at 297 [hereinafter “Canadian Courts”]; and R St J Macdonald, “The Relationship Between International Law and Domestic Law in Canada” in R St J Macdonald, G L Morris and D J Johnston (eds), Canadian Perspectives on International Law and Organization (Toronto, University of Toronto Press, 1974) at 88. 12 Board of Governors of Seneca College of Applied Arts and Technology v. Bhadauria (1981), 124 DLR (3d) 193 (SCC) [hereinafter “Bhadauria”].

Grounding a Cause of Action for Torture 377

2 THE ADVANTAGES OF THE TRANSNATIONAL HUMAN RIGHTS FRAMEWORK OVER THE TRADITIONAL CONFLICT OF LAWS APPROACH

It may be possible for victims to bring an action in a Canadian court against state officials responsible for acts of torture through a traditional conflict of laws approach. In the choice of law analysis, the Canadian court must decide whether to apply Canadian tort law or the domestic law of the foreign state. Applying the general rule in Tolofson v. Jensen—that the lex loci delicti (the place where the tort was committed) should be the governing law—to a case where the torture occurred in the foreign state would result in the application of the foreign law.13 However, Justice La Forest stated in Tolofson that the court retained the discretion to apply Canadian law when “a rigid rule on the international level could give rise to injustice.”14 Thus, if the court finds that the law of the foreign state limits the action in a way that gives rise to injustice, Tolofson suggests that the court may use its discretion to apply Canadian tort law. The choice of law determination thus far seems capable of vindicating the rights of the victims to obtain legal redress and compensation. However, it presents problems of legitimacy that arise to a lesser extent in the transnational approach. La Forest J. justified the lex loci delicti rule in Tolofson by drawing on the principle of comity, “the informing principle of private international law, which has been stated to be the deference and respect due by other states to the actions of a state legitimately taken within its territory.”15 He stated: “. . . [I]t seems to me self evident, for example, that State A has no business in defining the legal rights and liabilities of citizens of State B in respect of acts in their own country . . . and it would lead to unfair and unjust results if it did.”16

La Forest J.’s justification of the lex loci deliciti is based on the idea that courts should respect state sovereignty. In most cases, it does seem unfair and illegitimate for foreign courts to interfere with a state’s determination of the legal rights and liabilities of its citizens for acts committed within its territory. A question to be asked is: when would this not be illegitimate or unfair? As noted above, La Forest J. indicated that it would be legitimate to apply the law of the forum state if applying the law of the foreign state would lead to injustice. A problem with this exception is that the court may be open to the criticism that it is imposing its own sense of justice upon another legal system. La Forest J. suggests another basis on which a court may decide not to apply the foreign state’s law, namely in situations in which the foreign state is in breach of an overriding norm of international law: 13

Tolofson v. Jensen; Lucas v. Gagnon [1994] 3 SCR 1022 [hereinafter “Tolofson”]. Ibid. at 1054. 15 La Forest J. cites this definition of comity from Morguard Investments Ltd. v. De Savoye, [1990] 3 SCR 1077 at 1095 [hereinafter “Morguard”]. Tolofson, supra n. 13 at 1048. 16 Tolofson, ibid. at 1052. This statement considers territorial jurisdiction over the act as the primary determinative factor for choice of law. 14

378 S Raponi “On the international plane, the relevant underlying reality is the territorial limits of law under the international legal order. The underlying postulate of public international law is that generally each state has jurisdiction to make and apply law within its territorial limit. Absent a breach of some overriding norm, other states as a matter of ‘comity’ will ordinarily respect such actions and are hesitant to interfere with what another state chooses to do within those limits.”17

Hence, La Forest J. leaves open the possibility that a Canadian court may decide against applying the laws of the foreign state if such laws violate international human rights norms that override the principle of comity. The issue is which international human rights norms may override the principle of comity? The reference to some “overriding norm” seems intended to draw a specific link to the idea of peremptory or jus cogens norms—norms accepted and recognised by the international community of states as norms from which no derogation is permitted.18 According to some sources, the prohibition of torture has come to be recognised as a jus cogens norm.19 La Forest J.’s reference to an “overriding norm” provides a more legitimate and principled basis on which a court may decide not to apply the lex loci delicti rule because the determination is not based solely on the court’s sense of justice, nor is it simply results-orientated. At a structural level, the very foundation of international human rights law challenges the principle of territoriality that underlies the principle of comity. In matters involving human rights, a state’s treatment of its citizens and a state’s actions in its territory is no longer purely a domestic matter; it may also concern the international community as a whole.20 Human rights norms that are considered at least to be obligations erga omnes— obligations owed to the international community as a whole—directly challenge the principle of comity because these obligations are owed by all states, to all states. As such, all states have an interest, and perhaps even an obligation, in 17

Tolofson, supra n. 13 at 1047 (emphasis added). The definition of a peremptory norm is provided in article 53 of the Vienna Convention on the Law of Treaties, (1969) 1155 UNTS 331 Vienna Convention on the Law of Treaties, adopted 23 May 1969, U.N. Doc. A/CONF.39/27 (1969), 1155 U.N.T.S. 331 (entered into force 27 Jan. 1980), reprinted in 8 ILM 679 (1969). According to Ian Brownlie, the least controversial examples of jus cogens are the prohibition of aggressive war, the law of genocide, the principle of racial nondiscrimination, crimes against humanity and the rules prohibiting slavery and piracy. I Brownlie, Principles of International Law (Oxford, Clarendon Press, 1979) at 513. 19 The Human Rights Committee, the treaty body which oversees compliance with the ICCPR, referred to torture as an example of a peremptory norm, stating that states may not reserve the right to engage in torture. General Comment No.24 (52), U.N. Doc. CCPR/C/21/Rev.1/Add.6 (1994). See also infra n. 22 and the sources cited infra n. 47. 20 In his critique of the arguments used to reject the transnational human rights claim in TelOren, Anthony D’Amato argues: “If human rights means anything in international law, it means that traditional state-based jurisdictional exclusivities must give way to a more fundamental realization that the rights of the people count for more than the rights of states . . . The new law of human rights, in short, calls for a change in world view. The interest that a country has in its nationals is expanded, under the law of human rights, to include an interest in non-nationals, especially where basic human rights are threatened.” A D’Amato, “Universal In Rem Jurisdiction for Human Rights Cases?” in International Law Studies: Collected Papers Volume Two (The Hague, Kluwer Law International, 1997) at 206 [hereinafter “Universal”]. 18

Grounding a Cause of Action for Torture 379 enforcing and promoting obligations erga omnes.21 In such cases, State A not only has a legitimate interest in enforcing the legal rights and liabilities of citizens of State B in respect of acts in their own country; it may even have an obligation to do so if basic human rights would otherwise be violated. A right or norm may fit within La Forest J.’s notion of an “overriding norm” if it is sufficiently basic as to be erga omnes (owed to all states by all states), even if it is does not rise to the level of a jus cogens norm. It is generally accepted that the prohibition of torture is such a basic human right that it is the concern of all states.22 Even if a court applies international human rights law in its choice of law analysis to decide against applying the foreign law, the court will still have a few options regarding which body of law it should apply. First, the court may solely apply Canadian domestic law. Second, the court may apply international human rights standards directly as the rules of decision—or, perhaps more accurately, as the basis for such rules of decision.23 The second option presents less of an offence to comity because the court applies the international law that governs both states instead of imposing its own localised law onto the other state. As Daniel Bodansky states, “[s]o long as a court, when reviewing a human rights 21 In Concerning Barcelona Traction, Light & Power Co, Ltd. (Belgium v. Spain), [1970] ICJ Rep.3 at 32 [hereinafter Barcelona Traction], the Court stated that obligations erga omnes “are the concern of all States. In view of the importance of the rights involved, all States can be held to have a legal interest in their protection.” At that time, the Court stated that they included “the outlawing of acts of aggression, and of genocide, as also from the principles and rules concerning the basic rights of the human person, including slavery and racial discrimination”. Note that while jus cogens norms will also be erga omnes, there may be obligations that run to the world and which thus qualify as erga omnes obligations which do not also rise to the level of having the constitution-like overriding or peremptory quality of a jus cogens super-norm. For example, in the Barcelona Traction formulation, the “basic right” not to be enslaved is almost certainly jus cogens, while the right to be free from racial discrimination may “only” be erga omnes (at least where that discrimination is not part of an apartheid-like context). 22 The former U.N. Special Rapporteur on Torture, Peter Kooijmans, has stated in his reports to the Commission on Human Rights that the prohibition of torture is clearly recognised by the international community as both an obligation erga omnes and a jus cogens norm. In the 1993 report he stated:

“The Special Rapporteur’s mandate is a reflection of the fact that the international community has come to the conclusion that the prohibition of torture is an obligation for all States, whether or not they have ratified a treaty which explicitly contains this prohibition. This conclusion is based, inter alia, upon the view of the International Court of Justice which in 1970 stated that the obligation to respect the basic human rights, to which the right not to be tortured belongs beyond any doubt, is an obligation erga omnes for each and every State, an obligation which a State has vis-à-vis the community of States as a whole.” Kooijmans added that “[i]t is now generally accepted that the prohibition of torture is a peremptory norm which under no circumstances may be encroached upon.” P Kooijmans, Question of the Human Rights of All Persons Subjected to any Form of Detention or Imprisonment, in Particular: Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, Report of the Special Rapporteur, Mr. P. Kooijmans, pursuant to Commission on Human Rights Resolution 1992/32, U.N. Doc. E/CN.4/1993/26, para. 10. See also his 1987 report, infra n. 96. 23 For example, American courts have applied international law as the rules of decision for actions brought under the ATCA. That being said, international human rights law will not come ready-made with rules on civil liability. It may need to be applied mutatis mutandi (making necessary modifications to reflect the civil liability, as opposed to state responsibility, context).

380 S Raponi violation in another country, is able to apply international law standards as the rules of decision, then it need not impose its own substantive standards on the other state . . .”24 Under the American model of transnational litigation, courts have generally determined the substantive principles to be applied by looking to international law. American courts have applied domestic law to procedural issues, such as standing and statutes of limitation, and where necessary, they have used American domestic law to “guide” their interpretation of international law.25 It seems inevitable that when a court interprets and applies international principles and rules to determine civil liability for a human rights violation, elements of its own domestic law will affect the court’s interpretation and application of the international rules and principles. Although the direct application of international human rights standards seems to be more legitimate than the sole application of Canadian law, applying international human rights law as it has been developed by international organs may still constitute an imposition of “Western” legal norms that do not respect the legal and cultural norms of states that have neither participated in the formulation of international human rights law, nor recognised the authority of the various international human rights bodies and treaties.26 Even from the international human rights perspective that challenges state sovereignty, comity should be still be valued because it accords respect to the laws, traditions and cultural norms of other nations. A transnational human rights action that defines general international standards by taking into account the context, legal norms, and cultural norms of the other state would be more in accordance with the principle of comity. Such a version of the transnational human rights model can claim greater legitimacy because, rather than imposing its own domestic laws to determine legal rights for acts committed in a foreign state by foreign citizens, courts will apply international human rights norms that govern both the forum and the foreign state, and they will interpret and apply these norms in light of the laws and the context of both the forum and the foreign state.27 24 Daniel Bodansky, “Human Rights and Universal Jurisdiction”, in Mark Gibney (ed.), World Justice? U.S. Courts and International Human Rights (Boulder, Colo., Westview Press 1991) at 3. 25 That is, to assist them in applying international human rights law mutatis mutandi. See the discussion on how international law has been applied by American courts under the ATCA in M Swan, “International Human Rights Tort Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines”, chapter 3 in this volume. 26 Further analysis of this issue is beyond the scope of this chapter. See some of the discussion in B Wells and M Burnett, “Cultural Challenges: Injunctions in Australian Courts and the Right to Demand the Death Penalty under Saudi Arabian Law”, chapter 23 in this volume, and W Adams, “In Search of a Defence of the Transnational Human Rights Paradigm: May Jus Cogens Norms be Invoked to Create Implied Exceptions in Domestic State Immunity Statutes?”, chapter 9 in this volume. 27 The consideration of the foreign state’s domestic laws and social, cultural and political context becomes particularly important to the legitimacy of these actions when there is universal consensus on a general right, but there are different interpretations on what the content of that right should be. See A An-Na’im, “Toward a Cross-Cultural Approach to Defining International Standards of Human Rights: The Meaning of Cruel, Inhuman, or Degrading Treatment or Punishment”, in A An-Na’im, (ed.), Human Rights in Cross-Cultural Perspectives (Philadelphia, University of Pennsylvania Press, 1992).

Grounding a Cause of Action for Torture 381 A second advantage of the transnational human rights model is that it avoids problems that would be created if a court were to decide that Canadian law is the governing law. There are applicable principles and concepts in international law, such as the principles of indirect state responsibility in general international law and individual command responsibility in international humanitarian law, which are not adequately covered by Canadian negligence or intentional tort rules.28 In transnational human rights litigation, courts can directly draw upon the existing body of rules and norms of international law that determine both state and individual responsibility. Furthermore, by selecting only the Canadian rules of negligence and intentional tort, the traditional choice of law approach is limited in its ability to vindicate the values at stake. A finding of negligence or battery for acts that would otherwise constitute torture under international law may not adequately characterise the gravity of the harm caused. The characterisation of the act as a violation of the international norms against torture is not only important from a retrospective, corrective justice perspective.29 It also limits the prospective value of the judgment. Koh argues that plaintiffs who sue the government or government officials in domestic courts for international wrongs seek not only compensation but also norm-enunciation through judicial declarations that the state has transgressed universally recognised norms of international law.30 This element of norm-enunciation also contributes to the development of principles of accountability in transnational law, both civil liability and state responsibility, and to the interpretation of the content of international human rights norms.31 Third, transnational human rights litigation unifies the choice of law question with the determination of jurisdiction. Applying a standard conflict of laws 28 For example, in Velásquez Rodriguez v. Honduras, (1988) 9 HRLJ 212, the Inter-American Court of Human Rights found that a state is responsible for human rights violations if it fails to comply with its duty to take reasonable steps to prevent human rights violations and to use the means available to carry out a serious investigation, identify those responsible and impose the appropriate punishment. Although the principle of indirect responsibility was applied against a state in Rodriguez, the principle may still have relevance as a background norm for developing the law on the liability of state officials. On using international criminal law principles of individual accountability by analogy, see V Oosterveld and A Flah, “Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability”, chapter 16 in this volume. 29 Corrective justice values proportionality between the crime and the punishment, or the tort and the remedy. Since certain criminal offences or types of torts carry more stigma or societal condemnation than other offences or torts, a corrective justice perspective might hold that the characterisation of the crime or tort should be proportional to the nature of the act. 30 Koh states: “Although their announced aim has been retrospective redress, the underlying focus of their actions has been prospective: to provoke a judicial declaration that the conduct of United States or foreign government officials violates a norm of international law. To the extent that plaintiffs have won such judicial declarations, they have sought to use them primarily as political constraints upon the defendants’ future conduct.” Koh, supra n. 4, at 2368–9. 31 Kenneth Randall, in advancing his thesis that the domestic adjudication of Filártiga-like cases is justified by a partial paradigm shift from one that is based on decentralised sovereign states (the Westphalian paradigm) to a human rights paradigm, argues that, in turn, such domestic efforts also represent an institutional method of advancing the human rights paradigm. K Randall, Federal Courts and the International Human Rights Paradigm (Durham, N.C., Duke University Press, 1990) at 206.

382 S Raponi analysis would mean that, independently of a court’s decision to apply Canadian law or the law of the other state, the court must determine whether it may take jurisdiction according to the “real and substantial connection” test.32 In a standard conflict of laws case, states are considered to have an interest in the matter based on territoriality or the nationality of the victim or defendant. With the erga omnes components of international humanitarian and human rights law, all nations have an interest based on the nature of the act; the offender is considered to be the “hostis humani generis, an enemy of all mankind”.33 Bodansky argues persuasively that when a domestic court attempts to enforce a norm of international law prohibiting certain conduct, it should necessarily have universal jurisdiction; universal jurisdiction should not be viewed as a separate question.34 Hence, a transnational human rights action would not require the same degree of connection with the act or the parties in order to exercise jurisdiction or apply international human rights law.35

3 CHALLENGES TO TRANSNATIONAL HUMAN RIGHTS TORTS

There are significant objections to transnational human rights litigation in part due to the challenge that these actions pose to the traditional sovereignty model of international law. Some of the more significant obstacles to transnational human rights litigation were presented in Hanoch Tel-Oren v. Libyan Arab Republic.36 In particular, Judge Bork’s sole opinion challenged the existence of a private cause of action for torture, contrary to the Filártiga holding. First, Judge Bork justified the court’s position of judicial restraint in terms of both judicial competence and the separation of powers principle. Second, he argued that a private cause of action could not be inferred from international law unless it was expressly granted by an international treaty or by customary international law. He rejected the action because he concluded that international customary and conventional law do not in fact create a private cause of action for human rights violations.37

32

Hunt v. T. & N. [1993] 4 SCR 289 and Morguard, supra n. 15 at 1096. Filártiga, supra n. 6, at 890. 34 Bodansky, supra n. 24, at 14. 35 In Filártiga, the court alluded to the principle of universal jurisdiction when it declared that “the torturer has become—like the pirate and slave trader before him—hostis humani generis, an enemy of all mankind.” Supra, n. 6. 36 Supra n. 10. The plaintiffs brought an action against the PLO for the terrorist attack on a tourist bus in Israel. 37 Judge Bork’s judgment has received strong criticism. See A D’Amato, “Judge Bork’s Concept of the Law of Nations is Seriously Mistaken”, (1985) 79 Am J Int’l L 92; and J Quigley, “Judge Bork is Wrong: The Covenant is the Law”, (1993) 71 Wash U L Q 1087. 33

Grounding a Cause of Action for Torture 383 The Competence and Authority of Domestic Courts in Adjudicating Transnational Law In appealing to the separation of powers principle, Judge Bork stated that the act of state doctrine and the political question doctrine would have barred the action even if the court had found a cause of action.38 He concluded that there was a potential for illegitimate interference with the political branches’ functions on the basis that foreign affairs has traditionally been within the exclusive powers of the political branches, and that the determination of the content of international principles involves a “politically-sensitive” controversy.39 The concern with judicial competence and separation of powers may present a greater challenge for Canadian courts because, unlike US federal courts who derive their authority from the American Alien Tort Claims Act (ATCA) and the Torture Victims Protection Act (TVPA), there is no specific legislative authorisation for Canadian courts to apply international human rights law as the basis for a private cause of action.40 In addition, the reason that Canadian courts have refused to give direct effect to international treaties without their incorporation through domestic legislation is due to a concern about maintaining the separation of powers between the executive and legislative branches of government. For example, in a recent Supreme Court of Canada case involving the issue of whether unimplemented international treaties should be used as an aid to interpreting domestic law, Iacobucci J. (disagreeing with the majority judgment on this issue) stated: “In my view, one should proceed with caution in deciding matters of this nature, lest we adversely affect the balance maintained by our Parliamentary tradition, or inadvertently grant the executive the power to bind citizens without the necessity of involving the legislative branch.”41

Koh argues that a court following customary international law without legislation like the ATCA or the TVPA would nonetheless not raise separation of powers concerns.42 However, in order for courts to do this, they would first have to recognise themselves as having the authority and competency to apply 38

Tel-Oren, supra n. 10, at 803. Ibid. at 801–2, 808. 40 Torture Victims Protection Act, 28 U.S.C. §1350 [hereinafter “TVPA”]. 41 Due to Iacobucci J.’s concern with maintaining the balance of power, he rejected the majority’s decision that unimplemented international human rights treaties can be used as an aid in interpreting domestic law. He stated: “[T]he result will be that the appellant is able to achieve indirectly what cannot be achieved directly, namely, to give force and effect within the domestic legal system to international obligations undertaken by the executive alone that have yet to be subject to the democratic will of Parliament.” Baker, supra n. 11, at para. 80. Since Baker did not deal with the interpretive use of international human rights law to create common law causes of action, Iacobucci J.’s worries may prove more persuasive to his colleagues on the Court outside the context of statutory interpretation. 42 Koh, supra n. 4, at 2386. 39

384 S Raponi customary international law and to perform the law-creating function of giving interpretative content in the process of application.43 A particularly difficult challenge is determining the content of international customary law. According to Jessup, the difficulty in creating and applying transnational law is a problem that “lies in the minds of men”—something which can be improved by creating an understanding amongst students, scholars, lawyers and judges.44 He argues that the application of international law by a domestic court is not as indeterminate as it appears because such determinations are constrained by the authorities which specify transnational law: domestic statutes, treaties, UN General Assembly resolutions, as well as domestic courts themselves.45 Although interpreting transnational law from such sources involves a creative process, it shares common elements with developing the common law or interpreting statutes in domestic law. Jessup argues: “Judges in national courts do not have the same freedom in determining what law to apply [as some international tribunals], but their process of making new case law is not fundamentally very different from the international judge’s resort to the general principles of law, even though they label their source as the ‘wisdom of the founding fathers’ or the ‘spirit of the constitution’ or ‘the rule of reason.’ ”46

Transnational adjudication also shares some of the same creative processes used in a standard conflict of laws case since in the latter situation, the court must choose between different bodies of law and it must be able to interpret and apply a foreign body of law. Hence, judges are not being asked to do something that is beyond their capabilities. Furthermore, within the field of international human rights law, judges have a number of sources that aid in defining, interpreting and applying the rules and principles. In the case of torture, there seems to be a sufficient amount of support among international scholars, international human rights bodies, and other domestic legal systems for the idea that there presently exists an international customary norm prohibiting torture.47

43

Koh, supra n. 4, at 2386. Jessup, supra n. 5, at 108–9. D’Amato similarly expresses concern for the state of ignorance regarding international law that is reflected from the bench and by the curricula of American law schools. “Universal”, supra n. 20, at 217. 45 Jessup, ibid. at 113. 46 Ibid. at 96. 47 The Human Rights Committee included torture in its proposed list of customary international human rights norms in its General Comment No.24 (52), supra n. 19. Many international scholars consider the prohibition against torture to be customary law. D’Amato argues that multilateral conventions containing prohibitions against torture constitute evidence of customary law binding upon all states and not just the parties thereto. International Law: Process and Prospects (United States, Transnational Publishers, Inc., 1995) at 180. With respect to other national legal systems, American courts have stated that the prohibition of torture constitutes an international customary norm. See Filártiga, supra n. 6. In addition, torture falls within the list of customary international human rights law provided in the Third Restatement of the Foreign Relations Law of the United States (St. Paul, Minn., American Law Institute Publishers, 1987) at §702. 44

Grounding a Cause of Action for Torture 385 Koh refers to cases that demonstrate that American courts, long before the modern period, routinely applied international law as the law of the state.48 He argues that these cases indicate a conception of international law that included public and private international law, and that governed the relations between individuals as well as between states. For example, Blackstone described the law of nations as: “. . . a system of rules, deducible by natural reason, and established by universal consent among the civilised inhabitants of the world . . . to insure the observance of justice and good faith, in that intercourse which must frequently occur between two or more independent states, and the individuals belonging to each.”49

As revealed by Blackstone’s statement, judges were regarded as capable of interpreting the law of nations because of the natural law idea that the principles for establishing justice between states, and the individuals belonging to each state, could be systematically deduced through “natural reason”.50 Blackstone also suggests a more empirical basis on which to determine the rules to be applied— the consent of nations. Some modern cases have affirmed that courts have a duty to apply international law whenever it arises in a case. Justice White expressed this view in his dissent in Banco National de Cuba v. Sabbatino: “Principles of international law have been applied in our courts . . . because they represent a consensus among civilised nations on the proper ordering of relations between nations and the citizens thereof. Fundamental fairness to litigants as well as the interest in stability of relationships and preservation of reasonable expectations call for their application whenever international law is controlling in a case or controversy.”51

This judgment retains Blackstone’s idea that the principles of international law are based on a consensus among nations without requiring—as Blackstone seems to do—that these principles be deducible by natural reason.52 The more 48 In Hilton v. Guyot, 159 U.S. 113 (1895) at 163, Justice Gray stated that “[i]nternational law, in its widest and most comprehensive sense . . . is part of our law, and must be ascertained and administered by the courts of justice, as often as such questions are presented in litigation.” 49 W. Blackstone, Commentaries on the Laws of England (Chicago, University of Chicago Press, 1979) vol.4 at 66; cited by Koh, supra n. 4, at 2351. 50 In a manner similar to Blackstone’s idea of natural reason, Alan Brudner argues that human rights principles which are self-evidently contained in the notion of the common good, such as the equal dignity of persons and the prohibition of genocide, are part of the law of nations solely by virtue of their being a priori principles of practical reason. Based on this idea, he argues that such principles bind all states and should automatically be part of municipal law. A Brudner, “The Domestic Enforcement of International Covenants on Human Rights: A Theoretical Framework”, (1985) UTLJ 219 at 251 [hereinafter “Domestic Enforcement”]. Brudner’s natural law approach claims to avoid the separation-of-powers problem because a rule of customary law is considered to be an expression of common human reason, rather than the expression of the executive will that requires confirmation by the general will through the legislative assembly. Ibid. at 222, 230. 51 (1964), 376 U.S. 398 at 453. 52 Unfortunately, the judgment makes the same problematic reference as did Blackstone to “civilized nations”. Marttii Koskenniemi argues that determining rules of international law based on natural reason “is not open to a modern lawyer or court, much less an international court, established for the settlement of disputes between varying cultures, varying traditions, and varying conceptions of reason and justice. Such conceptions seem to be historically and contextually

386 S Raponi recent cases of Filártiga and Kadic v. Karadzic have stated that courts must interpret international law as it exists among the nations of the world today by consulting the works of jurists, writings on public law, the general practice of nations and judicial decisions recognising and enforcing that law.53 These modern cases demonstrate that judges may still see themselves as capable of applying international law in ways that necessarily involve the creativity of interpretation without relying on the contentious claim that the rules of international law are deducible by natural reason. Furthermore, Koh reveals that the history of the “law of nations” was not limited to the law governing relations between states (the “law of states”), but also included the law governing individual responsibility (for example, the law of piracy) and transnational commercial transactions (lex mercatoria).54 Lex mercatoria is an example of delocalised transnational law because it was developed from the mercantile customs recognised by trading nations.55 Koh states that the explosion of transnational commercial litigation in the 1970s helped to engender the new generation of transnational public law cases because it has forced the re-evaluation of comity, separation of powers, and judicial incompetence— the three obstacles used to support judicial restraint in Tel-Oren: “The persistent question arose, ‘if contracts, why not torture?’ If American courts could subject the commercial conduct of foreign sovereigns to legal scrutiny without offending comity, why should comity immunise that same sovereign from judicial examination of its egregious public conduct? . . . If a court could hold a foreign sovereign defendant in violation of a commercial contract without usurping the executive function, why couldn’t it hold the same defendant in violation of a human rights treaty, or a clearly defined jus cogens norm against torture? . . . If a court could evaluate the ‘reasonableness’ of the extraterritorial assertion of American regulatory laws . . . why couldn’t the same court construe far less ambiguous terms in a human rights treaty?”56

The ability of transnational commercial litigation to overcome the obstacles presented in Tel-Oren indicates that there is nothing inherently problematic in having “transnational law” developed, interpreted and applied by domestic courts.57 conditioned, so that imposing them on a non-consenting state seems both political and unjustifiable as such.” “The Pull of the Mainstream”, (1990) 88 Mich LR 1946 at 1947. 53 Kadic v. Karadzic, 70 F.3d 232 at 238 (2d Cir. 1995); Filártiga, supra n. 6 at 880–1. 54 Koh, supra n. 4 at 2356. Piracy is a classic example of how the law of nations was applied to individuals. Pirates are outside the municipal laws of various states and are subject directly to duties imposed by international law. D’Amato, “Universal”, supra n. 20, at 214. 55 Lon Fuller refers to lex mercatoria as an example that demonstrates that judges can apply nonstate customary norms directly, without first requiring that these norms be converted into statemade law: “Where the custom of merchants was taken by judges as the standard of decision, the judges spoke of this custom as having the force of law and they applied it as something binding on them as well as on their parties.” Anatomy of the Law (New York, The New American Press Library, 1968) at 73. 56 Koh, supra n. 4. at 2365–6. 57 For further discussion of the “if contract, why not tort?” argument, see R Wai, “The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legalism”, chapter 8 in this volume.

Grounding a Cause of Action for Torture 387 An important development in Canadian law is the increasing use of international human rights treaties as an aid for interpreting domestic law, including the Charter of Rights and Freedoms (the Charter).58 In addition, although courts will not give direct effect to treaties that are not incorporated into Canadian law through domestic legislation, courts have given some indirect effect to international treaties through a common law presumption that legislation is meant to be consistent with Canada’s international obligations. Even before the 1999 Baker judgment, the Supreme Court of Canada had recognised for some time “a rule of construction that Parliament is not presumed to legislate in breach of a treaty or in a manner inconsistent with the comity of nations and the established rules of international law”.59 The use of this presumption is also supported by statements made by government representatives in the international fora.60 Instead of regarding international instruments merely as optional interpretive aids, this rule of construction seems to oblige courts to consider international customary law and international treaties to which Canada is a party in their interpretation of domestic law. In some cases, the use of the presumption has been limited to cases where the court first finds an ambiguity in the legislation; if there is an ambiguity in the legislation, then the interpretation that is most consistent with international law is to be preferred.61 Anne Bayefsky observes that whereas the pre-Charter cases applying the presumption required an ambiguity before the court could consider international treaties, in the post-Charter era, the Supreme Court has frequently referred to international treaties without an effort to first look for an ambiguity.62 The leading majority decision by the Supreme Court of Canada supporting the consideration of international human rights law in the interpretation and the application of the Charter is Slaight Communications Inc. v. Davidson.63 Citing his dissent in Re Public Service Employee Relations Act (Alta.), Chief Justice Dickson stated that “the Charter should generally be presumed to provide protection at least as great as that afforded by similar provisions in international human rights documents which Canada has ratified.”64 This statement implies that international documents not only aid the court in interpreting the 58 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c.11 [hereinafter “Charter”]. See W A Schabas, International Human Rights Law and the Canadian Charter (Carswell Thomson Professional Publishing, 1996) and A F Bayefsky, International Human Rights Law: Use in Canadian Charter of Rights and Freedoms Litigation (Toronto, Butterworths, 1992). 59 Daniels v. The Queen, [1968] SCR 517 at 541. Baker, supra n. 11 and infra text at n. 66. 60 During Canada’s presentation of its Second Report to the Human Rights Committee, Canadian representative Martin Low told the Committee that although the ICCPR was not formally incorporated into Canadian law, it could be used as an aid to interpreting ambiguous provisions of domestic law, as well as the Charter. U.N. Doc. CCPR/C/SR.559, section 23. 61 Schavernoch v. Foreign Claim Commission et al. [1982] 1 SCR 1092; Capital Cities Communications v. C.R.T.C. [1978] 2 SCR 141. 62 Bayefsky, “Canadian Courts”, supra n. 11, 317–18. 63 [1989] 1 SCR 1038 [hereinafter “Slaight Communications”]. 64 Re Public Service Employee Relations Act (Alta), [1987] 1 SCR 313 at 349–50; cited in Slaight Communications, ibid. at 1056.

388 S Raponi Charter when the application of a provision is ambiguous, but more importantly, these instruments provide a minimum standard of human rights protection which the Court must meet in its interpretation and application of the Charter. The statement implies that courts have an obligation to consider international human rights instruments in Charter cases to ensure that their interpretation and application of the Charter is consistent with these instruments. Recently, the Supreme Court of Canada considered the application of international human rights norms to a judicial review of an immigration officer’s deportation decision under Canada’s Immigration Act. The majority in Baker stated: “. . . [The Convention on the Rights of the Child] has not been implemented by Parliament. Its provisions therefore have no direct application within Canadian law. Nevertheless, the values reflected in international human rights law may help inform the contextual approach to statutory interpretation and judicial review.”65

Justice L’Heureux-Dubé (writing for the majority) justified this use of international law in terms of the interpretative presumption: since the legislature is presumed to respect the values and principles contained in customary and conventional international law, interpretations that reflect these values and principles are preferred.66 Although the majority decision affirms the legitimacy of using international human rights instruments to inform statutory interpretation and judicial review, L’Heureux-Dubé J.’s formulation does not go so far as to establish that the courts are obligated to consider these instruments.67 Despite the lack of a clear duty upon courts to consider Canada’s international obliga65

Baker, supra n. 11, at para. 70. L’Heureux-Dubé J. also justified the use of international law by referring to the important role of international human rights law as an aid in interpreting domestic law in other common law countries, including India and New Zealand. She referred finally to the important critical influence that international human rights law has had on the Supreme Court of Canada’s interpretation of Charter rights. Ibid. 67 That being said, L’Heureux-Dubé J. does source her reference to international human rights law in the principle of statutory interpretation, a principle judges have a duty to apply. In his analysis of the increasing use of international human rights instruments by English courts, Murray Hunt argues that the time has come to recognise that domestic courts have an obligation to construe domestic law in conformity with a state’s obligations under international human rights treaties. M Hunt, Using Human Rights Law in English Courts (Oxford, Hart Publishing, 1997) at 297–8. It is possible that the majority’s use of “may” is descriptive rather than normative. Depending on the statute and interpretation being put forward, international human rights law may (or may not) be relevant. Where it is relevant, then arguably, the judges must have regard to the relevant values and principles. This would seem the only way to reconcile the use of “may” with the endorsement of an interpretive presumption that is best understood as not leaving courts the option of whether or not to apply it. Further, it may be that shared “values and principles” becomes the litmus test. Some international treaties entered into by the executive may not accord sufficiently with existing values and principles to justify placing a duty on judges to give indirect effect through interpretation. For an argument that Baker’s conception of “values and principles” means international human rights treaties are distinct from other treaties in how they relate to the interpretive presumption, see C Scott, “Canada’s International Human Rights Obligations and Disadvantaged Members of Society: Finally Into the Spotlight?”, (1999) 10 Constitutional Forum 97 at 101. 66

Grounding a Cause of Action for Torture 389 tions in the interpretation and application of domestic law, the increasing use of international human rights norms by Canadian courts indicates that courts do have the capacity and the authority to interpret and apply international law. Not only have Canadian courts considered the interpretation of international human rights norms developed by international scholars and treaty-bodies, they have also developed their own interpretation and application of these rights within the Canadian context.

Private Cause of Action Under International Human Rights Law: The Relationship Between Private and Public Law A second obstacle presented by Judge Bork’s judgment in Tel-Oren is his argument that a private cause of action cannot be inferred from international human rights law unless it is expressly granted by an international treaty or customary international law.68 Judge Bork stated that individuals do not have a private cause of action under the International Covenant on Civil and Political Rights (ICCPR) because it does not explicitly create a private cause of action. By this he meant that there was no self-executing provision which would, by US constitutional law, be given direct effect by US courts. However, he did seem to accept that states have an obligation to create civil liability in order to comply with the treaty’s requirements with respect to giving domestic effect to its provisions.69 Second, Judge Bork argued that even if the ICCPR suggests that a private cause of action should be recognised in a state’s municipal courts without the need for further legislative transformation, it can only be against that state itself, and not against foreign states or foreign citizens.70 These statements raise issues relating to the relationship between public and private law, and the relationship between domestic law and international law. Judge Bork’s argument that a private cause of action cannot be created or inferred from international human rights treaties presents a particular problem for Canadian courts. As discussed above, Canadian courts will not give direct effect to international treaties; only international customary law is directly incorporated into Canadian domestic law. Despite this principle, Canadian courts have, as already discussed, given indirect effect to international human rights treaties by using these treaties as an aid to interpreting legislation and the Charter. In the area of private common law, international human rights norms 68

Tel-Oren, supra n. 10, at 808–10. Judge Bork stated: “[T]he covenant does not itself say individuals can sue; rather, it leaves to states the fulfilment of an obligation to create private rights of action.” Ibid. at 819 (emphasis added). 70 Ibid. at 819. D’Amato argues that according to Bork’s reasoning, no rule of international law would be available to an individual suing in tort and, as a result, the ATCA would be made meaningless. D’Amato argues that a litigant has a cause of action under the ATCA when he refers to recognised legal rights that he claims have been invaded by the defendant. “Universal”, supra n. 20, at 213. See also Quigley supra n. 37. 69

390 S Raponi have been used to determine public policy.71 This relationship between domestic and international law in Canada suggests two possibilities. First, is it possible to derive a private cause of action from customary international law? Second, to the extent that international human rights conventions can be used to inform the development and the application of the common law, can this process be used to recognise a new cause of action within the Canadian common law that draws upon international human rights principles and rules? Rather than tackle these two issues in their full complexity, I will primarily deal with one obstacle faced by both options: the obstacle created by Canadian jurisprudence on the relationship between public and private law within Canada. According to one strand of this case law, represented by the Saskatchewan Wheat Pool case, a violation of a penal or regulatory statute will not give rise to an independent right of action for damages if the statute is silent on the matter of civil remedies.72 In Board of Governors of Seneca College v. Bhadauria, a case in which the plaintiff argued that a tort of discrimination should be recognised, this meant that civil liability could not arise directly from a breach of the Ontario Human Rights Code, a statute which deals primarily with private sector discrimination on grounds such as race and gender.73 The Ontario Court of Appeal recognised the tort, but did so in a way that avoided finding that a civil cause of action arose directly upon breach of the Code. Instead, Wilson J.A. recognised a new civil cause of action, a common law duty not to discriminate, by reference to the policies and values reflected in the statute.74 If the approach of Wilson J.A. were to be applied to the case of torture, it would mean that a court would not have to determine whether civil liability arises directly from an international treaty or an international customary norm. Rather, a court would be able refer to the norms and principles prohibiting torture in various treaties and in international customary law to inform a new cause of action. In this 71 In Re Drummond Wren [1945] 4 DLR 674 at 679 (Ont CA), Justice Mackay stated that courts may look to various public law statutes, as well as international treaties, as an aid in determining principles relevant to public policy. Although he concluded that a discriminatory restrictive covenant was void because it was offensive to the public policy of Ontario, he stated that the his conclusion was “reinforced” by the wide official acceptance of international policies and declarations against the type of discrimination perpetuated by the covenant. By contrast, the Ontario Court of Appeal in Re Noble and Wolf, [1949] 4 DLR 375 at 399, concluded that the obligations set forth in the United Nations Charter and other international covenants cannot be used to expand the principles of public policy in Canada because these instruments had not been made part of Canadian law. More recently, in Canada Trust Co. and Ontario Human Rights Commission (1990), 69 D.L.R. (4th) 321 at 351–2 (Ont CA), Tarnopolsky J.A. considered the guarantee of equality in international instruments as one source among others (including provincial and federal statutes, and the Charter) in determining public policy. He held that the discriminatory provisions in a charitable trust were void on the grounds of public policy. 72 The Queen in the Right of Canada v. Saskatchewan Wheat Pool, [1983] SCR 205. See A Brudner, “Tort—Civil Liability for Breach of Statutory Duty Abolished”, (1984) 62 Can Bar Rev 668 at 668, 673. 73 Bhadauria, supra n. 12. 74 Bhadauria v. Board of Governors of Seneca College of Applied Arts and Technology (1979), 105 DLR (3d) 707 at 715 (Ont CA).

Grounding a Cause of Action for Torture 391 manner, the court could recognise a common law duty not to violate customary international human rights norms, including the prohibition of torture. However, the Supreme Court of Canada overturned Wilson J.A.’s judgment and held that, when the legislature creates a comprehensive statutory regime to deal with a certain body of law, these public bodies and procedures “cover the field” in such a way as to exclude and preclude private or civil actions in that area. Thus, a common law tort of discrimination was foreclosed in Bhadauria. Two years later, in the above-noted Saskatchewan Wheat Pool case, the Court added a further barrier to human rights torts by holding that statutory torts cannot be implied. Not long afterward, a third Supreme Court case, Dolphin Delivery, raised the bar even higher by holding that the Charter does not create a constitutional private law of human rights.75 In Dolphin Delivery, McIntyre J. held that the Charter does not apply to private litigation or to the common law in the absence of some governmental action, primarily based on the view that the Charter was intended to regulate the relationship between the government and the individual. He held that the Charter would only apply to private litigation where one of the parties invokes or relies upon some governmental juridical act, such as a statute or regulation, which produces an infringement of Charter rights. Underlying the court’s determination seems to be a desire to limit the scope of the Charter to government conduct and to the remedies provided by the Charter. One reason the Court gave for such a limitation of the Charter’s scope of application was the prudential concern that constitutionally-derived torts would undermine or outflank the Bhadauria and Saskatchewan Wheat Pool doctrines. For example, McIntyre J. cites the following passage from an article by Anne McLellan and Bruce P. Elman: “[T]he better view is that the Charter applies only to government action. To hold otherwise would be to increase the scope of the Charter immeasurably. In cases involving arrests, detentions, searches and the like, to apply the Charter to purely private action would be tantamount to setting up an alternative tort system. In the area of private discrimination, an entirely new system of civil liability in competition with the dispute resolution mechanisms fostered by human rights legislation would result.”76

These arguments suggest how deeply runs the rationale which we might refer to as the “covering the field” rationale. In Bhadauria, the “covering the field” argument was clearly used to exclude civil actions based on human rights legislation. Laskin C.J.C. held that because the Ontario Human Rights Code creates a comprehensive system to adjudicate and enforce breaches of the Code, it forecloses any civil action arising directly 75 Retail, Wholesale and Department Store Union, Local 580 v. Dolphin Delivery Ltd. [1986] 2 SCR 573 [hereinafter “Dolphin Delivery”]. 76 Dolphin Delivery, supra n. 75 at 597. A McLellan and B P Elman, “To Whom Does the Charter Apply? Some Recent Cases on Section 32”, (1986) 24 Alta L Rev 361 at 367.

392 S Raponi upon a breach thereof. He also concluded that the Code excludes any common law action based on an invocation of the public policy expressed therein because the Code itself sets out procedures for the vindication of that public policy.77 Whereas Wilson J.A.’s judgment at the Court of Appeal suggests the idea that civil actions complement the Code’s procedures by promoting the goals of the legislation, Laskin C.J.C.’s judgment seems to be based on the belief that civil litigation would undermine the legislative intent that discrimination cases be addressed exclusively by an administrative body. It can thus be seen that, based on Dolphin Delivery and Bhadauria, a Canadian court may similarly find that human rights treaties have established a comprehensive regime of bodies, procedures and remedies for addressing human rights violations that exclude resort to other means of settling disputes or providing remedies. Furthermore, there is also some support for this view in international law.78 In the Tehran Hostages case, the International Court of Justice (ICJ) asserted that the rules of diplomatic law constitute “a selfcontained regime which [is] entirely efficacious”.79 More problematically, in the Nicaragua case, the ICJ stated that “where human rights are protected by international conventions, that protection takes the form of such arrangements for monitoring or ensuring respect for human rights as are provided for in the conventions themselves [and in any event] the use of force could not be the appropriate method to monitor or ensure such respect.”80 77

Bhadauria, supra n. 12, at 194–5, 203. Theodor Meron states that the principal proponent of the view that the remedies established under human rights treaties exclude other means of settlement is Jochen Abr. Frowein. T Meron, Human Rights and Humanitarian Norms as Customary Law (New York, Oxford University Press, 1989) at 229. Meron cites J A Frowein, “The Interrelationship between the Helsinki Final Act, the International Covenants on Human Rights, and the European Convention on Human Rights” in T Buergenthal (ed.) Human Rights, International Law and the Helsinki Accord (New Jersey, Allanheld, Osmun & Co. Publishers, 1977) at 71, 79. It is important to note that, in contrast to the Bhadauria judgment, which found that the remedies established under the Human Rights Code exclude other means of settlement because of the comprehensiveness of the Code’s adjudicative and enforcement system, Frowein’s argument is not based on a claim that the system of remedies established under human rights treaties such as the ICCPR is comprehensive. On the contrary, Frowein recognises that “the system of remedies found in the covenants is obviously inadequate” due to the fact that state interests are at stake in these matters and, secondly, that the ICCPR does not answer the question of how exclusive its system of supervision is intended to be. Frowein argues: “What seems more dangerous, however, than the mere absence of efficient remedies is the possible consequence of these provisions on the right of states inter se to take up the matter of the implementation of the International Covenants in the context of the states’ other bilateral or multilateral relations.” Frowein, ibid. at 78–9. Frowein’s concern seems to be that a state will regard itself as directly authorised by the ICCPR to take its own formal action against another state that has breached the covenant’s provisions, rather than using the covenant’s ineffective interstate complaint procedures. It is unclear whether Frowein’s argument would also reject the development of domestic judicial mechanisms that allow individuals to bring actions against foreign states based on the customary international norms or international principles that are reflected in the ICCPR, rather than directly under the Covenant’s provisions. 79 United States Diplomatic and Consular Staff in Tehran (United States v. Iran), 1980 ICJ Rep. at 3, 40 [hereinafter the Iranian Hostages case], cited by Meron, Human Rights, ibid. at 229. 80 Military and Paramilitary Activities In and Against Nicaragua (Nicaragua v. US), 1986 ICJ Rep. at 134 [hereinafter the Nicaragua case]. As Meron notes, the ICJ did not assert generally that 78

Grounding a Cause of Action for Torture 393 Thus, in order to establish a cause of action for breach of a human right within the common law by drawing on international human rights norms, a critical analysis that limits the effect of both Bhadauria and the ICJ dicta is needed. Perhaps the most significant weakness in Laskin C.J.C.’s judgment is the unsubstantiated assumption concerning the comprehensiveness of the Ontario Human Rights Commission.81 Laskin C.J.C. seems to have assumed that the human rights system would be more effective in addressing discrimination than the courts had been in the past. However, the Ontario Human Rights Commission has not been able to live up this promise; an overwhelming majority of the complaints are not brought forth by the Commission and there is an excessive backlog problem. Hence, the argument based on effectiveness and comprehensiveness is undermined. Secondly, since the Code itself does not explicitly exclude private actions, Laskin C.J.C. did not justify why individuals could not choose between a human rights complaint or instead a tort action. He did not explain why this would undermine the legislature’s intentions in creating the Commission, or why it would restrict the Commission’s ability to carry out its functions. Nor did he address why a kind of competition from private law may not be a welcome way to force the Code system to be more efficient in its responses to claims. Such competitive pressure could include forcing the government to fund the Code’s institutions adequately in order to make good on Bhadauria’s basic premise of an effective statutory mechanism to address private sector discrimination. Based on this analysis of Bhadauria, determining whether a human rights treaty precludes private litigation involves two questions: whether the protection provided is comprehensive; and second, whether private litigation would undermine, or be incompatible with, the treaty’s goals or the effectiveness of its treaty remedies are exclusive. Ibid. at 230. The ICJ did not assert that where human rights are protected by international conventions, these rights must be exclusively monitored and enforced by the procedures and institutions provided by the conventions themselves. Rather, the concern seems to be with the form of the measures used to protect these rights. The section of the judgment cited here suggests that states may take other measures than those explicitly provided for in the conventions, but these measures must take “the form of such arrangements for monitoring or ensuring respect for human rights as are provided for in the conventions themselves”. The ICJ indicated that the use of force, particularly the use of force at issue in the case, is not an appropriate method to monitor or ensure such respect. However, to the extent that a human rights convention provides a complaint procedure between states, a complaint mechanism for individuals, a body to adjudicate complaints, a provision guaranteeing compensation or redress for victims of human rights abuses and other similar measures, then allowing individuals to sue state officials for human rights violations in order to seek compensation and a formal declaration of a human rights violation would seem to be an acceptable form of ensuring respect for the human rights protected in the conventions. In Human Rights and Humanitarian Norms as Customary Law, Meron does not consider whether transnational human rights litigation would be an appropriate measure, compatible with international conventions; rather, he considers whether uman rights treaties exclude countermeasures that do not involve the use of armed force and he concludes that they do not. 81 N Gupta, “Reconsidering Bhadauria: A Re-examination of the Roles of the Ontario Human Rights Commission and the Courts in the Fight Against Discrimination” (L.L.M. Thesis, University of Toronto, 1993) and I Hunter, “The Stillborn Tort of Discrimination”, (1982) 14 Ottawa L R 219 at 224.

394 S Raponi procedures. Meron similarly argues that, in determining whether a treaty excludes recourse to other remedies than those provided in the treaty, one must consider the comprehensiveness and the effectiveness of the dispute-settlement procedures, as well as the integrity of the treaty.82 Whereas Laskin C.J.C. began with a presumption in favour of interpreting Ontario’s human rights legislation as contemplating only statutory remedies, Meron and Louis Henkin argue against such a presumption in interpreting human rights treaties. They argue that nothing in the character of human rights agreements suggests any intention to eliminate the ordinary legal consequences of international undertakings and the ordinary remedies for their violation.83 As Meron proposes, “[w]hether a particular human rights treaty excludes remedies dehors the treaty depends, of course, not on abstract legal theory but on a good faith interpretation of the terms of the treaty in light of their context and the object and purpose of the treaty.”84 In the Nicaragua case, the problem was not that there existed a comprehensive and self-contained procedural system such that other remedies would interfere with the integrity of the system. Rather, the problem was that the remedy at issue—the use of force—was considered to be incompatible with the form of the measures provided by human rights conventions for monitoring and ensuring the protection of human rights, and contrary to the shared goals of these conventions—the goal of ensuring respect for human rights. In contrast, the judicial remedies provided by transnational human rights litigation are far more compatible with the goals and the form of the measures provided by treaties such as CAT and the ICCPR.85 In assessing whether a human rights treaty precludes a transnational human rights litigation, the first question is whether the treaty establishes a comprehensive and effective procedural system. The procedures and remedies provided by international human rights conventions, such as the ICCPR and CAT, are largely ineffective in providing individuals with redress or a remedy. Although individuals can bring complaints under these two conventions, the procedure is only available if states agree to recognise the Committee’s authority to hear individual complaints. Even then, the Committee’s decision is not binding. The procedures and remedies provided by human rights conventions cannot be considered comprehensive or effective since they largely depend upon state consent and state enforcement.86 Human rights conventions leave it open to states to 82 As an example, he states that the comprehensiveness and effectiveness of the disputesettlement procedures established by the EEC Treaty supports the generally accepted view that the European Union’s legal system constitutes a self-contained regime that excludes the resort to other remedies. Supra n. 78, at 231–2. 83 Meron, supra n. 78 at 230–1, and L Henkin, “Human Rights and ‘Domestic Jurisdiction’ ” in T Buergenthal (ed.), Human Rights, International Law and the Helsinki Accord, supra n. 78 at 31. 84 Meron, ibid. at 231. 85 See supra n. 80. 86 Meron argues: “Because procedures for the settlement of disputes and remedies recognised by human rights treaties are often weak and based on optional acts of acceptance, which are not commonly resorted to, to endorse the exclusivity of treaty remedies would intensify the fragility and ineffectiveness of such treaties’ norms. Supra n. 78, at 230.

Grounding a Cause of Action for Torture 395 develop various possible procedures and remedies to enforce guaranteed rights. In fact, these conventions obligate states to take domestic legislative and judicial measures to promote the rights guaranteed by the treaty and to provide effective remedies for individuals whose rights have been violated. Even if the procedures and mechanisms created under a human rights convention are not comprehensive or effective, we must still ask whether transnational human rights litigation is an appropriate solution to fill the gap. In order to assess whether private litigation complements the procedures provided by a human rights treaty, we must first consider how their functions differ. As set out in the first section of the chapter, a transnational human rights action may serve a variety of purposes. From a corrective justice perspective, individuals are able to obtain redress and compensation. Individuals may also seek to obtain an internationally-recognised declaration that their state or its officials either violated their fundamental human rights, or failed to provide protection or an effective remedy. From a distributive justice perspective, such a declaration may promote more effective human rights protection at home, and internationally. International conventions, such as the ICCPR, largely depend upon state consent. In agreeing to become a party to a convention, a state obligates itself to respect and ensure the rights guaranteed by the convention. As part of the duty to ensure these rights, a state party is also obligated to take legislative, judicial and other measures to provide individuals with effective remedies. The primary procedural mechanism provided under these conventions is the reporting system. States submit periodic reports on the measures they have taken to implement the convention. The reporting mechanism creates a constructive relationship between states and supervisory bodies that aims at gradually developing the national protection of human rights. An advantage of having a treatybody that makes recommendations, as opposed to having an adjudicative body that imposes sanctions or orders remedies, is that other states will be encouraged to become a party to the treaty. Allowing individuals to enforce these rights against their state, in a foreign court no less, may seem incompatible with this feature of the reporting procedure and the method by which it seeks to make remedies more effective in national legal systems. On the other hand, the fact that some treaty regimes such as those of the ICCPR and CAT include procedures for individual communications indicates that individual actions against one’s state does not itself undermine such treaties. In addition, the fact that the ICCPR requires that individual complainants exhaust local remedies indicates that the Human Rights Committee is not regarded as the exclusive or most appropriate forum whereby individuals can obtain remedies for the violation of their rights under the ICCPR. Rather, procedures and remedies provided by national legislation and domestic courts seems to be preferred by international human rights conventions; the Committee is the forum of last resort that is triggered only when national systems have failed in this regard. Consequently, in order to be compatible with the structure of lodging communications under the ICCPR, foreign courts would

396 S Raponi also have be the court of last resort. Hence, if the plaintiff is to obtain effective redress in the jurisdiction where the human rights violation was suffered, the foreign court may have to refuse jurisdiction.87 While domestic courts can serve as the missing link between the promulgation and the realisation of international human rights norms in their own states and while foreign courts can play a supportive role by taking jurisdiction on this last resort basis, the remaining question is whether foreign courts should serve this function by adjudicating a transnational human rights action when the state in which the torture was committed is unwilling or unable to provide adequate redress for victims.88 This question addresses Judge Bork’s second argument, that even if the ICCPR does permit a private cause of action in municipal courts, the action can only be against that state itself and not against foreign states. Farooq Hassan strongly argues against the use of foreign courts. He argues that, through Filártiga-like cases, local legal systems are illegitimately circumvented by handing adjudication over to a foreign court. He warns that using domestic machinery to address a particular international human rights abuse in the absence of an international machinery may do more harm than good.89 Instead he argues: “Efforts therefore should be made . . . to make international human rights law a part of domestic law by appropriate legal developments thereby making the home government answerable before domestic legal tribunals. To be able essentially to prosecute a foreign state in a rare and isolated case . . . for its wrongdoing under customary international law is to open a Pandora’s box which ultimately might do very little for what we are trying to accomplish.”90

Hassan raises significant potential problems with transnational human rights litigation, and hence his concerns should be taken seriously. In part, his arguments emphasise the need to use foreign domestic courts cautiously, particularly since such abuses often arise from a complex social and political context. In cases in which victims are unable to obtain redress from their own states, foreign courts may be used as a last resort—not to circumvent local systems, but to try to make them more responsive to human rights abuses. Hassan also argues that allowing foreign domestic courts to enforce international human rights is contrary to the statist philosophy through which the international protection of human rights has developed. For example, international human rights tribunals can only adjudicate violations based upon prior state consent to the applicability of a claims procedure. However, supporters of 87 See McConville’s discussion of forum non conveniens and how it relates to the exhaustion of local remedies: A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality”, chapter 6 in this volume. 88 Bayefsky, “Canadian Courts”, supra n. 11, at 295–6. 89 F Hassan, “A Conflict of Philosophies: The Filártiga Jurisprudence”, (1983) 32 Int’l & Comp L Q 250 at 257. 90 Ibid. at 257–8.

Grounding a Cause of Action for Torture 397 transnational litigation reject this statist conception of international human rights law. Randall, for example, argues that the emergence of international human rights has resulted in at least a partial shift from a paradigm of sovereign states to a new paradigm of individuals and human rights.91 Koh sees this development as a re-emergence of the inclusion of individuals under international law that existed before the twentieth century. He argues: “Tokyo and Nuremberg pierced the veil of state sovereignty and dispelled the myth that international law is for states only, re-declaring that individuals are subjects not just objects of international law.”92 If the individual is considered to be a subject, rights-holder and beneficiary in international human rights law, then transnational human rights actions may be regarded as complementing international procedures by allowing individuals to make claims and acquire remedies when they would otherwise be unable to do so.93 However, it is arguable that more normative support is needed in order to establish the authority or legitimacy of foreign courts in adjudicating actions brought by individuals against foreign state-officials or other foreign individuals, for acts committed in a foreign state. One possible justification in the case of obligations erga omnes is that all states and their courts have a legitimate interest in adjudicating these violations. In Barcelona Traction, the ICJ stated that obligations erga omnes include “the outlawing of acts of aggression and of genocide . . . the principles and rules concerning the basic rights of the human person, including protection from slavery and racial discrimination.”94 Torture seems to fit within these obligations. The court in Filártiga alluded to the erga omnes doctrine when it asserted that “the torturer has become—like the pirate and slave trader before him—hostis humani generis, an enemy of all mankind.”95 As noted earlier, the former UN Special Rapporteur on Torture, Peter Kooijmans, has suggested in his reports to the Commission on Human Rights that the prohibition of torture constitutes an obligation erga omnes: “[A]ll states have a legal interest in compliance with the prohibition of torture . . . [T]he transgressor . . . is responsible to the international community as a whole and, in principle, other States may bring a claim as representative of that

91

Randall, supra n. 31. Koh, supra n. 4, at 2358–9. 93 Randall states: “Since individuals may not bring any claims before the International Court of Justice, the assumption of domestic jurisdiction over human rights and terrorism cases is especially necessary. While domestic courts are horizontally arranged in relation to the world legal order, they may appropriately enforce norms that are emitted by centralized institutions . . . The absence of a truly global civil tribunal does not foreclose federal court jurisdiction over human rights and terrorism claims; it suggests only that, institutionally, the world has not yet fully provided human rights and antiterrorism enforcement machinery. In the meantime, federal courts should act as ‘double agents’ of both the United States legal system and the world legal order . . . at least when fundamental humanitarian norms are at issue . . . In fact, the modern treaties and general world order trends may even obligate federal courts to represent the international system.” Supra n. 31, at 205. 94 Barcelona Traction, supra n. 21. 95 Filártiga, supra n. 6 at 890. 92

398 S Raponi community.”96 In this sense, allowing a victim access to foreign courts could be seen as closely analogous to the power of foreign states to bring claims against transgressing states. In addition to having such a legitimate interest and associated power to adjudicate actions brought by foreign citizens for acts of torture committed in a foreign state, states may actually have an international obligation to provide redress and compensation in such cases through their courts. Article 14 of the CAT states: “Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation . . .”97 Although it may be argued that this provision was intended to create an obligation for a state to provide redress and compensation for acts of torture committed or condoned by a state against persons within its territorial jurisdiction, the language of the provision does not limit the scope of the article in this manner. Interpreting this article to include any victim of torture who is unable to obtain adequate redress or compensation through his or her own national legal system or through international mechanisms can be justified in the case of torture because it generates obligations erga omnes. A significant objection which may be raised is that the recently-adopted Statute of the ICC will establish a comprehensive and effective system of enforcing violations of international criminal law (including certain acts of torture) and will thus preclude transnational litigation from achieving the same ends once the ICC regime is functional. Although the entry into operation of the ICC will mean that torturers can be prosecuted before an international adjudicative tribunal, the circumstances in which the Court will be able to obtain jurisdiction are significantly limited. If neither the state in which the act was committed nor the state of the nationality of the person accused is a party to the Statute, or gives its consent to the prosecution, then the Court will not be able to obtain jurisdiction on its own; it may only do so with the authorisation of the Security Council.98 In addition, although article 13 of the ICC Statute does enable victims and their representatives to seek redress before the Court by allowing them to provide information to the Prosecutor from which the Prosecutor can initiate an action, the Prosecutor may decide not to pursue the investigation or the case may be blocked by the above noted limitations concerning jurisdiction. In such cases, victims may legitimately wish to be able to turn to private litigation as an alternative. Because article 75 of the ICC Statute enables the court to order reparations to victims, such a private action that seeks compensation instead of punishment may be viewed as not only compatible with the ICC’s functions and goals, but also in active harmony with them. Another relevant feature of the 96 P Kooijmans, Question of the Human Rights of All Persons Subjected to any Form of Detention or Imprisonment, in Particular: Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, Report of the Special Rapporteur, Mr. P. Kooijmans, pursuant to Commission on Human Rights Resolution 1992/32, U.N. Doc. E/CN.4/1987/13 at 14–15; cited by Meron, supra n. 78, at 200. See also supra n. 22. 97 CAT, supra n. 2. 98 Articles 12 and 13 of the ICC Statute, supra n. 3.

Grounding a Cause of Action for Torture 399 ICC is that the Court can only investigate or prosecute a case if the state is unwilling or unable to do so itself. Hence, prosecution through national systems is regarded as the primary mechanism to enforce international criminal law. This element of the Statute indicates that actions brought against torturers at the national level would not be usurping the jurisdiction or purpose of the Court. On the contrary, this provision makes the ICC the forum of last resort. Hassan raises the important question of what transnational human rights litigation hopes to achieve and whether or not it is able to do accomplish this. He argues that instead of bringing Filártiga-like actions in foreign courts, efforts should be made to develop human rights law at home.99 Koh seems to advocate transnational public litigation as a way in which individuals can develop human rights protection in their home societies. He argues that transnational public law litigation serves the prospective aim of obtaining a judgment that articulates norms of transnational law, announces that a transnational norm has been violated and promotes a political settlement.100 However, with respect to the goals such as these that transnational litigation hopes to achieve, the “covering the field” argument still presents legitimate concerns that need to be addressed by the transnational model on a case-by-case basis. When a court adjudicates a civil action for a violation of a human rights treaty or a customary international human rights norm, the court should consider whether the action might, in the prevailing state of affairs, be incompatible with existing mechanisms that also address those same violations. For example, allowing private litigation for violations which are being addressed through truth and reconciliation commissions may be incompatible, and may even frustrate the achievements and goals of such commissions.101 A benefit of conceptualising the action within a “transnational law” paradigm is that it supports, and may even require, the consideration of other foreign and international mechanisms, procedures and standards for addressing particular human rights violations.

4 CONCLUSION

In his critique of the Tel-Oren case, D’Amato concludes: “Perhaps our judiciary is not ready for a case such as this one, which may be somewhat ahead of its time. Common law proceeds incrementally, and perhaps a lot of backing and filling was necessary in the human rights field before a claim such as the one in Tel-Oren would get a proper hearing.”102

Although transnational human rights litigation provides the potential to promote the recognition and protection of international human rights within a 99 100 101 102

Hassan, supra n. 89, at 257–8. Koh, supra n. 4, at 2349. In such cases, the court may draw upon international norms governing amnesty. “Universal”, supra n. 20, at 217.

400 S Raponi broader context, the observation of D’Amato and the obstacles discussed above suggest a major counter-impulse. A Canadian court may well refuse to adjudicate a claim explicitly structured on the “transnational model”, at least without legislative support. Litigants may accordingly prefer to take the path of least judicial resistance by staying within the traditional conflict of laws analysis, using international human rights norms in relatively non-radical ways. Two examples may be given. International human rights law could provide normative content to the public policy doctrine when that doctrine is invoked as justification for avoiding a choice of law result. Also, the universal legal interest in the observance of erga omnes norms could be used to inform analysis as to why a court should retain jurisdiction or even take jurisdiction in the first place. Despite the use of international human rights norms in these two components of the analysis, if the substantive law applied is wholly or primarily localised Canadian tort law, then the scope of a favourable judgment will be much more limited in its ability to signal that the defendant has violated international law, to affect the political process, and to contribute to the development of international human rights law. As stated in the Introduction, in order for Canadian courts to recognise and adjudicate transnational human rights actions, they must gradually go beyond the existing conceptual barriers between domestic and international law, and between public and private law. One way to move towards this would be for Canadian courts to apply international human rights norms on a regular and consistent basis in interpreting and developing Canadian domestic law (tort and conflict of laws, combined) in accordance with Canada’s international obligations.

15

International Human Rights Law and the Tort of Torture: What Possibility for Canada? EDWARD M. HYLAND 1

“. . . [H]uman beings as such are under the direct protection of international law.” —F.V. Garcia-Amador2 “On the one side there were the timorous souls who were fearful of allowing a new cause of action. On the other side there were the bold spirits who were ready to allow it if justice so required.” —Lord Denning3

1 INTRODUCTION H E P U R P O S E O F this chapter is to examine whether one and one equals two. Can the views of Professor Garcia-Amador and Lord Denning be combined to arrive at the conclusion that, within the common law of Commonwealth countries, private legal remedies are—or should be—available to individuals who have been tortured? Is there a basis for recognising an international tort of torture? In order to give shape to the analysis, the specific focus will be the received doctrine and judicial traditions of Canada. More specifically, does international human rights law, as received in some fashion by Canadian law, provide victims with a right of action against their torturers in Canadian courts? In what follows, I shall first examine the trajectory of the development of international law’s prohibition of torture. Then, in Section 3, I shall look at the debate in the United States about whether international law itself can be

T

1 Of the Ontario Bar, M.A. (Toronto), S.T.B. (Centre Sèvres, Paris), S.T.M. (Regis College, Toronto), LL.B. (Toronto), in practice with the firm Iler Campbell, in Toronto. 2 F V Garcia-Amador, “State Responsibility: International Responsibility”, (1956) 2 YBILC 173 at 203. 3 Candler v. Crane Christmas & Co. [1951] 1 All E.R. 426 at 432 (C.A.); Cf. Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd. [1963] 2 All E.R. 575 (H.L), in which the House of Lords embraced Lord Denning’s view in Candler, and recognised an action in negligence for professional or business misrepresentation resulting in economic loss.

402 E M Hyland understood to provide a right of action against torturers. Section 4 of the chapter will explore the intersection between international law and Canadian law, in light of how Canadian law has treated the impact of public law on private law. In section 5, I will examine developments in, and conceptions of, both public and private international law in order to assess the implications for Canadian courts of recognising an international tort of torture.

2 THE INTERNATIONAL HUMAN RIGHTS PROHIBITION ON TORTURE : STATUS AND MEANS OF ENFORCEMENT

Torture represents a singular paradox at the heart of the international human rights system. Universally condemned and outlawed, it is also a widespread practice among states throughout the world. Against the backdrop of the two world wars, in 1948 the members of the United Nations included the prohibition of torture in the Universal Declaration of Human Rights.4 Article 5 reads: “No one shall be subjected to torture or to cruel, inhuman or degrading punishment.” Less than twenty years later, the international community again expressed a prohibition of torture in the International Covenant on Civil and Political Rights (ICCPR). In language reminiscent of the Universal Declaration, the ICCPR states in its article 7: “No one shall be subjected to torture or to cruel, inhuman or degrading treatment or punishment. In particular, no one shall be subjected without his free consent to medical or scientific treatment.”5 In a General Comment, the UN Human Rights Committee concluded that article 7 “relates not only to acts that cause physical pain but also to acts that cause mental suffering to the victim.”6 Moreover, “. . . [the] aim of article 7 . . . is to protect both the dignity and the physical and mental integrity of the individual. It is the duty of the State party to afford everyone protection through legislative and other measures as may be necessary against the acts prohibited by article 7, whether inflicted by people acting in their official capacity, outside their official capacity or in a private capacity.”7 4 Universal Declaration of Human Rights, adopted 10 December 1948, GA Res. 217A (III), UN GAOR, 3d Sess. Resolutions, part 10, at 71, UN Doc. A/810 (1948), reprinted in (1949) 43 Am J Int’l L 127 (Supp.) article 5 [hereinafter “UDHR”]. For an analysis of the idea of torture in international law see: S Ackerman, “Torture and other Forms of Cruel and Unusual Punishment in International Law”, (1978) 11 Vand J Trans L 653; Y Dinstein, “The Right to Life, Physical Integrity and Liberty” in L Henkin (ed.), The International Bill of Rights—The Covenant on Civil and Political Rights (New York, Columbia University Press, 1981) 121 at 122–6. 5 International Covenant on Civil and Political Rights, adopted 16 December 1966, GA Res. 2200 (XXI), UN GAOR, 21st Sess., Supp. No. 16, UN Doc. A/6316 (1966), 999 UNTS 171 (entered into force 23 March 1976), article 7(1) [hereinafter “ICCPR”]. 6 UN Human Rights Committee, General Comment 20 Article 7 (Forty-fourth Session, 1992) para. 5, reprinted in (1994) 1 IHRR 26. 7 Ibid. at para. 2. The Committee goes on to state: “Article 7 should be read in conjunction with article 2, paragraph 3, of the Covenant . . . The right to lodge complaints against maltreatment prohibited by article 7 must be recognised in the domestic law. Complaints must be investigated

What Possibility for Canada? 403 The importance of article 7 is underscored in article 4 of the ICCPR, which prohibits states from derogating from their obligation not to torture, even in times of public emergency. Among regional human rights instruments, the European Convention on Human Rights and the American Convention on Human Rights also prohibit the practice of torture,8 and they do so absolutely and without exception.9 Despite the adoption and entry into force of these instruments, torture continued to be widespread. As a result, the international community pressed for additional mechanisms to eliminate torture. In 1975, the UN General Assembly adopted the Declaration on the Protection of All Persons from Being Subjected to Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment.10 Among the steps forward, the Declaration defined torture for the first time:11 “For the purpose of this Declaration, torture means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted by or at the instigation of a public official on a person for such purposes as obtaining from him or a third person information or confession, punishing him for an act he has committed or is suspected of having committed, or intimidating him or other persons.”

The Declaration meshed with the line of legal doctrine that viewed the prohibition of torture as absolute and without exception.12 Significantly for the themes of this chapter and volume, it also required that “victim[s] shall be afforded redress and compensation in accordance with national law.”13 Like the Universal Declaration of Human Rights, the torture Declaration was not in itself legally binding on states. Still, it constituted an important step that helped make possible the UN General Assembly’s subsequent drafting and adoption of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (the Convention or Convention against Torture) in 1984.14 promptly and impartially by competent authorities so as to make the remedy effective . . . States may not deprive individuals of the right to an effective remedy, including compensation, and such full rehabilitation as may be possible,” at paras 14 and 15. 8 European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 UNTS 221, Europ. T.S. No. 5 (entered into force 3 September 1953), reprinted in (1994) 34 ILM 943, article 3 [hereinafter “ECHR”]; American Convention on Human Rights, OASTS No. 36, OAS Off. Rec. OEA/Ser. L/V/IL 23, Doc. 21 Rev. 6 (entered into force 18 July 1978), reprinted in (1970) 9 ILM 673, article 5(2) [hereinafter “ACHR”]. See also, African Charter on Human and Peoples’ Rights, concluded at Banjul 27 June 1981 OAU Doc. CAB/LEG/67/3 Rev. 5 (entered into force 21 October 1986), reprinted in (1982) 21 ILM 58, article 5 [hereinafter “ACHPR”]. 9 ECHR, supra n. 8 at article 15(2); ACHR, supra n. 8 at article 27(2). 10 UNGA, Resolution 3452 (XXX), 9 December 1975. 11 Ibid. at article 1. 12 “No State may permit or tolerate torture or other cruel, inhuman or degrading treatment or punishment. Exceptional circumstances such as a state of war or a threat of war, internal political instability or any other public emergency may not be invoked as a justification of torture or other cruel, inhuman or degrading treatment or punishment,” UNGA, Resolution 3452 (XXX) supra n. 10 at article 3. 13 Ibid. at article 11. 14 Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 December 1984, GA Res. 39/46, UN GAOR, 39th Sess., Supp. No. 51, UN

404 E M Hyland The Convention builds upon the definition of torture in the Declaration on Torture.15 The threshold for characterising acts as torture is the infliction of severe pain or suffering. Only acts of a certain gravity are considered torture, in contradistinction to cruel, inhuman or degrading treatment.16 On the other hand, the decision not to include the criterion of systematic infliction of pain, which was suggested during the travaux préparatoires, means that a single act is sufficient to constitute torture.17 For there to be torture, an act must be intentional and inflicted for purposes tied to the state and its organs, in the sense that a finding of torture is made only when public authorities can be held responsible, directly or indirectly by acquiescence, for the act. Article 1(2) of the Convention also stipulates that wider international or domestic definitions of torture are not to be read down by the force of the definition in article 1(1). The significance of this gains in perspective in light of the Convention’s restriction of the definition of torture to acts for which public authorities can, in some measure, be held responsible. The UN Human Rights Committee has interpreted the scope of article 7 of the ICCPR more broadly, it seems, stating that “[t]he aim of the provisions of article 7 . . . is . . . to afford everyone protection . . . against the acts prohibited by article 7, whether inflicted by people acting in their official capacity, outside their official capacity or in a private capacity.”18 That being noted, some interpretive alignment of article 1(1) of the Convention with article 7 of the ICCPR may be possible by reading the word “acquiescence” in article 1(1) broadly. The phrasing of article 1(1)— Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 ILM 1027 [hereinafter “CAT”]. Canada signed the Convention on 23 August 1985, and ratified it on 24 June 1987. It came into force for Canada on 24 July 1987. In addition, the following international and regional instruments prohibit torture and cruel and unusual treatment or punishment: Standard Minimum Rules for the Treatment of Prisoners, approved by ECOSOC, Res. 663C (XXIV) (1957) and amended by ECOSOC Res. 2076 (LXII) (1977), article 31, reprinted in C Gane and M Mackarel (eds), Human Rights and the Administration of Justice (The Hague, Kluwer Law International, 1997) 385; Code of Conduct for Law Enforcement Officials, GA/Res. 34/169 (1979), article 5, reprinted in Gane and Mackarel, 359; Principles of Medical Ethics Relevant to the Role of Health Personnel, Particularly Physicians, in the Protection of Prisoners and Detainees against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment, GA/Res. 37/194 (1982); European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment, concluded at Strasbourg 26 November 1987, (1987) E.T.S. No. 126 (entered into force 1 February 1989), reprinted in (1987) 27 ILM 1152; Inter-American Convention to Prevent and Punish Torture, OEA/Ser.P, AG/doc. 2023/85 rev. 1, 12 March 1986, reprinted in (1986) 25 ILM at 519. 15 “For the purposes of this Convention, the term ‘torture’ means any act to which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity,” CAT, supra n. 14 at article 1(1). 16 See, for example, the judgment of the European Court of Human Rights in the case Ireland v. United Kingdom, ECHR (1978), Series A, No. 25, at paras 167–8. 17 J H Burgers and H Danelius, The United Nations Convention against Torture: A Handbook on the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (Dordrecht, Martinus Nijhoff Publishers, 1988) at 117–18. 18 Supra n. 6, at para. 2.

What Possibility for Canada? 405 “inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity”—conceivably could include a private person who has tortured under cover of the public authorities’ general failure to prevent “private” violence of this sort.19 While there is interstate agreement imposing international obligations to prevent torture, there is also consensus among international lawyers that the prohibition of torture has crystallised into a rule of customary international law.20 As such, the prohibition applies equally to all states, irrespective of whether they are parties to any of the conventions. The American Institute of Law’s Third Restatement of the Foreign Relations Law of the United States includes the prohibition of torture within the scope of customary international human rights law.21 Moreover, there is consensus that the rule against torture is a peremptory or jus cogens norm within the meaning of article 53 of the Vienna Convention on the Law of Treaties.22 The preamble of the Convention against Torture highlights the agreement against torture, situating it in the context of making “. . . more effective the struggle against torture and other cruel, inhuman or degrading treatment or punishment throughout the world.”23 The purpose of the Convention, then, is to reinforce existing prohibitions.24 The Convention’s main value is to sharpen the focus on states’ obligations in the fight against torture and for the protection of its victims. Principal among the obligations under the Convention is the state’s duty to take effective measures to prevent torture from occurring on its 19 The dominant interpretation of article 1(1) would not allow such a broad approach. As Burgers and Danelius point out, “Another question which caused some discussion during the travaux préparatoires was whether or not an act of the kind referred to in article 1 should be regarded as torture irrespective of who committed the act. The conclusion was, however, that only torture for which the authorities could be held responsible should fall within the article’s definition,” supra n. 17, at 119. See also O Schachter, International Law in Theory and Practice (Boston, Martinus Nijhoff Publishers, 1991) at 341: “[S]ome prohibitions based on human rights are expressly directed only at governmental action. An example is the prohibition against torture and other ‘cruel, inhumane or degrading treatment’. Yet even in a case of this kind, the State’s responsibility to prevent private acts of torture may be implied under a duty of due diligence.” See also I Brownlie, System of the Law of Nations: State Responsibility, Part I (Oxford, Clarendon Press, 1983) at 161–2. 20 Schachter, supra n. 19 at 338. 21 American Law Institute, Rest. 3rd, Restatement of the Foreign Relations Law of the United States (St. Paul, American Law Institute Publishers, 1987) Vol. 2, at §702. 22 (1969) 1155 UNTS 331. “For the purposes of the present Convention, a peremptory norm of general international law is a norm accepted and recognized by the international community of States as a whole as a norm from which no derogation is permitted and which can be modified by a subsequent norm of general international law having the same character.” See, for example, UN Human Rights Committee, General Comment No. 24 (52), U.N. Doc. CCPR/C/21/Rev.1/Add. 6 (1994), “Reservations that offend peremptory norms would not be compatible with the object and purpose of the Covenant . . . Accordingly, a State may not reserve the right to engage in slavery, torture. . . .” at para. 8. Neither the United Nations nor any of its legal organs has adopted an authoritative or exhaustive list of principles for determining jus cogens norms, cf. International Law Commission, “Draft Articles on the Law of Treaties with Commentaries”, (1966) II YBILC 187 at 247–9. 23 Supra n. 14. 24 Burgers and Danelius, supra n. 17 at 1.

406 E M Hyland territory.25 An indication of the peremptory character of the prohibition of torture is that states must give effect in their criminal law to the universal jurisdiction to prosecute acts of torture.26 States are required to prosecute alleged torturers found within their jurisdiction, even if these persons are not nationals of the forum state and the torture occurred abroad. While the first twelve articles of the Convention shine the light on states parties, articles 13 and 14 shift the focus and provide victims of torture with two important rights: the right to complain if they have been tortured, and the right to seek fair and adequate compensation. Significantly, the right to adequate and fair compensation imposes an obligation on states to satisfy that right without limitation to where the torture occurred: “Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation . . .”27 Moreover, the right extends to the dependants of a victim of torture who has died as a result of torture.28 It should be clear from the preceding discussion that the international prohibition of torture is more than an expression of principle. It is a fundamental norm, expressing a definable, universal and obligatory rule prohibiting certain acts against individuals. As we saw above, the history of international human rights law has eroded this exclusive view of the role of the state in international law. Individuals are directly under the protection of international law, and do 25 “1. Each State Party shall take effective legislative, administrative, judicial or other measures to prevent acts of torture in any territory under its jurisdiction. 2. No exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political instability or any other public emergency, may be invoked as a justification of torture. 3. An order from a superior officer or a public authority may not be invoked as a justification of torture,” supra n. 14 at article 2. 26 Ibid. at articles 5–7. Universal jurisdiction is a principle of customary international law in which states are allowed jurisdiction over non-nationals for certain crimes occurring outside the territory of the state claiming jurisdiction. While long accepted for the offences of piracy and slave trading, it has come to cover other offences such as hijacking, war crimes and crimes against humanity as well as, perhaps, trafficking in narcotics. Cf. Brownlie, supra n. 19 at 304–5; Universal Jurisdiction over Drug Offences Case [1987] 74 ILR 166 (Fed. Rep. Germany, Fed. Sup. Ct.); W W Bishop, “General Course of Public International Law”, (1965) II Recueil des Cours 151 at 323–4. 27 CAT, supra n. 14 at article 14. There is no limitation in the language of the article restricting the right only to individuals who have experienced torture in the jurisdiction of the state in which they are claiming compensation. The precise scope of this article has not been the subject of consideration, either by the Committee against Torture or judicially in Canada. However, it does impose upon a state an obligation of means—ensuring that its legal system allows victims of torture to obtain redress. While the language leaves obscure the meaning of “legal system”, the characterization of the requirement as an obligation of means (and not result) is the result of interpretation. Thus, article 14 could be read in tandem with the traditional rule of customary international law which obliges a state to provide foreigners with access to its courts, thus creating for states parties an obligation to provide foreign victims of torture with a private right of action before their courts. For a discussion of different types of obligations under treaties, see International Law Commission, “Report of the International Law Commission on its Twenty-ninth Session: State Responsibility”, (1977) II:2 YBILC 5 at 16. For a discussion of the traditional customary international rules regarding the treatment of foreigners, see E Jiménez de Arechaga, “International Responsibility” in M Sørensen, (ed.), Manual of Public International Law (New York, St. Martin’s Press, 1968) 531 at 553–7; Schachter, supra n. 19 at 240–1. 28 CAT, supra n. 14 at article 14(1).

What Possibility for Canada? 407 not (and should not have to) depend on states to vindicate their rights. However, the fact is that the capacity of individuals to seek redress for rights violations is limited still by the principle of state sovereignty; logically this need not be the case. One possibility is for torture victims to be able to sue their torturers in Canadian courts. There is, however, a powerful objection to this idea: international law does not expressly grant victims such a right of action. Nowhere has this question been so intensely debated as in the United States, which has been the forum for lawsuits for torture and other human rights violations that have occurred in other countries.

3 THE UNITED STATES ’ EXPERIENCE

The threshold for this type of civil action was crossed in Filártiga v. Peña-Irala.29 On appeal to the US Court of Appeal, Second Circuit, the Paraguayan plaintiffs sought damages against a former Paraguayan police officer for the torture and death of their son in Paraguay. They sued under the Alien Tort Claims Act.30 The issue was whether the Alien Tort Claims Act granted the court jurisdiction over the matter. The court held that where an alleged torturer is found and served with process within the United States, the Act provided federal jurisdiction. Judge Kaufman also concluded that not only jurisdiction but also a cause of action could be inferred from §1350 of the Act itself, insofar as the plaintiffs were pleading a violation of international law.31 However, he gave no reasons to support his conclusion. At most, Judge Kaufman established an individual right in international law to be free from torture. But from this right, who owes obligations to the victim and what remedy follows? On what basis could a victim of torture go into a US federal court and seek a remedy in tort from another individual? One is left to wonder what subject matter a federal court is in fact granted power to adjudicate. The problem, as Casto has concluded, is that “there is serious doubt . . . whether international law, unassisted by domestic law, creates a tort remedy that may be invoked in domestic courts by private individuals.”32 29

630 F.2d 876 (2d Cir. 1980). 28 U.S.C. §1350 (1982). The section states: “The [federal] district court shall have original jurisdiction over any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.” 31 Filártiga, supra n. 29 at 887: “[W]e believe it is sufficient here to construe the Alien Tort Statute, not as granting new rights to aliens, but simply as opening the federal courts for adjudication of the rights already recognized by international law.” The lack of reasoning here has been the subject of critique: W R Casto, “The Federal Courts’ Protective Jurisdiction over Torts Committed in Violation of the Law of Nations”, (1986) 18 Conn L Rev 467 at 473–8. According to Casto, §1350 does not create a cause of action, but is purely jurisdictional. Therefore, the question as to the basis for the cause of action remains open. In Casto’s view, to claim that the cause of action is “recognized by international law”, is of itself insufficient to found a right of action, at 479. However, D’Amato argues that §1350 does provide a statutory cause of action, in addition to jurisdiction, A D’Amato, International Law Studies: Collected Papers (Boston, Kluwer Law International, 1997) vol. 2 at 205–17. 32 Casto, supra n. 31 at 475. 30

408 E M Hyland Four years later, in the lead opinion in Tel-Oren v. Libyan Arab Republic, Judge Edwards agreed with Judge Kaufman in holding that §1350 requires only a violation of the law of nations to establish a cause of action.33 The main debate in Tel-Oren was between Judge Edwards and Judge Bork, the latter holding that §1350 only granted jurisdiction and that the plaintiffs had to demonstrate that they had a right of action directly under either a treaty of the United States or the law of nations.34 Both judges agreed, however, that customary international law does not create rights to sue; nor do treaties, unless they make explicit provision for rights of action. For Judge Edwards, a right of action was available to the plaintiffs only because of §1350’s statutory grant of such a right. The conflicting approaches in Tel-Oren were taken up anew in the Northern District of California Federal District Court, in Forti v. Suarez-Mason.35 Using the Alien Tort Claims Act, Argentinian nationals sued a former general of Argentina, alleging torture, murder, and prolonged arbitrary detention. In addressing the defendant’s motion to dismiss on the ground, inter alia, that §1350 of the Alien Tort Claims Act only granted jurisdiction and not a right of 33 726 F.2d 774 at 779 (1984). The case involved an action for damages against the Palestine Liberation Organisation, Libya and other Arab groups, arising out of a 1978 attack on a bus in Israel. The District Court of the District of Columbia dismissed the action for lack of jurisdiction (517 F. Supp. 542). The court unanimously dismissed the appeal, with each panel member giving different reasons for judgment. For his part, Judge Edwards also advanced an alternative method for grounding a right of action: §1350 would permit a plaintiff to bring a domestic tort claim provided he or she had met the jurisdictional threshold. This would require establishing that the tort constituted a violation of a treaty of the United States, or of customary international law, at 782–8. 34 Ibid. at 801. The main difference between the two opinions is expressed in the following terms, by Judge Bork: “The lack of clarity in, and absence of consensus about, the legal principles involved by appellants [plaintiffs], together with the political context of the challenged actions and the PLO’s impingement upon American foreign relations, lead to the conclusion that appellants’ case is not the sort that is appropriate for federal court adjudication, at least not without an express grant of a cause of action,” at 808. Judge Bork’s argument is twofold. He reads §1350 as jurisdictional only, and as requiring a separate substantive right of action in order to begin a lawsuit. He rejects the plaintiffs’ broad reading of §1350, insofar as treaties and customary international law stand in the same position in the section. According to Judge Bork, treaties do not provide such a right of action, unless they are self-executing. In other words, individual rights recognized in treaties become part of domestic law, and therefore are enforceable in domestic courts, only through implementing legislation. In the absence of implementing legislation, there is no cause of action available to individuals (at 808–10). To give §1350 a broad reading would, in his view, render meaningless the traditional view—that treaties that provide no cause of action cannot be sued on without statutory authorisation (at 812). Finally, Judge Bork concludes that the very idea of a cause of action— “whether a plaintiff is a member of the class of litigants that may, as a matter of law, appropriately invoke the power of the court” (at 801)—precludes a right of action available to individuals. This is so because the interests protected at international law are properly those of states and not of individuals (at 817). Therefore, unless there is an express grant of a cause of action to individuals, none should be inferred from the simple fact of a violation of a right protected under international law (whether treaty or custom), since the “proper class of litigants” at international law consists of states. As D’Amato has convincingly argued, Judge Bork’s opinion is based on a serious misunderstanding of international law, A D’Amato, “What Does Tel-Oren Tell Lawyers? Judge Bork’s Concept of the Law of Nations is Seriously Mistaken”, (1985) 79 Am J Int’l L 92. (For a contrary view, see A P Rubin, “Professor D’Amato’s Concept of Jurisdiction is Seriously Mistaken”, ibid. 105). 35 672 F.Supp. 1531 (N.D. Cal. 1987).

What Possibility for Canada? 409 action, and that the plaintiffs had to establish a private right of action either under treaty or customary international law, Judge Jensen held: “There appears to be a growing consensus that §1350 provides a cause of action for ‘certain international common law torts’. It is unnecessary that plaintiffs establish the existence of an independent, express right of action, since the law of nations clearly does not create or define civil actions, and to require such an explicit grant under international law would effectively nullify that portion of the statute which confers jurisdiction over tort suits involving the law of nations. Rather, a plaintiff seeking to predicate jurisdiction on the Alien Tort Statute need only plead a ‘tort . . . in violation of the law of nations’.”36

Judge Jensen rejected the view that §1350 is purely jurisdictional and that a plaintiff must establish a right to sue under international law. With Judge Bork (and Judge Edwards), he accepted that international law does not of itself grant private rights of action. Therefore, for §1350 to be effective, it must not only grant jurisdiction, but also recognise a cause of action. A plaintiff would thereby be able to plead violations of internationally-protected rights as torts, either under a treaty to which the US is party or under customary international law. Judge Jensen clarified that an “international tort” is one that must be definable, obligatory, and universally condemned.37 In other words, the international norm must be sufficiently mature in its expression, in order that its violation be judicially determinable. This requires universal consensus prohibiting the act in question, determining the nature of the duties owed and to whom they are owed and establishing criteria that will allow a court to ascertain whether given conduct amounts to the prohibited act and is thus a violation of the norm. Moreover, the prohibition must be non-derogable. Equally clear is that such universal norms do not, in and of themselves, give rise to a right of action. In their amicus brief to the Filártiga court, the US Departments of State and Justice failed to find any cases in which judgments against torturers were based directly on customary international law.38 More recently, Professor Nadine Strossen concluded that relatively few US courts have directly enforced rules of customary human rights law.39 While some lower US federal courts have relied on customary human rights law as a rule of decision, the judgments have not been affirmed on appeal.40 On the contrary, where 36 Forti v. Suarez-Mason, supra n. 35, at 1539. Cf. also: Amerada Hess Shipping Corp. v. Argentine Republic 830 F.2d 421 at 424–5 (2d Cir. 1987), rev’d on other grounds, 488 U.S. 428; Xuncax v. Gramajo 886 F.Supp. 162 at 179 (D.Mass. 1995). 37 Forti, supra n. 35 at 1540. 38 United States, Memorandum for the United States Submitted to the Court of Appeals for the Second Circuit in Filártiga v. Peña-Irala, reprinted in (1980) 19 ILM 585 at 604 FN 45. 39 N Strossen, “Recent U.S. and International Protection of Individual Rights: A Comparative Legal Process Analysis and Proposed Synthesis”, (1990) 41 Hastings L J 805 at 823. 40 Fernandez v. Wilkinson, 505 F.Supp. 787 (D. Kan. 1980) (the applicant, a Cuban illegal, petitioned for habeas corpus on the basis that he was being illegally detained. The court held that the detention was contrary to customary international law prohibiting arbitrary detention), aff’d 654 F.2d 1382 (10th Cir. 1981) (on appeal the decision was affirmed on the basis of statutory interpretation); Garcia-Mir v. Meese, 788 F.2d 1446 (11th Cir. 1986) (application for habeas corpus by

410 E M Hyland appellate courts have resorted to international human rights law as the rule, they have done so under a cause of action granted by statute.41 On the question of what is required to found a cause of action, whether simply the violation of the international norm, or, instead, a specific grant under international law, a consensus seems to have formed around the former position. The various courts that have considered the question have embraced the rationale of Judge Kaufman:42 “In the twentieth century the international community has come to recognise the common danger posed by the flagrant disrespect of basic human rights and particularly the right to be free of torture. . . . In the modern age, humanitarian and practical considerations have combined to lead the nations of the world to recognise that respect for fundamental human rights is in their individual and collective interest. Among the rights universally proclaimed by all nations . . . is the right to be free of physical torture. Indeed, for the purposes of civil liability, the torturer has become—like the pirate and slave trader before him—hostis humani, an enemy of all mankind.”

As inspirational as these words are, in the end US courts have taken refuge in the statutory provision of the Alien Tort Claims Act, both to establish jurisdiction and to recognise a right of action. More difficult is the question of whether, and if so, how, a right of action for an international tort might be established at common law—that is, absent a statutory authorisation. 4 THE INTERSECTION OF CANADIAN AND INTERNATIONAL LAW : WHAT POSSIBILITY IS THERE OF RECOGNISING A TORT OF TORTURE AT COMMON LAW ?

The former Chief Justice of the Supreme Court of Canada, Bora Laskin, expressed the challenge underlying the question contained in this subsection’s heading in his dissent in Harrison v. Carswell:43 “It seems to me that the present case involves a search for an appropriate legal framework for new social facts which show up the inaptness of an old doctrine developed detained Cubans dismissed on the ground that international law prohibiting arbitrary detention has been displaced by the controlling action of the executive and judiciary); Siderman de Blake v. Republic of Argentina, 965 F.2d 699 (9th Cir. 1992) (if violations of jus cogens committed outside the US are to be exceptions to sovereign immunity, then Congress must make them so); Bradvica v. I.N.S., 128 F.3d 1009 (7th Cir. 1997) (customary international law is not applicable in domestic courts where there is a controlling legislative act). 41 As the review above reveals, the grant has been inferred under the Alien Tort Claims Act. On the other hand, the Torture Victim Protection Act, Pub.L. 102–256, March 12, 1992, 106 Stat. 73 s. 2 expressly provides for a right of action: “An individual who, under actual or apparent authority, or color of law, of any foreign nation (1) subjects an individual to torture shall, in a civil action, be liable for damages to that individual; or (2) subjects an individual to extra-judicial killing shall, in a civil action, be liable for damages to the individual’s legal representative, or to any person who may be a claimant in an action for wrongful death.” 42 Filártiga, supra n. 29 at 890. See also, In re Estate of Ferdinand Marcos, Human Rights Litigation, 25 F.3d 1467 (9th Cir. 1994), cert. denied 115 S.Ct. 934; Alvarez-Machain v. U.S., 107 F.3d 696 (9th Cir. 1996). 43 [1976] 2 SCR 200 at 209–10.

What Possibility for Canada? 411 upon a completely different social foundation. The history of trespass indicates that its introduction as a private means of redress was directed to breaches of the peace or to acts likely to provoke such breaches. Its subsequent enlargement beyond these concerns does not mean that it must be taken as incapable of further adaptation but must be applied on what I can only characterise as a level of abstraction which ignores the facts.”

As Laskin C.J. recognised, the search for an “appropriate legal framework for new social facts” entails addressing squarely the relationship between courts and legislatures: “[Do the courts have] a balancing role to play, without yielding place to the Legislature, where an ancient doctrine, in this case trespass, is invoked in a new setting to suppress a lawful activity supported both by legislation and by a well-understood legislative policy[?]”44 His Lordship thought that the courts possess such a balancing and creative role, a view the majority did not share. Whether courts will allow a right of action in tort for torture turns on answers to the following questions: (i) what is the curial attitude toward the development of new torts? (ii) what is the judicial attitude toward the relationship between public human rights law (and statute law, more generally) in Canada and the common law of torts? (iii) to what extent are Canadian courts willing to rely on international human rights law, conventional or customary, as a guide in the development of new common law torts? (iv) how does or should the Canadian Charter of Rights and Freedoms affect the translation of the international right to be free from torture into a common law private right of action?

Curial Attitude Toward the Development of New Torts The availability to individuals of a common law right of action depends on courts recognising legally protected interests. As Professor Prosser has observed, the novelty of a claim or its failure to fit within existing nominate torts is not a bar to founding a claim: “There is no necessity whatever that a tort must have a name. New and nameless torts are being recognised constantly, and the progress of the common law is marked by many cases of first impression, in which the court has struck out boldly to create a new cause of action, where none has existed before . . . [T]he law of torts is anything but static, and the limits of its development are never set. When it becomes clear that the plaintiff’s interests are entitled to legal protection against the conduct of the defendant, the mere fact that the claim is novel will not of itself operate as a bar to the remedy.”45

A classic example in this regard is Donoghue v. Stevenson, in which the House of Lords extended the manufacturer’s duty of care to include the ultimate 44

Ibid. at 202. W L Prosser, Handbook of the Law of Torts, 4th ed. (St. Paul, Minn., West Publishing Co., 1971) at 3–4. 45

412 E M Hyland consumer of products.46 In his speech, Lord MacMillan put the matter succinctly: “The criterion of judgment must adjust and adapt itself to the changing circumstances of life. The categories of negligence are never closed.”47 Courts will typically look to precedent, public policy, the requirement of doing justice in a given case, and the values of certainty and coherence in the development of the law in determining whether to recognise new causes of action. For example, in an Australian decision, a judge refused to strike out a claim alleging a duty in tort upon an insurer to use good faith in processing a worker’s compensation claim, despite the fact that the duty already existed as a term of the insurance contract.48 Also, courts in some Commonwealth countries are currently developing rights to be protected from unjustified invasions of privacy and from harassment.49 Part of the debate concerns defining precisely the nature of the interest protected, the relationship between the duty and the protected interest, what criteria of remoteness are engaged, whether there are any defences, whether injury is required, and what standard of care is required and on what criteria of reasonableness. In describing judicial law-making, Dickson J. characterised the common law method of analysis in a manner that recognised, at least implicitly, the challenge that Laskin C.J. had laid down in Harrison v. Carswell: “Courts undoubtedly legislate in a certain sense. The process of the common law is analytically inductive, seeking for re-application in the future, general premises which are derived from particular facts litigated in the past. In extracting the general from the particular, judges make law for the future as well as apply it in the present. The principles which are developed far transcend the particular facts at hand.”50

However, as Dickson J. also recognised, “where [the] line is to be drawn between judicial activity which works acceptable changes into the law and that which oversteps the bounds ascribed to judicial decision-making, does not admit of easy demarcation.”51 Where the line gets drawn depends on the judiciary’s willingness to remain open to the development of the common law, to embrace (and create, as neces46

[1932] AC 562 (HL). Ibid. at 619. 48 Gibson v. Parkes District Hospital (1991) 26 NSWLR 9 (“[T]he tort may arise where the nature of the relationship brought about by the contract, as distinct from the terms of the contract, is such as to impose a duty to act in good faith. On that basis, the duty is a true tort duty, not a contractual duty and the existence of a contractual term is not a necessary foundation for it”, at 35). But see Banque Keyser Ullmann S.A. v. Skandia (UK) Insurance Co. Ltd., [1991] 2 AC 249 (HL) (House of Lords confirmed the Court of Appeal’s decision that an insurer’s duty of full disclosure to its client did not give rise to a cause of action in tort); Gimson v. Victorian Workcover Authority [1995] 1 VR 209 (Sup Ct of Victoria) (the court rejected the holding in Gibson and held that there is no duty to deal fairly and in good faith in processing a claim for compensation, breach of which would give rise to a cause of action). 49 Bradley v. Wignut Films Ltd. [1993] 1 NZLR 415 (HC) (although the court did not find a cause of action for breach of privacy on the facts, it did recognise it as an emerging tort, at 423); Khorasandjian v. Bush [1993] QB 727 (CA) (recognizing a tort of harassment). 50 B Dickson, “The Role and Function of Judges”, (1980) 14 L Soc Gaz 138 at 181. 51 Ibid. at 180–1. 47

What Possibility for Canada? 413 sary) new legal frameworks able to recognise and vindicate the changing content of personal rights and corollary duties. As societies change, due to changing economic, social, political and cultural conditions, the substance and scope of rights and duties will inevitably evolve to give expression to new underlying values. The challenge for the courts will be (as it has been) to fashion the intellectual tools to identify those interests that deserve legal protection, as well as the conduct that will attract legal sanction for failing to respect those interests. The “line of demarcation”, to borrow from Dickson. J. is most obscured in the relationship between public law principles and values, and private law rules.

Canadian Public Human Rights and Tort Law The backdrop for considering the nexus between human rights law and tort law, which serves as a proxy for the link between public and private law, is the increasing statutory definition of the rules and regulations governing private relationships. Historically, common law principles of property, trust, contract, and tort law governed these relationships. In the past fifty years, however, we have seen the emergence of the administrative state to regulate relationships such as commercial ventures, industrial activity, labour relations, the family, and charitable activities.52 This shifting ground raises the question of the precise distinction between tort law and other statute-based rules, and whether breaches of these rules give rise to a cause of action in tort. A detailed discussion of courts’ treatment of a breach of a statutory duty and its civil consequences is beyond the scope of this chapter. It is sufficient to observe that it was contradictory and confused, based as it was on an attempt to impute to the legislature the intention of conferring civil liability for breach of a statute. The situation led one writer to remark, “[i]n effect the judge can do what he likes, and then select one of the conflicting principles stated by his predecessors in order to justify his decision.”53 In 1983, the Supreme Court of Canada finally resolved the confusion in Canada. It rejected the idea of a nominate tort of statutory breach giving rise to a right of recovery merely upon proof 52 For example, the political purposes doctrine in charitable trust law, as expressed in McGovern et al. v. Attorney General [1981] 3 All ER 493 (Ch D), is now entirely regulated by the Income Tax Act, RSC 1985, c.1 (5th Supp.), ss. 149.1(6.1) and (6.2). In the area of workers’ compensation, in 1913 then Ontario Chief Justice Meredith recommended a system based on no fault. It would be administered by a state agency, with payments for the duration of the disability financed by means of a collective employer liability fund. It would eliminate any rights of action with respect to accidents covered by the fund. That system exists to this day in Ontario and other provinces, R C B Risk, “ ‘This Nuisance of Litigation’: The Origins of Workers’ Compensation in Ontario” in D H Flaherty, (ed.), Essays in the History of Canadian Law, Vol. 2 (Toronto, University of Toronto Press, 1983) 418. The Supreme Court of Canada left workers covered by a collective agreement without any remedies at common law against the exercise of management discretion in unilaterally changing the terms of employment, once the collective agreement has lapsed and a new one has not been negotiated, CAIMAW v. Paccar of Canada [1989] 2 SCR 983. 53 G Williams, “The Effect of Penal Legislation in the Law of Tort”, (1960) 23 Mod L Rev 233 at 246.

414 E M Hyland of breach of a statutory duty and resulting damages.54 Writing for the Court, Dickson J. concluded that a breach of statute, where it has an effect on civil liability, should be considered in the context of the general law of negligence. At most, a breach could be considered evidence of negligence, with the statutory duty serving to limit the required standard of care and define the parameters of reasonableness, for the purpose of determining liability in negligence. The Court’s judgment in Saskatchewan Wheat Pool represented a clear repudiation of what several authors had characterised, and which the Court accepted, as the worst kind of “judicial legislation”.55 The Court was reluctant to engage in what it considered to be the illusory task of discerning the legislature’s intention from a statute’s silence on whether a breach of duty gives rise to a cause of action. What carried the day in Saskatchewan Wheat Pool was the view that statutes were increasingly addressing issues of civil responsibility and individual compensation, and therefore the role of tort liability in compensation and the allocation of loss was becoming less important.56 In the end, Dickson J. eschewed the “will-o’-the-wisp” effort to discern legislative intention, holding that if the legislature was silent on the issue of civil liability, then it could mean only one of two things: either the legislature had not turned its mind to the question, and therefore there was no legislative intent one way or the other; or, it deliberately omitted the possibility of civil remedies. Curiously, he did not consider that the legislature might be legislating on the basis that there is nothing inherently incompatible between a common law private remedy and statutory remedies, and therefore no comment is required in the statute. In other words, statutory remedies for injured interests could be interpreted as coexisting with extant (or to be developed) common law remedies, with the former excluding the latter only through an express provision.57 Within a human rights context, the relationship between statutory breaches and private rights of action erupted in the Ontario Court of Appeal’s decision in Bhadauria v. Board of Governors of Seneca College of Applied Arts and

54 The Queen v. Saskatchewan Wheat Pool [1983] 1 SCR 205. The judgment has stood the test of time, as courts throughout the country have accepted its holding (cf. Hercules Managements Ltd. v. Ernst & Young [1997] 2 SCR 165, for the most recent Supreme Court of Canada consideration of Saskatchewan Wheat Pool). A nominate tort of statutory breach is a separate tort for breaching a statutory provision, where a statute imposes penalties for violation of its provisions but is silent with respect to civil liability. See Thornton v. Kirklees Metropolitan Borough Council [1979] QB 626 (CA): “To my mind counsel for the plaintiff is right in his submission that on the true construction of the Act Parliament intended that there should be a cause of action for damages . . .”, at 642. For an analysis of Saskatchewan Wheat Pool, see A Brudner, “Civil Liability for Breach of Statutory Duty Abolished”, (1984) 62 Can Bar Rev 668 (Brudner welcomed the certainty created by Saskatchewan Wheat Pool, as an antidote to the tremendous unpredictability that existed in Canada regarding the legal effect of a breach of a statutory duty). 55 Saskatchewan Wheat Pool, supra n. 54 at 216. 56 Saskatchewan Wheat Pool, supra n. 54 at 224. 57 R Sullivan, Driedger on the Construction of Statutes, 3rd ed. (Toronto, Butterworths, 1994) at 370. As Sullivan states, it is presumed that the legislature does not intend to abolish or limit the rights of subjects (and, arguably, remedies available to subjects) without an express statement.

What Possibility for Canada? 415 Technology.58 In what Laskin C.J. later described as “. . . bold . . . and [to] be commended as an attempt to advance the common law”,59 the Ontario Court of Appeal found that a person who can establish that she was refused employment because of her ethnic origin, and thus suffered damages, has a common law action for discrimination against the person who refused her employment. In point of fact, Wilson J.A., for the Court of Appeal, found only that Seneca College was in breach of a common law duty not to discriminate, which was sufficient to dispose of the appeal.60 Consequently, she did not address the question of civil liability that might arise from a breach of the duty not to discriminate under the Human Rights Code. However, at a minimum her judgment implied that the existence of the statutory prohibitions and remedies did not preclude parallel development of the common law. On appeal, the Supreme Court of Canada reversed the Court of Appeal’s judgment and rejected its reasoning, Laskin C.J. penning the reasons despite the above-quoted praise for the Court of Appeal’s boldness. Laskin C.J. held that the Human Rights Code not only excluded any right of action at common law for the tort of discrimination, but also that the comprehensiveness of its administrative remedies and enforcement mechanisms foreclosed any private right of action based on a breach of its provisions.61 On this latter point, though, the Chief Justice was somewhat equivocal, recognising that there had been penal statutes giving rise to civil actions. In contrast, he concluded that this particular statute, the Code, was regulatory in the sense that responsibility for enforcement under a comprehensive scheme was given to an administrative agency.62 Because of the Code’s comprehensiveness, Laskin C.J. held that the legislature intended to foreclose the operation of the common law through the court system. Why this is so, his Lordship did not clearly explain. As Professor Klar has pointed out, the Code did not state that all possible common law remedies for discrimination were abolished, so it is unclear on what basis the Court came to this conclusion.63 Perhaps it was a way of side-stepping the thorny issue of specifying the circumstances under which a court should treat unlawful conduct that 58

(1979), 105 DLR (3d) 707 (Ont CA). The Board of Governors of the Seneca College of Applied Arts and Technology v. Bhadauria [1981] 2 SCR 181 at 194. Despite these words of admiration, Laskin C.J., for the Court, concluded: “In my opinion, however, this [attempt at recognising a tort of discrimination] is foreclosed by the legislative initiative which overtook the existing common law in Ontario and established a different regime which does not exclude the courts but rather makes them part of the enforcement machinery under the [Ontario Human Rights] Code,” at 194–5. Neena Gupta has concluded that the Supreme Court’s decision in Bhadauria was motivated by two factors: confidence in the capacity of administrative agencies to deal with human rights complaints, and fear that courts would continue to prove hostile to human rights issues, N Gupta, “Reconsidering Bhadauria: A Re-examination of the Roles of the Ontario Human Rights Commission and the Courts in the Fight against Discrimination” (LL.M. Thesis, Faculty of Law, University of Toronto, 1993) at 168. 60 This was a new cause of action, as the matter was res integra before the Court. 61 Bhadauria, supra n. 59 at 195 (SCC). 62 Bhadauria, supra n. 59 at 188–9. This ambivalence was finally laid to rest two years later in Saskatchewan Wheat Pool, supra n. 54, and accompanying text. 63 L N Klar, “Developments in Tort Law: The 1980–81 Term”, (1982) 3 Sup Ct L Rev 385 at 397. 59

416 E M Hyland harms another person as tortious, which the Court of Appeal had not addressed. By saying simply that there was no cause of action at common law because of the comprehensiveness of the enforcement regime, the Court spared itself the task of determining whether a cause of action in tort could be made out at all. The Supreme Court’s decisions in Saskatchewan Wheat Pool and Bhadauria illustrate the tension between judge-made law and the law of elected officials. The tension is between the legislative decision to deal with public and private disputes through administrative procedures, which provide for relatively fast and inexpensive means for resolving conflict and serve to advance public policies, and the traditional idea of the rule of law, with its notions of individual justice and the equitable resolution of cases. As Professor Stone put it, “Legislatures create administrative agencies with the desire and expectation that they will perform efficiently the tasks committed to them . . . It is an aim so obvious as to make unavoidable the conclusion that the function which courts are called upon to perform, in carrying into operation such administrative schemes, is constructive, not destructive, to make administrative agencies, wherever reasonably possible, effective instruments for law enforcement, and not to destroy them.”64

In other words, courts should tread lightly in the face of the legislature’s decision to provide administrative mechanisms to resolve disputes. In the English case of Nagle v. Feilden, Salmon L.J. expressed forcefully the rule-of-law side of the equation: “One of the principal functions of our courts is, whenever possible, to protect the individual from injustice and oppression. It is important, perhaps today more than ever, that we should not abdicate that function.”65 The case involved a woman horse trainer who was denied a licence to practise her profession. Although the United Kingdom’s Sex Disqualification (Removal) Act, 191966 prohibited such discrimination and provided administrative remedies, the Court of Appeal found that the woman had nonetheless disclosed a private cause of action in tort. In contrast, the Supreme Court of Canada has limited private rights of action in tort for wrongful conduct where a statutory scheme occupies the field. In one area of human rights law (anti-discrimination law), where regulation of private relationships is covered by a codified scheme, the courts are seemingly prevented from granting individuals any private rights of action. However, this conclusion needs to be qualified in two respects. First, Bhadauria was decided before the advent of the Charter. What possible influence might the Charter have on the development of the common law in the field of human rights? Second, 64 H F Stone, “The Common Law in the United States”, (1936) 50 Harv L Rev 4 at 18. For a review of the involvement of Canadian courts in labour relations, see P Weiler, “The ‘Slippery Slope’ of Judicial Intervention: The Supreme Court and Canadian Labour Relations: 1950–1970”, (1971) 9 Osgoode H L J 1. 65 [1966] 2 QB 633 at 654 (CA). See Watson v. Prager and another [1991] 3 All E R 487 at 504–5 (CA) Here, the Court of Appeal held that restrictions on a boxer’s ability to make arrangements regarding his career were subject to the common law restraint of trade doctrine. In so holding, the Court of Appeal adopted the approach in Nagle. 66 1919, c. 71 (UK).

What Possibility for Canada? 417 Bhadauria turned on the Court’s view that the Code provided a complete and effective set of remedies to deal with discriminatory conduct. It was on this basis that Laskin C.J. concluded that recourse to the common law was foreclosed.67 The question arises, therefore, whether the common law is inapplicable in the absence of a complete and effective legislated remedial system.

International Human Rights Law, Torture and the Common Law: A Right of Action? (a) The state of affairs in Canada and surrounding context As we have seen, one of the distinguishing characteristics of international human rights litigation in the United States is that it has proceeded under the authority of legislation, which the courts have interpreted not only as providing adjudicative jurisdiction but also as granting rights of action to victims of torture and of other human rights abuses. According to the emerging consensus, US courts are recognising a private right of action for human rights violations contrary to customary international law. It is sufficient for a plaintiff to plead a violation of international law in order to ground a cause of action. There is no requirement to go further and show that international law itself creates a cause of action. In contrast, such a statutorily-authorised right of action does not exist in Canada, perhaps with one exception. In 1988 the Yukon Territories passed the Torture Prohibition Act, which makes it a tort for a public official or anyone acting at the instigation of a public official to inflict torture on another person.68 Its enactment was in response to Canada’s ratification of the Convention against Torture. In its second report to the Committee against Torture, the Canadian government cited this legislation as an example of its compliance with article 14 of the Convention, which imposes on states the obligation to provide mechanisms of redress for victims of torture.69 While the Yukon provisions have not been judicially considered, it seems clear that they are only intended to provide a right of action to persons who have been the victims of torture at the hands, or with the acquiescence, of Yukon officials. This territorial limitation contrasts with the universal civil jurisdiction available under the United States’ 67

This is not to say that Laskin C.J.’s conclusion was empirically correct. S Y 1988, c.26: “1. Every public official, and every person acting at the instigation of or with the consent or acquiescence of a public official, who inflicts torture on any other person commits a tort and is liable and renders his or her employer liable to pay damages to the victim of the torture,” section 1. 69 Multiculturalism and Citizenship Canada, Outlawing an Ancient Evil: Torture, Second Report of Canada (April 1, 1988–December 1991) (Ottawa, Minister of Supply and Services Canada, 1992) at paras 208–9; cf. Torture Prohibition Act, supra n. 68, Preamble: “Recognising that Canada is a party to the United Nations Convention against Torture and Other Cruel, Inhuman, or Degrading Treatment or Punishment, The Commissioner of the Yukon Territory, by and with the advice and consent of the Legislative Assembly, enacts as follows: . . .” 68

418 E M Hyland Torture Victim Protection Act (TVPA) and the Alien Tort Claims Act.70 Thus, apart from the Yukon’s Torture Prohibition Act with its limited reach, there is no legislation in Canada granting individuals a right of action for a violation of international human rights law prohibiting torture.71 There is increasing international effort at providing victims with more effective means of securing remedies and adequate reparations in the face of human rights violations. For example, the United Nations Sub-Commission on Prevention of Discrimination and Protection of Minorities has developed basic principles and guidelines outlining the right of victims of gross human rights and humanitarian law violations to reparation and redress. According to these principles, every state is under an obligation to ensure the availability of appropriate remedies to any person claiming that his or her rights have been violated. This includes the right of access to national procedures for protection, and it imposes upon states the obligation of ensuring that their legal systems provide for prompt and effective administrative, civil, and criminal procedures of redress and reparation.72 Although ambiguous on this point, the SubCommission’s Basic Principles and Guidelines on the Right to Reparation can be read as requiring universal jurisdiction for civil claims where the underlying 70 According to section 5 of the Torture Prohibition Act, “ ‘public official’ includes a peace officer and any person in the public service of the Yukon (a) who is authorised to do or enforce the doing of any act or thing or to exercise any power, or (b) upon whom any duty is imposed by or under any act,” supra n. 68. 71 Note that the federal government has jurisdiction over the Yukon and was thus in a position to legislate directly in implementation of the Convention. It does raise the issue of whether the Federal government is implicitly saying that the absence of equivalent legislation in each of the ten provinces of Canada results in non-compliance with article 14. 72 UNCHR, Sub-Commission on Prevention of Discrimination and Protection of Minorities, “Basic Principles and Guidelines on the Right to Reparation for Victims of Gross Violations of Human Rights and Humanitarian Law”, E/CN.4/Sub.2/199617. “Every state shall ensure that adequate legal or other appropriate remedies are available to any person claiming that his or her human rights have been violated. The right to a remedy against violations of human rights and humanitarian norms includes the right to access to national and any available international procedures for their protection,” at article 4. “The legal system of every State shall provide for prompt and effective disciplinary, administrative, civil and criminal procedures so as to ensure readily accessible and adequate redress, and protection from intimidation and retaliation. Every State shall provide for universal jurisdiction over gross violations of human rights and international humanitarian law which constitute crimes under international law,” at article 5. “Civil claims relating to reparations for gross violations of human rights and international humanitarian law shall not be subject to statutes of limitations,” at article 9 (emphasis added). In the case of Velásquez Rodriguez (1990) 11 HRLJ 127, the Inter-American Court of Human Rights held that: “It is a principle of international law, which jurisprudence has considered ‘even a general principle of law’, that every violation of an international obligation which results in harm creates a duty to make adequate reparation,” at para. 25. The Court further held that article 63(1) of the American Convention of Human Rights, supra n. 8, required the Court to look to the Convention itself and applicable principles of international law to determine the appropriate level of indemnity. As such, the Court was not bound by the internal laws of Honduras, the state found in breach of the Convention. In its reasoning, the Court relied on views of the UN Human Rights Committee and decisions of the European Court of Human Rights in support of the view that international human rights law requires indemnification for human rights violation, at para. 29. The Court again adopted this view in Aloeboebe v. Suriname (1992), 13 HRLJ 140.

What Possibility for Canada? 419 acts also constitute crimes under international law.73 Within the context of torture, the Convention against Torture already imposes on states parties the obligation to “. . . ensure in [their] legal system[s] that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation . . .”74 From the point of view of international law, this obligation extends to a state’s judicial branch, which is considered part of the state apparatus.75 There is accordingly some argument to be made that the Sub-Commission’s view of general international law on gross violations of human rights could, with time, come to interact interpretively with article 14 of Convention against Torture, with Canadian judges being under a duty to consider this interaction as a live option. It should finally be recalled that one crucial aspect of the context in Canada is that Canadian courts have moved to limit private rights of action at common law where the legislature has legislated in the field. On the one hand, breaches of duties arising under a legislative scheme do not give rise to a cause of action, unless expressly granted. On the other hand, the comprehensiveness of a regulatory scheme is inversely related to the likelihood of the courts recognizing a parallel common law tort action. (b) The relationship between international law and Canadian law Answering the question of whether in Canada there might be a basis for a common law right of action for international human rights violations ultimately requires a consideration of the role that international human rights law plays within domestic law. As Anne Bayefsky points out, the principles governing the relationship between Canadian domestic law and international law reflect the attempt to reconcile a number of (competing) policies, including “. . . protection of national sovereignty, protection of the supremacy of Parliament and the Legislatures from the powers of the executive, satisfaction of Canada’s international obligations, [and] realization by individuals of the benefits of international norms.”76 Has the legal protection of international human rights so encroached upon the sovereignty of states as to subject the latter’s law to the requirements of the former, as expressed in either treaty or custom? Canadian case law supports the view that customary international law is directly and immediately adopted into domestic law, even in the absence of a legislative enactment, although no Supreme Court of Canada case appears to 73

See the emphasized words, ibid. CAT, supra n. 14 at article 14. 75 Cf. International Law Commission, “Report of the International Law Commission on its Twenty-ninth Session: State Responsibility”, supra n. 27 at 13; M Hunt, Human Rights Law in English Courts (Oxford, Hunt Publishing, 1997) at 313–15; Brownlie, supra n. 19 at 144. 76 A Bayefsky, “International Human Rights Law in Canadian Courts” in B Conforti and F Franconi, (eds), Enforcing International Human Rights in Domestic Courts (The Hague, Martinus Nijhoff Publishers, 1997) 295 at 296. 74

420 E M Hyland have said so directly even if the case appears to have proceeded on this assumption.77 However, a good number of lower court cases have adopted the view that customary international law is part of Canadian common law.78 That being said, no case has yet been decided in which customary international law has been invoked as the sole basis for a private law cause of action, in contrast with cases in which it has been invoked by way of some jurisdictional or substantive defence to proceedings being brought against the person invoking custom.79 Despite the thin record to date, at least one scholar has concluded after careful study that “customary international human rights law can be directly invoked as part of the law of the land, and itself provide the basis for a remedy.”80 If customary international law is directly assimilated into the common law, it is otherwise for treaties. In Canada, treaty-making is a prerogative of the federal executive and is not subject to Parliamentary consent.81 Because of the policy of protecting the legislative supremacy of Parliament and the provincial legislatures, the Supreme Court of Canada has held that a treaty will have direct domestic legal effect only if it is implemented by legislation.82 This decision of 77 Reference Re Power of Municipalities to Levy Rates on Foreign Legions and High Commissioner’s Residences [1943] SCR 208 at 213–14, Duff, C.J.; 232, Rinfret J.; 241, Hudson, J.; Reference Re Whether Members of the Military or Naval Forces of the United States of America are Exempt from Criminal Proceedings in Canadian Criminal Courts [1943] SCR 483 at 501–2, Kerwin, J.; 517, Taschereau, J. 78 Re Alberta Union of Provincial Employees et al. and the Crown in Right of Alberta (1981), 120 DLR (3d) 590 at 607 and 620 (Alta QB); Orelien v. Canada [1992] 1 FC 592 at 608–9 (CA); Carrato v. The United States (1982), 40 OR (2d) 459 at 461 (Ont HC); Re Regina and Palacios (1984), 45 OR (2d) 269 at 276 (CA). There are many old cases stating that customary international law is incorporated into and forms part of the law of England: Triquet v. Bath (1764), 3 Burr 1478; Heathfield v. Chilton (1767), 4 Burr 2015; Emperor of Austria v. Day and Kossuth (1861) 3 De GF&J 217. For a review of the situation in New Zealand see, R Harrison, “Domestic Enforcement of International Human Rights in Courts of Law: Some Recent Developments”, (1995) NZLJ 256. See also R St J MacDonald, “The Relationship Between International Law and Domestic Law in Canada” in R St J MacDonald, G L Morris and D J Johnston, (eds), Canadian Perspectives on International Law and Organization (Toronto, University of Toronto Press, 1974) 88 at 111. 79 Immunity cases, both diplomatic and state, are good examples of custom providing a common law defence (in contrast with cause of action), prior to these areas of law having been largely codified in Canada by statute: see, for example, the two 1943 references to the Supreme Court of Canada cited supra n. 77. There is also case law which invoke various norms from the law of the sea such as hot pursuit cases: see for example, The Ship “North” v. The King (1906), 37 SCR 385. 80 Bayefsky, supra n. 76 at 300. Bayefsky goes on to note: “The right of action has not been affected by the Charter [due to section 26 of the Charter].” Section 26 reads: “The guarantee in this Charter of certain rights and freedoms shall not be construed as denying the existence of any other rights or freedoms that exist in Canada,” Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), c.11. 81 See P W Hogg, Constitutional Law of Canada, 4th ed. (Toronto, Carswell, 1997) vol. I at 11–15. 82 Re Arrow River and Tributaries Slide & Boom Co. Limited [1932] SCR 495. “The treaty in itself is not equivalent to an Imperial Act and without the sanction of Parliament, the Crown cannot alter the existing law by entering into a contract with a foreign power . . . [T]hese rights and privileges are, under our law, enforceable by the Courts only where the treaty has been implemented or sanctioned by legislation rendering it binding upon the subject . . . In the absence of affirming legislation this provision of the treaty cannot be enforced by any of our Courts whose authority is derived from municipal law,” at 510–11. Cf. The Parlement Belge (1878–9), 4 PD 129: “Blackstone must have known very well that there were a class of treaties the provisions of which were

What Possibility for Canada? 421 the Supreme Court anticipated a subsequent judgment of the Privy Council, which held that the power to implement a treaty was subject to the division of powers in the Constitution.83 This ruling reinforced the provinces’ legislative sovereignty in the face of the federal government’s treaty-making power. Thus, for a treaty-based human right to have direct legal effect in Canada and be enforceable by the courts, it must be incorporated domestically through a legislative enactment of the appropriate legislature. However, a powerful interpretive presumption is available to the courts that will give some degree of indirect legal force to international treaties whatever the degree to which they have or have not been incorporated in legislation: to the extent possible, domestic legislation should always be interpreted in a manner that is consistent with international treaty obligations, on the presumption that Parliament and the legislatures do not intend to violate those obligations.84 The recent Supreme Court case of Baker solidifies this interpretive presumption of statutory compliance with international law in unequivocally holding that this presumption applies to international human rights treaties, such as the Convention on the Rights of the Child at issue in that case. However, it remains an open issue, unaddressed in Baker, how much this presumption applies to the inoperative without the confirmation of legislature; while there were others which operated without such confirmation. The strongest instance of the latter, perhaps, which could be cited is the Declaration of Paris in 1856, by which the Crown in exercise of its prerogative deprived this country of belligerent rights, which very high authorities in the state and in the law had considered to be of vital importance to it. But this declaration did not affect the private rights of the subject; and the question before me is whether this treaty does affect private rights, and therefore required the sanction of the legislature,” at 150. 83 A.-G. Canada v. A.-G. Ontario (Labour Conventions) [1937] AC 326 (PC). With the ratification of human rights conventions and of international and regional trade agreements, there have been more calls for a reconsideration of the Labour Conventions case by the Supreme Court, or at least questioning the wisdom of that decision. See, for example, J Claydon, “The Application of International Human Rights Law by Canadian Courts”, (1981) 30 Buff L Rev 727 at 735; T H Strom and P Finkle, “Treaty Implementation: The Canadian Game Needs Australian Rules”, (1993) 25 Ott L Rev 39 at 58; R E Sullivan, “Jurisdiction to Negotiate and Implement Free Trade Agreements in Canada: Calling the Provincial Bluff”, (1986) 24 U West Ont L Rev 63. For a strong contrary view, see A L C de Mestral, “Treaty Power, and More on Rules and Obiter Dicta”, (1983) 61 Can Bar Rev 856. While Labour Conventions remains, to this day, the statement of law with respect to the division of treaty powers, in obiter various justices of the Supreme Court have not ruled out the possibility of reconsidering the decision, cf. Francis v. The Queen [1956] SCR 618, Kerwin C.J.; MacDonald et al. v. Vapor Canada Ltd. [1977] 2 SCR 134: “In my opinion, assuming Parliament has power to pass legislation implementing a treaty or convention in relation to matters covered by the treaty or convention which would otherwise be for provincial legislation alone, the exercise of that power must be manifested in the implementing legislation and not be left to inference,” MacDonald at 171, Laskin J. (emphasis added). 84 Baker v. Canada (Minister of Citizenship and Immigration) (1999), 174 DLR (4th) 193 (SCC). See discussion in W Adams, “In Search of a Defence of the Transnational Human Rights Paradigm: May Jus Cogens Norms be Invoked to Create Implied Exceptions in Domestic State Immunity Statutes?”, chapter 9 of this volume. See also Daniels v. The Queen [1968] SCR 517 at 541; American Farm Bureau Federation v. The Canadian Import Tribunal (1990) 74 DLR (4th) 449 at 482 (SCC). However, if it is not possible to give domestic legislation an interpretation that is consistent with a treaty obligation, and there is a conflict between international law and the domestic statute domestic law, then the latter will prevail: Daniels, ibid. at 541; Capital City Communications et al. v. CRTC, [1978] 2 SCR 141 at 173.

422 E M Hyland interpretation of law in Canada as a whole. Even assuming a broad application, would it permit—indeed require—Canadian courts also to approach the common law as presumptively in conformity with Canada’s international law obligations, including those found in treaties? Would such an interpretive use outside the context of statutory interpretation constitute an unjustified end-run around the basic principle that treaties acquire the force of law in Canada through legislative enactment?85 We turn now to one legal theorist’s approach that, implicitly, addresses this issue. (c) Jus cogens and Canadian common law Alan Brudner has identified three ways for enforcing international human rights law in domestic courts: using international human rights norms as guides for constitutional and statutory interpretation; using these norms as guides for judges in their development of common law principles; and the direct creation by international norms of domestic rights and obligations.86 His argument with respect to the last method rests on a natural law view that jus cogens norms are deducible from natural reason, and therefore are directly assimilable into domestic law. Jus cogens norms possess a degree of self-evidence such that they depend on their inherent rationality and not on the consent of nations for their legal force.87 However, in a retreat of sorts, Brudner then limits the function of jus cogens norms within the domestic legal system to that of principles guiding and constraining the operation of rules of decision.88 Because of their level of generality, Brudner considers jus cogens norms incapable of providing, directly, rules of decision within and for domestic systems. What is curious, and left unexplained in Brudner’s analysis, is how this account of the direct enforcement of jus cogens norms differs functionally from the other two roles he sees international human rights norms playing in domestic law. On Brudner’s view it seems that international norms are only able to act as heuristic principles for determining domestic legal rules of decision. At the level of decision, he maintains the parallel universes of domestic and international law. What is the difference between the role played by international law when it functions as a principle for constitutional and statutory interpretation, a guide for determining the limits of the common law, and a principle for distilling and then applying a rule of decision in a particular case? None, it would seem—at first blush. However, it is arguable that Brudner’s theory does add some conceptual support to the argument that international human rights treaties, at least core norms that are candidates for jus cogens status, can be legitimately invoked as 85

See W Adams, ibid. for the view that this would constitute unacceptable judicial legislation. A Brudner, “The Domestic Enforcement of International Covenants on Human Rights: A Theoretical Framework”, (1985) 35 UTLJ 219. 87 Ibid. at 251. 88 Ibid. at 251–2. 86

What Possibility for Canada? 423 powerful background principles in articulating the common law. It may be replied that this does no additional work given that any norms in treaties that are jus cogens will also be part of general international law and, if this is so, it is general international law, not treaty law, which melds with the common law. There is something in this riposte. At minimum, however, Brudner’s focus on substance over form provides one reason why it would be unacceptable for a defendant to rely on the fact that a right is in a treaty and not expressly incorporated by domestic legislatures as justification for the argument that the common law cannot be taken as providing a cause of action for a universal tort of torture.89 (d) The Canadian Charter and the common law: constitutional torts? Rather than rely on philosophical accounts of the force of jus cogens norms, would we not be better off seeking to ground a tort of torture in the fundamental rights that now lie at the centre of Canada’s constitution? The Supreme Court of Canada has established that the Canadian Charter of Rights and Freedoms “should generally be presumed to provide protection at least as great as that afforded by similar provisions in international human rights documents which Canada has ratified.”90 Thus, the norms of the Convention against Torture are to be taken as protected by the Charter. To the extent the Convention places an obligation on states parties to provide for civil redress for torture, how easily does that translate into Charter-based cause of action for the tort of torture? Here the Slaight Communications presumption must contend with what the Supreme Court had previously established as the relationship between the Charter and private rights in Canadian common law. In RWDSU v. Dolphin Delivery, the majority of the Supreme Court held that private parties do not owe each other constitutional duties and therefore cannot found a private cause of action based on the violation of a Charter right.91 Reliance on the common law, in the absence of government action, does not violate the Charter. However, the majority did allow that the courts should develop the common law in a manner consistent with Charter values. For the majority, McIntyre J. put the matter as follows:

89 Significantly, this recruitment of Brudner’s argument into the position of playing defence is consistent with the savings clause in article 1(2) of the Convention against Torture which reads: “This article is without prejudice to any international instrument or national legislation which does or may contain provisions of wider application,” supra n. 14. 90 Slaight Communications Inc. v. Davidson [1989] 1 SCR 1038 at 1056–7 per Dickson C.J. For cases which have applied the Slaight Communications approach in the context of deportation that risks leading to torture, see e.g. Suresh v. R (1998), 38 OR (3d) 264 esp. at 274 and 278 (Gen Div) and Said v. Canada (1999), 48 Imm LR (2d) 9 at 19 (Fed Ct, Trial Div). 91 [1986] 2 SCR 573; Hill v. Church of Scientology [1995] 2 SCR 1130; M.(A.) v. Ryan [1997] 1 SCR 157; Dobson (Litigation Guardian of) v. Dobson [1999] 2 SCR 753 at para. 84 (minority reasons of McLachlin J., concurred in by L’Heureux-Dubé J.).

424 E M Hyland “Where, however, private party ‘A’ sues private party ‘B’ relying on the common law and where no act of government is relied upon to support the action, the Charter will not apply. I should make it clear, however, that this is a distinct issue from the question whether the judiciary ought to apply and develop the principles of the common law in a manner consistent with the fundamental values enshrined in the Constitution. The answer to this question must be in the affirmative. In this sense, then, the Charter is far from irrelevant to private litigants whose disputes fall to be decided at common law. But this is different from the proposition that one private party owes a constitutional duty to another, which proposition underlies the purported assertion of Charter causes of action or Charter defences between individuals.”92

In the absence of a reversal of position by the Supreme Court, it is clear that the Charter does not create a private cause of action for torture qua violation of section 7 or section 12.93 Rather, the question is whether the Charter strongly encourages courts to develop the common law to allow for the creation of a tort of torture as a function of the interplay between international law and the Charter in accordance with the Slaight Communications interpretive presumption. The answer to the question depends, in part, on the degree to which judges will be moved to develop the common law to ensure that it conforms to the values of the Charter. In R v. Salituro, Iacobucci J., for the Supreme Court, outlined the approach that courts ought to adopt, an approach that the Court has subsequently confirmed: “Judges can and should adapt the common law to reflect the changing social, moral and economic fabric of the country. Judges should not be quick to perpetuate rules whose social foundation has long since disappeared. Nonetheless, there are significant constraints on the power of the judiciary to change the law. As McLachlin J. indicated in Watkins [ [1989] 2 S.C.R. 750], in a constitutional democracy such as ours it is the legislature and not the courts which has the major responsibility for law reform; and for any changes to the law which may have complex ramifications, however necessary and desirable such changes may be, they should be left to the legislature. The judiciary should confine itself to those incremental changes which are necessary to keep the common law in step with the dynamic and evolving fabric of our society. . . . Where the principles underlying a common law rule are out of step with the values enshrined in the Charter, the courts should scrutinize the rule closely. If it is possible to change the common law rule so as to make it consistent with Charter values, without upsetting the proper balance between judicial and legislative action . . . then the rule ought to be changed.”94 92

RWDSU v. Dolphin Delivery [1986] 2 SCR 573. Section 7 reads: “Everyone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice.” Section 12 states: “Everyone has the right not to be subjected to any cruel and unusual treatment or punishment.” 94 [1991] 3 SCR 654 at 670 and 675 (the Court found that the common law rule of spousal testimonial incompetence in criminal cases was inconsistent with developing social values and the values of equality in the Charter); cf. Hill, supra note 91, at 1169; R v. Swain [1991] 1 SCR 933 (the Court found that the common law rule that allowed the Crown to raise the defence of insanity, 93

What Possibility for Canada? 425 Taking Iacobucci J.’s statement as a weathervane, it is apparent that the Supreme Court, despite stating that courts are the custodians of the common law and should develop it in a way that is consistent with society’s emerging needs and Charter values,95 has adopted a very cautious view of the degree of change that is desirable. As Cory J. put it for the majority in Hill, “far-reaching changes to the common law must be left to the legislature.”96 Indeed, it would seem that these two judges, in their partial dissent in Baker, have signalled that this cautious view may, for them at least (or, rather, for Iacobucci J. given that Cory J. has now retired from the Court), carry over to the question of the interpretive relevance of international human rights treaties on the development of the common law. In their opinion in Baker, Iacobucci and Cory J.J. took issue with the majority’s robust use of the Convention on the Rights of the Child for purposes of statutory interpretation: “It is a matter of well-settled law that an international convention ratified by the executive branch of government is of no force or effect within the Canadian legal system until such time as its provisions have been incorporated into domestic law by way of implementing legislation. . . I do not agree with the approach adopted by my colleague [L’Heureux-Dubé J. for the majority], wherein reference is made to the underlying values of an unimplemented international treaty in the course of the contextual approach to statutory interpretation and administrative law, because such an approach is not in accordance with the Court’s jurisprudence concerning the status of international law within the domestic legal system. In my view, one should proceed with caution in deciding matters of this nature, lest we adversely affect the balance maintained by our Parliamentary tradition, or inadvertently grant the executive the power to bind citizens without the necessity of involving the legislative branch. I do not share my colleague’s confidence that the Court’s precedent in Capital Cities survives intact following the adoption of a principle of law which permits reference to an unincorporated convention during the process of statutory interpretation. Instead, the result will be that the appellant is able to achieve indirectly what cannot be achieved directly, namely, to give force and effect within the domestic legal system to international obligations undertaken by the executive alone that have yet to be subject to the democratic will of Parliament.”97

If one puts this view together with the views Cory and Iacobucci J.J. have expressed on the relationship between the common law and the Charter, it would seem a fortiori that these two judges would see recognition of a common law tort based on international law as an even greater disruption of the balance in the parliamentary tradition. Yet, caution is needed in drawing such a conclusion because these two judges also appeared to offer a strong endorsement of the despite the objections of the accused, was inconsistent with s. 7 of the Charter. For the majority, Lamer C.J. was of the view, “Given that the common law rule was fashioned by judges and not by Parliament or a legislature, judicial deference to elected bodies is not an issue. If it is possible to reformulate a common law rule so that it will not conflict with the principles of fundamental justice, such a reformulation should be undertaken,” at 978). 95 Cf. Salituro, supra n. 94 at 678. 96 Hill, supra n. 91 at 1171. 97 Iacobucci J. in Baker, supra n. 84 at 234–5.

426 E M Hyland Charter as a more legitimate vehicle for bringing international human rights treaty law to bear on interpretation of domestic law than by way of a direct relationship between these treaties and non-constitutional law: “[T]he result may well have been different had my colleague [L’Heureux-Dubé J.] concluded that the appellant’s claim fell within the ambit of rights protected by the Canadian Charter of Rights and Freedoms. Had this been the case, the Court would have had an opportunity to consider the application of the interpretive presumption, established by the Court’s decision in Slaight Communications and confirmed in subsequent jurisprudence, that administrative discretion involving Charter rights be exercised in accordance with similar international human rights norms.”98

Thus, the caution suggested by Salituro and Hill should not be taken as meaning that a tort of torture could not be grounded in the double influence of, first of all, article 14 of the Convention against Torture on the Canadian Charter and, second, of the Canadian Charter on the common law. Iacobucci and Cory J.J. in Salituro and Hill did not have in mind, it seems safe to say, a situation where the Charter can be said to have interpretively received a positive obligation on the state of Canada to provide channels for torture victims to bring claims for civil redress. Quite apart from the role of article 14, it is also important to note that it is perfectly consistent with Iacobucci J.’s cautious Salituro approach to build on the existing torts of assault and battery as a way to give common law recognition to the cluster of Convention-derived constitutional rights in relation to torture. It remains, however, a challenge to arrive at a Charter-informed tort of torture that applies not only extraterritorially but also to conduct for which there is no consent or acquiescence by the Canadian government. There are signs, though, that of all Charter norms, the prohibition on torture is the most likely to enjoy a form of extraterritorial application. The Court had its first occasion to reconsider the fundamental values underlying s. 12 of the Charter in Kindler v. Canada (Minister of Justice).99 Kindler had been found guilty of murder and sentenced to death in the United States, and had escaped to Canada. He was arrested in Canada and ordered extradited. The Canadian government failed to seek assurances from the US government that the death penalty would not be imposed. Kindler appealed the extradition on the grounds that it violated sections 7 and 12 of the Charter. Although the majority of the Court upheld the extradition to face the death penalty as not violating the Charter, La Forest J. did acknowledge that some punishment or treatment in a foreign country would be so contrary to the values of Canadian society, and such an attack on the dignity of the individual concerned that extradition in those circumstances would have to be refused. He specifically cited, as an example, the situation of someone being deported who would face torture upon his or her return. In the Court’s view, there were some circumstances, such as the likelihood of torture 98 99

Iacobucci J. in Baker, supra n. 84 at 235. [1991] 2 SCR 779.

What Possibility for Canada? 427 as part of the punishment a person faced upon his or her return, in which Charter values would prevail and be applied in a way to affect the legal process in another country.100 At the very least, the Kindler dictum suggests the legitimacy of a common law tort of torture that could be brought against officials of the government of Canada for their involvement in the decision to deport in the face of knowledge of a substantial risk that the deportee will be tortured by a foreign state. Even if the torturers in Sri Lanka or Iran might not easily come within the scope of the common law tort, at the very least Canadian public officials who “acquiesced” and who were indeed complicitous in the torture fall squarely within the combined logic of civil redress under article 14 of the Convention and extraterritorial reach of the Charter in relation to expulsion of persons to face torture abroad. On the other hand, the Supreme Court has resisted application of the Charter extraterritorially in the sense of application to conduct which occurs outside Canadian territory (as opposed to conduct that occurs in Canada, such as a decision to deport, and which effectively facilitates rights-violating conduct abroad). Very significantly, the Court, in Cook, has recently allowed for the extraterritorial application of Charter protections where to do so would not interfere with the sovereign authority of the foreign state.101 The issue in Cook was whether the Charter applied to the taking of the appellant’s statement by Canadian police in the United States in connection with their investigation of an offence committed and to be prosecuted in Canada. In allowing the appeal and ordering a new trial because of a breach of the appellant’s Charter right to counsel, the majority of the Court held that the Charter did apply to the Canadian police officers’ action in the United States on the jurisdictional basis of the officers’ nationality (Canadian) and because the application of Charter standards did not conflict with any obligation of the officers under the territorial jurisdiction of the foreign state, in this case the United States. It has been suggested above why the Charter can be the legitimate vehicle for recognising a common law tort of torture in Canada in at least three circumstances: (1) torture occurring in Canada, either by public officials or acquiesced in by public officials; (2) decisions by Canadian officials to deport a person to a 100 Ibid. at 832. The Court may soon have the opportunity to consider the issue of deportation to torture head on as a result of two judgments of the Federal Court of Appeal in Suresh v. Canada (Ministry of Citizenship and Immigration), [2000] Doc. A-415-99 (FCA) and Ahani v. Canada (Ministry of Citizenship and Immigration) [2000] Doc. A. 413-99 (FCA), handed down in January 2000, in which a unanimous bench of three judges held that, while it would infringe section 7 of the Charter to deport a person to a country where there would be substantial risk of torture, such deportation could be justified under section 1 of the Charter as a reasonable limitation on the right not to be tortured. Contrast this attitude of the Federal Court of Appeal regarding the dispensability of rights protection with the more rights-sensitive approach in Philippines (Republic) v. Pacificador (1999) 60 CRR (2d) 126 at 154 (Ont Ct Gen Div). In Pacificador, the court stayed the deportation of a Philippine national on the ground that there was a possibility that he would be deprived of a speedy trial in the Philippines, where he was accused of murder. The court said that to return him to such a situation would be a violation of s. 7 of the Charter. 101 R v. Cook [1998] 2 SCR 597.

428 E M Hyland situation where there is a substantial risk of torture occurring in a foreign state; and (3) conduct abroad by Canadian state agents that constitutes torture. The Supreme Court’s approach both to the relationship between the common law and the Charter and to the extraterritoriality of Charter norms suggests that a universal tort of torture will not emerge as a matter of course. But a Charterinformed analysis of the common law can credibly get to this result, helped along by the doctrinal reconceptualisation of the nature of state sovereignty that is underway and by the development of specific norms in international human rights law that call for redress for torture and similarly-serious violation to be made possible in domestic legal systems. Although arising in the criminal law context and concerned with the particular legal rights in sections 7 to 14 of the Charter, it is arguable that the logic of Cook should allow for the development, under the influence of the Charter, of a common law tort of torture in Canadian courts where Canadian agents are the ones who are responsible for the torture abroad. An obvious example is the torture of Shidane Arone, in Somalia, by members of the Canadian armed forces serving there as peacekeepers. As Bastarache J. put it, in his minority opinion (Gonthier J. concurring) in Cook, non-human rights principles of international law (including the principle of sovereign authority) should not be applied to circumscribe and limit the application of the Charter. He said:102 “The Charter . . . is an expression of the fundamental relations between the inidvidual and the state which is the [principal] legal organ of the society of which that individual is a part. That an interpretation of such rights might place the state in violation of its international law obligations should be accorded less weight than, in the case of a mere statute, an expression of the legislative will of the state as it may exist from time to time. . . . [E]ven if s. 32(1) [the application clause in the Charter] were to conflict with some aspect of with international law, I emphasise here that the presumption of statutory interpretation that Parliament intended to legislate in conformity with international law must be applied with great care in the Charter context. The Charter is the fundamental expression of the minimum obligations owed to individuals in our society; I would not be inclined to accept that Canada’s international law obligations could truncate rights defined by the Charter.”

Not only should Charter rights not be so truncated, but the same conclusion also applies to the development of the common law under the influence of Charter values. In the next and final section, we will see how developments in public international law on the duty to provide (or provide for) reparations and in private international law on the relationship between comity and the international human rights order may prove conducive to a form of universal tort of torture 102 R v. Cook, supra n. 101 at paras 147–8. This dictum of Bastarache J. was not part of his reasoning on the extraterritoriality application and effects test, and was passed over in silence by the majority. Arguably, it remains open to the Court to revisit the question raised by Bastarache J. in his comments, and to adopt his analysis.

What Possibility for Canada? 429 in Canadian common law that is consistent with the opinion expressed by Bastarache J.

5 CONVERGENT TRAJECTORIES OF PUBLIC AND PRIVATE INTERNATIONAL LAW

To return, then, to the matter of torture, the question that remains is whether the individual victim of torture has a common law right of action against his or her torturer. A more modest question could be asked: is there such a cause of action at least against torturers who act in Canada? In other words, is there a common law analogue to the Yukon’s statutory implementation of article 14 of the Convention against Torture? If this more modest question is first asked, we can separate out the jurisdictional question of the geographical scope of the norm’s application from its existence as a norm within Canada’s own “private” law. If there is a duty to have a system of civil claims for redress within domestic law according to public international law and if private international law would permit one state’s domestic torture-redress norms to be applied to extraterritorial conduct in some contexts, then there may be no need to ground a universal tort of torture in any given system of law.

The Evolving and Deepening Public International Law on Reparation for Torture If the obligations on states under international human rights law are not compatible with the individual right to seek redress, then in effect we are talking about a state’s power to disperse and limit the locus of accountability to the abstract structure of the state (state responsibility under international law and public law judicial review under domestic law) and thereby shield persons who torture, directly or indirectly, for the state. In the year 2000, can public law’s relationship to private law truly be said to be of a sort that was rejected over a half-century ago at Nuremberg as an unacceptable elevation of the abstraction of the state over the fact of individual humans carrying out acts in the name of the state? As it turns out, international human rights law has yet to embrace fully and firmly the logic of Nuremberg in the civil law sphere. The emergence of the individual as the subject of human rights at international law is the source of the international community’s recognition of the requirement to provide victims with more effective remedies following human rights violations.103 Accordingly, states are coming to admit their obligation to 103 As Higgins puts it, “Human rights are held simply by virtue of being a human person. They are part and parcel of the integrity and dignity of the human being. They are thus rights that cannot be given or withdrawn at will by any domestic legal system,” R Higgins, Problems and Process: International Law and How We Use It (Oxford, Clarendon Press, 1994) at 96. It is on this basis that obligations at international law are owed directly to individuals.

430 E M Hyland ensure that their legal systems provide for prompt and effective administrative, civil, and criminal procedures of redress and reparation. In the case widely considered to be the leading pronoucement on a duty to repair as part of the general duty to ensure all human rights, Velásquez Rodriguez v. Honduras, the InterAmerican Court of Human Rights expressed the view that it is a general principle of international law that violation of a right that results in harm creates an obligation to make adequate reparation.104 The Court’s claim is broad, suggesting that the obligation arises irrespective of a specific provision of a human rights convention. However, the Court’s dictum is also limited in that the Court left open the precise means of reparation, at least as a matter of principle. Thus, Velásquez Rodriguez does not stand for the proposition that a violation of an international human right gives rise to a private right of action. While customary human rights law might impose on states an obligation to provide reparation for a violated human right, it does not follow that the form—or even a form—of such reparation must be a private right of action. However, as Jordan Paust has demonstrated, and even Judge Bork conceded in Tel-Oren, customary international law has provided private rights of action in some areas of law.105 The issue is under what conditions will the international community recognise as a matter of general international law to extend such a remedy to the human rights violation of torture? And closer to home, under what conditions will a Canadian court take such a step? In his reasons in TelOren, Judge Edwards refused to take it on the evidence before him about the state of international law that such a private cause of action existed in international law for torture without “guidance from the Supreme Court.”106 As for treaty law, even the Convention against Torture is not free of ambiguity. While article 14 imposes on a state the obligation to ensure that its legal system provides a victim of torture with an enforceable right to fair and adequate compensation, the Convention does not prescribe the precise modalities of the right or the way in which the obligation is to be discharged, whether a private civil right or, for example, a statutory right claimed within an adminstrative law regime. It appears to be left by article 14 for each state to determine the means by which the obligation for civil redress will be fulfilled. Yet, article 14’s reference to an “enforceable right” cannot but be understood as requiring some kind 104

Supra n. 72. J J Paust, “Litigating Human Rights: A Commentary on the Comments”, (1981) 4 Hous J Int’l L 81 at 87. Tel-Oren, supra n. 33 at 813–16 (e.g., violation of safe-conducts, infringements on the rights of ambassadors, and piracy). And note: “The law administered by the Prize Court is international law which originates in the practice and usage long observed by civilised nations in their relations with each other or in express international agreement.” Halsbury’s Laws of England, 4th ed. (London, Butterworth & Co. (Publishers) Ltd., 1982) vol. 37 at para. 1306. See also, The Zamora [1916] 2 AC 77 at 91 (PC); France Fenwick Tyne and Wear Co. Ltd. v. H.M. Procurator General, The Prins Knud [1942] AC 667 at 678 (PC). See also a review of US case law on the direct application of customary international law and private rights of action for breaches of international law, see Paust, ibid. See also R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No 3) [1999] 2 All ER 97 at 109 (Lord Browne-Wilkinson). 106 Tel-Oren, supra n. 33 at 795. 105

What Possibility for Canada? 431 of legal structures and procedures within which a claim to compensation can be put forward. It is arguable that, at minimum, a common law right of action should be recognised at least until legislatures take the necessary steps to put statutory regimes in place that comply with article 14. With the exception of the Yukon Territory’s Torture Prohibition Act and perhaps some provincial criminal compensation schemes, there is currently no domestic legislative regime that provides adequate remedies to victims of torture. Thus, it would seem perfectly consistent with the wording of article 14 and the general purposes of the Convention against Torture for a Canadian court to develop a common law tort of torture applicable in the three circumstances listed earlier (territorial conduct, decisions to deport and acts of agents of Canada abroad) as a provisional way to ensure Canada is in compliance with its international—and Charter— obligations until necessary and sufficient legislation covers the field.

Comity in Private International Law and an Argument for Principled Instrumentalism in Choice of Law Analysis Our concern involves persons, interests and policies that have significant connections with more than one sovereign community. There is the foreign national who has been tortured, the accused torturer, the state in whose territory and with whose sanction the torture occurred, and Canada’s relations with this state (and possible others). Where the torture victim and the torturer find themselves together in a third country, the choice is often between proceeding before the forum court or not proceeding at all. It will be assumed that, from a strictly public international law perspective, there is no basis for the forum court having general jurisdiction over the defendant in a civil action, much less a unified body of law on civil liability for torture by which to decide the case. There is, though, another way of examining this dilemma. By definition, private international law involves courts in geographically complex legal problems that transcend the limits of territorial sovereignty and where there is no supranational authority whose responsibility it is to resolve these controversies. In wrestling with the traditional questions that make up the conflict of laws paradigm—whether or not to take jurisdiction, whether or not to recognise and enforce a foreign judgment, which law should function as the rule of decision—courts are forced to confront issues of government and political power, resource allocation, the structure of the international economy, the expectations and interests of the parties, values of fairness, justice and certainty and the role and function of national legal systems, including the courts, within the international community. Inevitably, courts are thrown into a balancing exercise where the reference points are polycentric. They are not faced with the binary choice of international versus municipal law, of subordinating municipal law to universal international norms. More specifically, the discipline of choice of law analysis helps the courts break away from a purely public international

432 E M Hyland law frame of reference, allowing them to recast the question facing them in terms of identifying the state interests involved and assessing them in light of a broader ambition to do justice in the circumstances of the case. This is not to say that Canadian courts’ understanding of private international law sits easily with the characterisation found in the preceding sentence. Most Canadian judges would likely see a cause of action for torture committed abroad as an interference with the sovereign authority of a foreign state and inconsistent with the notion of comity that is at the heart of Canadian private international law. La Forest J. accepted comity, for the purpose of defining a contemporary Canadian approach to conflicts of law, as “the recognition which one nation allows within its territory to the legislative, executive or judicial actions of another nation, having due regard both to international duty and convenience, and to the rights of its own citizens or of other persons who are under the protection of its laws.”107 Comity, however, is not an end in itself, but a doctrinal corollary of the structure of the international order which must be understood in light of the evolution of the international order itself and especially by reference to the principles of order and fairness which the Supreme Court of Canada has held to be the underlying values of a modern system of private international law.108 Significant for our purposes is the Court’s recognition that conflicts rules must be shaped and developed in light of changing international (and federal) circumstances. For example, the structural problems of creating and maintaining a federal economic union, all the while integrating the ongoing legal fact of jurisdictional sovereignty of the Canadian provinces, drove the Court’s judgment in Tolofson. A variation on the comity theme, La Forest J. rejected the positivistic rule-oriented approach developed early in the twentieth century, and championed one that looks to the underlying realities of the legal order as the basis for developing a rational and workable system of private international law.109 More 107

Morguard Investments Ltd. v. De Savoye [1990] 3 SCR 1077 at 1096. Morguard Investments Limited, supra n. 107, at 1097; cf. Tolofson v. Jensen; Lucas (Litigation Guardian of) v. Gagnon [1994] 3 SCR 1022 at 1058. Although comity is not an end in itself, certainly in Tolofson it played a decisive role in the outcome of the case and the adoption of a strict rule of lex loci delicti as the basis for choice of law in torts. In his reasons for judgment, La Forest J. expressed the view that order is a precondition for the achievement of justice, and it was on this basis that he held that the rule of decision to be applied in a torts case is that of the lex loci delicti: “From the general principle that a state has exclusive jurisdiction within its own territories and that other states must under principles of comity respect the exercise of its jurisdiction within its own territory, it seems axiomatic to me that, at least as a general rule, the law to be applied in torts is the law of the place where the activity occurred,” at 1049–50. Later in his judgment, in response to concerns about the rigidity of the rule he had formulated, his Lordship countered, “True, it may be unfortunate for a plaintiff that he or she was the victim of a tort in one jurisdiction rather than in another and so be unable to claim as much compensation as if it had occurred in another jurisdiction. But such differences are a concomitant of the territoriality principle. While, no doubt, as was observed in Morguard, the underlying principles of private international law are order and fairness, order comes first. Order is a precondition to justice,” at 1058. 109 Tolofson, ibid. at 1047; Hunt v. T&N PLC [1993] 4 SCR 289 at 322. See J Herbert, “The Conflict of Laws and Judicial Perspectives on Federalism: A Principled Defence of Tolofson v. Jensen”, (1998) 56 U T Fac L Rev 3. 108

What Possibility for Canada? 433 concretely, the Court has given pride of place to a view of the world shaped by the exigencies of commerce and the phenomenon of globalisation.110 Again, in the words of La Forest J., “[t]he business community operates in a world economy and we correctly speak of a world community even in the face of decentralised political and legal power. Accommodating the flow of wealth, skills and people across state lines has now become imperative [for private international law rules].”111 However, more than commerce is driving changes in international conditions. Indeed, hand in hand with the establishment of international legal regimes equipped to deal with the phenomenon of economic globalisation are efforts to inject human rights and environmental concerns into the equation. One need only think of the North American Agreement on Labour Cooperation (NAALC) and the North American Agreement on Environmental Cooperation (NAAEC), two side agreements accompanying the North American Free Trade Agreement.112 There is also the ongoing debate about the inclusion of a social charter clause in the World Trade Organisation (WTO) or the coordination, in some fashion, of International Labour Organisation norms with the WTO order. The examples could continue. The point is that the changes to the world’s economic order also mean changes to the social conditions in which people work, live and relate to the environment. These changes are increasingly being governed by multilateral agreements, reflecting a growing world-wide concern for human and environmental rights.113 We thus are seeing an emerging teleological and contextual method in conflict of laws which gives pride of place to underlying changing social and economic conditions and presumably also to overarching international legal developments. It is in this context that a choice of law analysis of the proper rule of decision to apply in cases of international torture might yield different results for the possibility of an action in tort than if we were to assimilate it to the lex loci delicti general rule for tort laid down in Tolofson.114 Recall that the discipline of choice of law analysis involves identifying the relevant interests and values that will allow a court to do justice in a situation where a single legal system cannot obtain the right solution. This analysis would not necessarily diminish the 110 At times in considerable tension with La Forest J.’s problematic use of the notion of territorial sovereignty, and even “absolute” sovereignty, as the governing idea of the international legal order. 111 Morguard, supra n. 107 at 1098. 112 North American Agreement on Labour Cooperation (1993) 32 ILM 1499; North American Agreement on Environmental Cooperation (1993) 32 ILM 1480. 113 For a discussion of recent attempts by the international community to articulate the obligations of member countries, see C Elwell, “World Social Policy Conferences as Rule-Making: Implications for Canadian Federalism in a Decentralised Environment”, (1997) 4:3 Canadian Foreign Policy 83. See also D Crerar, for an analysis of the weight to give human rights in the context of Canadian conflicts analysis, “A Proposal for a Principled Public Policy Doctrine PostTolofson”, (1998) 8 Windsor Rev Legal & Social Issues 23. 114 For a detailed discussion of Tolofson and how it was not completely faithful to La Forest J.’s articulation of his own choice of law method, see J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11 in this volume.

434 E M Hyland peremptory normative weight of the international prohibition against torture. Instead, it would provide judges with a different (and more familiar) analytical framework for deciding whether a victim of torture that occurred abroad might have a cause of action in a Canadian court. Comity is about nations respecting each other’s sovereignty. However, such respect is not unconditional. It must include the way in which sovereignty is treated within the international legal order. Comity, it bears emphasising, concerns mutual sovereign respect and not the mutual respect of states divorced from how international law, notably international human rights law, constructs the evolving legal meaning of sovereignty. In Tolofson, the majority of the Court constructed what seemed a strict rule based on the idea of comity only to turn it into a general rule capable of some flexibility. The Court recognised the possibility of overriding norms whose effect would be to suspend the operation of comity, thus leaving the forum court free to rely on other rules to promote the supranational values and goals at issue in a given case.115 My argument similarly relies on the presence of overriding—peremptory—norms in the international legal order, but is different as to the role of those norms in choice of law analysis. Taking the international law of human rights seriously requires moving to a choice of law analysis characterised by a principled instrumentalism, an approach I will now seek to develop, at least schematically. Where a court is faced with the task of vindicating a torture victim’s right to be free from torture and where the existing law of the forum court renders it unable to provide justice (with the result of impunity for the torturer), the court should look to international human rights law both as embedded in the notion of comity and as a candidate rule of decision for the case. Put in other terms, there is no comity, in general, in the absence of the enforcement of international obligations and, in particular, without the respect for human rights as a condition of justice. Here, it is important to observe that respect for and enforcement of international human rights law does not somehow stand above the logic of reciprocity that underlies so much of international law. A scheme of reciprocity is present even in international human rights law at least with respect to the most basic rights. In the Barcelona Traction case, the International Court of Justice held that certain human rights obligations are owed erga omnes, by all states to all states.116 Thus, the purpose of the core of international human 115 Tolofson, supra n. 108 at 1047: “On the international plane, the relevant underlying reality is the territorial limits of law under the international legal order. The underlying postulate of public international law is that generally each state has jurisdiction to make and apply law within its territorial limit. Absent a breach of some overriding norm, other states as a matter of ‘comity’ will ordinarily respect such actions and are hesitant to interfere with what another state chooses to do within those limits” (emphasis added). See discussion of this passage in J Orange, supra n. 114, and J Llewellyn, “Just Amnesty and Private International Law”, chapter 22 of this volume. 116 Decision in the Case of Barcelona Traction Light and Power Company, Limited (Belgium v. Spain) (1970) 9 ILM 227 at 259 (I.C.J.): “[A]n essential distinction should be drawn between the obligations of a State towards the international community as a whole, and those arising vis-à-vis another State in the field of diplomatic protection. By their very nature the former are the concern of all States. In view of the importance of the rights involved, all States can be held to have a legal

What Possibility for Canada? 435 rights law is to impose on all nations obligations for the protection of individuals and thus to give all states the requisite legal interest, and standing, to use all peaceful means available to them in order to seek respect for those rights— whether the right to engage in interstate diplomatic espousal of claims, which was at issue in Barcelona Traction, or to provide the judicial machinery that facilitates interpersonal espousal of claims by victims of human rights violations against the alleged perpetrators. The notion of erga omnes human rights norms thus shapes the principle of sovereignty in international law so as to convert what might be presented as illicit, extraterritorial intervention into a foreign state’s domestic affairs into a sanctioned power to try to enforce international law. In this way, we can see as inherent in the international human rights system the interest of all states in the protection, promotion and enforcement of human rights. Each nation is a member of the international community, which has forged and continues to deepen a consensus on the conditions for legally protecting the dignitary interests of persons throughout the world. As a limit on the principle of sovereignty in public international law, international human rights law also functions, then, as a constitutive element of the notion of comity in private international law. We have witnessed the demonic logic of the apotheosis of the territorial principle—the Holocaust, Cambodia, Rwanda, the terrible atrocities in the former Yugoslavia, the barbarities occurring in Chechnya as I write, the list could go on. Even without the conceptual framework provided by the legal notion of erga omnes norms, our experience should make acutely clear the interests of the forum court in providing a means for achieving justice with respect to an incident of torture halfway around the world. A private international law analysis which allowed itself to be shaped by the underlying values and interests of international human rights law would thus see the principle of comity as informed and conditioned by those interests and values. Comity would no longer stand for the principle of territoriality alone. Instead, a court would understand comity as membership in a wider moral and legal community whose history includes this ever-deepening commitment to the legal protection of the dignity of the human person. There is no sovereignty without this membership, and the values of the international community give full meaning to sovereignty. To be sovereign is to respect, promote and assist the international community in respecting and promoting human rights. Courts would have to consider the underlying legally protected interest of persons to respect for and protection of their human rights as a factor in choosing the rule of decision in an action for torture. A court would consider the international law prohibiting torture as a possible rule of decision, alongside the law of the jurisdiction in which the torture occurred and the law of the forum. In such an analytical framework, international law would not supplant domestic interest in their protection; they are obligations erga omnes.” On erga omnes norms in international human rights law, see especially T Meron, Human Rights and Humanitarian Norms as Customary Law (Oxford, Clarendon Press, 1989).

436 E M Hyland law, but exist alongside it. Critical for the choice, though, would be that rule that was most consistent with the forum jurisdiction’s membership in the international community, and which would promote the solidarity of that community in respecting and protecting human rights. An example of such an analysis is the District Court of New York’s judgment on remand in Filártiga.117 Engaging in an interest analysis, Judge Nickerson concluded that the interests of the global community transcend those of any one state and that he should look to international law for the substantive rules of decision in the case.118 The court concluded that the nature of the cause of action was more than simply assault and battery, that torture is “. . . an act so monstrous as to make its perpetrator an outlaw around the globe.”119 As David Burger has remarked, the consequence of engaging in this type of analysis would be three possible outcomes: the choice of the law of the forum, the choice of the law of the territory in which the tort occurred, or some third legal regime.120 From the perspective of Canadian choice of law jurisprudence, this form of analysis would take place while still respecting the principle of comity so central to the Supreme Court’s conception of private international law. In a tort action for torture, assuming that a court would not decline jurisdiction, the judge would first consider the lex loci delicti and its incorporation of the international law against torture. The court might examine whether there is a domestic prohibition of torture, whether the foreign jurisdiction provides for a right of action for torture in its domestic law, and the nature and extent of remedies available in the foreign law. Respect for the foreign law, on the basis of comity, would be limited by the extent to which the foreign law has itself observed comity in complying with its international obligation to provide adequate legal redress to the victims of torture. In the measure that the foreign law did not satisfy its obligations, comity would not require the forum court to choose the foreign law as the rule of decision. Similar considerations would also apply in the court’s consideration of its jurisdiction’s own law. The point is that a conflicts analysis would provide to courts considering whether to recognise a tort of torture a framework consisting of the underlying values of international human rights law. These values are no less important than the commercial values that the Supreme Court identified and elevated in Tolofson.121 The values of human dignity, of protecting the physical and psy117

Filártiga v. Peña-Irala, 577 F.Supp. 860 (1984). Ibid. at 863. 119 Ibid. 120 D Burger, “Transnational Public Policy as a Factor in Choice of Law Analysis”, (1984) 5 N Y L Sch J Int’l & Comp L 367 at 388. 121 Tolofson, supra n. 108 at 1047. In his reasons, La Forest J. stated, “I have already indicated, of course, that I view the lex loci delicti rule as the governing law. However, because a rigid rule on the international level could give rise to injustice, in certain circumstances, I am not averse to retaining a discretion in the court to apply our own law to deal with such circumstances. I can, however, imagine few cases where this would be necessary,” at 1054. Establishing a private right of action for torture might be just such an exceptional circumstance. Given the fact that domestic legal systems are so anaemic in their incorporation and adoption of international human rights law (whether 118

What Possibility for Canada? 437 chological integrity of the individual person, have crystallised into an internationally recognised legal interest to be free from torture. These are the values that a court should look to in determining which law has the most significant relationship between the occurrence of torture and the parties involved.122 At the heart of whether there is a tort of torture in geographically complex cases is fundamentally a choice of law question, but one based on substantive principles of selection as opposed to the traditional substance-neutral principles: which legal system will recognise and protect an individual’s interest to be free from torture? Which legal system will advance the forum jurisdiction’s interests as a member of an international community committed to eliminating torture and providing victims with effective remedies? Considerable further work needs to be done to flesh out this embryonic conception of value-laden choice of law in which an instrumental method is fused to an intrinsically-valuable purpose. In following the choice of law principle of “choose that system of law which best integrates compensation for torture into its approach to civil liability,” international human rights law would play a role at two sequential stages of analysis. First, recourse to international human rights law would be as the source of normative purpose which provides the criterion for determining which of the candidate national legal systems should be chosen where torture is the underlying harm a tort claim seeks to have remedied. A second recourse to international human rights law would see a court apply it, mutatis mutandi, as the rule of decision where none of the candidate national legal systems adequately promotes the torture-reparation goal; in so doing, domestic judges would in effect be fashioning a transnational common law that is a hybrid of public law norms and private law principles of civil liability.123 Is this normative analysis sufficient to give rise to a right of action within Canada, at common law? The answer, in the final result, may depend on one very simple factor: judicial attitudes toward creating causes of action and using not just public law prescriptions but public international law principles as reasons for doing so. Here, we may have no choice but to appeal to, and place faith in, the evolution of a more cosmopolitan judicial culture in Canada. In the final customary or conventional), perhaps it is precisely in the circumstances of human rights violations that courts should be most flexible in adopting the rule which will leave the least chance of rendering an unjust decision. 122 This principle of flexibility was affirmed by the Privy Council in Red Sea Insurance Co. Ltd. v. Bouygues SA and others [1994] 3 All ER 749 at 762 (PC). 123 Judge Rosalyn Higgins’ approach to prescriptive jurisdiction and “extraterritoriality” discourse could be of great assistance in developing the approach I am suggesting: see Higgins, supra n. 103 at 76–7. Higgins argues that the exercise of jurisdiction in certain circumstances (e.g. to fight crime) should be based on whether international solidarity would be helped or harmed. She goes on to state that, as a matter of principle, the key to understanding the issue of extraterritoriality lies in the protection of common values rather than in the invocation of state sovereignty for its own sake. On this basis, then, an argument could be made that international solidarity in the effort to eliminate torture and the commitment to respect and to protect the inherent dignity of every person are grounds for extraterritorial jurisdiction, in the sense of a combined extraterritoriality of adjudicative and prescriptive jurisdiction. Foreign domestic judges would apply norms that are not, or not solely, grounded in the law of the state in whose territory the torture occurred.

438 E M Hyland analysis, the issue is about doing justice, and about Canada’s obligation to the international community and to individuals to see that torture is eliminated and that its victims are able to vindicate their rights. Surely, this is at the heart of our courts’ role.

16

Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability VALERIE OOSTERVELD 1 and ALEJANDRA C. FLAH 2

1 INTRODUCTION

domestic laws have long recognised forms of derivative liability. Under international law, superiors can be held responsible for crimes committed by their subordinates through the doctrine of command responsibility. In many national jurisdictions, employers can be held liable for the activities of their employees under the doctrine of respondeat superior. In all cases, the focus of the liability is control. That is, liability falls on those who had control over the actor and who failed to exercise that control to stop the harm. Command responsibility and respondeat superior are structurally analogous: they provide ways through which a person in authority can be held liable for deeds done by another. This chapter will examine whether these doctrines could provide a useful foundational approach in a tort action brought in Canada by a foreign victim in order to hold a superior responsible for torture committed by a subordinate.3 First, the parameters of command responsibility are explored. Command responsibility has long been recognised in international humanitarian law. When its conditions are met, commanders may be held liable for failing to prevent, repress, investigate or punish criminal subordinates. The doctrine of command responsibility has been applied by international criminal tribunals and national criminal and military courts, and recently also in American Alien Tort Claims Act cases. Respondeat superior, and its specific application in enterprise liability, is considered next. Under the common law doctrine of respondeat

B

OTH INTERNATIONAL AND

1

Legal Officer with the Canadian Department of Foreign Affairs and International Trade. Associate, McCarthy Tétrault, Toronto. 3 We begin from the assumption that a tort claim for torture by and against a foreign national can be brought in Canada. 2

442 V Oosterveld and A Flah superior, principals are liable for the torts committed by their agents within the scope of the agents’ employment, regardless of whether the principal personally breaches a duty, or authorises the agents to breach a duty, towards a third party. Respondeat superior is a domestic tort law doctrine. A discussion follows as to how command responsibility and respondeat superior could be translated into a framework of analysis in Canada for foreign tort claims.

2 COMMAND RESPONSIBILITY

The Doctrinal Development of Command Responsibility Under international law, there are two distinct forms of responsibility for violations of human rights or humanitarian law: state responsibility and individual criminal responsibility. Individual criminal responsibility can be subdivided into direct responsibility and indirect responsibility. Direct individual responsibility applies to those who “planned, instigated, ordered, committed or otherwise aided and abetted in the planning, preparation and execution” of a crime,4 and these people are held individually responsible for the crime. Indirect individual responsibility applies to those who are held responsible not for committing the crime itself but for failing to stop another from committing the crime. The doctrine of command responsibility is a form of indirect responsibility, under which the commander or superior is held criminally responsible for crimes committed by his or her subordinate. It is distinct from holding a commander responsible for ordering crimes, which is considered direct individual responsibility. The doctrine of command responsibility, which is also known as superior responsibility, is a well-established norm of customary and conventional international law.5 It applies to all those in a position of authority, whether military, paramilitary or civilian, who fail to act when they have a positive legal duty to prevent or repress violations of international humanitarian law (such as genocide, crimes against humanity or war crimes) by others who are within their control.6 Command responsibility is not, however, responsibility for all crimes

4 Statute of the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, UN Doc. S/25704 at 36, annex, art. 7 (1993) and S/25704/Add.1 (1993), adopted by the Security Council on 25 May 1993, UN Doc. S/RES/827 (1993). 5 Command responsibility traditionally refers to responsibility of military or paramilitary commanders, while superior responsibility refers to the responsibility of a civilian superior. However, the term “command responsibility” is often used as a shorthand to refer to both military and civilian superior responsibility: we will use the term “command responsibility” to refer to both concepts unless explicitly indicated. 6 See Prosecutor v. Delalic, Mucic, Delic and Landzo, ICTY Case No. IT-96-21-T (16 November 1998) at para. 334.

Command Responsibility and Respondeat Superior 443 committed by subordinates. There must be some personal dereliction of duty by the commander in order for derivative criminal responsibility to arise.7 A form of command responsibility has existed for hundreds, if not thousands, of years.8 However, the doctrine remained, to a large extent, undefined until World War II. Following that war, the International Military Tribunal based in Nuremberg, the International Military Tribunal for the Far East based in Tokyo, and national tribunals set up by the allies applied and developed the doctrine. The result was detailed discussion on when a commander may be held liable, and the extent of his or her liability. The first case on the issue of derivative liability was Yamashita, in which the trial commission found that Yamashita, a general in command of all Japanese forces in the Philippines in 1944–5 and military governor, failed in his duty to properly monitor his troops, who were accused of executing, raping and torturing numerous civilians. The commission stated that where “there is no effective attempt by a commander to discover and control the criminal acts, such a commander may be held responsible, even criminally liable, for the lawless acts of his troops . . .”.9 In addition, the trial commission held that Yamashita knew or must have known about the atrocities, as the crimes committed by subordinates were so “extensive and widespread, both as to time and area, that they must either have been willfully permitted by the accused, or secretly ordered by the accused”.10 In the High Command case decided at Nuremberg, General Wilhelm von Leeb was accused of implementing Hitler’s orders calling for the murder of Russian political officers and the mistreatment of Russian civilians. Von Leeb claimed that he was completely unaware of the atrocities of his subordinates and that he had done everything to oppose Hitler’s illegal orders except outright refusal to obey. Von Leeb was deemed guilty of certain charges and was acquitted of others. The tribunal found that a commander is not liable for every crime committed by subordinates under his command and control, only the ones for which his act or omission could be linked to the crime: “Criminality does not attach to every individual in this chain of command from that fact alone. There must be personal dereliction. That can occur only where the act is directly traceable to him or where his failure to properly supervise his subordinates constitutes criminal negligence on his part.”11

A second significant Nuremberg trial was known as the Hostage case. The accused were all high-ranking German military officers, charged with the 7 United States v. Wilhelm von Leeb (High Command Trial), UN War Crimes Commission, 12 Law Reports of Trials of War Criminals (1948) at 76. 8 W H Parks, “Command Responsibility for War Crimes”, (1973) 62 Mil L Rev 1 at 2–20; W D Burnett, “Command Responsibility and a Case Study of the Criminal Responsibility of Israeli Military Commanders for the Pogrom at Shatila and Sabra”, (1985) 107 Mil L Rev 71 at 77–84. 9 The Yamashita Case 1945–6, Decision of the United States Military Commission at Manila, reprinted in L Friedman (ed.), The Law of War: A Documentary History (New York, Random House Publishers, 1972) at 1596 at 1597. 10 Ibid. at 1596. 11 Supra n. 7, at 76.

444 V Oosterveld and A Flah murder and deportation by their troops of thousands of civilians from Greece, Yugoslavia, Norway and Albania. The relevant defendants claimed that they had no knowledge of the crimes and that this was in part because they did not read reports of the ongoing events—reports which subordinates were expected to make to their commanders. The tribunal held that a commander cannot escape liability for his or her own failure to obtain knowledge of events: “If he fails to require and obtain complete information, the dereliction of duty rests upon him and he is in no position to plead his own dereliction as a defense.”12 This approach was confirmed by the International Military Tribunal for the Far East, in the case of Admiral Toyoda: “[I]f this accused knew, or should by the exercise of ordinary diligence have learned, of the commission by his subordinates, immediate or otherwise, of the atrocities . . . or of the existence of a routine which would countenance such, and, by his failure to take any action to punish the perpetrators, permitted the atrocities to continue, he has failed in his performance of his duty as a commander and must be punished.”13

Each of these World War II cases assisted in the development of the doctrine of command responsibility. After the Yamashita case had confirmed the existence of both duty and responsibility for breach of duty in broad brushstrokes, the High Command, Hostage and Toyoda cases provided more explicit detail.14 The International Military Tribunal for the Far East also found that civilian leaders, such as members of government and government officials, were responsible in the same manner as military commanders for dereliction of duty. Other World War II cases added to the understanding of certain aspects of the doctrine, for instance by confirming that command responsibility was not a form of strict liability.15 Following World War II and the resulting trials, the 1949 Geneva Conventions were adopted. While these Conventions did state that a commander who ordered unlawful acts was therefore liable for those acts, they did not mention the responsibility of commanders who fail to prevent, or put an end to, a breach of the Conventions. This was due in part to lingering doubts as to whether Yamashita and the other cases might have gone too far in articulating the law of command responsibility.16 By 1977, however, the precedential value of the postWorld War II cases was clear. Additional Protocol I included two provisions on command responsibility which seem to have come close to the exacting postWorld War II principles. Article 86, entitled “Failure to act”, sets out the standards by which knowledge and failure to act are to be judged: 12 United States v. Field Marshal Wilhelm List, UN War Crimes Commission, 11 Law Reports of Trials of War Criminals (1948) at 1271–2, (the Hostage case). 13 United States v. Soemu Toyoda, Official Transcript of Record of Trial at 5006, quoted in Parks, supra n. 8, at 72. 14 Parks, supra n. 8, at 63. 15 International Military Tribunal for the Far East Judgment (November 1948), reprinted in Friedman, supra n. 9 at 1038–9. Japanese Foreign Minister Hirota, Prime Minister Tojo and Foreign Minister Shigemitsu were found guilty of inaction amounting to criminal negligence. 16 Burnett, supra n. 8, at 138.

Command Responsibility and Respondeat Superior 445 “The fact that a breach of the Conventions or of this Protocol was committed by a subordinate does not absolve his superiors from penal or disciplinary responsibility . . . if they knew, or had information which should have enabled them to conclude in the circumstances at the time, that he was committing or was going to commit such a breach and if they did not take all feasible measures within their power to prevent or repress the breach.”17

The knowledge test contained in the words “had information which should have enabled them to conclude” does not seem, on its face, to go quite as far as the Hostage case’s duty to obtain information, but the difference between the two formulations is probably not all that great. Protocol I’s article 87 is entitled “Duty of commanders”, and it requires that military commanders prevent, suppress, punish and report breaches of the Convention and the Protocol.18 Recent case law of the International Criminal Tribunal for the Former Yugoslavia has reaffirmed that articles 86 and 87 are firmly embedded in international law.19 The most recent iteration of the doctrine of command responsibility is found in article 28 of the Statute of the International Criminal Court, which, to a large extent, reflects the basic tenets of articles 86 and 87 of Protocol I, but with two specific mens rea standards: a “knew or should have known” standard for military commanders—which seems to reiterate the Hostage approach—and a “consciously disregarded information which clearly indicated” standard for civilian commanders.20

Proving Command Responsibility How does one prove command responsibility? A prosecutor must present convincing evidence on four elements: (1) proof of the atrocity, (2) link to the commander, (3) mens rea of the commander, and (4) actus reus of the commander. The first element is proof that a subordinate committed a violation of international humanitarian law. The second element, demonstrating the link between the subordinate and the commander, has two sub-elements: proof of command and proof of control. Proof of command can be demonstrated by 17 Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts, 8 June 1977, 1125 U.N.T.S. 3, reprinted in 16 I.L.M. 1391 (1977), art. 86. These articles should be considered as reflecting customary law: W J Fenrick, “Some International Law Problems Related to Prosecutions Before the International Criminal Tribunal for the Former Yugoslavia”, (1995) 6 Duke J Comp & Int’l L 103 at 119. 18 Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts, 8 June 1977, supra n. 17 at article 87. 19 Delalic, supra n. 6, at para. 393; Prosecutor v. Blaskic, ICTY Case No. IT-95-14-T (3 March 2000) at paras 300 and 325–9. 20 Rome Statute of the International Criminal Court, UN Doc. A/CONF.183/9 (17 July 1998), article 28. The difference in mens rea standards reflects the different levels of control existing within a military hierarchy and a civilian hierarchy. See discussion of linkages between articles 86 and 87 and the International Criminal Court Statute’s article 28 in I Bantekas, “The Contemporary Law of Superior Responsibility”, (1999) 93 AJIL 573 at 589–90.

446 V Oosterveld and A Flah showing that the military or civilian hierarchy links the commander and the subordinate in a relationship where the commander in principle has command over the subordinate. The term “commander” applies to a wide range of people in positions of authority: military, paramilitary, political and bureaucratic superiors, including heads of state, senior governmental officials and both civilian and military heads of detention camps.21 Therefore, command responsibility not only implicates the highest commanders, including Presidents and Ministers of Defence, but also those further down the command chain, even though their subordinates might be few in number.22 Different commanders have different reaches of command. For example, a higher degree of responsibility is assigned to military commanders responsible for an occupied territory, referred to as “occupational commanders”. An occupational commander has broad executive and legal authority, and is responsible for peace and order within a territory.23 He or she is therefore responsible not only for all units in the area even though they may not be operationally subordinate to him and even though they may in fact be troops from other parts of the military or from other countries.24 The command responsibility of a “tactical commander” is narrower than that of an occupational commander. A tactical commander is criminally liable only for directly-subordinated units. Sometimes, a tactical commander does not have the power to dismiss or punish subordinates. Therefore, the responsibility of these tactical commanders will be limited to the measures that he or she can take within the limits of his or her power under the circumstances to prevent or stop war crimes.25 The second sub-element is proof of control. If the commander does not have actual control, then it is difficult to argue that he or she could have prevented the crime. While proof of command creates a presumption of proof of control, in some situations it will be impossible to prove that someone was officially in command, for example in situations where a person takes control but does not have the de jure authority to do so.26 It will then be necessary to prove that the person was in control. Proof of control may come from evidence that subordinates followed orders and that they considered the de facto leader their superior.27 The third element of command responsibility is proof of mens rea, or knowledge of the commander. Actual knowledge of the commander may be shown by direct evidence, such as proof that the commander witnessed the crime or was 21 Fenrick, supra n. 17, at 111. See also International Military Tribunal for the Far East Judgment, supra n. 15, at 1038–9. 22 Burnett, supra n. 8, at 143. 23 Parks, supra n. 8, at 83. 24 Burnett, supra n. 8, at 110. 25 Parks, supra n. 8, at 84. This may mean reporting the conduct of the subordinate to a superior in the military hierarchy who does have the power to punish the subordinate. 26 Bantekas, supra n. 20, at 580–4. 27 Prosecutor v. Kayishema and Ruzindana, Case No. ICTR-95-1-T (21 May 1999) at paras 217–23; Delalic, supra n. 6, at paras 364–78.

Command Responsibility and Respondeat Superior 447 informed shortly thereafter.28 If actual knowledge cannot be shown, it is necessary to prove that the military commander “should have known” or the civilian superior “consciously disregarded information which clearly indicated” crimes.29 Such proof is often provided through circumstantial evidence such as the fact that the commander had executive authority over a territory in which war crimes were widespread.30 Also, if there is evidence that reports of crimes were made to the commander’s headquarters, he or she may be presumed to know the contents of these reports.31 This evidence may be rebutted by showing that the commander was absent from his or her command at the time of the offence or reports of the offence, or was ill. However, this rebuttal is only temporary because, upon resumption of command, knowledge will again be presumed.32 Whether the knowledge is imputed or actual, there are three levels of awareness in relation to ongoing or future violations. The first is that the commander discovers that a specific crime, particularised by time, place, perpetrator and type is happening or planned by subordinates.33 The second type of knowledge is that the commander learns that a subordinate group or unit is engaged in a criminal policy or organised routine, for example, by killing captured prisoners of war.34 Third, the commander becomes aware that the subordinate’s crimes are likely to occur in the future, which creates a risk of future war crimes.35 Thus, the commander does not always have to know of the exact crime committed by the subordinate before command responsibility arises. Evidence of knowledge may be found by examining details such as: the rank, position, power and responsibilities of the commander; reporting systems of the commander’s chain of command; the physical proximity of the commander to the location of the subordinate’s crime; the training, age and experience of the subordinates; the size and training of the commander’s staff; and the particular context in which the offence was committed.36 Parks notes that these are necessarily subjective criteria, as the circumstances of command are too great to be considered in one objective test.37 Others have added additional indicia of a 28 Pre-violation knowledge goes to duties to prevent and repress while post-violation knowledge goes to duties to investigate, punish and prevent recurrence. 29 This language is taken from article 28 of the Rome Statute of the International Criminal Court, supra n. 20. 30 The mens rea standard is set out in the Rome Statute of the International Criminal Court, supra n. 20, at art. 28, which is generally regarded as codifying customary law in this regard. In the past, however, the exact nature of the mens rea standard for command responsibility was the subject of much debate—with arguments ranging from “should have known”, “must have known”, “had reason to know” to “had information which should have enabled them to conclude in the circumstances”: Parks, supra n. 8, at 95. See also W H Parks, “A Few Tools in the Prosecution of War Crimes”, (1995) 149 Mil L Rev 73 at 75; Burnett, supra n. 8, at 132; Statute of the International Criminal Tribunal for the Former Yugoslavia, supra n. 4, at art. 7; and Delalic, supra n. 6, at paras 379–93. 31 Hostage case, supra n. 12, at 1271–2; Burnett, supra n. 8, at 132–3. 32 Burnett, supra n. 8, at 133; Parks, supra n. 8, at 103. 33 Burnett, supra n. 8, at 133. 34 Ibid. 35 Ibid. 36 Parks, supra n. 8, at 90–5. 37 Ibid. at 90.

448 V Oosterveld and A Flah commander’s knowledge to this list, including: the number of illegal acts; the type of illegal acts; the scope of illegal acts; the time during which the illegal acts occurred; the number and type of troops involved; the logistics involved, if any; the geographical location of the acts; the widespread occurrence of the acts; the tactical tempo of operations; and the modus operandi of similar illegal acts.38 An examination of these knowledge elements will assist in judging how far up the chain of command the liability may reach for a particular incident. There is a “sliding probability ratio” whereby the more extensive the activity involved in the offence and/or size of the unit involved, the higher in the chain of command may knowledge be subjectively imputed.39 The fourth element of command responsibility is that of actus reus: the failure to prevent, investigate, report or punish a subordinate’s illegal act. The omission may be either intentional or through neglect of duty. This element raises the question of what the commander is expected to do once he becomes aware, or must have become aware, of the crimes committed by the subordinate. The customary international law standard is that the commander’s duty is “simply to take such measures as are within his power and appropriate to the circumstances.”40 These measures may include: issuing orders to cease such behaviour or ensuring that previous orders are reiterated and enforced; investigating reports of criminal activity; punishing the perpetrators; reporting the act to superiors; instituting more training for subordinates on the laws of war; or, if the commander is severely limited in his command authority or has exhausted other remedies, resigning or otherwise strongly indicating his disagreement with the activity and actively working to stop the illegal behaviour.

Command Responsibility in US Alien Tort Cases In the United States, domestic legislation allows tort actions to be brought by foreign nationals against foreign nationals for violations of certain human rights. Under the aegis of the Alien Tort Claims Act (ATCA), many cases have been brought against commanders or superiors for failing to stop their subordinates from torturing the complainant.41 The most significant case involving command responsibility for torture is that of Xuncax v. Gramajo.42 Gramajo was Guatemala’s Vice Chief of Staff and director of the Army General Staff 38 Second Interim Report of the Commission of Experts Established Pursuant to Security Council Resolution 780 (1992), 26 January 1993, submitted to the Security Council by Secretary-General on 5 October 1993 (S/26545) at para. 58. Cited in Delalic, supra n. 6 at para 386. 39 Parks, supra n. 8, at 92. 40 Burnett, supra n. 8, at 134. 41 E.g., Forti v. Suarez-Mason 672 F.Supp. 1531 (1987) and 674 F.Supp. 707 (1988) (ND California); Xuncax et al. v. Gramajo, 886 F.Supp. 162 (1995) (D Mass) at 172.; Kadic et al. v. Karadzic, 70 F.3d 232 (2d Cir 1995); Mushikiwabo v. Barayagwiza 1996 WL 164496 (SDNY 1996) and Abebe-Jiri v. Negewo 72 F.3d 844 (11th Cir 1996). Of these cases, Xuncax specifically relied on the command responsibility doctrine by name. 42 Xuncax v. Gramajo, supra n. 41.

Command Responsibility and Respondeat Superior 449 from March 1982–93, commander in 1982 of the military zone in which the plaintiffs resided, and Minister of Defence from 1987–90. The plaintiffs alleged that he “was personally responsible for ordering and directing the implementation of the programme of persecution and oppression that resulted in the terrors visited upon the plaintiffs and their families”.43 As phrased, the claim seemed primarily to raise allegations of direct individual responsibility, but the nature of the violations were such that specific orders and directions were either unlikely or impossible to prove, making command responsibility relevant. One plaintiff, Teresa Xuncax, claimed damages for the execution of her husband, who was stripped, bound and masked and forced to walk naked around his village by forces under Gramajo’s command before being killed.44 Another plaintiff, Dianna Ortiz, a nun and a citizen of the United States, claimed damages for her kidnapping, torture and multiple rapes by personnel under Gramajo’s command. Rather than requiring proof of the alleged orders and directions, the Court instead applied the doctrine of command responsibility, citing Yamashita.45 The Court found that “Gramajo was aware of and supported widespread acts of brutality committed by personnel under his command resulting in thousands of civilian deaths” and that he “refused to act to prevent these atrocities”.46 There was evidence that Gramajo strictly controlled his military subordinates.47 As the plaintiffs often relied on multiple levels of hearsay to demonstrate Gramajo’s responsibility, the Court used the statements only to show that “Gramajo had reason to know of widespread atrocities and, therefore, [had] to take steps to end them.”48 The Court was satisfied that all aspects of the command responsibility doctrine were satisfied: the atrocities alleged took place; Gramajo was a commander in full command and control of his personnel; he knew of the crimes committed by his subordinates and supported such crimes; and he took no action to stop the brutality. The Court found Gramajo liable in tort under the command responsibility doctrine for, among other crimes, the torture of Xuncax’s husband and Ortiz.49 Xuncax v. Gramajo clearly illustrates that the tort need not have been committed directly by the defendant, and that the defendant is liable under the doctrine of command responsibility. The Court in Xuncax accepted that a leader may be responsible under the doctrine if “he knew or should have known that [a pattern of war crimes was] going on but failed to prevent or punish them”.50 In such circumstances, a leader is responsible when he is aware of and fails to seek to prevent widespread acts of brutality committed by personnel under his 43 44 45 46 47 48 49 50

Xuncax v. Gramajo, supra n. 41 at 171. Ibid. at 169–70. Ibid. at 171–2. Ibid. at 172–3. Ibid. at 173, 174. Ibid. at 173, n. 4. Ibid. at 176, 198. Ibid. at 172.

450 V Oosterveld and A Flah command.51 While the Xuncax case does set out a “should have known” test for the mens rea element, the Court did not discuss the various elements of command responsibility in much detail. This leaves much room for development in future decisions on the issue.

3 RESPONDEAT SUPERIOR 52

US Law on Agency Relationships and Respondeat Superior Another area of derivative liability distinct from command responsibility is found within domestic common law, notably as developed in the United States. The law of agency has long been employed as an instrument to hold employers vicariously liable for the torts of their employees through the doctrines of respondeat superior and a particular version thereof called “enterprise liability”.53 In the United States, the American Law Institute’s Restatement (Second) of Agency describes this form of liability as follows: “A master or other principal may be liable to another whose interests have been invaded by the tortious conduct of a servant or other agent, although the principal does not personally violate a duty to such other or authorise the conduct of the agent causing the invasion.”54

The common law tort doctrine of respondeat superior, developed through socalled “master and servant”—employer–employee—cases, is based on the premise that “[a] master is subject to liability for torts of his servants committed while acting in the scope of their employment.”55 It should be noted that agency-based respondeat superior can be applied to employer-defendants who are natural persons56 although it has been most commonly applied where the defendant is some form of associational actor such as a corporation or municipality. To the extent the doctrine can apply to natural persons, it provides a close parallel to command responsibility: in both cases, the presence of a legal actor, such as a corporation, for whom the human employer works does not 51

Xuncax v. Gramajo, supra n. 41 at 172. This chapter does not discuss the concept of directors’ liability through the responsible corporate officer (“RCO”) doctrine in light of the fact that, while it holds individual superiors liable, it is a criminal liability doctrine and thus not applicable to a tort liability analysis for human rights violations. 53 Enterprise liability will be discussed in the next subsection. 54 Restatement (Second) of Agency (1958) [hereinafter Restatement (Second)] §216. 55 Ibid. at §219. 56 The respondeat superior doctrine has been applied in the United States under section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. s. 78j(b) (1994) which states that it is unlawful for “any person”, directly or indirectly, by the use of any means to employ any manipulative or deceptive device or contrivance in contravention of the rules and regulations of the Securities Exchange Commission, where “person” means a natural person, company, government, or political subdivision, agency, or instrumentality of a government. 52

Command Responsibility and Respondeat Superior 451 preclude the human actor from being the one found liable for acts of a person subordinate to both the human and the legal actors. A number of factors are used to determine whether an employee’s actions fall within the scope of his or her employment, such as: whether the employee’s actions were taken to further the employer’s interests; whether the employer could have foreseen the employee’s wrongful acts; when and where these wrongful acts occurred; and the employee’s authority over the tort victim.57 Sections 257 to 261 of the Restatement further expand on the liability of employers for the acts of their employees in situations where the employee acts under apparent authority. Section 257 subjects a principal to liability “for loss caused to another by the other’s reliance upon a tortious representation of a servant or other agent, if the representation is (a) authorised; (b) apparently authorised; or (c) within the power of the agent to make for the principal.”58 In turn, section 261 imposes vicarious liability on a principal “who puts a servant or other agent in a position which enables the agent, while apparently acting within his authority, to commit fraud upon third persons . . .”59 The respondeat superior doctrine has been developed to some extent through United States securities cases. The general trend in American courts has been to resort to respondeat superior to prevent controlling persons from relying on the statutory defences found in securities statutes. As such, the court in SEC v. Management Dynamics, Inc., restricted the application of the statutory defences to those situations in which an action lies outside the scope of agency.60 The Fifth Circuit in Paul F. Newton & Co. v. Texas Commerce Bank further clarified that restriction of the application of vicarious liability in the area of securities law would “contradict the pervasive application of agency principles in nearly all other areas of the law”.61 In tort law, the United States Supreme Court has found principals strictly liable for intentional torts committed by their agents in a number of human rights contexts. For example, in General Building Contractors Association v. Pennsylvania, the Court found the defendants vicariously liable for intentional discrimination in violation of the Civil Rights Act of 1964 on a respondeat superior basis, in spite of the absence of bad intent on the part of the principals.62 Similarly, in Meritor Savings Bank v. Vinson, the Court urged the resort to respondeat superior in establishing whether sexual harassment had occurred contrary to the Civil Rights Act.63

57 See R E Lutner, “Employer Liability for Sexual Harassment: The Morass of Agency Principles and Respondeat Superior”, (1993) 3 U of Ill L Rev 589. 58 Restatement (Second), supra n. 54, at §257. 59 Ibid. at §261. 60 515 F. 2d 1277 (2d. Cir. 1973). 61 630 D. 2d 1111 at 1118 (5th Cir. 1980). 62 458 U.S. 375 (1982). Civil Rights Act, 42 U.S.C. §1981 (1994) 63 477 U.S. 57 (1986).

452 V Oosterveld and A Flah Beyond Agency to Enterprise Liability The California Supreme Court expanded on the doctrine of respondeat superior in the more recent case of Mary M. v. City of Los Angeles. The police officer stopped Mary M. as a result of her careless driving and her sobriety test indicated that she had been drinking. As a result of her pleas not to send her to jail, the officer, who was driving a marked police car and was wearing his uniform, with badge and gun, drove the complainant to her home in his car. He entered her home and, after telling her that he expected payback for his not having sent her to jail, he proceeded to rape her. In addition to the police officer’s criminal conviction for his conduct, the complainant brought a civil suit against the city.64 The California Supreme Court held the defendant city vicariously liable for a sexual assault committed against the complainant by an individual police officer.65 It reasoned that its decision to impose vicarious liability did not depend on either a finding of fault by the city or on whether the latter could have taken further precautions. Instead, the court based its judgment on the premise that it would be “unjust for an enterprise to disclaim responsibility for injuries occurring in the course of its characteristic activities”.66 In dismissing the city’s argument that vicarious liability should not be imposed because the police officer was not acting on behalf of the city, but was instead pursuing his own selfinterest, the court stated that employers may be held liable even if the employee’s conduct is prohibited by the employer, is intentional or malicious, and does not benefit the employer.67 In contrast to command responsibility, it can be seen that a standard of strict liability was thus involved. With the California case of Liu v. Republic of China, we move to a tort case very close in its issues to those arising in foreign tort cases.68 The issue was the vicarious liability of a foreign state for a wrongful death in the United States. A member of the government of the Republic of China (Taiwan) had ordered a number of individuals to murder Liu in California.69 The Ninth Circuit Court of Appeals first ruled that it had jurisdiction over the matter under the Foreign Sovereign Immunities Act of 1976 (FSIA) as a result of the noncommercial tort exception to immunity under section 1605(a)(5) thereto.70 On the substance of 64 For a case comment see “Respondeat Superior—Vicarious Liability—California Supreme Court holds police department vicariously liable for rape committed by on-duty police officer”, (1992) 105 Harv L Rev 947. 65 814 P.2d 1341 (1991). 66 Ibid. at 1343. 67 Case comment, supra n. 64, at 948. 68 642 F. Supp. 297 (ND Cal 1986), unreported order, 1987 WL 49413 (ND Cal Aug. 27, 1987), rev’d, 892 F. 2d 1419 (9th Cir 1989), cert. dismissed, 111 S. Ct. 27 (1990). 69 For detailed facts of the case, see Brent E. Christopher, “State Responsibility and Noncommercial Torts Under the Foreign Sovereign Immunities Act: Searching for the Applicable Respondeat Superior Doctrine through Liu v. Republic of China”, (1992) 27 Texas Int’l L J 137 at 143–4. 70 28 U.S.C. §§1330, 1332(a)(2)–(4), 1391(f), 1441(d), 1602–11 (1988). Section 1605(a)(5) of the FSIA provides as follows: cont.

Command Responsibility and Respondeat Superior 453 the claim, the Court then held the government of the Republic of China vicariously liable for the wrongful death of Liu based on the orders to kill having been perpetrated by an official of that government within the scope of his employment. Here, the Court relied on the concept of “enterprise liability”. Similar to the theory employed in Mary M., the doctrine of “enterprise liability” consists of a variation of respondeat superior which broadens the “scope of employment” test in two ways: (1) the liability of a superior will not only extend to the actual or possible control over his employees, but also to the risks inherent in or created by the enterprise;71 and (2) the employee’s conduct need not be for the benefit of the superior in order for the latter to be found liable.72 In Liu, the Ninth Circuit stated that the application of this broader version of the respondeat superior doctrine in California furthers three policy rationales: (1) an employer is better suited to spread the risks of its enterprise, even when done through public funds; (2) accident prevention; and (3) certainty of compensation.73 It is important to note at this point that the actor found to bear derivative liability involved through the doctrine of “enterprise liability” will be, as the label suggests, an enterprise such as a corporation or a unit of government. Neither of the two California cases just canvassed dealt with holding corporate or governmental officials responsible. For example, it was not Liu’s superior in Taiwan’s security services or the overseeing Cabinet minister who was sued, but Taiwan itself. The reason liability could be found against a foreign state in Liu was of course because a significant element of the tort had occurred in the US, to which acts state immunity did not apply under the FSIA. This same result of non-immunity would have also occurred under Canada’s and the UK’s state immunity legislation, each of which also have domestic tort exceptions.74 Should the law on state immunity develop to the point that a state can no longer claim immunity before foreign courts when sued for certain kinds of human rights harms wherever they occur, then Liu will provide a precedent for derivative liability of the state. Liu as a domestic tort law case creates an important link between private law liability and the basic principle of state responsibility

“A foreign state shall not be immune from the jurisdiction of courts of the United States . . . in any case . . . in which money damages are sought against a foreign state for personal injury or death, or damage to or loss of property, occurring in the United States and caused by the tortious act or omission of that foreign state or of any official or employee of that foreign state while acting within the scope of his or her office or employment.” 71 72 73 74

Liu, supra n. 68, at 1427. Ibid. at 1428. Ibid. at 1427–30. For example, the Canadian State Immunity Act, c. S-18 states in section 6 as follows:

“A foreign state is not immune from the jurisdiction of a court in any proceedings that relate to (a) any death or personal injury, or (b) any damage to or loss of property that occurs in Canada.”

454 V Oosterveld and A Flah that a state is responsible, for the most part, for harm caused to other states by acts of its officials even when those acts are ultra vires the official’s powers.75

Recent Developments in Canada: The Enterprise Risk Doctrine The Supreme Court of Canada recently adopted a doctrine that resembles the “enterprise liability” doctrine applied by the California courts in Mary M. and Liu, both in the tests it applies and the policy rationales on which it is grounded. In the companion cases of Bazley v. Curry (Bazley) and Jacobi v. Griffiths (Jacobi), the Supreme Court elaborated and applied what may be termed an “enterprise risk” doctrine of employer’s vicarious liability.76 The facts of these cases are similar. In both Bazley and Jacobi, the defendants were non-profit organisations devoted to children’s programmes. The Children’s Foundation in Bazley operated two residential homes for children with behavioural disorders, where it employed childcare counsellors that acted as substitute parents for the children. As such, the Foundation charged its counsellors with caring for the children “physically, mentally and emotionally.” Further , “[t]he employees were to do everything a parent would do, from general supervision to intimate duties like bathing and tucking in at bedtime.”77 As for the Vernon Boys’ and Girls’ Club in Jacobi, it operated a drop-in centre that conducted children’s recreational activities and relied on volunteers for their personnel. In both cases, employees of the organisations were accused of sexually assaulting children under their care. Despite these employees having been discharged by their employers, the complainants brought civil actions for damages against the employers for the intentional tort committed by the employees. While a unanimous Court found the defendant vicariously liable in Bazley, a 4–3 majority of the same Court distinguished Bazley from Jacobi on their facts (the different roles played by the employees in each case) in order to find in favour of the defendant.78 The court began from the traditional common law “scope of employment” test for the vicarious liability of employers. This test is defined in Salmond and Heuston on the Law of Torts as follows: 75 See International Law Commission, Draft Articles on State Responsibility, Yearbook of the International Law Commission at its Fifty-first Session, 1999, Chapter II, Part One, Article 10— Attribution to the State of the conduct of organs acting outside their competence or contrary to instructions concerning their activity:

“The conduct of an organ of a State, of a territorial governmental entity or of an entity empowered to exercise elements of the governmental authority, such organ having acted in that capacity, shall be considered as an act of the State under international law even if, in the particular case, the organ exceeded its competence according to internal law or contravened instructions concerning its activity.” 76

Bazley v. Curry [1999] 2 S.C.R. 534; Jacobi v. Griffiths [1999] 2 S.C.R. 570. Bazley, supra n. 76, at para. 2. 78 Since the test was initially developed by McLachlin J. in Bazley the analysis will cite that case for reference. 77

Command Responsibility and Respondeat Superior 455 “A master is not responsible for a wrongful act done by his servant unless it is done in the course of employment. It is deemed to be done if it is either: (i) a wrongful act authorised by the master, or (ii) a wrongful and unauthorized mode of doing some act authorised by the master.”79

McLachlin J. described the second branch of the preceding test as referring to “unauthorised acts so connected with authorized acts that they may be regarded as modes (albeit improper modes) of doing an authorised act”.80 In the case of an intentional tort as opposed to a case of negligence, this test did not, however, prove useful for distinguishing between an employee’s wrongdoing that is “an unauthorised ‘mode’ of performing an authorised act that attracts liability, and an entirely independent ‘act’ that does not”.81 As a result, the Supreme Court in Bazley developed a new test of employer’s liability for an employee’s intentional wrongful conduct in which the following analytical steps are set out: “First, a court should determine whether there are precedents which unambiguously determine on which side of the line between vicarious liability and no liability the case falls. If prior cases do not clearly suggest a solution, the next step is to determine whether vicarious liability should be imposed in light of the broader policy rationales behind strict liability.”82

McLachlin J. found that there were no useful precedents in Canada for cases of intentional misconduct that would satisfy the first step of the test. She found that the Bazley facts could be grouped in neither of the following three existing categories of cases: those based on the rationale of “furtherance of the employer’s aims”; those based on the employer’s creation of a situation of friction; and the dishonest employee cases.83 However, McLachlin J. found that the common theme underlying the preceding three categories that had driven previous courts to find employers vicariously liable was that the employee’s conduct [was] “closely tied to a risk that the employer’s enterprise has placed in the community . . .”.84 Since the Court was unable to rely on precedent, McLachlin J. proceeded to consider the policy reasons behind the imposition of vicarious liability against the background of this common theme. Similarly to the Ninth Circuit’s decision in Liu, the first policy rationale considered by the Court in Bazley was that of effective compensation for the victim. To this end, McLachlin J. reasoned that it was fair to hold an employer who employs others to advance its own economic interests responsible for the consequent risks it imposes on the community.85 Second, the Court in Bazley held that the imposition of vicarious liability contributes to the goal of deterrence of 79 R F V Heuston and R A Buckley, Salmond and Heuston on the Law of Torts (London, Sweet & Maxwell, 1987) at 521. 80 Bazley, supra n. 76, at para. 7. 81 Ibid. at para. 11. 82 Ibid. at para. 15. 83 Ibid. at para. 17. 84 Ibid. at para. 22. 85 Ibid. at paras 30–1.

456 V Oosterveld and A Flah future harm, since employers are in a better position to reduce accidents and intentional wrongs through efficient organisation and supervision, even when the employer is not guilty of negligence in a particular case.86 The assessment of the preceding policy rationales are then filtered through a consideration of the extent to which the employer has materially enhanced the risk of the employee’s wrongdoing. Accordingly, McLachlin J. stated the test in terms of community risk, thereby uniting the “common theme” of the existing cases with a limiting principle for the policy rationales favouring liability: “Where the risk is closely associated with the wrong that occurred, it seems just that the entity that engages in the enterprise (and in many cases profits from it) should internalize the full cost of operation, including potential torts. . . . On the other hand, when the wrongful act lacks meaningful connection to the enterprise, liability ceases to flow . . .”87

In determining the extent of the connection between the risk created or enhanced by the employer and the employee’s intentional tort, McLachlin J. set out a number of factors to be considered by the courts. Her non-exhaustive list included the following: the opportunity that the enterprise afforded to the employee to abuse his or her power; the extent to which the wrongful act may have furthered the employer’s aims (and hence be more likely to have been committed by the employee); the extent to which the wrongful act was related to friction, confrontation or intimacy inherent in the employer’s enterprise; the extent of power conferred on the employee in relation to the victim; and the vulnerability of potential victims to wrongful exercise of the employee’s power.88 McLachlin J. considered the facts in Bazley in light of the above factors and held that the Children’s Foundation was vicariously liable for its employee’s sexual assault of one of the children under his care. Despite the Foundation’s non-profit status and its provision of benefits to the community, it had still placed the complainant in a situation that materially enhanced the risk of abuse. As such, the Court rejected the call to carve an exception from vicarious liability for non-profit organizations by stating the following: “The suggestion that the victim must remain remediless for the greater good smacks of crass and unsubstantiated utilitarianism. Indeed, it is far from clear to me that the ‘net’ good produced by non-profit institutions justifies the price placed on the individual victim, nor that this is a fair way for society to order its resources.”89

86 87 88 89

Ibid. at para. 32. Ibid. at para. 38. Ibid. at para. 41. Ibid. at para. 54.

Command Responsibility and Respondeat Superior 457 Summary The common law of agency holds masters responsible for the wrongful acts of their agents in spite of the masters’ absence of fault. Furthermore, the doctrines of respondeat superior and “enterprise liability”, both derived from the common law of agency, provide useful tools to impose vicarious liability on superiors—whether employers, controlling persons, organisations, or governments—for the intentional wrongful acts of those considered to be their agents, regardless of whether the superior in question prohibited the wrongful conduct. Courts will not be concerned with whether the conduct in fact benefited the superior. Under the “enterprise risk” doctrine developed in Canada, the notion of connection to a risk introduced by an enterprise into a community provides a principled way to think about the possible relationship between respondeat superior and tort liability of foreign officials for the human rights abuses of their subordinates. However, in view of the tendency of respondeat superior to be invoked via-à-vis enterprise defendants including units of government, bolstering principles will be needed in order to justify finding not only foreign states liable for acts of subordinates, but also foreign officials, including in situations which would not be sufficient to trigger derivative liability based on an extension of command responsibility to the civil liability context.

4 INTEGRATING THE COMMAND RESPONSIBILITY AND RESPONDEAT SUPERIOR FRAMEWORKS IN CANADIAN TORT LAW

Introduction: Sudan as Case Study The doctrines of command responsibility and respondeat superior can provide important frameworks for analysis in tort cases based on human rights harms committed abroad. To demonstrate the applicability of these approaches, we shall focus on a hypothetical example from Sudan. The human rights situation in Sudan is extremely grave. While the UN Human Rights Commission has recently welcomed the “expressed commitment of the Government of Sudan to respect and promote human rights and the rule of law”, it has also expressed its deep concern at a long list of human rights violations, from the large numbers of internally-displaced people caused by a lengthy civil war to the use of child soldiers, the abduction of women and children, forced labour, the use of weapons such as landmines against the civilian population, and severe restrictions on the freedom of religion and peaceful assembly.90 Most notably for present purposes, the Commission also decries the “widespread occurrence of torture . . . in particular of political opponents”.91 90 “Situation on Human Rights in the Sudan”, Commission on Human Rights Res. 1999/15, UN Doc. E/CN.4/RES/1999/15, 23 April 1999. 91 Ibid.

458 V Oosterveld and A Flah This latter concern has arisen from numerous reports from human rights organisations that torture of detainees suspected of plotting against the government by government security officials and police is common.92 Often, suspected political opponents are held incommunicado, either in prison or in secret detention centres notorious for torture, known as “ghost houses”.93 Various means of torture are used in these facilities, including beatings, burning, forcing detainees to stand for long periods in the sun, dousing detainees with cold water and locking detainees in a deep freeze.94 The Special Rapporteur on the Situation of Human Rights in Sudan indicated that he has become convinced that torture is practised widely by the security forces in most cases of detention.95 These practices led the Commission on Human Rights to urge the government of Sudan to “release all remaining political detainees, cease all acts of torture and cruel, inhuman or degrading treatment, close down all clandestine or unacknowledged detention centres and ensure that all accused persons are held in ordinary police or prison custody where family members and lawyers can visit them.”96 Take the hypothetical example of a Sudanese anti-government activist who has been subjected to torture in a “ghost house” and survived, and who has fled to Canada. She finds that her torturer’s superior is also living in Canada and brings a claim in tort against him.97 Canadian courts can turn to the two frameworks of analysis presented in Sections 2 and 3 as relevant analogies for finding derivative liability of superiors within foreign state structures. It is important to note at the outset that the plaintiff in a tort action would have to overcome a less rigorous burden of proof based on a “balance of probabilities”, as opposed to the criminal standard of proof “beyond a reasonable doubt”.

The Convention against Torture Before considering each doctrine in turn, it is important to note that the persuasiveness of command responsibility and respondeat superior as analogies for the tort of torture will be enhanced to the extent that the evolving lex specialis of 92 Amnesty International, AI Report 1999: Sudan, available at www.amnest.org/ailib/aireport/ ar99/afr54.htm; Amnesty International, AI Report 1998: Sudan, available at www.amnesty. org/ailib/aireport/ar98/afr54.htm, and Amnesty International, AI Report 1997: Sudan, available at www.amnesty.org/ailib/aireport/ar97/afr54.htm. 93 AI Report 1998: Sudan, supra n. 92. 94 AI Report 1999: Sudan, supra n. 92 and AI Report 1998: Sudan, ibid. 95 United Nations Commission on Human Rights, “Report of the Special Rapporteur on the Question of the Violation of Human Rights and Fundamental Freedoms in any Part of the World, Situation of human rights in the Sudan”, E/CN.4/1999/38/Add.1 17 May 1999. 96 “Situation on Human Rights in the Sudan”, Commission on Human Rights Res. 1998/67, UN Doc. E/CN.4/RES/1998/67, 21 April 1998. 97 This chapter does not deal with another form of redress: the possibility of punishment through the application of the doctrine of command responsibility in criminal proceedings undertaken by the Crown in Canada.

Command Responsibility and Respondeat Superior 459 torture itself already recognizes the concept of indirect individual responsibility. Here, it would be necessary to determine how the Convention against Torture (the Convention) applies not just to the hands-on torturer but also to those higher up in the state structure. It is clear from article 1(1) of the Convention that “consent or acquiescence” by an official is enough to satisfy a kind of state conduct component of torture, but it is not specifically stated whether the official who acquiesces is also guilty of torture or whether this characterisation is reserved for the hands-on torturer. The Convention does not set out complicity in torture as a separate form of conduct. Therefore, the interpretive issue is whether such complicity (through consent or acquiescence) is to be styled as article 1(1) “torture” or instead as “other acts of cruel, inhuman or degrading treatment or punishment which do not amount to torture as defined in article 1” which, per article 16, are not subject to the same onerous treaty obligations as torture.98 Similarly, article 2(3) of the Convention makes clear that superior orders (“an order from a superior officer or a public authority”) are not a defence for the hands-on torturer, but, again, it is not specifically stated whether the issuer of the orders is also guilty of (and liable for) “torture.” It is not the purpose of this chapter to present an account of the jurisprudence and doctrinal commentaries which have grown up around the Convention, but it will be assumed for present purposes that the prevailing understanding is that both ordering torture and “consent or acquiescence” in torture are themselves sanctionable as torture.99 Of course, principled arguments by analogy from the doctrines of command responsibility and respondeat superior can themselves provide interpretive reasons for construing the Convention to include indirect individual accountability as falling within the definition of torture.

Sudanese Officials and Command Responsibility A court could examine the hypothetical Sudanese case in terms of the doctrine of command responsibility. The first element, proof of torture constituting a war crime, crime against humanity or crime of genocide was committed against the plaintiff, could be established through the victims’ own testimony and the use of maps, medical observations and documented reports from such authoritative sources as the United Nations and international human rights organisations. As described, incommunicado torture in Sudan is widespread and well-recognised, and therefore the combination of testimony and reports would likely allow the plaintiff to meet the first test. The second element of command responsibility is proving the link between the perpetrator of the crime and the commander, which entails both proof of command and proof of control. The question in our hypothetical case is 98

For example, the try-or-extradite provisions do not appear to apply, nor does article 14. See, for example, Prosecutor v. Furundzija, ICTY Case No. IT-95-17 (10 December 1998) at paras 250–7. 99

460 V Oosterveld and A Flah whether the torturer and his commander were from the Sudanese military, police or some other organisation. The answer will affect how the link between the perpetrator and the commander is proven. It may be easier to prove a command relationship in a military setting, where the hierarchical structure is often clearer, sometimes even documented in military manuals. In addition, the scope of responsibilities of the defendant superior can often be inferred from the position the person occupies in the military hierarchy. In the case of the police, the structure is often also clear and similar to a military hierarchy. However, in the case of “secret police” or militias, the command link is usually harder to prove. In these cases, lines of command are deliberately obfuscated due to the secret nature of the activities (in “ghost houses”, for example). In addition, the Sudanese government sometimes uses “proxies” to carry out human rights abuses, therefore making the link between perpetrator and commander even more difficult to demonstrate.100 Finally, a command relationship may be more difficult to establish in a case where the plaintiff seeks redress against a bureaucratic official. Although it may not be difficult to demonstrate that the perpetrator and the superior are linked in a government hierarchy and therefore prove that “command” linkage, it is often harder to demonstrate that the superior had any control over the perpetrator due to the less rigid nature of civilian hierarchy as opposed to military hierarchy. However, in this situation, the court could look to cases involving civilian superiors from the Nuremberg and Tokyo Tribunals, as well as the International Criminal Tribunals for the Former Yugoslavia and Rwanda, for guidance on the kinds of cases in which civilian command responsibility has been found.101 Evidence for all types of command and control may come from humanitarian aid workers, media, NGO representatives and United Nations organisations—in other words, those who are likely to come into contact with the Sudanese power structures in order to carry out their missions. The third element of command responsibility is proof of mens rea. Here, the issue of knowledge is central. In a tort action, a Canadian court would likely apply the mens rea standard for command responsibility found in Canadian criminal legislation, and which in turn is reproduced from the International Criminal Court Statute.102 While the command responsibility doctrine applies regardless of whether the superior is military or civilian, military commanders are held to a higher mens rea standard (“knew or should have known”) than are civilian superiors (who need to have “consciously disregarded information which clearly indicated” crimes). In order to prove the mens rea of the superior, 100 See, for example, John Harker, Human Security in the Sudan: The Report of a Canadian Assessment Mission (January 2000), available at www.dfait-maeci.gc.ca. Harker discusses the links between the Government of Sudan and the murahleen militia with respect to abduction from 31–43. 101 E.g. Kayishema, supra n. 27. 102 See sections 5 and 7 of the Canadian act, adopted on 29 June 2000: Act respecting genocide, crimes against humanity and war crimes and to implement the Rome Statute of the International Criminal Court, and to make consequential amendments to other Acts (short title: Crimes Against Humanity and War Crimes Act).

Command Responsibility and Respondeat Superior 461 the plaintiff could rely upon several sources: witness testimony of his presence at the scene of the torture, utterances made by him and reported by a witness or the press, the fact that certain documentation (whether military, press, United Nations or human rights reports) was submitted to him, or that family members or others brought the situation of the plaintiff—or of others in the plaintiff’s position—to the attention of the defendant. The post-World War II cases,103 as well as the civil cases brought in the United States under the ATCA make it clear that a court may rely on both direct and circumstantial evidence to establish that the defendant knew that the atrocities were taking place. In addition, the court may rely upon an objective mens rea standard whereby the plaintiff would be required to prove that the defendant has constructive knowledge because a reasonable military commander of the defendant’s rank ought to have known of the torture being committed by his subordinates. Once proof of the torture, the command link and the superior’s mens rea have been established, the element of actus reus becomes difficult to negate since it can be demonstrated not only by positive acts but also, and primarily, by omissions. Here, what is appropriate or feasible will depend on, and flow from, the nature and timing of the knowledge possessed by, or imputed to, the defendant.

Sudanese Officials and Respondeat Superior The foregoing schematic application of the doctrine of command responsibility shows that it can provide a step-by-step framework for proving the liability of a superior in tort, especially to the extent the international law surrounding torture has already come to embrace the notion of indirect individual responsibility. There may be cases, however, where the various elements of command responsibility may be too difficult to prove, including in cases involving paramilitaries. In these cases, the doctrine of respondeat superior, and its application as “enterprise liability” or “enterprise risk” doctrines, also constitute useful legal tools to provide redress in Canadian courts for Sudanese victims of torture. The advantage of seeking redress through respondeat superior and its variables is the fact that, once the requisite elements of the acts of torture are satisfied, strict liability follows, such that there is no need to prove either intention on the part of the principal defendant or the same degree of knowledge. Risk assessment replaces the kind of fault inquiry involved in command responsibility analysis. In order to find liability under the respondeat superior doctrine, the plaintiff is required to establish that the defendant master is subject to liability for torture committed by servants under his control while acting in the scope of their employment. In the case of the Sudanese defendant in our hypothetical 103 Yamashita, supra n. 9; High Command Trial, supra n. 7; Hostage case, supra n. 12; Admiral Toyoda, supra n. 13.

462 V Oosterveld and A Flah example, it would then be a question of fact whether subjecting Sudanese citizens who oppose the government’s policies to torture would result in the furtherance of the defendant official’s interest in eliminating the opposition and maintaining power in Sudan. Other factors would be less difficult to establish, namely: whether the torture was committed in the scope of the perpetrators’ authority, whether the employer could have foreseen the wrongful acts, when and where those wrongful acts occurred, and the perpetrators’ authority over the victim. Reports by the United Nations and human rights organisations could be used to argue that the foreseeability, timing and location of the torture would have been known to the principal; and the authority exercised by the perpetrators over the victims would be established by the fact that the former are members of government forces while the latter are not. In our example, the Canadian court may wish to rely, by analogy, on the Canadian doctrine of “enterprise risk” in order to find the Sudanese official vicariously liable for torture. Applying the test developed by the Supreme Court of Canada in Bazley, the first step would require the court to determine “whether there are precedents which unambiguously determine on which side of the line between vicarious liability and no liability the case falls”.104 It may be argued that this case could be categorised under the “furtherance of employer’s aims” rationale. If that argument failed, the next step would require the court to establish whether the broader policy rationales of strict liability determine that vicarious liability should be imposed. It is clear that a Canadian court’s finding of liability of a Sudanese official for torture perpetrated against a Sudanese citizen would further a policy of effective compensation. At the very least, such a finding would result in a greater possibility of compensation for a victim who would otherwise have no redress in Sudan as a result of his or her political views or ethnic background. With respect to the policy of deterring future harm, it would in fact be questionable whether finding the defendant liable would further that objective. A finding of liability would at the very least serve as a kind of a “warning” to other superiors in Sudan and would, at least at the margins, be a consideration especially if there are fears of regime overthrow and a resulting need to flee abroad. Finally, as part of the Bazley enterprise risk analysis, the court would be required to consider the extent to which the defendant Sudanese official has materially enhanced the risk of his subordinates’ wrongdoing. At this stage, the plaintiff would be in a position to argue that the “enterprise” of the Sudanese government—including its specific goal of eliminating its opposition—afforded Sudanese military and police officials as well as paramilitary groups the opportunity to abuse their power. Moreover, if the plaintiff could demonstrate that the actual perpetrator was a member of the government forces, carried a weapon and had access to official infrastructure, these factors would be taken 104

Bazley, supra n. 76.

Command Responsibility and Respondeat Superior 463 to signal a significant degree of power conferred on the perpetrator in relation to his victim. The plaintiff would therefore be able to rely on the test developed by the Supreme Court of Canada in Bazley to argue that a particular defendant Sudanese official is vicariously liable for the torture committed against her by subordinates of that official by virtue of his connection to that perpetrator— which connection would, of course, have to be shown. In interpreting the parameters of “enterprise risk”, the court would likely look to the US case law discussed above. The broader doctrine of “enterprise liability” employed in California courts can be referred to in order to reject two possible arguments that a Sudanese official may make in his defence: firstly, that torture is prohibited under Sudanese law, such that his subordinates were acting illegally; and secondly, that the conduct of his subordinates could not have advanced Sudan’s or the Sudanese official’s interest in view of the sanctions and condemnation that has been imposed on Sudan by the international community as a result of such human rights violations. If the Canadian court accepted the approach of Mary M. v. City of Los Angeles,105 a Sudanese official would likely be precluded from making the foregoing representations. In that case, the California Supreme Court held that the liability of an employer would not be precluded by the fact that his employees’ conduct was prohibited by the employer, nor by the fact that it did not actually end up benefiting the employer.

5 CONCLUSION

Both international and domestic laws provide for forms of derivative liability of individuals. International law has the doctrine of command responsibility, under which commanders or superiors can be held liable for genocide, crimes against humanity and war crimes, including torture, committed by their subordinates. Command responsibility has been applied in domestic tort law in the United States, and has been incorporated into domestic Canadian criminal law.106 The doctrine of command responsibility provides an important framework for those who seek to prove, in tort, that a superior failed to prevent, report or punish crimes by a subordinate. In cases where the tortious wrong does not amount to genocide, crimes against humanity or war crimes, the doctrine of respondeat superior and its application as “enterprise risk” or “enterprise liability” may provide guidance. Under this doctrine, employers are liable for harmful conduct of employees towards third parties, even though the employer did not personally breach a duty or authorise the employer to breach a duty towards the third party. Thus, respondeat superior provides for a form of strict liability that command responsibility does not.

105 106

Mary M., supra n. 65. Crimes Against Humanity Act, supra n. 102.

17

Responsibility and Liability for Violations of Human Rights in the Course of UN Field Operations CHANAKA WICKREMASINGHE 1 and GUGLIELMO VERDIRAME 2

1 INTRODUCTION

United Nations (UN) field operations have increasingly involved UN, or UN-authorized, personnel in the exercise of significant governmental powers. The motivation for such operations is usually humanitarian in the broad sense, being for the maintenance of peace and security, humanitarian relief or, on occasion, directly for the maintenance of human rights. As such, in undertaking these activities the UN will have a direct impact on the lives of members of the local populations. However the regime governing the legal responsibility of the UN in the exercise of such powers is under-developed. Solutions to problems have tended to be found on an ad hoc, rather than a systematic, basis, and answers to a number of key questions remain complex or unclear. This chapter therefore seeks to examine the broad structure of legal responsibility in relation to UN field operations, rather than going directly to the more limited issue of transnational tort liability against the UN, on which there is little practice and for which, as will become clear, the prospects for success are at present limited. In doing so, an article of this type cannot deal comprehensively with all aspects of a highly complex legal situation. Rather, it seeks to outline some fundamental issues which arise under current conditions in the hope that it will contribute to the development of a more thorough and rigorous approach in the future.

R

ECENT DEVELOPMENTS IN

1 2

University of Bristol Merton College, Oxford

466 C Wickremasinghe and G Verdirame

2 THE NATURE OF UN FIELD OPERATIONS

It is axiomatic that human rights law and humanitarian law apply principally to authorities or institutions which exercise governmental or quasi-governmental functions. The focus of the present review of UN operations will be on those activities in which the UN exercises such “public” functions. The field operations which are the subject of the present study may broadly be fitted into the five-part categorisation which follows. However it might be noted that these are not entirely watertight categories, for example some complex operations combine different elements which may belong to more than one of these categories, whilst others have evolved between these categories over time. It might also be noted that these categories do not constitute an exhaustive coverage of UN activities in which questions of human rights might arise.3 International Administration of Territory (de jure) The clearest example of the UN exercising a broad range of governmental powers is where it is authorised to take over the administration of an area of territory.4 Although there are a limited number of examples of UN administration of territory, there are indications that this kind of institutional arrangement may be pursued more often in the future.5 It is clearly a complex and sensitive task to take over the administration of territory and its population, which will in most cases have recently undergone, or still be faced with, severe social upheaval or oppressive humanitarian conditions. Whilst placing such administration in the hands of an international body may overcome some difficulties, it may also raise others. Almost inevitably there are likely to be complaints by persons who feel aggrieved by actions of such a body. It is therefore to be expected that allegations of unfair or high-handed treatment will from time to time be made against the UN in such a context. The legal basis of such an international administration may be contained in a treaty, as was done in the case of the first experiment of this type, namely the 3 For example, the UN’s internal relations with its employees may also raise important issues of the relationship of individuals to the Organisation. In this context it might be noted that there has been at least one case in which it was argued that an international organization had infringed the human rights of one its employees (Mendaro v. World Bank, 717 F.2d 610 (D.C. Cir.1983). For fuller discussion see Singer, though it might be noted that he accepts that the UN should, exceptionally, retain absolute immunity (M Singer, “Jurisdictional Immunities of International Organisations: Human Rights and Functional Necessity Concerns” (1995) 36 Virg. JIL 53 at 87). 4 On the possession of sufficient personality on the part of the UN to administer territory, there seems to be little doubt (see E Lauterpacht, “The Contemporary Practice of the United Kingdom in the Field of International Law. Survey and Comment”, (1956) 5 ICLQ 405 at 411). 5 At present the UN is involved in the temporary administration of Kosovo, established under Security Council Resolution 1244 (1999) for an initial period of 12 months. It is also involved in a similar operation in East Timor, following the adoption by the Security Council of Resolution 1272 (1999).

Violations of Human Rights during UN Field Operations 467 administrative and legislative jurisdiction which the League of Nations enjoyed over the Saarland under the Treaty of Versailles.6 Alternatively the UN Security Council may impose international administration on an area of territory by means of a compulsory resolution under Chapter VII of the UN Charter. Thus for example the Security Council has recently established an international administration for Kosovo under resolution 1244 (1999).7 This resolution provides for the “deployment of international civil and security presences” in Kosovo, requesting that “both presences operate towards the same goals and in a mutually supportive manner”.8 It is also envisaged that some of the responsibilities of the “international security presence” will be progressively transferred to the “international civil presence”. Both the military and civilian administrations have been established for an initial period of 12 months at the time of writing. The powers of the UN Mission in Kosovo (“UNMIK”) are very wide, comprising, as the Secretary General has indicated, “all legislative and executive powers, including the administration of the judiciary”.9 In order to exercise 6 The League administered the region until 1935, when its unification with Germany was approved by plebiscite. Interestingly for present purposes, the administration of the Saar by the League was severely criticised on grounds which would now be considered as pertaining to human rights. Thus for example, the League was accused of backing French aspirations in the region and the process of “Gallicisation” of the territory before the referendum on the final status by discriminating against the use of German language in schools and in the workplace (W H Dawson, The Saar Territory: Its History, Population, Industry and Government by the League of Nations (London, 1934) at 18–20). Other examples of international administration established by treaty include the Agreement between the Netherlands and Indonesia concerning West New Guinea (West Irian) (1962), reproduced in (1963) 57 AJIL 493. It was on the basis of this agreement that the UN Temporary Executive Authority (UNTEA) was established, on which see P W van der Veur, “The United Nations in West Irian: A Critique”, (1964) 18 Int’l Org. 53 at 67. 7 There are other examples of international administrations of territory in contemporary practice. The UN Mission in Bosnia-Herzegovina, established under the Dayton Peace Agreements of 14 December 1995 (35 ILM 75) has wide administrative powers. The UN Transitional Authority in Cambodia, regulated by the Agreement on a Comprehensive Political Settlement in Cambodia of 23 October 1991 (31 ILM 183), exercised administrative powers in conjunction with the host government and other local actors (S Ratner, “The Cambodia Settlement Agreements”, (1993) 87 AJIL 1). The Security Council has recently approved the establishment of a UN administration in East Timor—see SC Res. 1264 (1999) and 1272 (1999). In other situations—for example in Haiti, Angola, Namibia, and, at the time of writing, in Sierra Leone—the UN may exercise some administrative powers as part of a mandate to prepare for and monitor local elections, or as part of peace-keeping functions. 8 Under Security Council resolution 1244, the responsibilities of the international security presence include: the enforcement of a ceasefire and deterring new hostilities; the demilitarisation of the Kosovo Liberation Army (KLA); securing public order and safety so as to facilitate the work of the international civil presence and the provision of humanitarian assistance; the supervision of de-mining; and conducting border-monitoring. The international civil presence is responsible for “performing basic civilian administrative functions where and for as long as required”; promoting the establishment of self-government and autonomy also by transferring administrative functions to local institutions, as they are established; the maintenance of law and order through international police forces pending the establishment of a local police force; the protection and promotion of human rights; and the assurance of safe and unimpeded return of refugees and displaced persons to their homes. 9 Report of the Secretary General on the United Nations Interim Administration Mission in Kosovo, 12 July 1999 (UN Doc. S/1999/779). See also Sec. 1, 1, Regulation No. 1999/1 (on the authority of the interim administration in Kosovo): “All legislative and executive authority with

468 C Wickremasinghe and G Verdirame these powers, three main offices have been established within the administration, the Police Commissioner, an Office for Civil Affairs and an Office for Judicial Affairs.10 In these cases the very purpose of the international administration is the establishment of lasting peace and good order for the societies in question. Therefore, in seeking to establish conditions in which human rights are respected, the UN and its personnel must, in all consistency, themselves be governed by internationally accepted standards of human rights when dealing with the civilian populations under their charge.

The Exercise by the UN of de facto Administrative Functions In the second category of operations, the UN exercises de facto administrative functions notwithstanding the absence of any formal act explicitly conferring such powers and functions on it. Refugee camps in Africa and Asia, in which a large percentage of the world’s refugee population is concentrated, are prime examples. Refugee camps are often administered by the offices of the United Nations High Commissioner for Refugees (UNHCR). Although there is no single administrative structure that is reproduced in all camps, a common feature is that UNHCR, alongside the non-governmental organisations to which it subcontracts the delivery of particular services, exercises varying degrees of control over camps, which in some cases amounts effectively to governmental control. UNHCR often establishes its own system for the maintenance of order in its camps parallel to that of the host State. This system is then enforced by hiring private security companies and/or refugees recruited as security guards, or by resort to the local police force. The question of maintaining order in refugee camps is particularly sensitive. The vulnerability of refugee populations hardly needs emphasising, consisting of dispossessed and disempowered persons, who find themselves within the territory of a foreign State which is often indifferent to their plight or is simply preoccupied with the care of its own population. In such cases they are frequently in the position of having to rely entirely upon international agencies for the provision of food and services essential for physical survival. Though the responsibility of these agencies to discharge their functions in a manner conforming to international human rights standards must, of necessity, be largely self-enforced, this should not disguise the applicability of those standards as the proper yardstick against which the agencies’ performance should be measured. respect to Kosovo, including the administration of the judiciary, is vested in UNMIK and is exercised by the Special Representative of the Secretary General” (UN Doc. UNMIK/REG/1999/1). The UN Transitional Administration in East Timor (UNTAET) has a similarly extensive mandate. 10 Report of the Secretary-General pursuant to Paragraph 10 of Security Council Resolution 1244 (1999), 12 June 1999 (UN Doc. S/1999/672).

Violations of Human Rights during UN Field Operations 469 The occurrence of human rights violations in refugee camps has been reported by various academics and by human rights organisations.11 In some cases, these violations are not directly related to acts of the administering agencies. However even in these cases it may be that, as part of the positive obligation to ensure human rights, the administering agencies have a duty to intervene, for example in order to prevent community leaders in the camps from violating the rights of other refugees. In other cases, however, it has been suggested that the administering agencies, i.e. both UNHCR itself and humanitarian nongovernmental organisations acting as agents for UNHCR, may be directly responsible for acts which violate the human rights of refugees.12

Military Operations under Chapter VII of the UN Charter In addition to these two forms of administration of territory, the UN is inevitably brought into direct contact with local civilian populations in military operations under Chapter VII of the Charter. Since the end of the Cold War Chapter VII operations have been mounted in a number of circumstances, but many of the more recent Chapter VII authorisations have been made to ensure the delivery of humanitarian relief. The legal structure of Chapter VII operations is complex and varied. Their distinguishing feature is that they are authorised by the Security Council using its compulsory powers of decision, in response to a breach of, or threat to, international peace and security, thus obviating the need for the consent of the State in whose territory the operations are to take place. As is well known, the United Nations does not have troops of its own at its disposal for Chapter VII operations (as had originally been envisaged under the Charter). Therefore, in some Chapter VII operations the Security Council has had to resort to direct delegation of enforcement powers to co-operating States or international organisations.13 In these cases the UN has in practice retained little control over the command and control of the operation. In other cases, a UN force may be established as an ad hoc subsidiary organ of the UN. Member States contribute troops to such an operation on an entirely voluntary basis. Though in the early years this type of structure was established for peacekeeping by consent (see next section), recent years have seen Chapter 11 African Rights, “The Nightmare Continues . . . Abuses Against Somali Refugees in Kenya” (1993); B E Harrell-Bond, Imposing Aid (Oxford, OUP, 1986) at 155 ff.; Human Rights Watch/Africa Watch, “Seeking Refuge, Finding Terror—The Widespread Rape of Somali Women Refugees in North Eastern Kenya” (1993); Lawyers Committee for Human Rights, African Exodus (1995); G Verdirame, “Human Rights and Refugees: The Case of Kenya”, (1999) 12 Journal of Refugee Studies 54 at 61–8; R Wilde, “Quis Custodiet Ipsos Custodes? Why and How UNHCR Governance of ‘Development’ Camps Should be Subject to International Human Rights Law”, (1998) 1 Yale Human Rights and Development Law Journal 5. 12 See Verdirame supra n. 11, and Wilde, supra n. 11. 13 See D Sarooshi, The UN and the Development of Collective Security (Oxford, Oxford UP, 1998).

470 C Wickremasinghe and G Verdirame VII powers added to the mandate of such UN forces. In these cases the UN—the Secretary-General, at the executive level and the Security Council, at the policy level—retains greater control over the operation. In this respect the SecretaryGeneral has identified “three levels of authority” within the UN in the course of military operations. First of all, the Security Council is responsible for the “overall political direction”; secondly, the Secretary-General is responsible for “executive direction and command”; and, finally, “command in the field . . . is entrusted by the Secretary-General to the chief of mission (special representative or force commander/chief military observer)”.14 Nevertheless, as will be explained, each national contingent remains subject to its own national disciplinary system. The threshold for the applicability of Chapter VII enforcement powers is that the situations should constitute a breach of, or threat to, international peace and security. Thus such operations will usually take place against the backdrop of an armed conflict in which members of a UN force may find themselves directly involved. In such circumstances important questions as to the applicability of international humanitarian law have arisen, as will be discussed in later subsections.

“Classic” Peacekeeping Operations A fourth category of operations is peacekeeping. As mentioned above, the “classic” form of peacekeeping by consent was developed by the UN during the Cold War period, in which stalemate in the Security Council meant that recourse to Chapter VII powers was extremely rare. The requirement of the consent of the host State to the presence of UN troops on its territory was considered a prerequisite, and in most cases a Status of Forces Agreement (SOFA) or an equivalent document was negotiated to this effect. In this kind of operation States also contribute troops on a voluntary basis, and will enter into agreements with the United Nations as to the terms on which their troops will serve. In this “classic” consensual form of peacekeeping, troops are intended to act as a barrier interposed between previously warring factions, and their task is to ensure that the latter maintain the terms of an agreed peace, or at least a cessation of hostilities. As such, peacekeeping troops are generally lightly armed, and only authorised to use force in self-defence. However, peacekeeping forces often have a significant impact on the local population and on local political and economic systems, the more so as a number of such operations have involved the deployment of troops over many years.15 14 Secretary-General, Supplement to an Agenda for Peace (1995), UN Doc. A/50/60, S/1995/1, at para. 38, where it is stressed that “(t)he distinctions between these three levels must be kept constantly in mind in order to avoid any confusion of functions and responsibilities”. 15 The impact of UN Peacekeeping Operations will obviously depend on a number of factors, including the mandate of the Operation, its scale, the modus operandi at field level and power

Violations of Human Rights during UN Field Operations 471 Other Humanitarian Operations of the UN Finally, it should be noted that, in other of its operations, the UN by itself, or in co-operation with State or other agencies, seeks to effect humanitarian objectives, is brought directly into contact with the local civilian population, and must therefore be sensitive to the potentiality of human rights claims made against it.16 Of particular importance within this residual category of UN operations is the provision of humanitarian assistance by UN agencies in disparate situations, and “post-conflict peace-building”.17 Whilst it is essential that the UN should continue its vital humanitarian work in situations throughout the world without being unduly impeded, it is equally important that the principles on which it provides such assistance are clear and adhered to, in the course of such operations.

3 THE INTERNATIONAL RESPONSIBILITY OF THE UNITED NATIONS

Having noted the breadth of UN operations in which it exercises significant governmental powers over the local population, it is necessary to consider whether, relations between different actors (UN and local actors). Recent examples of “classic” peacekeeping operations of some importance are: UNAVEM I, II and III, in Angola; ONUMOZ in Mozambique; UNOMUR in Rwanda before the 1994 genocide; and UNOMIL in Liberia. In other cases, such as Somalia, Haiti, the former Yugoslavia and Rwanda in 1994, elements of peace-enforcement characterized some, if not all, phases of the operations. A socio-legal study of the impact of peacekeeping on the host population is by M Heiberg “Peacekeepers and Local Populations: some comments on UNIFIL” in E J Rikhye and K Skjelsbaek (eds) The United Nations (London, Macmillan, 1994). For a critique of some of these operations, see: T Chopra, “Achilles’ Heel in Somalia: Learning from a Conceptual Failure”, (1996) 32 Texas Journal of International Law 31; F Olonisakin and E Kwesi Aning, “Humanitarian Intervention and Human Rights: The Contradictions in ECOMOG”, (1999) 3 International Journal of Human Rights 16; T G Weiss, “A Research Note about Military-Civilian Humanitarianism: More Questions than Answers”, (1997) 21 Disasters 95. 16 The types of difficulties which might arise are illustrated, for example, by the case of UN assistance in Afghanistan, which is particularly interesting, as it is arguable that the UN has in practice agreed to incorporate gender discrimination in its humanitarian assistance in Afghanistan. The Memorandum of Understanding concluded by the UN and Afghanistan states: “Art.13 The Authorities and the UN will make efforts to increase the participation of men and women in health, education—especially health education—and food security. Both parties acknowledge the economic difficulties and the specific cultural traditions that make this goal challenging. As a result, women’s access to and participation in health and education will need to be gradual.” (emphasis added) Clearly the situation in which the UN finds itself in its work in Afghanistan is very delicate, as, whilst it seeks to relieve clear humanitarian need, it is forced to co-operate with a particularly authoritarian Government. However if the UN came to be seen as assisting or participating in institutionalized discrimination in the delivery of humanitarian assistance on grounds of gender, it would raise important human rights questions, as well as constitutional questions for the UN itself. 17 The Secretary-General has observed that the involvement of the UN in post-conflict situations can include “demilitarisation, the control of small arms, institutional reform, improved police and judicial systems, the monitoring of human rights, electoral reform and social and economic development” (Secretary-General, supra n. 14, at para. 47).

472 C Wickremasinghe and G Verdirame as a matter of international law, the UN itself might be held responsible in relation to claims that it has infringed international humanitarian or human rights standards. The starting point for such an investigation is, of course, the legal personality of the UN in international law. The International Court of Justice has recognised the UN to be “a subject of international law and capable of possessing international rights and duties”.18 As a result of this finding the Court was able to establish that the UN had the capacity to espouse a diplomatic claim as against other subjects of international law. A necessary corollary of this capacity, which is borne out in the practice of the UN, is that claims may be made against it on the basis of its own responsibility for its international obligations. The responsibility of international organizations has not received the same rigorous and systematic study as the topic of State responsibility, which is the subject of a long-running and ongoing project of the International Law Commission. However, there seems to be broad agreement on the proposition that in general the principles and rules of the law of State responsibility should constitute, mutatis mutandis, the basis for the law of responsibility of international organisations.19 Accordingly the primary elements constituting a claim for international responsibility—(1) an unlawful act, (2) its imputability to the United Nations, and (3) resulting damage—will be considered in turn.20

An Unlawful Act In determining whether an act is unlawful or not, it is first obviously necessary to determine the legal standards which are applicable to the UN. In relation to UN military activities, whether enforcement actions under Chapter VII or peacekeeping activities, there has been considerable debate as to the applicability of international humanitarian law. Whilst the UN has for a long time accepted that its forces “shall observe and respect the principles and spirit of the general international conventions applicable to the conduct of military personnel”,21 the precise meaning of such statements has given rise to considerable

18

1949 ICJ Rep. 174, at 179. P Daillier, A Pellet and N Quoc Dinh, Droit International Public (Paris: LGDJ, 6th ed., 1999) at 756, who underscore that a fortiori customary rules on responsibility should apply to international organisations too; C F Amerasinghe, Principles of the Institutional Law of International Organisations (Cambridge: Cambridge UP, 1996) at 239–47; See M Arsanjani “Claims against International Organisations: quis custodiet ipsos custodes” (1981) 7 Yale Journal of World Public Order 131; C Eagleton “International Organisation and the Law of Responsibility” (1950) 76 Recueil des Cours 323; Garcia-Amador’s First Report on State Responsibility reprinted in [1956] YBILC Vol. 2, 173 at pp. 189–90. 20 See J Salmon “Les Accords Spaak—U.Thant du 20 février 1965” (1965) 11 AFDI 468. 21 Para. 28 of Model agreement between the UN and Member States contributing personnel and equipment to UN for peacekeeping operations (1991) UN Doc. A/46/185. In the practice similar provisions have also frequently appeared in status of forces agreements and forces regulations. 19

Violations of Human Rights during UN Field Operations 473 discussion. The better view now appears to be as follows.22 Clearly the UN is not a Party to the Geneva Conventions or the 1977 Protocols, and so is not bound by them as such. However, the UN as a subject of international law is bound by the customary international humanitarian law, of which a greater number of provisions appear to have crystallised in recent years, as reflected for example in the jurisprudence of the International Criminal Tribunal for the former Yugoslavia (ICTY).23 This view appears to be accepted by the UN itself in the recent Secretary-General’s Bulletin on “Observance by UN forces of international humanitarian law”, which inter alia makes explicit the obligation of UN forces to protect civilian populations in the area of the operations, and sets standards for the treatment of civilians and persons hors de combat as well as for the treatment of detainees.24 Nevertheless, a caveat might be added to this sanguine view of the applicability of international humanitarian law, which relies in large measure on criminal proceedings before national courts for its enforcement. Clearly, in the case of a UN force, this must be carried out primarily by the troop-contributing States in relation to their national contingents. However in a number of cases in which allegations of violations of international humanitarian law were made against troops serving in Somalia in the UNOSOM force, the relevant national courts found the Geneva Conventions not to apply.25 Beyond military operations and with respect to the administration of territory or the exercise of other “public” functions which the UN may carry out, the arguments for the applicability of international human rights are not dissimilar to those reviewed above in relation to international humanitarian law. Whilst the UN itself is not a party to any international human rights treaty, it must, nonetheless be subject to the customary international law of human rights when

22 See C Greenwood “International Humanitarian Law and UN Military Operations” (1998) 1 YB Int’l Humanit.L. 3 esp. at 14–23. 23 See for example the judgments in cases such as Tadic (IT-94-1) 7 March 1997, Appeals Chamber 15 July 1999 and Furndzija (IT-95-17) 10 December 1998. 24 UN Doc. ST/SGB/1999/13, which entered into force on 12 August 1999. For a background and discussion of its terms see D Shraga “UN Peacekeeping Operations: Applicability of International Humanitarian Law and Responsibility for Operations-Related Damage” (2000) 94 AJIL 406. 25 Both the Canadian Court and the Belgian Court found that there was no armed conflict in Somalia, within the meanings of the Geneva Conventions or the Additional Protocols, and thus they were not applicable; see The Queen v. Brocklebank Court Martial Appeal Court of Canada, 2 April 1996 (see K Boustany “Brocklebank: a questionable decision of the Court Martial Appeal Court of Canada” (1998) 1 YB Int’l Humanit.L. 371) and judgment of the Cour militaire de Bruxelles 17 December 1997 (1998 Journal des Tribunaux 286). In Italy a number of allegations are under judicial investigation following the publication in 1997 of the report of the Fact-Finding Commission of Enquiry set up by the Government, but no decision has been reported yet. However these allegations are being treated as offences under the Military Penal Code of Peace (Codice Penale Militare di Pace)—see N Lupi “Report by the enquiry commission on the behaviour of Italian peace-keeping troops in Somalia” (1998) 1 YB Int’l Humanit.L. 375. Problems which might arise where the effect of responsibility of the UN appears to bar, or cap, State responsibility are beyond the scope of this chapter, but see Boustany, ibid.

474 C Wickremasinghe and G Verdirame exercising governmental functions.26 Under the Charter, the UN is obliged to promote respect for human rights, this being one of its principal purposes set out in Article 1. As Arsanjani points out, “[b]ecause international organisations are created to promote public order, it would be perverse, even destructive, to postulate a community expectation that international organisations need not conform to the principles of public order”.27

Imputability Clearly, where the act in question has been committed by a UN agent or a member of a UN force which itself is a subsidiary organ of the organisation, in the course of his or her duties, there is little difficulty in imputing the act to the Organisation. Even where a soldier who is a member of such a UN force performs an illegal act in breach the relevant Forces Regulations, the UN will often be liable to third parties in respect of such acts. Hence, in a series of global settlements with the victims’ States of nationality, the UN settled claims for excesses committed by ONUC troops against persons and property in Congo.28 Imputing to the UN the wrongful conduct of troops in the course of peaceenforcement operations may be more problematic where there has been a delegation to States co-operating with the Security Council. In general where the UN does not have command and control of the troops in question, it will not consider their acts as its own. It will therefore not settle claims deriving from such acts, but rather expect the State of nationality of the relevant troops to do so.29 26 See generally T Meron (ed.) Human Rights and Humanitarian Norms as Customary Law (Oxford, Clarendon Press, 1989), ILA Human Rights Committee “Final Report on the Status of the Universal Declaration of Human Rights in National and International Law” Report of the SixtySixth ILA Conference (1994) 525–63. 27 See Arsanjani, supra n. 19 at 134. 28 Salmon, supra n. 20. However, in a contrasting development, Canada made payments to the clans of persons killed or injured by Canadian troops in Somalia, following the local custom of negotiations with clan elders, but without any formal admission of responsibility. Canadian Press, “Somali parents lose bid to sue over son’s death: Ottawa says it may still compensate family of teenager beaten to death by Canadian soldiers”, (14 July 1999) The Globe and Mail. The Canadian government has succeeded before the court of first instance in having dismissed a tort lawsuit brought against Canada by the family of an alleged torture-death victim, Shidane Arone: Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. The Attorney General of Canada (unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J.). See aso discussion in J Terry, “Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad”, chapter 4 of this volume; and C Scott, “Introduction to Torture as Tort: From Sudan to Canada to Somalia”, chapter 1 of this volume. As of midsummer 1999, “[a] [Canadian] federal [government] lawyer said negotiations are continuing with a lawyer representing the Arone family”; prior to that, “[t]he Defence Department paid $15,000 (U.S.) to the village clan as the regiment was leaving Somalia in 1993. But Mr. Bansie [the Arone family lawyer] has said the impoverished parents of Mr. Arone received none of that money.” Canadian Press, ibid. 29 Thus in Korea claims were not settled by the UN but by the Unified Command or the participating States—see F Seyersted “United Nations Forces—some legal problems” (1961) XXXVII BYIL 351 at 421. See also Amerasinghe, supra n. 19; and Shraga, supra n. 24 at 408.

Violations of Human Rights during UN Field Operations 475 It remains an important, but as yet untested, point as to whether, by delegating enforcement action to Member States, the UN can divest itself of responsibility for all that ensues.

Damage The requirement that damage must be shown, should not cause undue problems in most cases in which serious violations of human rights are claimed. In the Congo settlements the UN paid compensation for those injuries to persons and property which it deemed to be “unjustifiable damage”.30 Commentaries suggest that this refers to the UN’s position that only damage which did not arise from military necessity would be compensated.31 Of course, in a case in which the most serious forms of human rights violations (such as torture or the extrajudicial killing of detainees) were established, there could presumably be no grounds of justification. Finally it should be noted that claims against the UN at the international level are in the mould of classic diplomatic claims between States. The latter form of claim has proved to be of limited value in the promotion of human rights standards generally, largely because it is a remedy of the State rather than a remedy for victims. Most States reserve to themselves considerable discretion in deciding whether or not to espouse a diplomatic claim; the negotiation and settlement of claims is largely conducted in secret through the diplomatic channel; and there is often a discretion in the claimant State whether or not to pass on to the victim any reparation eventually paid.32 Claimants are therefore likely to prefer remedies which they are entitled to pursue in their own name, whether under an individual complaints mechanism at the international level or through civil litigation.

4 CIVIL LIABILITY OF THE UNITED NATIONS

The Status and Jurisdictional Immunities of the UN: The General Situation The question therefore arises as to whether the responsibility of the United Nations towards private persons can be directly engaged. Article 104 of the UN Charter states that “[t]he Organisation shall enjoy in the territory of each of its Members such legal capacity as may be necessary for the exercise of its functions and the fulfilment of its purposes”. The Convention on the Privileges and 30 See letter of Secretary-General to the Belgian Ministry of Foreign Affairs of 20 February 1965. annexed to Salmon supra n. 20. 31 See Salmon, supra n. 20; see also Shraga, supra n. 24 at 410–11. 32 Similar complaints arose in relation to the payment by Canada to the Arone family’s clan and not directly to the family and the assertion by the family that they have seen none of this compensation: Canadian Press, supra n. 28.

476 C Wickremasinghe and G Verdirame Immunities of the United Nations of 1946 (“the General Convention”) sought to spell out the details of the UN’s status as well as its privileges and immunities by stating in article 1: “The United Nations shall possess juridical personality. It shall have the capacity: (a) to contract; (b) to acquire and dispose of movable and immovable property; (c) to institute legal proceedings.”

Though reference is only made to the power to “institute” legal proceedings, it appears that in principle the capacities of the UN must include the capacity to be sued in national courts.33 The possession of juridical personality and the effective exercise of the other specifically enumerated capacities must imply such a potentiality. The practice of the Organisation appears to acknowledge in principle that such liability may arise, though in fact it will in most cases rely on its immunity from jurisdiction to resist the institution of proceedings in local courts.34 Immunity from jurisdiction for the UN is provided for in general terms in article 105 of the UN Charter, “[t]he Organisation shall enjoy in the territory of each of its Members such privileges and immunities as are necessary for the fulfilment of its purposes”. It is given further definition in article 2 of the General Convention which states “[t]he United Nations, its property and assets wherever located and by whomsoever held, shall enjoy immunity from every form of legal process except in so far as in any particular case it has expressly waived its immunity.” There are different views as to the extent of the immunity contained in these two provisions, which manifest themselves in the attitudes of States, national courts and the opinions of writers. The majority view appears to be that the UN enjoys wide immunity from suit, relying on the express words of the General Convention “immunity from every form of legal process”, and that the restrictive doctrine of State immunity should not be adopted in relation to the immunities of international organisations.35 The UN itself denies the application of the restrictive standard of immunity to international organisations on the basis that, whereas State immunity is 33 Thus, for example, the Convention contemplates that the UN may waive immunity and thus submit to the jurisdiction of local courts. However, in practice the general policy of the UN is not to waive immunity except in cases where it has insurance cover (e.g. in relation to road traffic accidents) and the insurers require a waiver in order to defend proceedings. 34 See, for example, Reports of the Secretary-General, “Administrative and Budgetary Aspects of the financing of the United Nations Peacekeeping Operations: financing UN peacekeeping Operations”, UN Docs A/51/389 (1996) and A/51/903 (1997). 35 See, for example, C W Jenks, International Immunities (London: Stevens, 1961); C Dominicé “L’immunité de juridiction et d’exécution des organisations internationales” (1984) 187 Recueil des Cours 145–238; and P H F Bekker The Legal Position of Intergovermental Organisations—a Functional Necessity Analysis of their Legal Status and Immunities (Dordrecht: Kluwer, 1994). Support for the minority view—i.e. that restrictive doctrine of sovereign immunity is applicable to international organisations—may be found in F Schroer “De l’application de l’immunité des états étrangers aux organisations internationales” (1971) 75 RGDIP 712.

Violations of Human Rights during UN Field Operations 477 based on concepts such as the sovereign equality of States, the grant of immunity to international organisations is based on the need to ensure the independent functioning of those organisations.36 It is said that, in the case of the UN, the provision of absolute jurisdictional immunity is functionally necessary and thus that article 2 of the General Convention should be construed accordingly. However, another view could just as easily be taken, namely that the extent of jurisdictional immunity is in fact limited by the concept of functional necessity. All that is required by article 105 of the Charter is that jurisdictional immunity should only be granted to the UN to the extent that it is functionally necessary, and that the provisions of the General Convention must thus be read in the light of the Charter provisions.37 Although this notion of limiting the extent of immunity by reference to functional necessity is on its face attractive (in that grants of immunity increasingly require justification in the light of human rights standards of due process and access to court), it is likely to raise some significant problems in application. Given the functionally-based personality and activity of international organisations, does the concept of functional necessity give a clear enough basis on which to limit immunity in a particular situation?38 If it is to be a meaningful limitation, who is to determine whether it is functionally necessary to extend immunity to a particular act of an organisation?39 Although there is some national jurisprudence in relation to other 36

See, for example, 1985 YBILC Vol. II pt. 1 at 161. In support of this contention it might be said that the General Convention must be interpreted consistently with the article 105 of the Charter, as it was concluded pursuant to article 105(3) of the Charter, in order to “determin[e] . . . the details of the application of paragraphs (1) and (2)” of Article 105. Thus the General Convention refers to Articles 104 and 105 in its Preamble. In any event if there was a conflict between provisions of the General Convention and the Charter it would probably have to be resolved in favour of the Charter, under article 103. However it should be said that in the one case against the UN where this argument was made a Belgian court of first instance and Court of Appeal rejected it on the basis that the Convention, the Charter and the General Convention had equal status and that the latter, being later in time, was limited by the former— Manderlier v. UN and Belgium (45 ILR 446, 69 ILR 139) This view might be criticised for its failing to give priority to the Charter in accordance with Article 103 (as has since been recognised in Article 30(1) of the Vienna Convention on the Law of Treaties 1969). 38 The point is made clearer by comparison with other regimes of functional immunity. For example, in relation to aspects of diplomatic immunity such determination is made easier by the fact that the Vienna Convention on Diplomatic Relations spells out in broad terms what the functions of a diplomat are. It is clear therefore that there will be some acts which the diplomat will perform outside his functions. Whereas in relation to international organisations which are only endowed with limited capacities to perform certain functions, it is not clear that it may perform any acts outside the performance of its functions. All its acts must in some way be connected to the furtherance of its functions. 39 See the rather nuanced findings of ICJ in its Advisory Opinion in the Cumaraswamy case (29 April 1999)—(for further discussion see C Wickremasinghe “Difference relating to immunity from legal process of a special rapporteur of the Commission of Human Rights” (2000) 49 ICLQ 724). In the context of restrictive State immunity national courts are faced with a problem which is to some extent comparable, when they seek to determine whether an act was jure imperii or jure gestionis. In relation to claims by a respondent State that although an act was commercial in character it was nonetheless in pursuit of sovereign purposes, national courts have insisted that the appropriate test is the nature of the act rather than the purposes for which it was performed, and have thus avoided determining what sovereign purposes may include. However, national courts appear to have no way of avoiding consideration of the purpose of an act when applying the 37

478 C Wickremasinghe and G Verdirame international organisations to support a functional limitation on immunity, it has been rejected on the only occasion on which the question has been raised in relation to the immunity of the UN.40 It is worth adding in the present context, however, that if the functional limitation on the extent of immunity can be maintained, Article 105 suggests that the UN shall only enjoy immunity in so far as it is “necessary for fulfilment of its purposes”. It should be remembered in this context that one of the principal purposes of the UN is “to achieve international cooperation . . . in promoting and encouraging respect for human rights and fundamental freedoms”.41 Therefore it might be possible to argue that if an act which it carries out, violates human rights and fundamental freedoms, that act would be contrary to the purposes of the UN and therefore could not be immune. While such an argument has not yet been raised in any reported case, an analogy might be drawn with decision of the House of Lords in the Pinochet case,42 to the extent that some Members of the Court held that acts such as torture could not be interpreted to fall within the concept of “official” acts for which a former head of State enjoyed immunity. On the other hand, it might be protested that were such arguments to succeed, they would involve a municipal court in determining a question of UN law (i.e. as to the scope of the purposes of the UN) which, in the UK at least, may be found to be non-justiciable on the municipal plane.43 It should be noted, however, that many aspects of the doctrine of non-justiciability remain controversial.44

functional test as a restriction on the extent of the jurisdictional immunities of international organisations. This may in turn lead to difficult questions of interpretation as to what the functions of the organisations properly entail, and which national courts may not be the appropriate forum to determine. 40 Although, here the courts did not actually enquire into the limits of functional necessity because they erroneously read the General Convention to have priority over the provisions of the UN Charter, on the basis of the lex posterior rule—see the Manderlier case, supra n. 37. 41 Article 1(3) of the UN Charter. 42 R v. Bow Street Stipendiary Magistrate, ex parte Pinochet Ugarte (No. 3) [1999] 2 All ER 97. 43 In Westland Helicopters v. AOI [1995] 2 All ER 387 the English High Court found that questions involving the application or interpretation of public international law, being the proper law of international organisations, were not justiciable before a municipal court. 44 For a critique, see M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of ‘Act of State’ in Anglo-Canadian Law”, chapter 13 in this volume; see also Higgins, Problems and Process— International Law and How We Use it (Oxford, Oxford UP, 1994) at 212–13, including a persuasive case against the findings of the House of Lords on non-justiciability in the International Tin Council litigation. On the specific question of a domestic court interpreting the scope of UN purposes, there are instances in other contexts where this is necessary in deciding some point of domestic law—see, for instance, the majority of the Supreme Court of Canada in Pushpanathan v. Canada (Minister of Citizenship and Immigration) 160 DLR (4th) 193 grappling with the meaning of “purposes and principles of the UN” as found in article 1F(c) of the Geneva Convention on Refugees, which article was incorporated into Canadian law.

Violations of Human Rights during UN Field Operations 479 Immunity and access to justice In the case of Manderlier v. UN and Belgium, an alternative argument was raised.45 The case concerned a claim by a Belgian national for compensation for damage to property held in Congo which was alleged to have been suffered as a result of acts of members of ONUC in 1962. He claimed, inter alia, that as a result of the immunity of the UN he was deprived of his right to fair hearing of his case by an independent and impartial tribunal, contrary to article 6 of the European Convention on Human Rights and contrary to article 10 of the Universal Declaration of Human Rights. The Court found that article 6 of the European Convention could not have the effect of subjecting the UN to the jurisdiction of the Belgian Courts contrary to the provisions of the General Convention.46 The decisions of the Belgian Courts in Manderlier may be of limited precedential value. Firstly, as has been discussed, their reasoning on the relationship of the UN Charter to the General Convention is contentious; second, the fact that they are judgments of the courts of a single State; and third because of their vintage. With respect to this last factor, it is important to note that human rights law has developed considerably in the period since 1969, and, in particular, regimes of immunity have come under increasingly close scrutiny by international human rights bodies.47 Thus in the recent cases of Beer and Regan v. Germany and Waite and Kennedy v. Germany, the European Court of Human Rights reviewed the immunities of an international organisation.48 The cases had first arisen as labour disputes between an international organisation and various of its employees, in which the organisation relied on its immunities to protect it from the jurisdiction of the local labour courts. The European Court found essentially that the immunity did not contravene the requirements of article 6 in these cases, because there was an alternative forum under the organisation’s internal administrative procedures which were available to claimants pursuing employment disputes.49 The UN is under a similar obligation to provide alternative remedies, by virtue of article VIII, section 29, of the General Convention, as follows: “The United Nations shall make provisions for appropriate modes of settlement of: (a) Disputes arising out of contracts or other disputes of a private law nature to which the United Nations is a party; or 45

45 ILR 446, and 69 ILR 139. See A Clapham, “Revisiting Human Rights in the Private Sphere: Using the European Convention on Human Rights to Protect the Right of Access to the Civil Courts”, chapter 19 in this volume, for contemporary usage of article 6 of the ECHR on the issue of immunity involving transnational access to courts. 47 Apart from the cases mentioned below, see Clapham, ibid. for discussion of the case of Osman and the application in Jaccard-Veloso. 48 Judgments of 18 February 1999. 49 For concerns in relation to this finding, especially on the question of the adequacy of the alternative remedies, see A Reinisch, “International Decisions: Waite and Kennedy v. Germany, Beer and Regan v. Germany”, (1999) 93 AJIL 933. 46

480 C Wickremasinghe and G Verdirame (b) Disputes involving any official of the United Nations who by reason of his official position enjoys immunity, if immunity has not been waived by the Secretary-General.”

In fulfilment of the obligation contained in paragraph (a), the UN will first seek to negotiate a settlement, but failing that, will often rely upon arbitration. Thus, many of its contracts provide for the settlement of disputes by arbitration. The applicability of paragraph (a) to allegations against the UN of human rights abuses would thus depend upon two factors. The first factor is whether the allegations could be re-clothed as “private law” actions in tort, an important question not explored in depth in the present chapter but canvassed elsewhere in this volume.50 The second factor is the willingness of the UN to enter into an agreement to arbitrate once the complaint has arisen, since, unlike contractual liability, such agreement cannot be made in advance for tortious liability. Even then, the adequacy of arbitration as an alternative to tort litigation involving a serious violation of human rights cannot be easily assumed—it must be assessed in relation to the principles of access to justice, as well as those of procedural and substantive justice that must be applicable to this species of claim. It may therefore be questioned whether the assertion of immunity in these circumstances would satisfy the recent jurisprudence of the Strasbourg Court.51

Status and Immunities of Non-military UN Field Operations The vast range of UN field operations means that generalisations are of limited value. However, it is usually the case that an operation will be undertaken by a subsidiary organ of the UN. These may be standing organs, for example UNHCR or UNICEF, or organs established ad hoc for the purposes of a particular operation, as in the case of various UN peacekeeping forces which have been established. Subsidiary organs may have certain capacities delegated to them by their parent organ. Thus, in relations with third parties, such organs may, for example, be able to enter into contracts with private persons in their own name. Despite this, however, they do not enjoy legal personality distinct from the UN itself. Thus, if it were necessary to institute legal proceedings, such organs would have to rely on the capacity of the UN to do so, and, conversely, if a private party purported to commence legal proceedings against such an organ, the suit could be struck out on a successful assertion of the jurisdictional 50 See e.g., in this volume, J Terry, “Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad”, chapter 4; J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11; and G Virgo, “Characterization, Choice of Law and Human Rights”, chapter 12. 51 Alternatively it might be argued that, as the substantive allegation envisaged in such cases is of a violation of human rights, the complainant who is deprived of a remedy by reason of immunity may be able to claim that this, in itself, is a further violation of his human rights, i.e. Article 13 ECHR and Article 3 ICCPR. The claim would, of course, be against the forum State in such cases.

Violations of Human Rights during UN Field Operations 481 immunity of the UN under Article 105 of the Charter and, where applicable, the General Convention.52 A UN field operation may be imposed by the Security Council regardless of whether the “host” State consents, if it is necessary for the maintenance or restoration of international peace and security under Chapter VII of the Charter. However, in most cases UN operations will take place on the basis of an agreement with the host State. Some UN organs have standard form agreements for their operations, which deal with questions of jurisdiction and disputes arising out of the operation. Thus, the revised model Agreement concerning the activities of UNICEF contains provision that the host State will be responsible for all claims resulting from the execution of the operations as planned in the Agreement. Accordingly the host State must “defend, indemnify and hold harmless UNICEF and its employees or agents against all liabilities, suits, actions, demands, damages, costs or fees on account of death or injury to persons or property” resulting from the operation as planned, in so far as they do not amount to reckless misconduct of employees or agents.53 UN Forces In relation to UN peacekeeping forces the legal situation is more complicated because such troops are not in the employment of the UN but are usually made up of a number of different national contingents. Thus a complex of interlocking international instruments may be applicable. The consent of the parties being a fundamental principle of peacekeeping forces, in most cases the UN will enter into a Status of Forces Agreement (SOFA) with the host State.54 In addition it will usually enter into participation agreements with the States which contribute forces in order to define the terms on which the forces are to serve in a UN operation. Further, in order to ensure the effective functioning of a force (particularly as it is made up of different national contingents), it may also be necessary for the UN Secretary-General to issue Forces Regulations. All of these instruments will have a bearing upon the standards of conduct to which an operation is subject, and on claims and assertions of immunity arising therefrom. Again, as the practice of the UN varies in accordance with the exigencies of each operation, generalisations are somewhat dangerous. However, some general observations are attempted here, though for reasons of space they are necessarily broad and subject to exception in specific cases.55 In relation to host 52

See F Morgenstern Legal Problems of International Organisations (Cambridge: Grotius, 1986). Of course, there is scope to argue that human rights abuses of the nature of torture will either be, or be closely connected to, “reckless misconduct”. However in relation to possible remedies against the Host State, see below Section 5. 54 It also appears that where a formal SOFA has not been negotiated Guidelines to similar effect have been established; see R Seikman National Contingents in UN Peace-Keeping Forces (Dordrecht: Martinus Nijhoff, 1988). 55 For a more detailed analysis see D W Bowett United Nations Forces (London: Stevens, 1964) and Seikman, supra n. 54 at 120–53. 53

482 C Wickremasinghe and G Verdirame State jurisdiction, SOFAs generally accord to the force privileges and immunities in relation to the jurisdiction of the host State based on those of the UN under the General Convention. In addition individual members of the forces will be granted privileges and immunities necessary for the independent exercise of their duties, including immunity from the civil and criminal jurisdiction of the host State, discussed further below. In accordance with the participation agreements and Forces Regulations (which are usually annexed thereto), peacekeeping troops remain subject to the military rules and regulations of their respective national States. In addition, the Forces Regulations contain additional rights and obligations in relation to their functions and conduct as members of an international force, including frequently an express obligation to respect and observe the spirit and principles of international humanitarian law.56 The Force Commander has overall responsibility for ensuring the good order of peacekeeping forces, but disciplinary action is the responsibility of the commander of each national contingent and, if necessary, the authorities of the relevant contributing State. Disciplinary action is exercised in this way both in respect of breaches of national military regulations, but also of the Forces Regulations themselves. In general, the practice has been that the respective national States continue to enjoy exclusive criminal jurisdiction over their own troops.57 Whilst the peacekeeping force as a whole will usually enjoy immunity from the civil jurisdiction of the host State, individual members of the force are generally only immune from civil jurisdiction in relation to their official acts.58 Thus, in principle the non-official acts of members of the force will be subject to the civil jurisdiction of the host State.59 In relation to claims for compensation by private parties for damage arising out of official acts of members of the force, the UN has recognised that its own responsibility is engaged. In relation to various peacekeeping forces it has established Claims Commissions or equivalent panels to deal with such claims. These 56 As has been discussed, it is now established that general international humanitarian law applies to international organisations no less than to States, at the very least in their spirit and principles if not also the specific rules. Thus, an express statement in Forces Regulations is not technically necessary to create obligations for the UN, although its inclusion has the advantage of making the position clear, and in practice is often accompanied by additional provisions on enforcement. 57 See, for example, the criminal law proceedings in various jurisdictions against soldiers who were part of the UN Operation in Somalia, supra n. 25. 58 An important exception to this is contained in “Agreement relating to the legal status, facilities, privileges and immunities of the United Nations Organisation in the Congo” (concluded 27 November 1961—414 UNTS 229), which in effect excludes almost all aspects of the jurisdiction of the local courts (Article 10). The removal of claims in relation to non-official acts of members of the force, as well as official acts, from local jurisdiction, appears to have been because of the breakdown of the local judicial system—see Bowett, supra n. 55 at 236–40 and Seikman, supra n. 54 at 145–7. 59 If there is any dispute as to whether an act was in the course of the respondent’s official duties or not, it is for the Commander of the force to certify its nature. However by analogy with the role of Secretary-General in relation to immunities of officials and experts of the UN under the General Convention, this may in fact only constitute a strong, but rebuttable, presumption—see the Advisory Opinion of the ICJ in the Cumaraswamy case, supra n. 39.

Violations of Human Rights during UN Field Operations 483 claims commissions are usually comprised of three members, one appointed by the UN Secretary-General, a second by the Government of the host State and the third by joint agreement (or failing that by the President of the ICJ). Commissions may determine claims and make awards of compensation, although the law which they must apply is not usually specified. Bowett suggests that the applicable law will be either the law of the host State as the lex loci delicti, or the general principles of law, and this is reflected in practice.60 The recent increases in the scope of peacekeeping activities, as well as in the size and types of claims made against the UN, have led the UN General Assembly to adopt a resolution which seeks to limit the liability of the UN in relation to claims in tort.61 The resolution seeks to establish both temporal and financial limits to such liability. There is a temporal limitation for the submission of claims against the UN, being six months from the time of damage from when it was known or could have been known to the claimant (and in any event no more than one year from the date on which the operation has concluded). The financial limitations apply to compensation for personal injury, illness and death, as well as damage to and loss of property. In relation to liability for personal injury, illness or death, compensation is assessed according to prevailing local standards, subject to a ceiling of $50,000. Compensation for nonpecuniary loss is excluded, thus a claimant may be able to claim for the costs of treatment, care and loss of earnings but not for pain and suffering. However, two important riders should be added. The first is that the UN does not accept liability for damage incurred by reason of operational necessity. Yet, it is hard to imagine that a serious violation of human rights could be justified as being operationally necessary. The second is that in cases of gross negligence and wilful misconduct the financial limitations to liability will not apply, thus potentially enabling the victim of a serious human rights violation to make a claim for full compensation.62 While such procedures would appear suitable to deal with straightforward private law claims, for example for personal injury, death, or damage to property resulting from road traffic accidents, it may be questioned whether the procedural regime is also appropriate to deal with cases in which allegations of serious violations of human rights are in issue, even though the conduct in question may be simultaneously characterised as falling within standard tort categories related to harm to persons.63 60 See Bowett, supra n. 55 at 244. In the case of ONUC a Claims Commission was not established, though the Special Representative of the Secretary-General was empowered to establish one. Instead where a case could not be settled by negotiation, arbitration was resorted to. In the one reported arbitration relating to non-contractual loss relating to damage to an aircraft, the applicable law under the compromis was Congolese law, Starways v. UN (1971) Rev. Belge du Droit International 451–68. 61 GA Res. 52/247 (22 June 1998); for commentary see Shraga, supra n. 24 at 409–12. 62 In such circumstances the UN will reserve its right to seek an indemnity from the relevant troop-contributing State—see Shraga, ibid. 63 On multiple characterisations of the same conduct, see Virgo, supra n. 50.

484 C Wickremasinghe and G Verdirame Thus, in order to determine whether the UN is responsible in relation to allegations of a violation of the human rights of a local civilian by a member of a peacekeeping force, much will depend upon the nature of the conduct alleged. If the conduct was simply criminal and did not arise from the official duties of the force (Bowett gives the examples of rape or theft),64 the responsibility of the UN is limited to a thorough investigation of the incident and an insistence that the relevant contributing State carry out disciplinary and/or criminal proceedings. If, however, the act could be said to be “official” the UN itself would be responsible, although, because of its immunity, civil proceedings could not be brought in national courts, and a direct claim would have to be made against it, where appropriate, to a Claims Commission.65 Bowett suggests that, where the act was criminal under the law of the host State, the responsibility of the UN would be satisfied by it fulfilling its duty to investigate, insisting that the contributing State exercise such criminal jurisdiction and making appropriate monetary compensation.66 He suggests that the UN would also be responsible in relation to claims for monetary compensation, if it can be established: firstly, that the acts were unlawful under the principles and spirit of general international conventions applicable to the conduct of military personnel (as recognized under the Forces Regulations); and, secondly, that such acts were attributable to the UN as official conduct or by reason of specific authorization or subsequent ratification by the Secretary-General or Commander. Clearly, the most serious violations of human rights law would be unlawful under the Forces Regulations, but the question as to whether they are also official is more difficult. If, for example, the allegation was of the torture of a suspected opponent of the force, it would clearly be in violation of the law and policy of the UN, and could not be an authorised act.67 It might, therefore, be problematic for the UN to adopt such an act by subsequent ratification. 64

Supra n. 55 at 246. In the case of Askir v. Boutros Ghali and others (933 F.Supp. 368) a property owner in Somalia, whose property was occupied by UNOSOM forces in 1992, brought proceedings for unauthorised or unlawful possession of property, against various high-ranking UN officials including the Secretary-General, before a US Federal Court. The Court had little difficulty in finding that the respondents enjoyed immunity, though in accordance with previous US jurisprudence it raised the question of whether the commercial exception (as applicable in relation to the doctrine of restrictive State immunity) was applicable to the immunity of the UN. However, it did not answer the question definitively, finding that in any event the deployment of troops in military operations was an act jure imperii, and that “occupation of property during such an operation to house troops, store and distribute supplies and ordinance, or manage the logistics and planning of peacekeeping and humanitarian relief efforts is part and parcel of such an operation”. Thus the Court stressed the “sovereign” or public nature of the operation as opposed to the private law claim in relation to property. If similar reasoning were applied to claims in tort for allegations of human rights abuses by UN troops, it would suggest a claim of immunity by the UN would be upheld in the US even if restrictive immunity were applied. 66 Supra n. 55 at 246. 67 As has been discussed, this raises an analogous question to the issue raised in the Pinochet case of whether an act of torture could be an official act in respect of which a former Head of State enjoyed immunity. 65

Violations of Human Rights during UN Field Operations 485 On the other hand the members of the force who carried out such an act may have done so, however unlawfully, in pursuance of the mission of the force. Also the victim himself may wish to establish the responsibility of the UN, as he may have a better chance of recovering compensation from it, than from an individual soldier or his national commanders. Pursuing a claim before a Claims Commission established locally is also likely to be a good deal easier than the complexities of litigation against the individual soldier or his commanders, claims which might potentially raise issues before a number of jurisdictions. Thus, as an alternative, it might be argued that for an act such as torture to occur, there must have been a failure of the system of discipline of the force, for which the UN is ultimately responsible. In that sense, in accordance with general principles of command responsibility and vicarious liability, it would seem that the responsibility of the UN could be established, without the need to make the difficult characterisation as to whether it was official or not.68 It should be stressed that as the law currently stands the immunity of the UN would prevent such claims from being brought before national courts. Thus claims must be submitted by the victim directly to the UN. If settlement can not be reached by negotiation, or by means of a Claims Commission or arbitration, then the victim’s State of nationality may be able to bring a claim by way of diplomatic protection. Thus, under the law as it currently stands, victims have no right to judicial protection in claims against the UN (although arbitration may subsequently be agreed between the parties).

5 HOST STATE LIABILITY

In cases where UN field operations take place on the basis of the consent of the host State and for reasons of immunity there is no possibility of bringing municipal law actions directly against the UN, it may be possible to argue that the host State should be liable for uncompensated loss and damage resulting from such operation. There are as yet no reported cases in which such claim was made against a UN field operation. However, an analogy may be drawn with the French case of Burgat v. Minister of Foreign Affairs,69 in which the Conseil d’Etat recognized that, as a matter of French public law, France as the host State of an international organisation, was responsible to indemnify a French citizen who was unable to obtain compensation for loss by reason of immunity granted under the host State arrangements. The case involved a property claim, but its reasoning would seem applicable to other forms of loss. Similarly, it may be possible to conceive of an access-to-justice claim before a Constitutional Court or an international human rights body. The claim would 68 See also V Oosterveld and A Flah, “Holding Leaders Liable for Torture by Others: Command Responsibility and Respondeat Superior as Frameworks for Derivative Civil Liability”, chapter 16 in this volume. 69 (1977) 104 JDI (Clunet) 630.

486 C Wickremasinghe and G Verdirame be that the grant of immunity by a State to a UN operation, without clear and adequate alternative provision for the settlement of disputes, engages the responsibility of the State granting such immunity.70 This approach appears consistent with the judgment of the European Court of Human Rights in Denise Matthews v. United Kingdom.71 In this case, the applicant submitted that her right to vote in elections, guaranteed under article 3 of Protocol No. 1 of the European Convention on Human Rights (ECHR) had been violated by the 1976 Act on Direct Elections of the European Community (EC), which excluded Gibraltar from the franchise for the elections of the European Parliament. The UK Government sought to defend itself on the ground that it could not be held responsible for an act of the EC, and that in order “to engage the responsibility of any State under the [European] Convention [on Human Rights], that State must have a power of effective control over the act complained of”.72 The Court rejected this argument and held that “the Convention does not exclude the transfer of competences to international organisations provided that Convention rights continue to be ‘secured’ ”.73 The Court stressed that the responsibility of the United Kingdom derived from its original decisions to enter into the relevant treaty commitments on its accession to the EC, which were made subsequent to, and, therefore, also subject to, its assumption of obligations under the ECHR.

6 LIABILITY OF STATES CONTRIBUTING TROOPS IN MILITARY OPERATIONS

As has been suggested, where enforcement action is delegated to participating member States, the UN is unlikely to accept responsibility for loss or damage suffered by third parties. Claimants may therefore seek compensation from the State whose nationals carried out the acts which are the subject of their complaints. There appears to be some practice in relation to Korea where the participating States agreed to settle claims from third States or their nationals.74 Where the troops form part of a UN force over which it has command and control, in general the UN will take responsibility for their acts in response to claims by third parties. Thus, although the contributing State remains responsible for maintaining the discipline of their national troops, it may not accept responsibility towards injured third parties. Thus, in the recent case brought against Canada by the family of a Somali allegedly tortured to death by Canadian troops in Somalia, the Canadian government appears to have succeeded in its argument before the judge at first instance that Canada owed no 70 Similar to the claims in Beer and Regan v. Germany and Waite and Kennedy v. Germany, supra n. 48. 71 Denise Matthews v. United Kingdom (No. 24833/94), Judgment of the Court, 18 February 1999. See also the discussion of Osman v. United Kingdom in Clapham, supra n. 46. 72 Denise Matthews v. United Kingdom, supra at para 27. 73 Ibid. at para 32. 74 See Seyersted, supra n. 19.

Violations of Human Rights during UN Field Operations 487 legal duty of care to the local population in Somalia, although the judge’s reasons were so limited that it is unclear whether this finding was based on the fact that the UN itself might bear responsibility.75 That being said, it is not necessarily axiomatic that because the UN itself may bear some legal responsibility for the conduct of troops involved in one of its operations, the troop-contributing States should be absolved from any responsibility. Support for a finding of liability on the part of a troop-contributing State can be found in the approach of the British House of Lords in AttorneyGeneral v. Nissan.76 That case involved a claim for damage to property in Cyprus arising from its occupation by British forces. These troops had initially been involved as part of an offer by Britain, Greece and Turkey (the parties to the Treaty of Guarantee of 1960), with the agreement of the Government of Cyprus, to assist in bringing an end to communal strife. However on the subsequent establishment of the UN Force in Cyprus (UNFICYP), the British soldiers who had been deployed became a contingent of this international force. The House of Lords found that the fact that the British troops had become part of UNFICYP made no difference to the liability of the Crown for the damage to the hotel in which the British forces were billeted. According to Lord Pearce, the Crown bore the primary responsibility for the acts of its forces even after the establishment of UNFICYP, for as along as British troops were billeted there.77 These findings were clearly premised on the fact that command and responsibility for discipline remained firmly with the contributing States.78

7 INDIVIDUAL LIABILITY

Individual Criminal Responsibility Finally, it is necessary to consider the position of the individual soldier or official whose conduct causes damage to third parties. In military operations of the UN, it is clear that individual soldiers remain subject to the exclusive criminal jurisdiction of their national States, both in relation to breaches of their national disciplinary codes and in relation to breaches of the Forces Regulations. As noted earlier, there have been some prosecutions of soldiers in connection with excesses committed during the UN operation in Somalia, although the poor 75 Arone, supra n. 28. The case was brought on behalf of the family of Shidane Arone, a youth who was allegedly beaten to death by Canadian troops during the UN Operation in Somalia. It was brought in the Canadian courts against Canada, but was dismissed for being commenced out of time and for failing to make out a cause of action against the federal Government. See the criticism of the reasoning in Terry, supra n. 28. An appeal is currently pending to the Ontario Court of Appeal. 76 [1969] 1 All ER 629. 77 Attorney-General v. Nissanm, supra n. 76 esp. at 647–88 78 It might be observed that in relation to substantive liability of the Crown towards the owner of the property in question, the fact that the latter was a British subject was a significant factor for some (Lords Reid and Wilberforce) but not all (Lord Morris) of their Lordships.

488 C Wickremasinghe and G Verdirame record of these prosecutions may undermine confidence that this is an effective deterrent. In relation to non-military UN operations, UN officials and UN experts on mission enjoy immunity from criminal and civil jurisdiction in relation to “all acts performed by them in their official capacity”.79 Thus, in general, if UN personnel can establish that their conduct was “official”, they will be able to assert immunity and avoid individual liability. However, this may require some qualification in the light of the recent decision of the House of Lords in the Pinochet case, which may be relevant by analogy.80 Given that the immunity of UN personnel is in principle a functional immunity ratione materiae,81 it would appear that where the conduct in question falls within the definition of torture, or amounts to a crime against humanity, an individual UN agent may not be able to avoid his or her own criminal responsibility by asserting immunity under the General Convention or relevant SOFA.82

Individual Civil Liability Similarly under the General Convention, officials of the UN and experts on mission enjoy immunity from civil jurisdiction for their official acts, whilst UN troops will usually enjoy a similar immunity by virtue of relevant provisions of SOFAs. Under the broad scheme of these immunities an injured party should thus not be left without a remedy in that, if the acts causing damage were “official” acts, they are acts of the UN for which it will be responsible. In contrast, if the acts were not official, then the responsible individual UN officials/agents will be liable and will not enjoy immunity from jurisdiction. However, given the potential complexities and practical difficulties of bringing a civil action against, for example, a foreign soldier whose conduct during a short tour of duty gives rise to a claim, it is important to ensure that cases should not “slip through the cracks”, leaving plaintiffs without redress. As has been discussed, in practice the UN has taken a fairly broad view, in the context of military operations, as to conduct for which it will offer compensation to third parties who suffer damage. Such conduct may include acts which are unlawful under national disciplinary law, under the Forces Regulations or under international humanitarian law. From the point of view of the injured party, the willingness of the UN to bear responsibility is likely to be decisive, as 79 Article V, section 18, General Convention on the Privileges and Immunities of the UN. In addition, experts on mission enjoy an apparently unlimited immunity from detention or arrest: Article VI, section 22(a). 80 R v. Bow Street Stipendiary Magistrate, ex parte Pinochet Ugarte (No. 3) [1999] 2 All ER 97. 81 i.e. it seeks to protect interests of the UN rather than benefiting the agent personally. 82 This potentially raises the inconsistency of limiting the immunity of an individual, but not that of the UN for which he is acting, not unlike the distinction that can arise between the immunities of a State and those of its officials—see Bühler, supra n. 44.

Violations of Human Rights during UN Field Operations 489 recovery of compensation is likely to be more effective and speedier against the UN than if they are forced to sue a particular individual. However, where the injured party seeks to sue an individual UN agent, the latter will only be able to claim immunity if they can show that their act was “official”. In this sense it is a functional immunity protecting the functioning of the UN, and in relation to which it is for the UN to assert or to waive it. Where the UN asserts immunity it will usually do so by the issue of a certificate, either by the Secretary-General in relation to UN officials and experts, or by the Force Commander in relation to peacekeeping troops. The effect of such a certificate has recently been considered by the International Court of Justice, which found that “When national courts are seised of a case in which the immunity of a United Nations agent is in issue, they should immediately be notified of any finding by the Secretary General concerning that immunity. That finding, and its documentary expression, creates a presumption which can only be set aside for the most compelling reasons and is thus to be given the greatest weight by national courts.”83

Thus, under international law, the national court has rather limited scope to depart from the finding of the UN that an individual’s act was official. However, it should be remembered that the other side of this coin is that where the immunity of an individual UN agent is successfully asserted, the UN is obliged under the General Convention to make provision for “appropriate modes of settlement” of the dispute.

83

See the Cumaraswamy Advisory Opinion of 29 April 1999, at para. 61.

18

Linking State Responsibility for Certain Harms Caused by Corporate Nationals Abroad to Civil Recourse in the Legal Systems of Home States MUTHUCUMARASWAMY SORNARAJAH 1

1 INTRODUCTION

is discussed in this chapter is whether a state incurs responsibility when its corporate national engages in torture or similar violations of human rights whilst operating abroad. It is a fact of power-centred international law that the focus of attention in issues relating to state responsibility involving nationals operating abroad has been on their protection from abuse by the host state. The law has by and large not sought to discuss the situation in which these nationals act in a manner that is harmful to the people of the host state. The law was developed in the context of flows of investments from developed to developing states. In that context, the focus has entirely been on the protection of the multinational corporation, which is often the vehicle of these investments, from the exercise of sovereign power of the host state. The fact that the modern multinational corporation is in itself a basis of global power and can hurt the interests of the host economy is seldom addressed in international law.2 The need to shift the focus away from the protection of the assets of multinational corporations to the protection of the people of the host state3 has

T

HE ISSUE THAT

1

Professor of Law, National University of Singapore, Singapore. It is evident that single individuals who control multinational corporations have considerably more power over world events than the heads of many developing countries. Yet, international lawyers do not even accept that multinational corporations have personality in international law. Such acceptance is necessary for the devising of norms relating to the control of their activities. In international relations literature, the role that multinational corporations play in world affairs is accepted and studied. S Strange, The Retreat of the State (Cambridge, Cambridge University Press, 1992); S Strange, Rival Firms and Rival States (Cambridge, Cambridge University Press, 1994). 3 The need for a creation of a people-centred international law for the future rather than retaining the state-centred international law of the past has been emphasised by some scholars and as fervently resisted by conservative groups in the international law establishment. The acceleration of concern with human rights and the environment will ensure that state-centric international law 2

492 M Sornarajah become apparent as a result of a series of recent incidents.4 These incidents have exposed the paucity of the reaction of power-centred international law to situations in which human rights and environmental norms are violated by multinational corporations operating in developing countries.5 Specific allegations of multinational corporations’ association with torture in host states continue to be made. There is a need for a policy response to this situation. Clearly the multinational corporation would be liable under the domestic law of the host state to the extent its conduct constitutes violation of that law.6 It is equally clear that there is jurisdiction for any domestic court of any state to prosecute the multinational corporation if the allegation relates to a violation of a jus cogens norm like the prohibition of torture.7 These propositions are now clearly established will suffer an eclipse, though the state will still have a role in providing regulation of activity of transnational actors in the globalising world. It is in the context of that role that the imposition of an active duty to control multinational corporations through the manipulation of the rules of state responsibility become relevant. 4 Starting with Bhopal, there have been an increasing number of major situations which have been recorded in the literature. For some other situations, see Human Rights Watch, The Price of Oil: Corporate Responsibility and Human Rights Violations in Nigeria’s Oil Producing Communities (New York, Human Rights Watch, 1999). And see G S Akpan, “Transnational Environmental Litigation and Multinational Corporations: A Study of the Ok Tedi Case” (Paper published by the Centre for Energy, Petroleum and Mineral Law, University of Dundee, Scotland, CP 11/98, 1998) and H M Osofsky, “Environmental Human Rights under the Alien Tort Statute: Redress for Indigenous Victims of Multinational Corporations”, (1997) 20 Suffolk Transnational Law Review 335. The issue is temptingly raised in F van Hoof, “International Human Rights Obligations for Companies and Domestic Courts: An Unlikely Combination?” in M CastermansHolleman, F van Hoof and J Smith (eds), The Role of the Nation State in the 21st Century: Essays in Honour of Peter Baehr (Boston, Kluwer, 1998) 47. 5 On the theme of power versus justice in contemporary international law, see M Sornarajah, “Power and Justice in International Law”, (1997) 1 Singapore Jl of International and Comparative Law 28. For early and more recent contributions to the increasing literature on human rights concerns with the operations of multinational corporations, see M Lippmann, “Multinational Corporations and Human Rights” in G W Shepaher and V Nanda (eds), Human Rights and Third World Development (Westport, Conn., Greenwood Press, 1985) and B A Frey, “The Legal and Ethical Responsibilities of Transnational Corprations in the Protection of International Human Rights”, (1997) 6 Minnesota J Global Trade 105. And further see D Weissbrodt and M Hoffman, “The Global Economy and Human Rights: A Selective Bibliography”, (1997) 6 Minnesota J Global Trade 189 and M A Geer, “Foreigners in Their Own Land: Cultural Land and Transnational Corporations—Emergent International Rights and Wrongs”, (1998) 38 Virginia J Int’l Law 331. For a denial of a linkage between human rights violations and multinational corporations, W H Meyer, “Human Rights and MNCs: Theory versus Quantitative Analysis”, (1996) 18 Human Rights Quarterly 368. A survey of the subject is to be found in M K Addo (ed.), Human Rights Standards and the Responsibility of Transnational Corporations (Boston, Kluwer, 1999). 6 Remedies under the local law, even if available, are at best tenuous as the regime which participates in the violations is unlikely to ensure that the victims receive relief even where remedies are available under local law. Though such remedies must be exhausted for an international claim to arise, denial of justice must be presumed where the local courts consist of appointees of the state which is hand in glove with the foreign investor. In the situation alleged in Doe v. Unocal, infra n. 60, it would be fruitless to seek redress through the domestic courts. 7 The Pinochet Case, [1999] 2 WLR 827. See in particular, the judgment of Lord Millett who argued for universal jurisdiction in situations of torture, even in the absence of any law incorporating conventional norms prohibiting torture in the domestic law. See also A Bianchi, “Immunity versus Human Rights: The Pinochet Case”, (1999) 10 EJIL 237.

State Responsibility for Harms by Corporate Nationals Abroad 493 and the principles involved have been adequately discussed in the other chapters in this book. What is argued in this chapter is that there is policy justification for holding the home state of the multinational corporation liable in such circumstances for not preventing its corporate national from engaging in such conduct, or after the event, for failing to provide redress through its courts. The effective imposition of responsibility on the home state will ensure that it will exercise the degree of control necessary to prevent its corporate national from abusing standards of international law.8 A law capable of preventing human rights abuses by multinational corporations will thus be created. Existing law premises protection of the corporate national itself on the linkage of nationality. This relationship is the basis of the rules of diplomatic protection for international wrongs to foreign nationals. The relationship assumes the capacity to control the corporate national and, in principle if not yet in law, a reciprocal obligation to ensure that the national acts in accordance with international norms.9 The fact that control over multinational corporations overseas is exercised by many home states is evident from their practice in extending their jurisdiction over certain types of activity through the extraterritorial application of their antitrust laws, export control laws and securities legislation.10 If such control exists, it is necessary to ensure that it is exercised to prevent harm to the people of the host state caused by the conduct of the multinational corporation.11 The nature of the control which attracts state responsibility in situation of liability for harm caused by nationals abroad has to be detailed. The notion of liability for extraterritorial harm is regarded as a new phenomenon, but in fact, as indicated below, there are old cases in which responsibility for the acts of nationals as well as others over whom a state had control in the territory of the state injured has been recognised. In some instances, the injury was to a third state. Clearly, such control exists where there is a link of agency12 or where there is knowledge in the home state of the injurious behaviour to the host state that 8 Control over the national remains in the home state even after he enters another state. R Y Jennings and A Watt (eds.), Oppenheim’s International Law, Vol. 1, 9th ed. (London, Longman, 1992) at 904 states this proposition as follows: “With his entrance into a state, an alien falls at once under its territorial supremacy, although he remains at the same time under the personal supremacy of his home state.” 9 The exercise of criminal jurisdiction on the basis of nationality is accepted in international law. Skirotes v. Florida, 313 US 69 (1941) (where Hughes CJ declared that the United States “is not debarred by any rule of international law from governing the conduct of its own citizen upon the high seas or even in foreign countries . . .”). Export control and other laws have been applied to the overseas conduct of corporate nationals. There is, however, selectivity in this exercise as the developed states have resisted efforts to have their environmental laws applied extraterritorially. 10 For the nature of the exercise of extraterritorial control, see generally, K von Meesen, Extraterritorial Jurisdiction in Theory and Practice (Boston, Kluwer, 1996). 11 M Gibney, K Tomasevski and J Vedsted-Hansen, “Transnational state responsibility for violations of human rights”, (1999) 12 Harvard Human Rights Journal 267. The possibility of suing states for sponsoring terrorism overseas was raised in Alejandre v. Cuba, 996 F.Supp 1239 (1997). 12 Zafiro Claim (1925), 6 RIAA 160 (where damage was caused to British property by a Chinese crew in the Phillippines under command of a US naval officer).

494 M Sornarajah the nationals had contemplated in the course of their ventures abroad.13 The element of control necessary in the situation of harm caused by the multinational corporation in its activity abroad is satisfied not only by the link of nationality and the ability to exert control through legislation, but also due to the fact that in modern times of fast communications, it is unlikely that the home state would not have had knowledge of the injurious activities, particularly if the home state is a powerful one with all the information networks within its command. Besides, there is the fact that the home state, to which the profits of the corporation are repatriated, stands to gain as a result of the activity and, hence, has some complicity in what takes place. The necessary degree of control to trigger responsibility can be satisfied in this situation. The need for such control to be exercised in the protection of host state interests is obvious. The home state permits flows of foreign investment into other countries with a view that profits made abroad will be repatriated. As the flipside of such expected benefits, it should owe an obligation to ensure that these profits are not made in violation of international standards or at the cost of harm to the host state, particularly a developing one.14 At the least, where profits are in fact made in violation of international law standards, the home state which profits from such deviation should owe a responsibility to the host state to pay reparation that bears some relation to those profits. The existence of such a responsibility will be an additional incentive to ensure that home states exercise the necessary control to prevent corporate violations of human rights norms. The home states of multinational corporations, being wielders of significant power, have the resources to ensure that the multinational corporations act in accordance with international norms in matters of human rights.15 The propositions advanced in the above paragraph require an argument that links existing features of the law on state responsibility with other newly developing features of international law. The notion of a link between the home state and the multinational corporations is an existing one, for it is on the basis of this link that diplomatic protection is offered by the home state to the multinational corporation when it suffers injuries at the hand of the host state. The link obviously involves control. Such control has not been denied by home states as there 13 The Alabama Claims Arbitration (1872), 1 Moore 495. The activity of building a vessel for use in war was of a private nature but the British argument that it could not control private activity was rejected. In both instances, the injury caused was extraterritorial. 14 Policy arguments, can be made, namely that the hindrance caused to economic development justify that the harm should be borne by the home state which is better able to withstand the harm, which are relevant to the situation. 15 The existence of a moral responsibility is evidenced by voluntary non-binding codes which associations of states like the OECD have adopted. OECD, “Guidelines for Multinational Enterprises”, (1976) 15 ILM 967. Corporations now make their own internal codes as well. But voluntary codes are commonly regarded as efforts to siphon off the steam generated in favour of making of binding codes recognising legal responsibility in multinational corporations. For the creation of personality and consequently responsibility in multinational corporations, see R Higgins, “A Multinational Corporation or an International Organisation?” in R Morgan (ed.), New Diplomacy in a Post-Cold War World: Essays for Susan Strange (New York, St Martins Press, 1993).

State Responsibility for Harms by Corporate Nationals Abroad 495 are obvious instances of such controls where foreign activity of the multinational corporation contravenes the home state’s national interests.16 The step that needs to be taken is to establish that there is a positive duty in international law to use this control to prevent activity by a multinational corporation which violates jus cogens norms involving violations of human rights or environmental standards. This step is easily taken. The jus cogens norms are binding on the home state. It therefore creates a duty to use the control it has over its citizens both real and corporate, to prevent the violation of the jus cogens norms, even while they are overseas. State responsibility arises where there is a failure to conform to this duty.17 Additionally, after the event, there is a duty on the part of the courts of the home state to provide relief to the victims of the violation. They have jurisdiction over the multinational corporation on the basis of nationality and, being an arm of the state which had a duty of control, they have a duty to reduce the damage that was done by providing prompt relief.18 The failure to fulfil this duty should give rise to state responsibility. There are sufficient policy justifications for the imposition of state responsibility on the home state for the conduct of its multinational corporations on the basis of the above propositions. An influential American school of international legal thought, the New Haven or Yale School, believes that the law must be shaped in order to further policy goals as common understanding of those goals evolves.19 The existing international law on state responsibility contains a sufficient reservoir of rules and principles that can be shaped through interpretation to respond to the growing knowledge about, and reality of, harm caused by multinational corporations. Power-based international law has been reluctant to give the law such shape, having been to date more concerned with developing rules on the protection of the multinational corporations and their assets. Another policy justification is the rise of the movement in international law away from fashioning state-centred rules towards the making of an international law that assures the protection of the rights of peoples. Though this idea is resisted by conservative forces,20 the vigour of the espousal of the rights of aboriginal peoples, the rise of the role of non-governmental organisations 16 The obvious example is presented by export controls laws which prevent even subsidiaries of multinational corporations operating abroad from exporting to enemy states. 17 The duty to control even groups over which there is control was recognized in older international law. Thus, in the Zafiro Claim, supra n. 12, the claim was made in respect of the crew of a private ship under the control of an American officer which damaged British property in the Philippines. Liability of the United States was based on the ground that there was a failure to exercise due diligence under the circumstances. The case is not on all fours with the situation under discussion as there was no civil authority within the Philippines at the time of the incident, but this would not have made a difference. The ability to control and the failure to exercise that ability is the main factor. 18 Much in the same way as the host state escapes state responsibility by providing relief to the alien through its domestic courts, the home state should also be afforded this opportunity. This is the basis of the local remedies rule. 19 For a work using the methodology in a modified form, see R Higgins, Problems and Process: International Law and How We Use It (Oxford, Oxford University Press, 1994). 20 See J Crawford (ed.), The Rights of Peoples (Oxford, Oxford University Press, 1990).

496 M Sornarajah and the recognition of environmental and social rights will accelerate the move of international law to a people-centred discipline.21 This movement will require the recognition of the duty of multinational corporations to desist from violations of international law in their operations in host states. But, in addition, it will require the recognition of the duty of their home states to ensure that they do operate in conformity with international standards in the human rights and environmental spheres. However, the impetus of the environmental and human rights movements has now shifted emphasis away from rules of protection of multinational investment which has so far been the focus of the law on state responsibility.22 Strong evidence for this claim can be found in the fact that the OECD’s proposed Multilateral Agreement on Investment (the MAI), which sought to create a regime for worldwide investment protection, ended in dismal failure. This was due to the concern of non-governmental groups which argued that the MAI did not adequately address environmental and human rights deprivations caused by multinational corporations. The protests by various bodies, taken up later by some governments, emphasised the need to build corporate obligations and not just corporate rights into the agreement. The massive protests at the Seattle Ministerial Meeting of the World Trade Organisation in December 1999 also evidenced similar concerns. It is thus an opportune time to address not only direct corporate responsibility but also the indirect responsibility of the home state for failing to prevent the harmful conduct of its corporate national in a host state. In particular, the link between the two needs to be explored, notably by elaborating the basis for placing an obligation on each home state to ensure that its domestic courts will provide relief to the victims of the multinational corporation’s conduct. Failure to allow for such a regime of civil liability may again involve the responsibility of the state beyond any obligation it owes by virtue of its failure to prevent conduct which it could have controlled. There are thus two separate issues to be discussed in this chapter. The first is the extent to which the home state of the multinational corporation may incur responsibility for the conduct of its corporate national in a host state. Such responsibility, it is argued, arises from the failure to exercise the power of control over its corporate national and avoid the harm to the host state.23 The second issue relates to whether there is an additional state responsibility which arises upon the failure of the domestic courts of the home state of the multinational corporation to provide a remedy for the victims of the human rights 21 C Grossman and D Bradlow, “Are We Being Propelled Towards a People-centered Transnational Legal Order?”, (1993) 9 American University Jl of International Law and Policy 125. 22 The major part of the law on state responsibility developed in the context of protection of alien property. Though the International Law Commission has striven mightily to draw attention away from this fact in its Draft Code on state responsibility, every undergraduate text devotes most of its pages to the protection of alien property. 23 The power of control arises because of the ability of the home state to exercise extraterritorial control over its nationals and also exercise control over its national at the time the national leaves the territory of the home state.

State Responsibility for Harms by Corporate Nationals Abroad 497 violations in which the multinational corporation had participated. The argument as to the second category is that it is precisely because state responsibility of the first kind could arise that there is a duty on the part of the domestic courts of the home state to provide a remedy against the multinational corporations in these circumstances. The two situations will be dealt with in turn. 2 STATE RESPONSIBILITY FOR FAILURE TO CONTROL THE MULTINATIONAL CORPORATION IN RELATION TO HUMAN RIGHTS VIOLATIONS

The Existing Rules on State Responsibility The right to protect nationals abroad, including (indeed, especially) corporate nationals, has been the focus of attention in the literature on international law on state responsibility.24 It has not focused on the protection of the national of the host state from the abuse of its power by the multinational corporation. This gap has existed in spite of the fact that, in historical terms, the need for protection in the latter situation was more evident. The British and Dutch East India Companies and their likes plundered and pillaged the natives of Asia and Africa with reckless abandon. The native people of Latin America, Australia and elsewhere were brought close to extinction so that their lands could be utilized for mining and other purposes by foreign investors. But international law, which the European states fashioned, was more concerned with the protection of the interests of these corporations and foreign investment interests than with the protection of their victims. An elaborate structure of rules permitting states to espouse the claims of the foreign investors and requiring the payment of prompt and full compensation was progressively built up.25 There is of course a power-based explanation for this situation.26 The powerful states fashioned rules in order to ensure the protection of the assets of 24 As noted supra n. 22, despite the efforts of the International Law Commission to shift the focus away to other areas, much of the law on state responsibility developed in the context of alien protection. Leading statements of the law on state responsibility for injuries to aliens can be found in C F Amerasinghe, State Responsibility for Injuries to Aliens (Oxford, Oxford University Press, 1965) and R B Lillich, International Law of State Responsibility (Charlottesville, University of Virginia Press, 1983). The earlier drafts on state responsibility prepared by Garcia-Amador concentrated on alien protection which was and is a divisive subject in international law. See F V Garcia-Amador Changing Law of International Claims (New York, Oceana Publications, 1984). 25 Its classic statement is in the Mavromattis Palestine Concessions Case [1924] PCIJ Rpts, Series A, No.2, p.12. The Court said: “By taking up the case of one of its subjects and by resorting to diplomatic action or international judicial proceedings on his behalf, a State is in reality asserting its own rights— its right to ensure, in the person of its subjects, respect for the rules of international law.” The converse of this is that a developing state should be able to assert its right of protection of its nationals when an alien causes damage in its state and its nationals are not provided relief in the home state of the alien in which his assets are situated and to which he has repatriated profits of his operations in the host state. 26 The role of power in rule formation in international law is coming to be studied in recent times. The work of the developing country international lawyers in this area predates the new inquiries that are being made. R P Anand, International Law and Developing Countries (1987); J T Gathii, “International Law and Eurocentricity”, (1998) 9 EJIL 184. M Sornarajah, “Power and Justice”, supra n. 5.

498 M Sornarajah multinational corporations and expatriate plantation owners in the weaker states of the world. No inquiry was made as to whether the exercise of this protection should be conditional on the conduct of the foreign citizen or corporation in the host state. An instructive case for purposes of illustration is the Schufeldt Claim.27 Here, the foreign national had extracted chicle28 in Guatemala using injurious techniques of bleeding the trees. The host state took the position that the trees would have been destroyed if the techniques were continued. The measure taken was for the protection of the industry. The host state advanced this as a reason for the termination of the concession agreement to extract chicle. When the matter went to interstate arbitration between Guatemala and the United States, the damage caused by the use of injurious techniques was disregarded even in the calculation of damages. Instead, the arbitrator granted damages, among other things, with touching solicitude for the “anxiety of mind” caused to the alien. There was much concern for the comfort of the foreign investor, none for the host state or its people. However, such instances of unconcern for the interests of the developing host state led to this power-based notion of state responsibility being resisted by the developing states in several ways. As a result, there has been significant erosion of the norms contended for by the developed states.29 Yet, the idea of the responsibility of the multinational corporation or its home state for the damage it may cause to the host state or its people has yet to be fully explored, even in a rhetorical fashion.30 It may be useful to reflect on the possibilities within current international law with some idea of how international law looked in times past. Prior to the ascendancy of the power-based notion of responsibility, it would have been easier to establish that there was a duty to ensure that a national did not act to the detriment of the host state whilst abroad. The right of protection of nationals abroad was conditional on their good conduct.31 There was a forfeiture of protection where a national engaged in censurable conduct abroad. Generally, states

27

(1930) RIAA 1079; (1930) 24 AJIL 799. Chicle is defined in the Oxford English Dictionary as “a gum-like substance obtained from the bully tree (Mimusops Globosa) largely used in the United States for the manufacture of chewing gum”. 29 In the area of foreign investment, many of the norms advanced by the developed states were dented by the forceful articulation of counter-norms by the developing states. For a study, see M Sornarajah, International Law on Foreign Investment (Cambridge, Cambridge University Press, 1994). 30 This is not to say that this has not been explored to some extent in the evolving responses to global environmental problems. See for example the Basel and Bamako treaties on transfer of hazardous wastes, infra n. 43 and text following. Yet, the treatment of state responsibility in the literature seldom addresses anything but conventional issues of state responsibility of the Trail Smelter variety. See e.g. K Zemanek, “State Responsibility and Liability” in W Lang, H Neuhold and K Zemanek (eds), Environmental Protection and International Law (London, Graham and Trotman, 1991) 187. 31 The older cases are discussed in E M Borchard, The Diplomatic Protection of Citizens Abroad (New York, Kraus Reprint Co., 1970; original edition 1915) 718–20; see also J Brierly, “The Theory of Implied State Complicity in International Claims”, (1928) 9 BYIL 42. 28

State Responsibility for Harms by Corporate Nationals Abroad 499 recognised this by not interfering in support of a citizen who had done wrong whilst abroad.32 This was a significant acknowledgement of a link between state responsibility and the conduct of the foreign national, but was there an active duty to prevent the nationals from engaging in censurable behaviour? There is, indeed, authority for the view in older international law that a state becomes responsible to the state which is harmed if the state responsible for the harm was aware of the intention of the national to commit injurious acts against another state and does not prevent him, her or it from doing so. So too, there was a duty on the part of the state to punish a national who committed a crime whilst abroad or to hand him or her over for punishment. Failure to do so amounted to a tacit approval of his, her or its conduct which triggered ex post facto responsibility of the state. Judge Jimenez de Arechega and A. Tanzi stated the position thus: “The State which becomes aware that an individual intends to commit a crime against another state or one of its nationals, and does not prevent it or the state which extends protection to the offender by refusing to extradite or punish him, gives tacit approval to his act. The State thus becomes an accomplice in his crime and establishes a link of solidarity with him: from such relationship the responsibility of the State arises.”33

Although the Arechega and Tanzi statement is made in reliance on Grotius, there is other, more positivistic, authority which also supports their argument. For example, the United States argued for such a rule in the well-known Alabama Claim in which it successfully claimed compensation in respect of a confederate warship built in Britain.34 There are also a considerable number of older arbitral awards which recognise similar claims based on responsibility of a state for not being diligent in controlling the acts of its nationals abroad.35 Borchard distils the following rule from these arbitrations: “A long line of cases has established certain qualification upon the non-liability of the government for the wrongful acts of private individuals. These consist in certain manifestations of the actual or implied complicity of the government in the act, before or after it, either by directly ratifying or approving it, or by an implied, tacit or constructive approval in the negligent failure to prevent the injury, or to investigate the case, or to punish the guilty individual, or to enable the victim to pursue his civil remedies against the offender.”36 32 Thus in the Pelletier Claim (1887) discussed in Borchard, supra n. 31 at 717, the United States refused to espouse the claim of a man shown to have been guilty of slave trading in Haitian waters. Slavery is subject to universal jurisdiction as is torture. The refusal of diplomatic protection and the trial of the national for the offence may be mandated in these circumstances by modern law. Where there was taking of property of an alien in execution of a criminal fine, this was considered a noncompensatory taking. Again, the idea is that the alien loses protection in these situations. 33 J de Arechega and A Tanzi, “International State Responsibility” in M Bedjaoui (ed.), International Law: Achievements and Prospects (Dordrecht, Martinus Nijhoff, 1991) at 359. They also suggest that “this theory of state complicity was adopted in several arbitral awards of the nineteenth century and by important writers on the subject”. 34 Supra n. 13. 35 These are stated in Borchard, supra n. 31, at 217. 36 Borchard, supra n. 31.

500 M Sornarajah This law as summarised by Borchard underwent a change as power-based and state-centred positivism gained sway over international legal thought. It came to be understood that the state could only be held responsible for the acts of its own organs and not for those of its nationals.37 In other words, the approach, as described by Borchard, of treating states as having a species of direct responsibility through implied approval disappeared as a general principle of attribution. The modern position is stated in Article 11 of the International Law Commission (ILC) Draft Articles on State Responsibility in the following terms: “The conduct of a person or a group of persons not acting on behalf of the State shall not be considered an act of the State under international law.”38

But, significantly, another draft article (article 8) does retain some notion of this older idea of the possibility of attribution of nationals’ conduct to the state itself. This article contemplates liability for the acts of persons if they were “in fact acting on behalf of the State” or where they were “exercising elements of governmental authority in the absence of official authorities and in circumstances which justified the exercise of those elements of authority.” It may be possible to argue that the rules stated by Arechega and Tanzi, and Borchard, could be interpreted as having been retained by articles 11 and 8 in combination, but there is no indication that the situation was even contemplated in the ILC drafting process with respect to those two draft articles. The ILC was quite content to go along with the power-based statement which was more concerned with the protection of aliens from injuries by the host state. That being said, the Draft Articles do not speak to substantive rules of law but, rather, only to framework principles of responsibility. Thus, it is generally considered that, even as the principle of attribution stated by Borchard was expunged from the new understanding of the general law of state responsibility, substantive international law continued to lay down positive duties with respect to the protection of foreigners in the classic type of situation in which there is mob violence directed at the alien which the host state did not prevent through due diligence.39 What we now tend to think of as this “classic” due diligence rule ceased being a general principle of attribution in the law of state responsibility (according to which negligent control and failure to provide a process for civil remedies was viewed as creating a kind of direct agency relationship between the home state and its national). It was transformed into a specific positive obligation of one branch of substantive law (the law dealing with protection of host states of foreign nationals). In the process, positive duties on the home state of these foreigners dropped out of the legal picture—at least in power-based, mainstream international legal thought. 37 See e.g. ibid. There are also cases which recognized the liability for the acts of brigands abroad on the basis that there was a duty on the home state to suppress brigandage. 38 ILC Report, Draft Articles on State Responsibility, (1996), GAOR., 51st Session., Supp. 10, p. 125. 39 The Iran–US Claims Tribunal dealt with many such situations. See Short v. Iran (1987) 16 IranUS CTR 76; Yeager v. Iran, (1987) 17 Iran–US CTR 92.

State Responsibility for Harms by Corporate Nationals Abroad 501 The unprincipled and indeed arbitrary nature of this transformation is apparent. That being so, there would seem to be good reason to argue that substantive international law should be susceptible in light of a policy-based understanding of international law to interpretative development by analogy.40 While we are quite distant from the old idea of a form of direct responsibility of states for the conduct of nationals abroad, there is no reason that present international law cannot be shaped to extend the substantive duties applicable to the host state of a corporation to the home state of the corporation where the requisite elements of knowledge and control are present. We should thus not be surprised to find that, recently, challenges to the notion of the one-sided diplomatic protection of the alien through international law despite the atrocious nature of his, her, or its conduct have (re)emerged. The idea is back on the table that the home state of a foreign citizen, quite apart from protecting the citizen or corporation, incurs responsibility if it does not prevent him, her, or it from engaging in injurious conduct whilst abroad.41 However, the creation of such responsibility has so far been effected either by way of treaties dealing with specific subject-matter (such as the Basel Convention on the Transfer of Hazardous Waste) or unilaterally accepted by some states concerned to regulate the conduct of their corporations abroad. The case for the existence of such liability in general international law is what this chapter seeks to establish.42 We can gain a better idea of how easily general 40 The inadequacies in the treatment of the ILC’s definition of attribution in the human rights area are dealt with in another context in R Lawson, “Out of Control, State Responsibility and Human Rights: Will the ILC’s Definitions of the Act of State Meet the Challenges of the 21st Century?” in M Castermans-Holleman, F van Hoof and J Smith (eds.), The Role of the Nation State in the 21st Century: Essays in Honour of Peter Baehr (Boston, Kluwer, 1998) 91. 41 Dicta in the Nicaragua Case, (1984) ICJ Reports p. 4, indicate responsibility of a state for acts of persons wholly unconnected with the state if there was control over these persons. Where there is such control, the persons would be equated with an arm of that state’s government. The Court said (para. 109): “What the Court has to determine is whether or not the relationship of the contras to the United States Government was so much one of dependence on the one side and control on the other that it would be right to equate the contras, for legal pursposes, with an organ of the United States Government, or as acting on behalf of that Government”. On the basis of this formulation, the existence of control over multinational corporations acting within the home state to the detriment of the interests of the host state or its people could engage the responsibility of the home state. In the human rights context, the principle was extended in Bosnia Genocide Case, ICJ Reports (1996) 595 where it was suggested by implication that a state becomes responsible if an individual who had committed a violation of the Genocide Convention was under the control of that state. For such a reading see S Rosenne, “State Responsibility and International Crimes: Further Reflections on Article 19 of the Draft Articles on State Responsibility”, (1997) 30 NYU J Int’l Law and Politics 145 at 159. And see the dicta in the judgment of the Internernational Tribunal for the Former Yugoslavia in the Tadic Case, (1997) 36 ILM 908 at 933 contemplating the Yugoslav state liability for atrocities committed by Serbs in Bosnia. 42 There is no doubt that where the state sends its nationals on a mission abroad to inflict damage on another territory, even if the damage is directed at a third party, there is liability. Rainbow Warrior Case, (1987) 26 ILM 1346. With respect to the Lockerbie case, the specific allegation has been that agents sent by Libya had caused the airliner to explode. In both instances, the incidents took place elsewhere but the acts were of agents of the state whose responsibility was alleged. Agency rather than nationality and the intention with which the act was done are the factors in the attribution of responsibility. See also the Nicaragua Case, supra n. 41, for a norm of near-agency for

502 M Sornarajah international law can embrace home state responsibility by noting some specific instances of the acceptance of such responsibility. Examples of treaty acceptance of such responsibility are to be found especially in the environmental sphere. Thus, for example, the treaties controlling the transport of hazardous waste impose a duty on states to prevent such transport to other states by nationals.43 Here, a clear treaty obligation arises not to permit citizens and corporate bodies to export hazardous waste to other countries.44 The law is no longer based on the premise of sovereignty that a developing state can choose to avoid the harm by refusing to accept the waste. It imposes an active duty on the exporting state to ensure that hazardous waste is not sent to other states.45 But these treaty obligations still seem to be territorially linked.46 The absence of territoriality, however, is not a problem in emerging international law of human rights which recognises its basic obligations, such as the prohibition of torture, as being owed to the whole of humanity.47 The meaningfulness of erga omnes obligations rests on the idea that state responsibility exists for the violation of any of those obligations which can be directly or indirectly traced to the acts or omissions of the state. The emergence of a duty to prosecute criminals in international law attests to the fact that the place of commission of the violation of the obligations is increasingly becoming insignificant.48 One may argue that the duty becomes more easily established in situations where the criminal is a national of the state. Correspondingly, there is an increasing recognition of a moral, if not yet a legal, duty to prevent the commission of harm injurious to the interests of other states and their peoples. Such a duty is often unilaterally assumed. purposes of direct attribution of conduct of rebels in one state to another state supporting those rebels. 43 The Basel Convention on the Control of Transboundary Movement of Hazardous Wastes and Their Disposal, (1989) 28 ILM 657 (in force, May 1992). See also K Kummer, International Management of Hazardous Wastes: The Basel Convention and Related Legal Rules (Oxford, Oxford University Press, 1995). For responsibility of flag states for discharge of waste by ships on the high seas, see B D Smith, State Responsibility and the Marine Environment (Oxford, Oxford University Press, 1988). 44 See generally R Risillo-Mazzeshchi, “Forms of International Responsibility for Environmental Harm” in F Francioni and T Scovazzi (eds), International Responsibility for Environemental Harm (Deventer, Kluwer Academic Publishers, 1991) and R Lefeber, Transboundary Environmental Interference and the Origin of State Liability (Deventer, Kluwer Academic Publishers, 1996). 45 See the Fourth ACP–EEC Convention of Lome (1989) and the International Atomic Energy Agency’s Code of Practice on the International Transboundary Movement of Radioactive Waste (1990). Texts of both instruments are in B Kwiatkowska and A H A Soons (eds), Transboundary Movement and Disposal of Hazardous Waste in International Law (Boston, Kluwer, 1993). 46 The obligation is on the state to prevent physical export of material from its territory, not to prevent export by a national who operates overseas. This is done through a licensing system which requires prior informed consent of the state to which the export is made to be obtained. 47 M Raggazi, The Concept of International Obligations Erga Omnes (Oxford, Oxford University Press, 1997). 48 G S Goodwin-Gill, “Crime in International Law: Obligations Erga Omnes and the Duty to Prosecute” in G S Goodwin-Gill and S Talmon, The Reality in International Law: Essays in Honour of Ian Brownlie (Oxford, Oxford University Press, 1999) at 199.

State Responsibility for Harms by Corporate Nationals Abroad 503 The best example of this would be the legislation recently enacted in some states to prevent the organisation of sex tours by their nationals to exploit child prostitutes in foreign states where the problem has become endemic and difficult to control. Here, the problem can only be controlled if both the state from which the tourists originate and those in which child prostitution exists take measures. States like the United Kingdom, Canada and Australia have voluntarily assumed the responsibility of preventing their nationals making such sex tours abroad. Though all these states’ statutes contemplate punishment after the event, some also include a duty to prevent the tours taking place through denial of passports and other means.49 The state has the means of preventing harm being caused to other states and persons living within other states. There is a duty to prevent such harm. The duty arises not only in terms of morality but in terms of law. Here, an extraterritorial situation is being controlled by regulating nationals’ conduct abroad. The moral responsibility to control such activity taking place abroad is accepted by the state, which uses nationality as the basis for imposing liability on its citizens.50 It may be possible to argue that there is not only a moral, but also a legal duty to take such measures. Beyond the harm caused to children through prostitution itself, the sex tourist is, in the context of sexually-transmittable diseases, like a circulating time bomb whose conduct can result in harming a wider circle of people than the children with whom he has “sex.”51 There is a duty on the part of his state, which alone has the best means of ascertaining his condition and his proclivities, to prevent the circulation of the person in other states that are least able to deal with the problem.52 The liability of tourist corporations which knowingly assist in such tours is already provided for in some legislation. The liability of airlines which knowingly carry such tourists must be contemplated. 49 Compare the situation of the rugby tours to South Africa during the apartheid regime. It was argued that the state should use its prerogative powers to prevent its national teams from travelling for this purpose. 50 Compare the refusal by the United States to issue passports to Mormon missionaries as they would preach polygamy abroad, noted in E M Borchard, Diplomatic Protection of Nationals (1904). 51 The link between torture and rape is increasingly being made. Sex with children is rape. There would then, in the context of a jus cogens principle, be a universal obligation to prevent sex tourism involving paedophiles. In terms of the seriousness of the harm to children, it may be useful, in light of the thematic focus of this book on torture as an example of serious human rights violations, to recall the recent confirmation by the European Court of Human Rights that rape constitutes a form of torture, at least for women in state custody: Aydin v. Turkey (European Court of Human Rights, judgment of 25 September 1997). Whether or not private acts of “sex” are to be considered torture under prevailing international law definitions of torture, the important point is that the kind of abuse of power and associated harm is closely analogous when one considers the conditions under which many children “choose” to prostitute themselves to foreign tourists. Not only does socioeconomic reality and their stage of mental development make their choice illusory but we should also not forget the jail-like circumstances of much prostitution. 52 The analogy with the reasoning in the Corfu Channel Case is clear: (1949) ICJ Reports, p. 4. In respect of the mining of the Corfu Channel and responsibility for the damage to vessels as a result of such mining, the ICJ held that Albania, the coastal state, was responsible under the law. See further M Sornarajah, “South East Asia and International Law—State Responsibility and Sex Tours in Asia”, (1997) 1 Singapore Jl of International and Comparative Law 414 at 422.

504 M Sornarajah But, what is more important is that states which enact such legislation accept that they have a responsibility to prevent the occurrence of such incidents which directly involve harm to the host state and its people. One may argue that this may indicate a moral responsibility and not a legal one, but in situations where the harm prevented is recognised as prohibited by a jus cogens principle, it is credible to argue that the responsibility is a legal one. The above examples53 clearly demonstrate that, in certain areas, the assumption of responsibility by states for the conduct of their nationals abroad is coming to be accepted. The argument may now plausibly be made that such responsibility must, as a matter of general international law, be recognized in respect of violations of jus cogens principles like torture, in two situations.54 The first is where a state knowingly permits its nationals to engage in violations of jus cogens principles whilst abroad. The second is where a state gives active assistance to those who are known to violate or are seen as capable of violating such jus cogens principles. There are conditions to be satisfied in the traditional law of state responsibility before any liability can be imputed to the state. The manner of the satisfaction of these rules in the situation of imposing responsibility for the acts of multinational corporations is now canvassed. With respect to the first situation, these rules require the existence of a duty and ability to control the national engaging in the proscribed behaviour and the failure to satisfy the duty. The elements of this duty to prevent are the subject of section 3 below. With respect to the second situation, it requires the existence of a link that makes attribution of the conduct of the multinational to the state possible. It will be argued in Section 4 below that the home state should be deemed to have implicitly adopted or ratified the conduct of the corporate national if it fails to provide for some system of civil remedies for the victims of corporate harm abroad. The lack of legal protection created by this failure should be considered as the requisite active assistance or link needed for attribution—the kind of “implicit, tacit or constructive approval” spoken of by Borchard. An alternative argument has responsibility of the home state being created on the basis of human rights law. There is a definite commitment on the part of all states to prevent the violation of human rights that take place not only within their territory, but anywhere. The statements of the rights in the major conventions are not territory-specific.55 They also provide for liability in states not only 53 There are other examples besides the one indicated here. Thus, the legislation to prevent bribery abroad by nationals may be a similar instance. Here again, there may be slow evolution of international norms prohibiting bribery which may make it incumbent on home states to act to prevent it. P M Nichols, “Regulating Transnational Bribery in Times of Globalisation and Fragmentation”, (1999) 24 Yale JIL 257 54 There is overwhelming authority for the view that torture is a violation of jus cogens principles. See e.g. references in S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, chapter 14 in this volume. 55 E.g. Article 3 of the International Covenant on Civil and Political Rights (ICCPR) creates a general duty to “ensure the equal right of the men and women to the enjoyment of all civil and political rights set forth in the present Covenant” whereas Article 2 contains a statement that some see as territory-specific, but which the body overseeing the ICCPR, the Human Rights Committee, has

State Responsibility for Harms by Corporate Nationals Abroad 505 in respect of violations which they sponsor, but also for violations by private persons whose acts have been condoned or encouraged by the states.56 The argument is possible that a state whose multinational corporations are known to participate in violations of human rights such as torture condone such violations as a matter of state policy, particularly if they do not take any measures to prevent them. A state that profits from the repatriation of the profits which the multinational corporation makes must be credited with the duty to ensure that such profits are made without mass violations of human rights.

3 THE DUTY TO CONTROL NATIONALS ABROAD

The principles that potentially apply to control of nationals’ conduct abroad are those relating to responsibility in relation to the acts of private citizens that the state was under a duty to control. As noted earlier, this category has grown up largely in the context of cases involving mob violence against a foreigner within the territory of the host state.57 The law relating to liability for agents sent out to commit acts of sabotage is also relevant. It is the combination of these two sets of rules which will provide an answer to the issue of home state liability for the acts of a multinational corporation which is associated with torture in a host state. Before getting to the question of state obligations to control nationals abroad, it should first be noted that jurisdiction (i.e. the power or liberty) of a state to control nationals abroad exists in international law. It is relatively common for some legal systems, notably those of civil law countries, to make it a violation of the country’s criminal code for a national to commit serious common crimes found is not limited to conduct of a state in its own territory and can include conduct in another state’s territory: see Communication No. 56/1979, Lilian Celiberti de Casariega v. Uruguay, Yearbook of the Human Rights Committee 1981–1982, Vol. II, 1989, pp. 327–9, esp. at paras 10.1–10.3 (arbitrary detention and torture by Uruguayan agents in Brazil). 56 See Reporters Notes to Article 702 in American Law Institute, Restatement of the Law (Third): The Foreign Relations Law of the United States (St Paul, Minn., American Law Institute Publishers, 1987) Vol. 2 at 167. The notes contemplate the possibility of avoiding responsibility by providing domestic remedies against violations. 57 This area of the law was recognized in modern times in the Rainbow Warrior Arbitration, supra n. 42 and the Nicaragua Case, supra n. 41. In the Rainbow Warrior Arbitration, a Dutch citizen was killed when French agents sank the boat named “Rainbow Warrior.” The damage was to an alien life and property by aliens sent into New Zealand as agents by France. A former New Zealand Minister said: “What New Zealand was saying to France on this matter was, in effect, that it was a political imperative that decent arrangements be made for compensation for damage suffered in New Zealand, but not by New Zealand.” France paid compensation to the family of the Dutch national killed. G Palmer, “Settlement of International Disputes: The Rainbow Warrior Affair”, [1989] Commonwealth Law Bulletin 585. In Letelier v. Chile, 488 F.Supp 665 (1980), state responsibility (of Chile) was recognised for the act of its agent in setting off a car bomb which killed a former Chilean ambassador in the US. In Al-Adsani v. Government of Kuwait, (1994), 100 ILR 465, the English court recognized that Kuwait could be liable for acts of its agents in Britain if they had caused personal injury to any British resident. In the result, sovereign immunity prevailed due to the alleged acts having occurred outside Britain.

506 M Sornarajah abroad. Common law jurisdictions tend, as a general rule, to prescribe criminal law rules on a territorial basis.58 However, it is far less common for states, whatever the nature of their legal system, to legislate extraterritorially outside the criminal law, with the US probably being the state most inclined to create legislated exceptions to the presumption of the territorial application of its law. But even the United States has been selective in the exercise of this jurisdiction. So far, it has exercised it when it has been advantageous to its interests.59 It has not exercised such jurisdiction on purely altruistic grounds. Claims before courts that environmental legislation applies to the operation of American companies abroad have uniformly failed, the courts interpreting the legislation as having only a territorial focus. It is true that the Unocal case establishes that judicial jurisdiction to entertain suits against corporations, alleging their participation in torture with a government against its own citizens, exists in domestic courts.60 But, in terms of the jurisdictional foundations of the Unocal case, it proceeded largely on the basis that there is universal jurisdiction over jus cogens violations of international law such as torture and slavery and not on the basis of nationality, although Unocal is an American corporation. The fortuitous existence of a statute according the US courts jurisdiction, the Alien Tort Claims Act, facilitated the court’s assumption of jurisdiction over corporate conduct abroad amounting to breaches of the “law of the nations.” Such legislation does not exist in other states which are home states of multinational corporations.61 The issue is whether a state’s capacity to control its nationals in respect of their activity abroad must be exercised by the state when it has knowledge that harm could eventuate to another state or its people. Phrased in this way, it can be seen that the only link that is missing is the absence of a clear positive duty to exercise the jurisdiction to control the conduct of nationals. It is clear that responsibility will arise if there are specific instructions given to nationals to 58 C Blakesley, “United States Jurisdiction over Extraterritorial Crime”, (1982) 73 J Criminal Law & Criminology 1109. 59 For example, the United States’ export control legislation applies to corporations operating overseas. Some tax legislation also applies, but the American courts have consistently held that most environmental legislation does not have extraterritorial application. 60 Doe v. Unocal, 963 F.Supp. 880 (1997); Doe v. Unocal, 27 F.Supp 2d 1174 (1998). The litigation involved claims that the imposition of forced labour, rape and violence on indigenous people in the remote regions of Burma through which Unocal was constructing a pipeline involved liability of the Burmese state as well as Unocal. The responsibility of the state was alleged on the basis that the army was involved in the atrocities, and that of Unocal flowed from complicity in the atrocities. See also for the description of litigation in respect of the massacre in East Timor brought in Boston against an Indonesian military officer who had come to the US to study management at Harvard; R Clark, “Public International Law and Private Enterprise: Damages for a Killing in East Timor”, (1996) 3 Australian Journal of Human Rights 21. And see the discussion of Todd v. Panjaitan, No. 92-12255, 1994 WL 827111 (D. Mass. 26 Oct. 1994) in J Terry, “Taking Filártiga on the Road: Why Courts Outside the United States Should Accept Jurisdiction Over Actions Involving Torture Committed Abroad”, chapter 4 in this volume, text at n. 18 et seq. The action was based on the Alien Tort Claims Act (1789) and the Torture Victim Protection Act (1992). For environmental litigation with a human rights dimension, see Beanal v. Freeport-McMoran, Inc., 969 F.Supp. 362 (1997). 61 A point made by the English courts in another context in Al-Adsani, supra n. 57.

State Responsibility for Harms by Corporate Nationals Abroad 507 engage in the harmful behaviour abroad.62 Such instructions in effect create an agency relationship such as contemplated by draft article 8 of the ILC Draft Articles on State Responsibility, resulting in the nationals acting on behalf of their state. But, where such instructions are absent, can a state’s mere knowledge of the harmful behaviour of its national be sufficient to trigger responsibility if the state fails to do what is necessary to cause the harmful conduct to cease?63 Finally, how far do the home state’s positive obligations go in situations where it is not alleged that the national’s employees are the direct agents of harm, but rather that the national is benefiting from a foreign state’s human rights violations which have been committed in order to advance a joint project involving both the national and that state? In other words, in a situation like the one in Unocal, is there a duty on the part of the United States to prevent Unocal acquiescing in the Burmese government’s torture and rape of local people as alleged in that case? The argument is that such a duty can be constructed under present international law. As already indicated, the intentional sending of nationals or agents abroad to cause harm involves state responsibility. To extend this to situations where there is actual or constructive knowledge that harm will be caused by nationals requires but a small leap in the law. This leap is justified by existing doctrines. There is a general duty in international law not to cause harm to other states. Where a state knows that its national’s activities will cause, or are causing, harm to other states or peoples, it is consistent with this duty that it should prevent such harm. As a matter of general principle, if the state has the right to have its nationals protected abroad, a concomitant duty to ensure that the nationals act in a manner consistent with international norms should be recognised. Additionally, in a situation such as alleged in Unocal, the home state of the foreign investor benefits from the foreign investment through the repatriation of the company’s profits. The profits are enhanced by the forced labour that is secured by torture and repression. The fact that the home state itself benefits in this way casts a duty on it to ensure that the company’s profits (which are in a sense also profits for the home state) are not secured through means that violate international norms. The home state has the ability to ensure that its nationals operating abroad as foreign investors act in a manner consistent with international norms through the exercise of jurisdiction on the basis of nationality. If tax and other legislation, such as antitrust legislation, can be extended in this manner, then the creation of a duty to exercise such jurisdiction to secure fundamental norms of 62 The Nicaragua Case, supra n. 41, provides obvious authority. There need not even be the link of nationality, the mere existence of control of a group sufficing. The ICJ held that requisite control over the Contras was absent and that state responsibility could not be imputed to the US. Nationality makes control easier to establish, particularly in circumstances where the national’s activities are known to the home state and the home state profits from them directly or indirectly. 63 And, further, is advance knowledge, not of actual harm, but of the risk of harm sufficient to trigger the duty to control the national in order to prevent the harm before it occurs? One would think that the knowledge of at least a high risk of harm triggers a duty of control.

508 M Sornarajah international law cannot be regarded as unjustified. Some states reached out in this manner to prevent corruption and bribery in international business by their national corporations even before agreement began to be reached under the auspices of regional and sectoral organisations like the Organisation of American States (OAS) and OECD on multilateral treaties.64 The prohibition of bribery has not nearly as much support in international policy as the prohibition of torture. Indeed, in the hierarchy of public policy norms, the prohibition of bribery comes well below the violation of human and environmental rights. The minimum effect of a state being unprepared to prevent the violation of international public policy norms is that it must lose its right of diplomatic protection of its nationals. The theory of foreign investment that was articulated in the past justified the protection afforded to business on the ground that the home state made sure the business practices of the multinational corporations it sent abroad were above-board, thereby entitling the state to give the corporate national its maximum protection. This policy fails in situations where the multinational corporation engages in practices that violate jus cogens norms against torture or engages in other practices which violate other fundamental norms which may not necessarily have risen to the level of jus cogens. There must be a satisfaction of the existing rules on state responsibility before a state is subjected to responsibility. As the International Court of Justice stated in the ELSI case, an allegation of state responsibility should not be lightly made.65 Hence, the rules on state responsibility must be clearly satisfied before an allegation of state responsibility for the conduct of nationals is raised. Knowledge must be attributable to the home state with respect to the acts of the nationals in question. This will ordinarily be difficult to satisfy as a state cannot know beforehand what its nationals who go abroad seek to do. But, where a multinational corporation, as in the Unocal situation, is alleged currently to be violating the human rights of peoples abroad, the situation will be sufficiently widely known for the state to be imputed with knowledge of the situation. There is not merely a jurisdictional power, but a duty to control the activities of the national control in these circumstances.66 The issue then becomes one of whether the state took reasonable steps to prevent the harm where advance knowledge of likely harm can be shown or to stop an ongoing violation of human rights. Returning to the parallel provided by the law on state responsibility for the activities of loosely-organised mobs which attack foreigners, the law has been developed on the basis that, if the state had knowledge of the situation, a duty 64 Foreign Corrupt Practices Act of the United States. The OECD Guidelines for Multinational Corporations (Draft Revised Text, March 2000, available on the Internet at http://www.oecd.org) requires multinational corporations not to offer bribes. 65 ELSI Case (1987), ICJ Reports, p. 15. 66 In the Sambaggio Case (1903) 10 RIIA 499, Arbritrator Ralston said: “Governments are responsible, as a general principle, for the acts of those they control.” Control through the nationality principle of the conduct of multinational corporations is possible as shown by the United States using such control in many areas, such as export of technology.

State Responsibility for Harms by Corporate Nationals Abroad 509 to protect the aliens is triggered. If a mob attack was foreseeable, then a duty of protection is owed to the aliens. State responsibility arises from the failure of the territorial state to provide such protection, liability flowing from the omission to act in a situation where the law creates a duty to act.67 Even if a standard of strict liability is not the basis of state responsibility, a due diligence fault requirement will be satisfied in these circumstances where there is the existence of the capacity of control and the failure to take the steps reasonable in the circumstances to exercise that capacity. Since a capacity to control exists in the home state with respect to a multinational corporation which operates abroad, the same rules can therefore be extended to render home states liable when they are aware of the conduct of their multinational corporate nationals and do not curb such conduct through the means available to them. International law requires that there must be an exercise of governmental authority before a state can become responsible. It may be argued that there is no governmental authority in the situation of a multinational corporation, whereas there is in the agent who is sent to commit sabotage abroad, as in the Rainbow Warrior situation in which French security agents were sent abroad to sabotage one of Greenpeace’s vessels while in port in New Zealand. But international law recognises very clearly that governmental authority may be exercised either by act or omission.68 Whether or not conduct by omission breaches international law depends on whether there is a positive duty placed by substantive international law on the state (i.e. a duty to act in a certain way and/or with a view to achieving a certain result). Again, the classic mob situation illustrates the way out of this supposed difficulty. The mob is by no means the agent of the government. The requirement of governmental authority is provided in these cases by the failure of the governmental authorities to act in the situation of danger to the alien posed by the mob.69 It is the failure to exercise due diligence to protect the alien from which governmental complicity in the episode arises. Likewise in the case of the multinational corporation, responsibility arises due to the failure of the state authorities to prevent it from engaging in human-rights-abusive conduct abroad. The theory of liability that has evolved primarily in arbitral case law on mob violence provides the necessary body of general principles for establishing liability of the home state for the torture and other human rights violations engaged in by multinational corporations. We would do well to remember that the application and development of these principles in some cases also involved the making of leaps in legal reasoning from general principles to concrete contexts no less creative than what is being contended for here. 67 Home Missionary Society Case (1920) 6 RIIA 20; R Pisillo-Mazzeschi, “The Due Diligence Rule and the Nature of the International Responsibility of States”, (1992) 35 GermanYIL 46. 68 It is generally recognised in the law of state responsibility that an omission to act to prevent harm entails responsibility. 69 Youmans Claim (1926) 4 RIIA 110; Zafiro Claim, supra n. 12, (where liability was on the basis that the naval officer did not stop the looting by the mob “with sufficient promptitude”).

510 M Sornarajah Some may argue that distance is a constraining factor in the placement of state responsibility on home states for corporate conduct abroad. In most instances of state responsibility, the state held responsible had caused injury to the alien resident within that state’s territorial limits. But this common factual situation of territoriality cannot be understood as having given rise to rise to any legal rule that the proximity to the act is necessary. Proximity has never been isolated as a limiting rule by any writer.70 In any event, it would be completely inconsistent with the willingness of international law to extend responsibility to states for extraterritorial acts of its nationals where there is some direct agency or similar kind of link. For example, in Short v. Iran,71 it was contemplated that if the Ayatollah Khomeini, who had yet to come to Iran, had instigated hostile acts towards the American nationals resident in Iran whilst still in exile in France, his incoming government would be responsible for the acts. And states have been held responsible for sending forth agents to engage in abduction abroad and for sending saboteurs into another state,72 while it is true that the existing principles of state responsibility will have to be extended creatively to cover state responsibility for corporate misconduct. But, as was argued earlier, if the need and will to achieve a pressing policy objective is to be placed at the centre of the evolution of international law, achieving it by adapting and extending existing law is a legitimate means to do so. Another challenge might take the form: why should there be state responsibility in these circumstances? This question has implicitly already been answered. The imposition of state responsibility on the home state is one means of providing for control over the conduct of a multinational corporation whose conduct otherwise remains unregulated in modern international law—and arguably is increasingly unregulated as economic globalisation pressures sap the will and capacity of states to regulate corporate conduct within their borders for fear of losing investment.73 The paradigm being argued for will throw the onus for control onto home states which are not only able to bear that burden, but which also have a moral duty to shoulder it because they and their societies profit from the overseas operations of their multinational corporations. It is clear that they have the ability to control such activity, as they have claimed wide extraterritorial powers of control over these corporations in several areas. Home state responsibility will help in the creation of an international law which prevents human rights abuses by these corporations which exert much power 70 Nor have the older cases required that the injurious acts should have taken place on the territory of the state held responsible. See e.g. the Zafiro Claim, supra n. 12, concerning responsibility of the United States in respect of acts of vandalism of a Chinese crew under the command of an American naval officer in a port city of the Philippines (which was, at the time, not under the control of any authority). The presence of the officer at the scene of the injury was held not to be necessary. 71 Supra n. 39. 72 Rainbow Warrior Arbitration, supra n. 42. For a finding of international human rights responsibility for abduction in another state, see Casariega, supra n. 55. 73 The voluntary codes which exist hardly provide the function of control.

State Responsibility for Harms by Corporate Nationals Abroad 511 and influence over the course of international relations without being subjected to any meaningful control. Finally, of specific interest to the civil liability theme of the present book is that an important aspect of such state responsibility is that the wrongdoing state then comes under a duty to provide a remedy to the victim. The failure to provide a remedy again leads to an independent basis of state responsibility. To this issue we now turn in the final section of this chapter.

4 STATE RESPONSIBILITY AND THE DUTY TO PROVIDE REMEDIES TO VICTIMS

It has been demonstrated that state responsibility arises from the breach of a duty to control the human rights violations of a multinational corporation abroad. Where such breach occurs, a duty on the part of the home state to provide a remedy to the victims of its misconduct then arises. More specifically, the contention is that in the Bhopal or Unocal-type situations, there is an obligation to provide a remedy through the domestic courts by the home state. This again is a contention that seems initially difficult to establish, but once it is conceded that there is a duty to control nationals from causing harm to people in other states the obligation can be laid down as a matter of law. The current paradigm of the duty to provide an effective local remedy is the situation of an alien who suffers injury at the hand of the host state. The duty to provide access to local courts in order to receive an effective remedy is already well-recognised in the law.74 It follows (at least at the level of consistency of the policy justifications underlying the law) that once the primary obligation to prevent harm abroad is accepted as applicable to the home states of multinational corporations, the law must be shaped to ensure that the home state provides a remedy. That is, at minimum where the home state itself has breached duties to control in situations of adequate knowledge of the corporate national’s conduct abroad, the duty to provide a remedy to the victims seems to follow as a matter of basic considerations of justice. Significantly, the existing law is not a blank slate on this, and in fact contains the kernel of a duty to provide access to the home state’s courts whether or not the home state itself bears responsibility for failure to control the corporate national prior to or during the harm. There is sufficient authority that a state which does not provide a remedy against the wilful harm caused to other states by a national becomes responsible to the state which suffered the harm.75 It must thus be noted that the duty to provide a remedy to victims may exist in general international human rights law, quite independently of any argument 74 C F Amerasinghe, Local Remedies in International Law (Cambridge, Grotius Publications, 1990). 75 The rule is usually discussed in the context of the requirement to either punish or hand over a national who has committed a crime abroad. There is no reason why the rule should not be applied to civil recourse in situations of human rights violations which involve acts like torture, which are considered criminal in most legal systems.

512 M Sornarajah relating to state responsibility for the state’s own involvement in or failure to prevent human rights harms.76 There is a generic duty to provide remedies for human rights violations.77 There is a credible argument to be made that not only is there a duty to provide a remedy, but that in the case of violations of jus cogens norms, like torture committed by nationals abroad, emerging international law recognises a positive duty to hold the offenders accountable.78 If such a positive duty exists, then it must follow that failure to perform it engages responsibility of states.

5 CONCLUSION

The globalisation of the world is supposed to be moving at a rapid pace but, historically, the process of globalisation has always been accompanied by the process of fragmentation which involves the withdrawal of people into their own cultures and values in order to face the onset of the global values. The process of fragmentation is seldom studied. The violence that is generated by the clash will be the concern of the future. The potential for violation of human rights in this clash has been insufficiently studied. Globalisation itself is a power-based process which seeks to secure the rights of business to the detriment of the rights of people. The positivist basis of the rules on state responsibility were formulated so as to enhance this process which stressed the rights of the foreign investor and eclipsed the earlier law which recognised the fact that responsibility could arise in the home state for the conduct of a national abroad. The Draft Articles of the ILC continues this trend by its emphasis on power-based solutions to state responsibility. Yet, the development of an international law of human rights is based on the competing notion of idealism in the law which requires the recognition of responsibility of the state for the violation of the more basic human rights giving rise to erga omnes obligations wherever they take place as long as they can be directly or indirectly attributed to that state. There is clear evidence of the impact of this competing objective. The progress towards the stabilisation of this norm may be incremental, but the signs are that once a breach is made the law will quickly develop to ensure this progress, notwithstanding the establishment views stated in the ILC’s Draft Articles. Work remains to be done in coalescing the different strands which are moving towards supporting the growth of such law. 76 The view is canvassed in R Pisillo-Mazzeschi, “International Obligations to Provide for Reparation Claims?” in A Randelzhofer and C Tomuschat (eds), State Responsibility and the Individual (1999) 149. 77 Velasquez Rodriguez v. Honduras, (1988) 9 HRLJ 212 (where the Inter-American Court of Human Rights recognised the duty of states to prevent violations and remedy or punish such violations after their commission). Recognition of human rights requires positive action by states, including access to courts for remedies. 78 G Goodwin-Gill, supra n. 48.

19

Revisiting Human Rights in the Private Sphere: Using the European Convention on Human Rights to Protect the Right of Access to the Civil Courts ANDREW CLAPHAM 1

1 INTRODUCTION H E B A S I C P R E M I S E of Human Rights in the Private Sphere was that we need to rethink human rights law in order to meet the challenges posed by private actors.2 Two approaches were suggested. The first proceeded from the principle that international human rights law demands that States protect individuals from private acts which threaten their rights. So, where a State has failed to take action that could have prevented such harm, that State may be found to have violated its international obligations. Failure to prevent, detect and punish such private actors may also lead to a finding that the State has failed to respect its international human rights obligations. It is further suggested that the European Convention on Human Rights places greater obligations on Contracting Parties than the simple State responsibility test under general international law. The second approach highlighted the fact that the difference between private and public acts may be a difficult distinction to draw and that excluding the private sphere from international concern leads to further disadvantage for the victims of human rights abuses in the private sphere.3 It is

T

1 Associate Professor of Public International Law, Graduate Institute of International Studies, Geneva. 2 A Clapham, Human Rights in the Private Sphere (Oxford, Oxford University Press, 1993). 3 Discussed by Professor Chinkin with regard to the specific discussions concerning the draft articles on State responsibility currently before the International Law Commission. C Chinkin, “A Critique of the Public/Private Dimension”, 10 European Journal of International Law (1999) 387–95. See also H Charlesworth “Worlds Apart: Public/Private Distinctions in International Law”, in M Thornton (ed.), Public and Private: Feminist Legal Debates (Melbourne, Oxford University Press, 1995) 243–60.

514 A Clapham suggested that, in some circumstances, human rights obligations may give rise to directly enforceable duties on the private actors themselves. This may not be the case for all rights in all circumstances.4 However, in some jurisdictions such direct enforceability is legally possible and politically appropriate in order to prevent the dangerous public/private distinction being used to shield private power from human rights scrutiny. In the context of the present volume, it seems useful to see whether these approaches can be developed so as to enhance the chances of torture victims being able to bring civil actions. We shall examine whether international human rights law not only demands that the State use its police and criminal justice system to protect individuals from private violence in the private sphere, but also whether this law obliges the State to grant victims the right to sue in tort where the State apparatus has failed in its duty to protect the victims or remedy the violation of their rights. As with several other contributions in this volume, this chapter takes the Pinochet imbroglio as one point of reference. It serves as a springboard for exploring the sorts of legal challenges that may be faced in the future as other officials find themselves accused of human rights abuses in

4 We might consider the analogous situation in constitutional law. The Constitution of Switzerland (in force on 1 January 2000), includes Article 35:

“Realisation of Fundamental Rights . . . 1. The fundamental rights shall be realised in the entire legal system. 2. Whoever exercises a function of the state must respect the fundamental rights and contribute to their realisation. 3. The authorities shall ensure that the fundamental rights also be respected in relations among private parties whenever the analogy is applicable.” Similarly the Constitution of South Africa of 1996 includes section 8: “(1) The Bill of Rights applies to all law and binds the legislature, the executive, the judiciary and all organs of the state. (2) A provision of the Bill of Rights binds a natural or a juristic person if, and to the extent that, it is applicable, taking into account the nature of the rights and of any duty imposed by the right. (3) In applying a provision of the Bill of Rights to a natural or juristic person in terms of subsection (2), a court—(a) in order to give effect to a right in the Bill, must apply or if necessary develop, the common law to the extent that legislation does not give effect to that right; and (b) may develop rules of the common law to limit the right, provided that the limitation is in accordance with section 36(1). (4) A juristic person is entitled to the rights in the Bill of Rights to the extent required by the nature of the rights and of the juristic persons.” The South African Constitution also provides in article 36(1): “The rights in the Bill of Rights may be limited only in terms of law of general application to the extent that the limitation is reasonable and justifiable in an open and democratic society based on human dignity, equality and freedom, taking into account all relevant factors including(a) the nature of the right; (b)the importance of the purpose of the limitation; (c) the nature and extent of the limitation; (d) the relation between the limitation and its purpose; and (e) less restrictive means to achieve the purpose.” For a discussion of the extent to which the UK Human Rights Act will allow for the Convention to be raised in the domestic courts in civil cases (where the defendant is not a public authority or the nature of the act is private) see M Hunt “The Horizontal Effect of the Human Rights Act”, (1998) Public Law 423–43 and A Clapham “The European Convention on Human Rights in the British Courts: Problems Associated with Incorporation of International Human Rights into Domestic Law”, in P Alston (ed.), Promoting Human Rights Through Bills of Rights (Oxford, Oxford University Press, 2000).

The European Convention on Human Rights 515 foreign jurisdictions.5 Thinking about this as a “private sphere” question may be helpful because, in a Pinochet-type situation, we are not dealing with a State’s direct responsibility for harm caused by its agents. Rather we are dealing with the State’s positive duties with respect to providing for the accountability of individuals actors. I shall explore what arguments might be used, in the context of evolving international human rights law, to show that there is an international legal obligation on a State to allow the victims of foreign torture to bring a suit in that State’s national courts, in the event that the State-of-custody authorities fail to prosecute or extradite a suspected torturer. Adapting the bifocused approach from Human Rights in the Private Sphere, two kinds of civil claim might be contemplated in the custody State. First, there might be a claim in negligence against that State’s authorities for their failure to prevent the human rights violation, or to prosecute or extradite the torturer; such a suit by victims of the acts of torture would seek to hold the authorities liable at the national level. Second, there could be a claim in tort (or a private prosecution, or a combination) brought against the actual torturer in the courts of the custody State. It will be suggested that in some circumstances the failure by the State to allow for such claims in the national legal order may represent a violation of international human rights obligations. We shall explore the implications of the judgment by the European Court of Human Rights in Osman v. United Kingdom.6 This case did not involve an accused torturer caught in a transnational jurisdictional tangle, however, it has important implications for the victims of human rights abuses as it emphasises that, under the European Convention on Human Rights, there is an internationally protected human right to sue the police authorities for negligence for 5 Tagged the “Pinochet Syndrome” by Barbara Crossette of the New York Times (electronic edition) in “Dictators face the Pinochet Syndrome”, 22 August 1999. There is already a sense of instability as various leaders find themselves the target of criminal complaints. The complaint filed in Austria against Izzaz Ibrahim al-Duri of Iraq is detailed in Robert Reid, “Saddam’s second in command leaves Austria”, (18 August 1999) Associated Press. For complaints filed in Paris against JeanClaude Duvalier of Haiti, see “Plaintes contre Bébé Doc pour crimes contre l’humanité”, (11–12 September 1999) Tribune de Genève at 5. See also the appeal by Human Rights Watch to the Justice Ministry in South Africa to press charges against Mengistu Haile Mariam for crimes against humanity and torture in Ethiopia, (2 December 1999) Pretoria News. Human Rights Watch suggests that despite the absence of national legislation, South Africa has jurisdiction under customary international law which is part of the law of South Africa under Article 232 of the Constitution. In the prelude to the establishment of the International Criminal Court, as more and more states introduce national legislation which allows for the prosecution of war crimes and crimes against humanity, it is likely that a number of criminal prosecutions may be attempted. It should not be assumed that this groundswell is an unalloyed good without significant trade-offs; for Ricardo Lagos and Heraldo Muñoz there is a real fear in countries that the result could be “disorder, confrontation, and the risk of endangering peaceful and successful democratic transitions”: “The Pinochet Dilemma”, (Spring 1999) Foreign Policy 26–39 at 27. Lagos was recently elected President of Chile, after this coauthored article was written. 6 ECHR, Judgment of 28 October 1998. The present author assisted Ben Emmerson who was counsel before the European Commission and Court of Human Rights. All the facts described and comment are based on the Council of Europe documents: Admissibility decision on Application 23452/94 of 17 May 1996, Report of the Commission, 1 July 1997, and the Judgment of the Court in Case 87/1997/871/1083.

516 A Clapham failing to protect the right to life in the context of a killing by a private person. The case is not about impunity in the traditional sense, as the murderer was immediately apprehended, tried and sentenced. The fact that the obligation goes beyond prohibitive laws and efficient prosecution brings the preventive aspect to the fore; therefore there are important implications for national authorities and national legal systems. The right to sue the authorities in negligence at the national level may come to be a vital tool in the hands of the victims of private and public violence.

2 HUMAN RIGHTS IN THE PRIVATE SPHERE : THE EVOLVING SCOPE OF INTERNATIONAL HUMAN RIGHTS LAW

In order to understand the logic of the Osman judgment it may be worth first recreating the building blocks in the argument, as the Osman case builds on a number of developments in international human rights law outside the Convention.

The International Covenant on Civil and Political Rights (1966) The International Covenant on Civil and Political Rights (ICCPR) has been interpreted as creating obligations for States parties even where the violence does not emanate from their own authorities. During the drafting various delegates took positions on this question. According to the travaux préparatoires: “While the view was expressed that the article should concern itself only with the protection of the individual from unwarranted actions by the State [E/CN.4/SR.90, p.9 (USA)], the majority thought that States should be called upon to protect human life against unwarranted actions by public authorities as well as by private persons [E/CN.4/SR.90, p.8 (GB), p.9 (IRL), p.10 (F), p.12 (DK) & (RCH); E/CN.4/SR.140, §2 & §38–39 (RCH), §10 (RL), §34–35 (F); E/CN.4/SR.144, §20–23 (RL), §27 (U); E/CN.4/SR.311, p.4 (F)] A/2929, Chapt VI, §4.”7

The United Kingdom (GB) and Lebanon (RL) took specific issue with the United States’ desire not to impose positive duties on States with respect to regulation of private sphere harm. The intervention by the United Kingdom is recorded in the summary records as follows: 7 At the 5th Session (1949), 6th Session (1950) and 8th Session (1952) of the UN Commission on Human Rights the phrase from Article 6(1) of the then draft International Covenant on Civil and Political Rights was discussed. As reproduced in M J Bossuyt (1987) Guide to the “Travaux Préparatoires” of the International Covenant on Civil and Political Rights (Dordrecht, Nijhoff) at p. 120, article 6(1) of the ICCPR reads: “Every human being has the inherent right to life. This right shall be protected by law. No one shall be arbitrarily deprived of his life.” International Covenant on Civil and Political Rights, adopted 16 Dec. 1966, G.A. Res. 2200 (XXI), U.N. GAOR, 21st Sess., Supp. No. 16, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171 (entered into force 23 Mar. 1976) [hereinafter “ICCPR”].

The European Convention on Human Rights 517 “Turning to consideration of the United States draft amendment to article 5 (E/CN.4/170), Miss Bowie said she could not support that proposal, as it would limit the article to a statement of the relationship between the State and the individual and would not safeguard the life of the individual from acts by other individuals or groups.”8

The representative of the United States, Mrs Roosevelt, suggested that: “the role of an international instrument should be to defend [individuals] from the unwarranted actions of the State. In general, actions committed by individuals or groups against other individuals were covered under existing codes of criminal law and it was not the task of the Commission on Human Rights to attempt to codify existing national legislation in that field. While the United States Government understood the considerations raised by the United Kingdom and Lebanon on that question, it nevertheless felt that the criminal law was essentially within the domestic jurisdiction of the State.”9

There was considerable debate at the time concerning the insertion of the word “arbitrarily” into article 6(1). Some drafters, such as Professor Cassin of France, were concerned that this could be interpreted to mean that the ICCPR would only deal with “relations between the State and individuals, disregarding the question of protection of life as among individuals”.10 He responded to the position articulated by Mrs Roosevelt by pointing out that: “the State had the duty not only of preventing its officials from killing its citizens but also of protecting each citizen from the untoward actions of other citizens or groups of citizens.”11 The “arbitrarily” qualification stayed in, but the fears of a limiting effect have not been borne out. The Human Rights Committee, which oversees the ICCPR, has interpreted the right not to be arbitrarily deprived of life as a right which is also applicable to killings in the private sphere. The Committee’s General Comment on article 6 states in part: “The Committee considers that States parties should take measures not only to prevent and punish deprivation of life by criminal acts, but also to prevent arbitrary killing by their own security forces.”12

8 Miss Bowie (United Kingdom) UN Doc. E/CN.4/SR 90, 20 May 1949, at p. 8. Fifth Session of the Commission on Human Rights, Lake Success, New York, 18 May 1949. 9 UN Doc. 20 May 1949, supra n. 8 at p. 9. 10 Mr Cassin (France) UN Doc. E/CN.4/SR.311, 12 June 1952, Eighth Session of the Commission on Human Rights, UN Headquarters, New York, 27 May 1952 at p. 4. The word “arbitrarily” was omitted from the text of the European Convention on Human Rights (1950). European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 221, Europ. T.S. No. 5 (entered into force 3 Sept. 1953), reprinted in (1994) 33 I.L.M. 943, 960, and in A Compilation of International Instruments, Volume II Regional Instruments United Nations publications ST/HR/1/Rev.5 (Vol.II) at 73 [hereinafter “ECHR”]. 11 UN Doc. 20 May 1949, supra n. 8 at p. 10. 12 General Comment No. 6, UN Doc. HRI/GEN/1/Rev.1 at page 6, para. 2 (emphasis added).

518 A Clapham The American Convention on Human Rights (1969)13 The American Convention on Human Rights has also been interpreted in a way that expands protection into the private sphere of actors whose acts are not always directly attributable to the State. Indeed, the Inter-American Court of Human Rights has delivered a key judgment on the positive obligations of States under the American Convention on Human Rights. States have a “duty to ensure” all rights in the treaty. The words of the Court are accordingly reproduced at some length: “174. The State has a legal duty to take reasonable steps to prevent human rights violations . . . 175. This duty to prevent includes all those means of a legal, political, administrative and cultural nature that promote the protection of human rights and ensure that any violations are considered and treated as illegal acts, which as such, may lead to the punishment of those responsible and the obligation to indemnify the victims for damages. It is not possible to make a detailed list of all such measures, since they vary with the law and the conditions of each State Party. Of course, while the State is obliged to prevent human rights abuses, the existence of a particular violation does not, in itself, prove the failure to take preventive measures. . . . 182. The Court is convinced, and has so found, that the disappearance of Manfredo Velásquez was carried out by agents who acted under cover of public authority. However, even had that fact not been proven, the failure of the State apparatus to act, which is clearly proven, is a failure on the part of Honduras to fulfill the duties it assumed under Article 1(1) of the Convention, which obligated it to ensure Manfredo Velásquez the free and full exercise of his human rights.”14

While the core of the Velásquez Rodriguez case concerned the duty to investigate disappearances, it remains equally authoritative for having confirmed that positive obligations exist for the State in the field of violence by private individuals and, most significantly for present purposes, that these positive obligations include not only prevention but also State compensation even where the violent actor was not acting on behalf of the State. States have adopted treaties in the context of the Organisation of American States (OAS) which build on these developments. The Inter-American Convention on the Forced Disappearance of Persons criminalised the act of forced disappearance and sets out a number of State obligations in the field of positive obligations.15 It also reaffirmed that the systematic practice of the forced disappearance of persons constitutes a crime 13 American Convention on Human Rights, adopted 22 November 1969 (entered into force 18 July 1978), reprinted in A Compilation of International Instruments, Volume II Regional Instruments (United Nations Publications ST/HR/1/Rev.5), at 14 [“Compilation, Regional”]. 14 Velásquez Rodriguez v. Honduras (1989) 28 ILM 291. 15 Adopted 9 June 1994, entry into force 28 March 1996. Reprinted in Compilation, Regional, supra n. 13 at 55; see also (1994) 33 ILM 1529.

The European Convention on Human Rights 519 against humanity. This criminalisation of certain human rights violations individualises the international violation and can extend to the private sphere.16 More recently the United Nations Security Council has emphasised “the responsibility of States to end impunity and to prosecute those responsible for genocide, crimes against humanity and serious violations of international humanitarian law”.17 There is no suggestion here that this responsibility only extends to acts committed within the territory of the State. In short, we are moving towards an obligation in international law on all States to prosecute all those for whom there is the requisite evidence that they have committed crimes against humanity, wheresoever and by whomsoever committed. Another OAS treaty, the Inter-American Convention on the Prevention, Punishment and Eradication of Violence Against Women boldly promises that “[e]very woman has the right to be free from violence in both the public and the private spheres.”18 Violence is defined in Article 1 as “any act or conduct, based on gender, which causes death or physical, sexual or psychological harm or suffering to women, whether in the public or the private sphere”. The Convention on the Elimination of All Forms of Discrimination Against Women (1979) and the Obligation under General International Law19 It is in the just-mentioned field of violence against women that it has now become more apparent that the State has positive duties under international human rights law to protect individuals from the violent acts of other individuals or groups. In its General Recommendation No. 19, the Committee on the Elimination of Discrimination against Women addressed the issue of “Violence against women”. It is worth quoting in full paragraph 9 of that general recommendation which makes reference to the state of “general international law” and not simply the treaty law of the Convention on the Elimination of Discrimination Against Women (CEDAW): 16 Although the Convention defines disappearances for the purposes of the Convention as an act perpetrated by State agents or acquiesced in by the State, the Rome Statute for the International Criminal Court, adopted 17 July 1998, UN Doc. A/CONF.183/9, has made it clear that disappearances can come within the category of crimes against humanity and that these crimes can be committed by individuals working with an “organisational policy” completely independent of the State. In order to be a crime against humanity under the Statute, article 7(1) states that the act of enforced disappearance of persons has to be committed “as part of a widespread or systematic attack directed against a civilian population and with knowledge of the attack”. An attack on a civilian population is defined according to article 7(2)(a) as “a course of conduct involving the multiple commission of acts referred to [above] . . . against any civilian population, pursuant to or in furtherance of a State or organisational policy to commit such attack”. The Statute is not yet in force. 17 UNSC Resolution 1265 of 17 September 1999, UN Doc. S/RES/1265 (1999). 18 Article 3. Adopted 9 June 1994, entry into force 3 May 1995. Reprinted in Compilation, Regional, supra n. 13 at 62; see also (1994) 33 ILM 1534. 19 Adopted 18 December 1979, G.A. Res. 34/180, UN GAOR, Supp. No. 46, U.N. Doc. A/34/46, at 193 (1979), UNTS vol. 1249, p. 13, (entered into force 3 September 1981), reprinted in A Compilation of International Instruments, Volume 1 (First Part) (United Nations Publications ST/HR/1/Rev.5) Vol.1, Part 1 at 150 [“Compilation, International”] ; see also (1980) 19 ILM 33.

520 A Clapham “9. It is emphasised, however, that discrimination under the Convention is not restricted to action by or on behalf of Governments (see articles 2(e), 2(f) and 5). For example, under article 2(e) the Convention calls on States parties to take all appropriate measures to eliminate discrimination against women by any person, organisation or enterprise. Under general international law and specific human rights covenants, States may also be responsible for private acts if they fail to act with due diligence to prevent violations of rights or to investigate and punish acts of violence, and for providing compensation.”20

For evidence of developments in the current state of general international law in this area, we might simply refer to the Declaration on Violence against Women adopted by consensus by the General Assembly on 20 December 1993 as Resolution 48/104. Article 4 provides: “States should condemn violence against women and should not invoke any custom, tradition or religious consideration to avoid their obligations with respect to its elimination. States should pursue by all appropriate means and without delay a policy of eliminating violence against women and, to this end, should: . . . (c) Exercise due diligence to prevent, investigate and, in accordance with national legislation, punish acts of violence against women, whether those acts are perpetrated by the State or by private persons . . .”

The similarly self-imposed obligation enshrined in the “Strategic Objectives and Actions” adopted at the Beijing World Conference on Women in 1995 further outlines the scope of States’ obligations and commitments: “224. Violence against women both violates and impairs or nullifies the enjoyment by women of human rights and fundamental freedoms. Taking into account the Declaration on the Elimination of Violence against Women and the work of Special Rapporteurs, gender-based violence, such as battering and other domestic violence, sexual abuse, sexual slavery and exploitation, and international trafficking in women and children, forced prostitution and sexual harassment, as well as violence against women, resulting from cultural prejudice, racism and racial discrimination, xenophobia, pornography, ethnic cleansing, armed conflict, foreign occupation, religious and anti-religious extremism and terrorism are incompatible with the dignity and the worth of the human person and must be combated and eliminated. Any harmful aspect of certain traditional, customary or modern practices that violates the rights of women should be prohibited and eliminated. Governments should take urgent action to combat and eliminate all forms of violence against women in private and public life, whether perpetrated or tolerated by the State or private persons.”

Conclusion Concerning International Human Rights Law Duties on States with Regard to Non-State Violence The interpretive comments of the UN human rights treaty bodies, the judgment of the Inter-American Court, and the adoption of new international instruments 20 Contained in UN Doc. GAOR, A/47/38, adopted January 1992 at the eleventh session of the Committee on the Elimination of Discrimination against Women (emphasis added).

The European Convention on Human Rights 521 point towards a set of international obligations for States even when the violence is committed by private individuals. Those obligations not only include duties to prevent and investigate violence, but also extend to a duty to compensate for failure to prevent or investigate it. In order for these duties to be meaningful, and in order to establish an effective preventative strategy, there needs to be effective access to domestic courts.21 If the facts and allegations are not aired before such tribunals, then the right to this protection from the State will be rhetorical rather than real, illusory rather than effective. In the following sections, I hope to show how very recent developments under the European Convention are forging connections between State obligations with respect to violence in the private sphere, and rights of individuals to access national court systems where the State has failed to act. 3 OSMAN , PRIVATE VIOLENCE AND ACCESS TO THE CIVIL COURTS

The facts of the Osman case are too complex and disputed to be usefully detailed here. The essence of the case was that Ahmet Osman’s teacher, Paul Paget-Lewis, developed an attachment to Ahmet who was at the time a fifteenyear-old British citizen living in London. The teacher was transferred to another school. Over the following three months there followed a number of attacks on the Osman family’s property. There was also a collision between the teacher’s car and a car carrying Ahmet Osman’s friend, Leslie Green. During this time there were various interviews between the teacher and the local authorities as well as between the police and the teacher. The car crash involving the teacher and Leslie Green was investigated by the police. Following an attempt to arrest the teacher for criminal damage, the teacher disappeared from his new school and travelled around England hiring cars under his newly-adopted name of Osman. The teacher had changed name by deed poll from Paul Paget-Lewis to Paul Ahmet Yildirim Osman. He stole a shotgun and returned to London where he was spotted by Leslie Green near the Osmans’ house. In the words of the European Court, the situation culminated as follows: “The Government accept that, on 5 March 1988, Detective Sergeant Boardman received a message which stated ‘phone Mrs Green’ but since there was no phone number on the note he did not connect the message with the mother of Leslie Green. On 7 March 1988 Paget-Lewis was seen near the applicants’ home by a number of people. At about 11.00 pm Paget-Lewis shot and killed Ali Osman and seriously wounded Ahmet. He then drove to the home of Mr Perkins [a deputy head teacher] where he shot and wounded him and killed his son. Early the next morning PagetLewis was arrested. On being arrested he stated ‘why didn’t you stop me before I did it, I gave you all the warning signs?’ ”22 21 See Golder v. the United Kingdom, ECHR (1975), Series A, No. 18 and Airey v. Ireland, ECHR (1979), Series A, No. 32. 22 Osman v. The United Kingdom, ECHR, Judgment of 28 October 1998, at paras 55–7. The judgment continues at para. 58: “Later that day Paget-Lewis was interviewed by the police. According

522 A Clapham For present purposes, two points of interpretation under the European Convention on Human Rights (the Convention) are relevant. First, the Court stated clearly that there were positive obligations on the State even where the violence emanates from a private individual or a State official acting outside official capacities: “It is thus accepted by those appearing before the Court that Article 2 of the Convention may also imply in certain well-defined circumstances a positive obligation on the authorities to take preventive operational measures to protect an individual whose life is at risk from the criminal acts of another individual.”23

The majority of the Commission and the Court held there was no violation of article 2 in the particular circumstances of this case.24 The Court’s majority judgment stated the standard of care in the following terms: “For the Court, and having regard to the nature of the right protected by Article 2, a right fundamental in the scheme of the Convention, it is sufficient for an applicant to show that the authorities did not do all that could be reasonably expected of them to avoid a real and immediate risk to life of which they have or ought to have knowledge. This is a question which can only be answered in the light of all the circumstances of any particular case.”25 to the record of the interview, Paget-Lewis said that he had been planning the attacks ever since he lost his job, and for the previous two weeks he had been watching the Osmans’ house. Although he considered Mr Perkins as his main target, he also regarded Ali and Ahmet Osman as being responsible for his losing his position at Homerton House. Paget-Lewis stated that he had been hoping in the back of his mind that the police would stop him. He admitted holding the family at gunpoint as they returned to the house, making Ali and Ahmet Osman kneel down in the kitchen, turning out the light and shooting at them. He denied that on earlier occasions he had damaged the windows of the Osmans’ house but admitted that he had let down the tyres of their car as a prank.” 23 Ibid. at para. 115. The Court went on at para. 116 to outline the scope of the obligation: “For the Court, and having regard to the nature of the right protected by Article 2, a right fundamental in the scheme of the Convention, it is sufficient for an applicant to show that the authorities did not do all that could be reasonably expected of them to avoid a real and immediate risk to life of which they have or ought to have knowledge. This is a question which can only be answered in the light of all the circumstances of any particular case.” 24 The Commission found by 10 votes to 7 that there had been no violation of Article 2. The majority considered that “[t]he failings to take additional investigative steps do not, in the Commission’s view, disclose any seriously defective response by the police to the threat posed by Paget-Lewis as perceived at the time” (para. 99). The minority led by the President of the Commission, Mr Trechsel, proposed a new test to determine the liability of the State where their duty to protect human rights is under examination, the “increase of risk test”: The minority said: “Responsibility is established, under this test, as soon as it can be said that the action called for would have considerably diminished the risk of the result, in other words, if the omission considerably increased the risk.” They rejected the majority approach which they categorised as a “causation” test pointing out that it is difficult to establish a link between omission and result. European Commission on Human Rights, Application 23452/94, Report adopted 1 July 1997. The Court found by 17 votes to 3 that there had been no violation of Article 2. In his separate opinion, Judge Lopes Rocha suggested that the test for omission should be “determined according to generally accepted rules. It has to be decided whether the assault originated from the failure to take a particular measure or measures where the assailant’s previous behaviour already pointed to a likelihood that he would act aggressively towards someone of whom he was particularly fond.” Osman, supra n. 22, at p. 59. 25 Osman, supra n. 22 at para. 116. However, in the circumstances of this case, the majority of the Court found there had been no violation of Article 2. The relevant part of Article 2 reads: “(1)

The European Convention on Human Rights 523 Second, the case turned on the fact that the Osman family could not sue the police for the tort of negligence in the English courts as the courts had, in the exercise of their common law role, developed a public policy immunity for the police when faced with negligence claims connected to the investigation or suppression of crime. The Osmans alleged a violation of article 6 of the Convention in that they had been denied access to a tribunal for a determination of their civil rights and obligations.26 While the Court found that there had been no violation of the article 2 duty to prevent violation of the right to life, it unanimously found that there had been a breach of article 6 of the Convention on the basis that “application of the exclusionary rule in the instant case constituted a disproportionate restriction on the applicants’ right of access to a court”.27 The issue at this stage of analysis was not the responsibility of the State for failing to exercise due diligence to prevent private violence. Rather, the State is responsible at the international level for having failed to provide in its legal system for a civil suit against the authorities in the wake of an incident of non-State violence resulting in the loss of life.

4 THE COMPLAINTS OF VICTIMS OF TORTURE IN CHILE CONCERNING ACCESS TO JUSTICE IN THE JACCARD - VELOSO ET AL . APPLICATION

The Context and the Claims The application to the European Court of Human Rights in Jaccard-Veloso asserted that “the Applicants or their next of kin have been the victims of acts of torture, disappearance, assassination or murder by General Augusto Pinochet in Chile”.28 Paulina Jaccard-Veloso, the First Applicant “is a Professor of Civil law at the University of Chile, having dual Swiss and Chilean nationality. Her husband Alexis Jaccard, a Swiss national, was abducted, tortured and murdered by the regime of General Pinochet. The acts perpetrated against the First Applicant’s husband are the subject of a request by Switzerland to the

Everyone’s right to life shall be protected by law. No one shall be deprived of his life intentionally save in the execution of a sentence of a court following his conviction of a crime for which this penalty is provided by law.” 26 Article 6(1) reads in relevant aspect: “In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law.” 27 Osman, supra n. 22 at para. 154. 28 The request for interim measures by Pauline Jaccard-Veloso and eighteen others against the United Kingdom was prepared for Human Rights Watch by Philippe Sands, Matthew Craven and the present author with the assistance of Hassiba Hadj Saharoui and submitted to the Court on 5 November 1998. Although not styled as a “case”, the request (Applic. 44191/98) will be referred to as Jaccard-Veloso. The request for interim measures was denied. See also Pradenas and others v. UK, Applic. 51287/99, denied provisional measures 17 January 2000.

524 A Clapham United Kingdom for the extradition of General Pinochet.”29 The other eighteen applicants have British or French nationality and their complaints relate to cases of torture and disappearances between 1973 and 1984. The Applicants contended, inter alia, that they risked being prevented from having access to court and to a remedy for the violation of rights by Pinochet. More specifically, they argued that, were the House of Lords to confirm the judgment of the English Divisional Court of 28 October 1998 (which had ruled that Pinochet benefited from Head of State immunity before UK courts), then the nineteen torture victims would be denied their rights under article 6(1) of the Convention, and the UK could be in violation of its obligations under articles 2, 3 and 6(1). It was argued that the obligation to ensure the prosecution or punishment of the perpetrators of acts of torture would be violated by the recognition of immunity in the UK’s jurisdiction. Let us consider in detail the various ways in which the potential denial of the right to bring civil proceedings raises questions under article 6(1). First, a potential violation of article 6(1) could result from the fact that the release of General Pinochet and his return to Chile through a failure to prosecute (in the UK or by way of extradition to another country seeking to prosecute him) would mean that there could be no hearing of the case before an “independent and impartial tribunal established by law” as required by article 6(1) of the Convention. The British applicants would not be able to bring civil proceedings before the English courts, the Swiss applicant would be denied a hearing in the Swiss courts, and the French applicants would similarly be prevented from exercising their rights in the French courts. The Jaccard-Veloso applicants thus asserted that their right to have their civil rights determined in a court were (or would be) violated by the action of the Attorney-General in refusing to allow a prosecution against Mr Pinochet,30 combined with the United Kingdom 29

From the original application on file with the Registry of the Court (para.2). The reasons for this refusal are contained in the examination of the UK state report before the UN Committee against Torture. See Summary record of the first part of the 355th meeting, 16/11/98. Consideration of the Report of the United Kingdom of Great Britain and Northern Ireland, UN Doc. CAT/C/SR.355, 06/10/99 Reply by the UK (Mr. Carter) at para. 23: “On 26 October 1998, in parallel with the extradition proceedings, lawyers acting for persons claiming to have been tortured in Chile during the Pinochet regime had, under section 135 of the Criminal Justice Act 1988, sought permission from the Attorney-General to bring proceedings for torture against Mr. Pinochet under section 134 of that Act. In deciding whether or not to agree to proceedings, the Attorney-General applied two tests laid down in the Code for Crown Prosecutors. The first was objective: was there sufficient admissible evidence to suggest that guilt could be established? If it was satisfied, the Attorney-General applied the second test: would proceedings be in the public interest? In the case of Mr. Pinochet, the Attorney-General had, after very thorough examination of the evidence, concluded that it was insufficient for there to be a chance of establishing guilt. He had therefore refused to consent to the opening of proceedings for torture against Senator Pinochet.” The concluding observations of the Committee against Torture addressed this point on 17/11/98, UN Doc. CAT/C/ UK: “E. Recommendations . . . (f) The Committee finally recommends that in the case of Senator Pinochet of Chile, the matter be referred to the office of the public prosecutor, with a view to examining the feasibility of and if appropriate initiating criminal proceedings in England, in the event that the decision is made not to extradite him. This would satisfy the State party’s obligations under 30

The European Convention on Human Rights 525 Government’s failure to provide any other civil or criminal remedy to the applicants through the British courts. Of course, the applications were made before the House of Lords’ decision that General Pinochet enjoyed no immunity with respect of the crime of torture with regard to allegations concerning acts of torture after 8 December 1988.31 The applicants therefore made sure to include a second argument: that, should the House of Lords find that immunity existed, then the United Kingdom would also be in violation of its obligations under the Convention. Here, Osman provided a powerful precedent. In Osman the European Court of Human Rights had held that the operation of the immunity for the police had to be proportionate to the aim claimed by the government. Any immunity granted by the United Kingdom law to Mr Pinochet similarly would have to be justified under the Convention. The immunity has to be justified as functional under the Convention. Any refusal to consider the applicant’s action in the national courts on grounds of formal grounds of immunity alone would therefore arguably violate the applicants’ right to access to court. The immunity has to be proportionate to the Government’s aim.

articles 4 to 7 of the Convention and article 27 of the Vienna Convention on the Law of Treaties of 1969.” See also the discussion in P Burns and S McBurney, “Impunity and the United Nations Convention against Torture: A Shadow Play Without an Ending?”, chapter 10 in this volume. 31 The date of ratification of the Torture Convention by the UK. Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), UNTS vol. 1465, p. 85, reprinted in (1984) 23 I.L.M. 1027 [hereinafter “CAT”]. R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No. 3) [1999] 2 All E.R. 97 (hereinafter Pinochet No. 3). This decision replaced the decision of the House of Lords of 25 November 1998, in R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No. 1), [1998] 4 All E.R. 897 (Pinochet No. 3) that was set aside on 15 January 1999 ([1999] 1 All E.R. 577) (Pinochet No. 2). According to the speech of Lord Browne-Wilkinson in Pinochet No. 1, the only extradition crimes in this context are those occurring after 29 September 1988, the date on which Section 134 of the UK Criminal Justice Act 1988 came into force and criminalised in UK law the act of torture (wheresoever committed). According to Lord BrowneWilkinson immunity was lost after 8 December 1988, this being the date when the CAT entered into force for the United Kingdom. Lord Browne-Wilkinson linked the loss of immunity ratione materiae to this date, as from this point it was clear that there was jurisdictional regime for the international crime of torture. (At 111) Cf. the speech of Lord Hope who preferred the date of 30 October 1988 as the date when Chile’s ratification took effect. At this date the obligations of customary international law were so great that Chile could not object on grounds of immunity ratione materiae. (At 152) See also the speeches of Lord Hutton (no immunity for crimes after 29 September 1988) (At 164) and Lord Millett (who saw no need for a time bar as “[I]nternational law cannot be supposed to have established a crime having the character of a jus cogens and at the same time to have provided an immunity which is co-extensive with the obligation it seeks to impose”) (At 179). The second and third applicants in the Jaccard-Veloso application were granted leave to intervene in the proceedings in the House of Lords. Pinochet was committed for extradition in the atutmn of 1999: The Kingdom of Spain v. Augusto Pinochet Ugarte, 8 October 1999, Unreported Judgment, Bow Street Magistrates’ Court (Pinochet No. 4). But it was later decided by UK Home Secretary Jack Straw that Pinochet should be permitted to return to Chile on the grounds of health problems that made him unfit for trial. Details of the relavent litigation are contained in Annex IV

526 A Clapham In sum, the Jaccard-Veloso applicants argued that the United Kingdom was (or would be, in the event a decision were taken releasing Pinochet from custody without a trial) in violation of its obligations under, inter alia, article 6(1) of the Convention as the applicants would be denied a determination of their “civil rights” by a tribunal.32 The question then arises do the applicants claims involve civil rights?

The scope of protected interests under article 6(1) The term “civil rights and obligations” has an autonomous Convention meaning according to long-standing Convention jurisprudence. However, the Court has struggled to keep its meaning within limits, and has sometimes emphasised ideas of private and/or personal rights (as contrasted to so-called “public law” rights). The concern of the Court has been to avoid becoming an appeal instance from national administrative decisions in issues such as social security. However, this private/public distinction has proved troublesome and various more expansive definitions have been suggested. Professor Fawcett (a former President of the European Commission on Human Rights) suggested in 1987 that various possibilities exist to circumscribe the scope of this term. His description of one expansive interpretation is instructive: “A broad construction of ‘civil rights and obligations’ in Article 6(1) would then cover all rights or obligations enforceable at law, regardless of whether the parties were individuals, corporations, or public authorities, or the State itself. It has some support in the fact that the expressions ‘civil rights’ or ‘civil and political rights’ have been adopted to describe the contents of the C.P.R. [Civil and Political Rights] Covenant, which corresponds of course at many points to Section I of the European Convention.”33

Under such an interpretation of “civil rights” the right not to be tortured and the right to life are civil rights and, as such, their determination has to be undertaken by an independent tribunal established by law (in accordance with article 6(1) of the Convention).34 This broader interpretation may imply moving beyond an assumption that some kind of right must already exist in national law in order to trigger the guarantees found in article 6(1). Under the European Convention it should not be 32 The arguments concerning article 6(1) were combined with claims relating to the Government’s obligations under article 2 (right to life), article 3 (torture and inhuman or degrading treatment or punishment) and article 13 (right to a remedy for a breach of Convention rights). 33 J E S Fawcett, The Application of the European Convention on Human Rights (Oxford, Oxford Universiy Press) Second Edition (1987) at p. 135. 34 Cf. Article 14(1) of the ICCPR which reads in part: “All persons shall be equal before the courts and tribunals. In the determination of any criminal charge against him, or of his rights and obligations in a suit of law, everyone shall be entitled to a fair and public hearing by a competent, independent and impartial tribunal established by law.”

The European Convention on Human Rights 527 necessary to demonstrate that the civil right already exists in domestic law; rather, the actual absence of a civil remedy in domestic law involves a violation of article 6(1). According to that provision, access has to be given to a court for the “determination” of a right. In this regard, it is important to recall the difference between the French and English texts of the Convention, with the Frenchlanguage version referring to “contestations”. As stated in the Moreira de Azevedo case by the European Court: “In the Court’s opinion the right to a fair trial holds so prominent a place in a democratic society that there can be no justification for interpreting Article 6(1) of the Convention restrictively. Conformity with the spirit of the Convention requires that the word ‘contestations’ [in the French language version] should not be construed too technically and that it should be given a substantive rather than a formal meaning. Besides, it has no counterpart in the English text of Article 6(1) (‘in the determination of his civil rights and obligations’; cf. Article 49: ‘dispute’—see, mutatis mutandis, the Le Compte, Van Leuven and De Meyer judgement of 23 June 1981, Series A, No. 43, page 20, para. 45.) In so far as the French word ‘contestations’ would appear to require the existence of a dispute, if indeed it does so at all, the facts of the case show that there was one.”35

It is worth noting that three judges in Osman—Judge Meyer, joined by Judges Lopes Rocha and Casadevall (concurring with the rest of the Court on the application of article 6(1) )—have specifically embraced the combined argument that article 6(1) applies to human rights (such as the right not to be tortured) and that there is no requirement that the national legal system already recognise such a right: “There was of course also a violation of the applicants’ right to a court, since the Osmans were denied any possibility to have their claims concerning the failures of the police properly examined by a tribunal. Whether or not they could rely on any substantive right thereto in domestic law is irrelevant, since they were asserting that they were the victims of a violation of fundamental (and therefore also civil) rights, which had to be secured to them under the Convention, notwithstanding anything to the contrary in domestic law or practice, and since their right to have their case heard in court was also such a right. It was likewise irrelevant whether the immunity of the police was or was not absolute, since the very principle of such immunity is not acceptable under the Rule of Law. The refusal to consider the applicants’ action was therefore an obvious denial of justice.”36

35 Moreira de Azevedo v. Portugal, ECHR (1990) Series A, No. 189, para 66. This passage was not actually necessary to the result in the case given the immediately following sentence: “In any event, the case concerned the determination of a right; the result of the proceedings was decisive for that right. (Ibid. at para 46).” However, it seems the Court intended to suggest a broader meaning of article 6(1) than might be suggested by the English text alone. 36 Emphasis added. Partly Dissenting, Partly Concurring Opinion of Judge de Meyer, Joined by Judges Lopes Rocha and Casadevall in Osman, supra n. 22, at p. 58. Cf. the Concurring Opinion of Judge Jambrek.

528 A Clapham In the context of the issues at play in this chapter, we need to note that, even if the determination of these civil rights is pursued as an adjunct to a criminal prosecution, this does not negate the fact that they are “civil rights”. A leading commentary on the Convention written by Harris, O’Boyle and Warbrick has put the matter in the following way: “A criminal prosecution brought by an applicant will involve the determination of his civil rights and obligations where such a prosecution is the remedy provided for in national law for the enforcement of a civil right, as for example in the case of some legal systems in connection with the right to a reputation. Article 6 also applies on the basis that civil rights and obligations are being determined when the victim of a crime joins a criminal prosecution as a civil party claiming compensation for injury caused by the crime.”37

In Helmers v. Sweden, one of the cases relied on by Harris et al., the European Court had to decide whether the private prosecution brought by Mr Helmers fell within the scope of article 6(1) of the Convention. The Court said: “The Court notes first, like the Commission, that although Article 6(1) does not guarantee a right for the individual to institute a criminal prosecution himself, such a right was conferred on the applicant by the Swedish legal system in order to allow him to protect his reputation.”38

In a Europe of forty-one Contracting Parties it is unlikely that the Court will attempt to draw up hard and fast rules between the protection of civil rights through the criminal law and the non-criminal law. Too many rights are protected through hybrid arrangements throughout Europe for any fair distinction to be drawn.

A Transnational Right to Access to Justice Under Article 6(1) of the European Convention The development of this notion of an article 6(1) right to access to justice as a transnational right to access to justice within Convention States needs a little more explanation. The reason that a decision by the UK authorities to release Pinochet could represent a violation of the Convention by the UK with respect to those applicants from France and Switzerland is because of the fact that the

37 D J Harris, M O’Boyle and C Warbrick, Law of the European Convention on Human Rights (1995) at 191–2 citing to Helmers v. Sweden, ECHR (1991) Series A, No. 212 with regard to the question of reputation. They also refer at the end of the passage quoted to Tomasi v. France, ECHR (1992) Series A, No. 241-A. Cf. Moreira de Azevedo, supra n. 35. 38 Helmers, supra n. 37 at para 29. Note the French Code de procédure pénale grants victims the right to pursue those who have committed violations outside French territory (Articles 1 and 689). Acts of torture as defined in the CAT are specifically included through Article 689(2). Under Title X concerning violations committed outside the territory of the Republic of France the Code includes terrorism, attacks on life, barbarous acts, as well as other relevant violations.

The European Convention on Human Rights 529 decision on prosecution or extradition (and immunity therefrom) would take place within the United Kingdom jurisdiction. The alleged future violations of human rights would, of course, happen in Switzerland and France where the applicants would have been denied access to Court. The link here may seem tenuous, but we need to recall the pathbreaking Soering judgment by the European Court of Human Rights.39 In Soering, the alleged direct violation (the subjection of the applicant to death row conditions in the US in the event that the UK authorities agree to extradite him) would have taken place in the United States. But, it was the decision as to whether to extradite that was within the jurisdiction of the United Kingdom, and it was the exercise of this jurisdiction that was found by the European Court to engage a form of indirect responsibility of one State for failing to prevent rights violations taking place in another State. The key to understanding the Jaccard-Veloso argument is that the UK would be acting in such a way as to preclude an available and willing court system from hearing the applicants’ case. Of course, in the Jaccard-Veloso context, it would not, as in Soering, be extradition, but rather the failure to extradite that would result in violation of article 6(1) rights: the release of General Pinochet would deny the Swiss and French applicants the chance to have their civil rights determined before a Swiss or French tribunal. The trigger for liability here is not, then, a decision to extradite, but rather a decision not to extradite. Moreover, the threat to the physical integrity of the victims takes a different form than in Soering (although it is worth noting that the Magistrate in the actual extradition hearing suggested that the families of disappeared persons may have been subjected to mental torture).40 Nevertheless, in broad terms, the decision taken by the authorities in the UK, whether judicial or executive, could lead to a denial of rights, albeit in another country. Arguably, therefore, the availability of a conception of a pan-European Conventional legal order would suggest that decisions in one country affecting access to court in another can raise questions under the Convention. On this point, another leading case dealing with the nature of States’ obligations under the Convention should be briefly recalled, Loizidou v. Turkey.41 In Loizidou, the Court emphasised the collective nature of the Convention guarantees and the fact that the Convention represents “a constitutional

39 Soering v. The United Kingdom, ECHR (1989) Series A, No. 161. Note also H.L.R. v. France, ECHR, Judgment of 29 April 1997, where the European Court of Human Rights did not rule out the possibility that article 3 of the ECHR might also apply where the danger emanated from persons or groups of persons who were not public officials. However, it would have to be shown that the risk was real and that the authorities of the receiving State were not able to obviate the risk by providing sufficient protection. 40 Pinochet No. 4, supra n. 31: “Whether the disappearances amount to torture; the effect on the families of those who disappeared can amount to mental torture. Whether or not this was intended by the regime of Senator Pinochet is in my view a matter of fact for the trial court.” 41 Loizidou v. Turkey (Preliminary Objections), ECHR (1995) Series A, No. 310.

530 A Clapham instrument of European public order (ordre public)”.42 The Court further recalled the “special character of the Convention as a treaty for the collective enforcement of human rights and fundamental freedoms”.43 Consistent with this Loizidou constitutional perspective, the applicants in Jaccard-Veloso submitted that, where future violations would take place in a Contracting Party (France or Switzerland) to the Convention, the case for a judgment which would prevent a violation is even stronger than in the Soering-type situation where the abuse of rights would have taken place in a non-Contracting Party to the Convention (the US).44 It seems fair to conclude that the Convention demands that Contracting Parties provide for access to Court for victims of torture. There are two straightforward routes open to victims according to the arguments outlined above. First, they have the right to initiate a civil suit against the authorities for failing to prevent, investigate or punish acts of torture or threats to the rights to life, by whomsoever committed—whether the violence emanates from the public or the private sphere. Second, victims may have a right to bring a case against their alleged torturer, as this also falls within their internationally protected right to access to Court. Where, the victims are primarily concerned about access to court in a different State from the State of custody, a third argument can be countenanced. In this situation the Convention arguably creates obligations which reflect the importance of greater European unity and an objective European system for the protection of human rights which is not based purely on direct responsibility for the acts of the public officials of that State.45 We have to stress that the European Court is not really applying inter-State obligations but rather operating as a sort of Constitutional Court for Europe: defining the scope of the human rights guarantees found in the Convention and articulating rules, principles and policies which are becoming part of the tapestry of European law (turning the Convention into a “constitutional instrument of European ordre public”46) at all levels, local, national, international and even in

42

Loizidou, supra n. 41 at para. 75. Ibid. at para 70. 44 Here again, Soering is relevant as one factor which motivated the Court in that decision was the presence of a less-rights-violative alternative than extradition to the US (namely, extradition to Germany which had also submitted an extradition request and which did not have the death penalty—and hence no death-row phenomenon). With Pinochet’s return to Chile, it would now seem that the case for trial in Europe is weaker as immunity in Chile has been lifted, Supreme Court, 8 August 2000. 45 For a challenge to this approach see R Lawson “Out of Control, State Responsibility and Human Rights: Will the ILC’s Definition of the ‘Act of State’ Meet the Challenges of the 21st Century?” in M Castermans, F Van Hoof and J Smith (eds) The Role of the Nation-State in the 21st Century (The Hague, Kluwer, 1998) 91–116 at 98–109. 46 Loizidou (Preliminary Objections), supra n. 41, at para. 75. Referred to by Lawson, supra n. 45 at 101, who goes on to quote and contrast the Court’s approach in the judgment on the merits where the Court asserts that its case law was “in conformity with the relevant principles of international law governing State responsibility”. Loizidou v. Turkey (merits), ECHR, Judgment of 18 December 1996 at para. 52. 43

The European Convention on Human Rights 531 the supranational European Court of Justice. At this point we should recall the last paragraph of the Preamble of the Convention which reads: “Being resolved, as the governments of European Countries which are likeminded and have a common heritage of political traditions, ideals, freedom and the rule of law, to take the first steps for the collective enforcement of certain of the rights stated in the Universal Declaration”.47

5 TORT CASES AGAINST STATE AUTHORITIES OR OFFICIALS AND THE DICEAN RULE OF LAW

The common feature of the Osman case and the Jaccard-Veloso applications is that, in both situations, the Convention State was not the perpetrator of the violence, but, by preventing access to the courts for public policy reasons, the State risked being found in violation of its international human rights treaty obligations. Alongside the logic of the arguments outlined above, it is suggested that there is a deeper theoretical principle at work. Denying an individual the right to sue public authorities (whether the police or military as general entities or specific officials in tort) offends against the principle of the rule of law. Following Dicey, the rule of law can be understood as structured by three elements, the second of which is particularly important in the present context: “It means . . . equality before the law, or the equal subjection of all classes to the ordinary law of the land administered by the ordinary Law Courts; the ‘rule of law’ in this sense excludes the idea of any exemption of officials or others from the duty of obedience to the law which governs other citizens or from the jurisdiction of the ordinary tribunals; there can be with us nothing really corresponding to the ‘administrative law’ (droit administratif) of the ‘administrative tribunals’ (tribunaux administratifs) of France. The notion which lies at the bottom of the ‘administrative law’ known to foreign countries is that affairs or disputes in which the government or its servants are beyond the sphere of the civil Courts and must be dealt with by special and more or less official bodies. This idea is utterly unknown to the law of England, and indeed is fundamentally inconsistent with our traditions and customs.”48

One does not have to embrace Dicey’s antipathy to administrative law to distil a basic point as valid now as then. The point is that, at least in the absence of a special system of administrative courts that deal with the conduct in question so as to provide reasonable effective remedies, it is essential that public officials be 47 The applicants submitted that, where future violations would take place in a Contracting Party to the Convention, the case for a judgment which would prevent a violation is even stronger than in the Soering-type situation where the abuse of rights would have taken place in a non-Contracting Party. The role of the Court under article 19 is to ensure the observance of the engagements undertaken by the High Contracting Parties. Article 19 reads in part: “To ensure the observance of the engagements undertaken by the High Contracting Parties in the Convention and the Protocols thereto, there shall be a European Court of Human Rights.” 48 A V Dicey, The Law of the Constitution, 8th ed. (London, Macmillan, 1915) at 120–1.

532 A Clapham accountable rather than immune from civil suit. Dicey elaborates on this principle and sees in it a thread which runs through English constitutionalism: “[T]here runs through the English constitution that inseparable connection between the means of enforcing a right and the right to be enforced which is the strength of judicial legislation. The saw, ubi jus ibi remedium [where there is a right there is a remedy], becomes from this point of view something much more important than a mere tautologous proposition. In its bearing upon constitutional law, it means that the Englishmen whose labours gradually framed the complicated set of laws and institutions which we call the Constitution, fixed their minds far more intently on providing remedies for the enforcement of particular rights or (what is merely the same thing looked at from the other side) for averting definite wrongs, than upon any declaration of the Rights of Man or of Englishmen.”49

The Convention is imbued with similar principles that have been gradually outlined in its jurisprudence. These principles include the importance of ensuring respect for the “rule of law” and the importance of making rights effective through “remedies” rather than leaving them as mere exhortations. In fact, since the entry into force of the Eleventh Protocol to the Convention with its establishment of direct access by individuals to the European Court, the Convention protection system is now founded more than ever on the twinned concepts of individual remedies and access for individuals to adjudication of their complaints about violations of their rights under the Convention. Most recently the case of Smith and Grady has highlighted how even the institution of judicial review of administrative decisions is inadequate under the Convention.50 Smith and Grady concerned the exclusion of lesbians and gay men from the armed services in the United Kingdom. The applicants had no opportunity to have a court review the human rights impact of the policy. The policy was merely reviewed for irrationality by UK courts and found to survive such scrutiny. The European Court held that there had been a violation of article 13 of the Convention as the UK courts had placed the threshold so high for a finding of irrationality concerning the Ministry of Defence policy regarding lesbians and gay men that the applicants had no effective remedy in relation to the violation of the respect for their private lives guaranteed by article 8 of the Convention.51 Following this judgment, we can say that the Convention demands that an individual must have a remedy at the national level so that decisions by the public authorities are reviewed in the same way that the European Court would review action alleged to violate a Convention right.

49

A V Dicey, The Law of the Constitution, 8th ed. (1915) at 118. Smith and Grady v. The United Kingdom, ECHR, Judgment of 27 September 1999. 51 Article 13 reads: “Everyone whose rights and freedoms as set forth in this Convention are violated shall have an effective remedy before a national authority notwithstanding that the violation has been committed by persons acting in an official capacity.” 50

The European Convention on Human Rights 533 Of course, Dicey’s point about the rule of law was the basic point that public officials should be subjected to suits in the courts in the same way that private persons are accountable for violations of the law. But a principled interpretation of the Convention case law has taken this point one step further and suggests that the authorities can be subject to a civil suit in tort even where the violence was not perpetrated by the authorities’ own agents. These elements of the rule of law imply that the State has a duty to allow suits not only against public officials but also against non-State actors. Crucially, for present purposes, this duty has a transnational dimension and must be taken to include non-State actors who are, or were, officials from other States.52 In some situations, officials from other States will remain immune from suit in the courts under the rules of international law. But, as we have seen in the context of the House of Lords judgment in the Pinochet case, this will not be the case for human rights violations that simultaneously amount to certain international crimes. Approached in this way, the term “non-State actors” refers to actors who are not actors of the State whose positive duties are at issue; thus, Pinochet while a State official of Chile, was and is a “non-State” person viewed through the lens of the UK’s responsibility in respect of his conduct. The UK’s (and other Convention Parties’) duties to provide access to their courts are still obligations with respect to conduct in the private sphere in the sense that they are not obligations which entail direct responsibility for violations committed by the UK’s own public officials. This “tortuous” reasoning may seem too tenuous to convince some readers that we are in the presence of clear-cut rights and remedies under the Convention system. But all that is being asserted is that a failure to allow access to court to complain about torture can give rise to obligations under human rights treaties such as the European Convention and not only under extraditeor-prosecute treaties such as CAT.53 Should the prosecutorial authorities fail to prosecute, the Convention rights become critical in the context of any decision to release the alleged offender.

6 FINAL REMARKS

It is hoped that the foregoing discussion has forced us to consider the possible scope of the State’s obligations under international human rights law when faced with an alleged foreign torturer. As we have seen, victims are already protected under the Convention from violence abroad in the context of deportation or extradition.54 52

See Human Rights in the Private Sphere, supra n. 2 at 124–5. Note article 7(1) of CAT reads: “The State Party in the territory under whose jurisdiction a person alleged to have committed any offence referred to in Article 4 is found shall in the cases contemplated in Article 5, if it does not extradite him, submit the case to its competent authorities for the purpose of prosecution.” 54 See Soering and H.L.R, supra n. 39. 53

534 A Clapham This short chapter has sought to argue that matters are beginning to progress further. Notably, the Convention demands access to a court for victims in order that there can be a determination of their civil rights in a tort case against State authorities even where the original act of violence was committed by a private individual. Any immunity for the authorities will have to be proportionate to a legitimate aim. (Osman). It remains possible, of course, that the European Court of Human Rights, or any other international human rights jurisdiction, will be prepared to hold that a State’s access-to-court obligations apply in the context of acts already committed outside the jurisdiction by individuals unconnected with the State party to whose courts access is sought. Where the acts were carried out by foreign agents abroad there is no reason automatically to presume that the Convention is inapplicable.55 A decision by a Convention State to release an accused torturer, either directly or indirectly by virtue of the operation of an immunity doctrine in its courts, denies the victims of torture the enjoyment of their Convention rights not only in the State of custody but also in other European Convention States. Considering the treaty obligation in article 19 of the Convention and its last preambular paragraph together, one would expect the European Court to ensure that, at least where other European States are ready to assume jurisdiction of an accused torturer, then the State of custody would have to consider the civil rights of victims of torture.

P OSTSCRIPT This chapter has attempted to show that the drafting, the wording and the subsequent interpretation of some of the leading general human rights treaties suggest that effective human rights protection demands that the obligations of States parties must extend beyond the State’s immediate obligations concerning its own actors and into the private sphere. If international human rights law is to be effective and to fulfil the promise of the rule of law, it must also extend to foreign torturers—in other words, into a transnational private sphere. A number of developments related to the pursuit of justice against Senator Pinochet have occured since the completion of this chapter. The publisher and editor of this volume have taken the view that these developments warrant a 55 It appears that the European Court will soon have the opportunity to rule on at least one aspect of this question. It will hear the Al-Adsani case in which an individual tried to sue Kuwait in UK courts for alleged acts of torture that occurred in Kuwait. The English Court of Appeal found that state immunity protected access to court for a civil remedy Kuwait from the lawsuit and the case was dismissed: Al-Adsani v. Kuwait, (1994) 100 ILR 465. The access to court for a civil remedy issue in Osman is thus raised with respect to acts by a foreign state outside the UK and the specific question will be whether the statutory immunity of foreign states for torts committed abroad is contrary to article 6(1). Even if the Court limits its ruling to the question of tort suits against foreign states for torture, similar issues arise where immunity is invoked by an individual who is sued in tort.

The European Convention on Human Rights 535 special mention, for the sake of making the coverage of the ‘Pinochet affair’ as up-to-date as possible before going to press. To that end, this postscript has been prepared in January 2001. It has been added as Appendix 4 at the end of the volume.

20

Civil Remedies for Torture Committed Abroad: An Obligation under the Convention against Torture? ANDREW BYRNES 1

1 INTRODUCTION

of torture by the community of States is reflected in numerous declarations, treaties, national constitutions and laws, and other forms of State practice. The prohibition against torture is generally accepted—at least so far as systematic State torture is concerned—as a norm of jus cogens and a crime in respect of which States may, but rarely do, exercise universal criminal jurisdiction under customary international law. The power of States to exercise jurisdiction in relation to extraterritorial acts of torture falling short of the systematic practice of torture has been expanded to include treaty obligations—in particular the obligations under the grave breaches provisions of the Geneva Conventions,2 and under the Convention against Torture (CAT).3

T

HE UNIVERSAL CONDEMNATION

1

Associate Professor, Faculty of Law, University of Hong Kong. Each of the Geneva Conventions contains a provision which identifies “grave breaches” of the Convention; these are breaches which are considered to be particularly serious violations of the Convention in question. They include willful killing, torture or inhuman treatment, including biological experiments, willfully causing great suffering or serious injury to body or health, and extensive destruction and appropriation of property, not justified by military necessity and carried out unlawfully and wantonly. Geneva Convention for the Amelioration of the Conditions of the Wounded and Sick in Armed Forces in the Field, 12 Aug. 1949, 75 UNTS 970, art. 50 [hereinafter “Geneva Convention I”]; Geneva Convention for the Amelioration of the Conditions of the Wounded, Sick and Shipwrecked Members of Armed Forces at Sea, 12 Aug. 1949, 75 UNTS 971, art. 51 [hereinafter “Geneva Convention II”]; Geneva Convention relative to the Treatment of Prisoners of War, 12 Aug. 1949, 75 UNTS 972, art. 130 [hereinafter “Geneva Convention III”]; Geneva Convention relative to the Protection of Civilian Persons in Time of War, 12 Aug. 1949, 75 UNTS 973, art. 147 [hereinafter “Geneva Convention IV”]. All States parties have an obligation to enact domestic legislation to permit the trial of persons suspected of having committed these offences, to search for such persons, and to put them on trial or to hand them over to another State party for trial: Geneva Convention I, art. 49; Geneva Convention II, art. 50; Geneva Convention III, art. 129; Geneva Convention IV, art. 146. 3 Convention against Torture and Other Cruel, Inhuman, or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, at 197, 2

538 A Byrnes The reality of the exercise of universal jurisdiction has fallen far short of the rhetoric, and of these clear treaty obligations. Although the 1990s have seen a number of cases involving the exercise of universal criminal jurisdiction over former rulers, such as Augusto Pinochet, or those alleged to have committed war crimes or crimes against humanity in the tragedies in the former Yugoslavia and Rwanda, this is a fairly recent phenomenon.4 The record has to date been sparse, even under the provisions of the Geneva Conventions which lay down a broad and unqualified obligation of States parties in relation to persons accused of grave breaches of those Conventions.5 These obligations have tended to focus exclusively on or, at least, to emphasize the taking of criminal sanctions against the alleged offenders. While most acts of torture as defined in the CAT would constitute a civil wrong of some sort under most legal systems, there has been relatively little resort by private individuals to civil remedies against those who may have committed torture outside a country but who come within the jurisdiction of the courts of those countries.6 Civil actions brought in the United States remain the distinct exception.7 U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), 1465 UNTS 85. See generally J Herman Burgers & H Danelius, The United Nations Convention against Torture (Dordrecht, Martinus Nijhoff, 1988) [hereinafter Burgers & Danelius]. 4 See, generally, International Law Association Committee on Human Rights Law and Practice, “The Exercise of Universal Jurisdiction in respect of Gross Human Rights Offences: Third Report”, prepared for the 69th Conference of the International Law Association, London, July 2000 (Rapporteur Menno T. Kamminga); International Law Association Committee on Human Rights Law and Practice, “The Exercise of Universal Jurisdiction in respect of Gross Human Rights Offences: First Report”, in Report of the Sixty-eighth Conference of the International Law Association, Taipei, May 1998, 563–76 (Rapporteur Menno T. Kamminga); Redress, Universal jurisdiction in Europe: Criminal prosecutions in Europe since 1990 for war crimes, crimes against humanity, torture and genocide (Redress, 30 June 1999), also available on-line at http://www. redress.org/unijeur.html. 5 Torture is addressed explicitly by a number of provisions in the 1949 Geneva Conventions, supra n. 2. These include not only the grave breaches provisions, but also common article 3 (Geneva Conventions I–IV, art. 3) and Geneva Convention I, art. 12; Geneva Convention II, art. 12; Geneva Convention III, art. 17; and Geneva Convention IV, art. 147. However, it is only in respect of grave breaches that States parties assume an express obligation to exercise universal jurisdiction. The Geneva Conventions do not define torture: for a discussion of that concept under the Conventions and the recent jurisprudence of the Yugoslav and Rwandan tribunals, see A Byrnes, “Torture and other offences involving the violation of the physical or mental integrity of the human person”, in G K McDonald and O Swaak-Goldman (eds), The Law and Enforcement of International Offenses— The Experience of International and National Courts (Dordrecht, Kluwer, 2000) 197. 6 For one of the few examples outside the United States, see Al-Adsani v. Government of Kuwait (1994) 100 ILR 465 (Eng CA) and (1996) 107 ILR 536 (Eng CA) (attempt to bring action in tort alleging torture unsuccessful on grounds of State immunity). 7 For a recent review, see J F Murphy, “Civil Liability for the Commission of International Crimes as an Alternative to Criminal Prosecution”, (1999) 12 Harvard HRJ 1 at 33–6. These civil actions have been primarily brought in US federal courts under the Alien Tort Claims Act of 1789, [ATCA] 28 USC §1350, and the Torture Victim Protection Act of 1991, Pub. L . No 102–256, 106 Stat. 73, 28 USC §1350 n (1992) [hereinafter “TVPA”]. The ATCA confers jurisdiction on federal courts over an action in tort brought by an alien where the tort is committed in violation of the law of nations or a treaty of the United States, and was given new life by the decision of the 2nd Circuit Court of Appeals in Filártiga v. Peña-Irala, 630 F. 2d 876 (2d. Cir. 1980), which held that an action could be brought relating to torture by one Paraguayan citizen against another in relation to acts which had occurred in Paraguay. The passage of the TVPA was in part a response to doubts raised

Civil Remedies for Torture Committed Abroad 539 Whether a civil action can be successfully pursued against a foreign perpetrator before the courts of a State (“the forum State”) depends primarily on the private international law of that State, including its choice of law and jurisdictional rules. Many States permit actions to be brought against a person in relation to acts which have occurred abroad. The forum State’s international obligations, including its duties to ensure respect for human rights, may affect directly or indirectly the content of those rules and the exercise of jurisdiction by the State. The purpose of this chapter is to explore aspects of the relationship between a State’s obligations under international human rights law and that State’s private international law regime. The principal question examined is whether a State party to the CAT is obliged not only to extradite or prosecute an alleged torturer found on its territory, but also to ensure that civil remedies are available under its national law for persons who have been subjected to torture in another State by persons for whom the forum State bears no international responsibility. The question is twofold: whether international law obligations in relation to the prevention and punishment of torture have any relevance to the jurisdictional and other private international law of a State, insofar as they may apply to cases in which the underlying civil claim that has been brought is in effect one of torture, and whether the CAT creates a specific obligation on States parties to provide a civil remedy with respect to torture committed abroad. At first sight, it appears attractive to suggest that the CAT, as part of its goal of pursuing alleged torturers wherever they may flee, obliges States not only to deploy their criminal justice systems against violators, but also to ensure that civil remedies can be sought against them. Civil actions can be more effective in providing redress for victims—not only in the form of awards of compensation, but also in terms of control of the action by those affected and the production of a more complete historical record.8 This chapter examines the arguments that can be made to support this expansive interpretation of the CAT. It concludes that, despite the fact that the objectives and language of the CAT provide some support for such an interpretation of it, the better view is that the CAT does not require that the resources of a State party’s civil law system be made available to persons who wish to pursue actions or other remedies for acts of torture which occurred outside that State—and for in other cases that the ATCA did in fact permit such actions to be brought, in particular the doubts expressed by Judge Bork in Tel-Oren v. Libyan Arab Republic, 726 F.2d 774 at 799 (D.C. Cir. 1984), cert. denied, 470 U.S. 1003 (1985). It provided that the federal courts had jurisdiction to hear claims by both aliens and US citizens in relation to torture and extrajudicial killings committed outside the United States. See generally M Swan, “International Human Rights Tort Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines”, chapter 3 in this volume, for an overview discussion of the TVPA; C Haffke, “The Torture Victim Protection Act: More Symbol than Substance”, (1994) 43 Emory LJ 1467, R Schwarz, “ ‘And Tomorrow?’ The Torture Victim Protection Act”, (1994) 11 Arizona JI&CL 271; and Y Gery, “The Torture Victim Protection Act: Raising Issues of Legitimacy”, (1993) 26 George Washington JIL&E 597. 8 J Alvarez, “Lessons of the Tadic Judgment”, (1998) 96 Michigan LR 2031, 2101–12, cited in Murphy, supra n. 7, at 48.

540 A Byrnes which that State is not otherwise responsible.9 Nevertheless, the CAT may have some impact on the law and practice in those States which allow actions relating to extraterritorial torture to be brought in their courts.

2 PRIVATE INTERNATIONAL LAW BACKGROUND

Although many of the private international law aspects of bringing a civil action in relation to torture that occurred abroad are extensively discussed elsewhere in this volume, it is useful to set out a number of propositions that are relevant background to the subsequent discussion in this chapter of article 14, the CAT provision on civil redress: • Acts of torture will almost inevitably constitute a civil wrong under the law of the forum State. In other words, if the acts had been committed in that State, they would give rise to a civil cause of action such as assault or battery. • Under many legal systems, it is possible to bring an action against a defendant in relation to acts committed outside the jurisdiction of the forum State. This involves the court’s rules of civil procedure according it subject-matter jurisdiction over certain events or activities abroad as well as personal jurisdiction over the defendant. This personal jurisdiction may be secured by service of process within the territorial jurisdiction or, in more narrowly-defined circumstances, outside it. • The exercise of jurisdiction over individual defendants will be subject to any personal immunities enjoyed by the defendant under national law, including any immunities under international law that have the force of law within the domestic legal system in question. • The law to be applied to the determination of the claim will normally be determined by the private international law rules of the forum. This law may include international norms directly applicable, statutorily or constitutionally incorporated into national laws, or relied on as interpretive aids. • Even when the courts of the forum State have the power to hear such a case, thus possessing basic adjudicative jurisdiction, they may have a discretion, or even a duty, not to proceed with it on the ground of forum non conveniens or equivalent doctrines. For example, under the common law doctrine of forum non conveniens, the courts of the putative forum State must consider whether there is an alternative available forum, and may dismiss or stay proceedings if they determine that another court system is not only available but also more appropriate for hearing the case. Thus, under many legal systems, it is already possible for a victim of torture to institute a civil action against an alleged torturer, even though the events in 9 For example, a person may be tortured abroad, in State X, by agents of State Y. In such a situation, State Y could not plead the extraterritorial location of the torture so as to escape its duty to provide redress for acts of torture for which it bears responsibility.

Civil Remedies for Torture Committed Abroad 541 question may have been committed abroad. What, then, would be the effect of concluding that the CAT obliges a State party to ensure this access? If there were such an obligation, the State would be obliged, first of all, to have or put in place procedural and substantive rules. These would permit the exercise of jurisdiction by the State’s courts over cases involving those civil actions that allege torture as defined in article 1 of the CAT. Secondly, the State would have to ensure access to other remedies, facilities or benefits available under national law, such as a criminal or civil compensation schemes, legal aid to assist a court action, and medical and other rehabilitation services, on the same basis as someone who has suffered a similar injury within the jurisdiction of the State.

3 EXTRATERRITORIALITY AND THE CAT : THE GENERAL SITUATION

One of the central goals of the CAT is to ensure that torturers should enjoy no safe haven from investigation, prosecution and punishment.10 In order to achieve this goal, the CAT imposes explicit obligations on States parties to ensure either that they are in a position to prosecute cases of alleged torture wherever they may have occurred, or to extradite alleged offenders to other States which have jurisdiction over the offender.11 As the number of States parties to the CAT grows, the net widens, so that as time passes, there will be fewer and fewer countries to which an alleged torturer can retreat and be safe from criminal legal process. The CAT contains a number of provisions which explicitly relate to extraterritorial torture, as well as a number which confine their operation to torture within the jurisdiction of a State party. The explicit extraterritorial dimension of the treaty is primarily reflected in its provisions governing the steps a State is obliged to take in the field of criminal law enforcement, in particular the obligation to try or surrender an alleged torturer. The only other provision of the treaty which has an explicit application to events outside a State party is article 3, which provides protection against return or expulsion of a person to a State where the person runs a substantial risk of being tortured. An example of another provision that could also be interpreted as having an operation in respect of extraterritorial torture is article 15 of the CAT. Article 15 provides that a State party is to ensure that any statements established to have been made as a result of torture should not be used as evidence in any proceedings. There would appear to be no reason why this should not be read as applying both to statements resulting from torture which have occurred in that State party and to those extracted in any other country.12 10

See, generally, Byrnes, supra n. 5. For a detailed discussion, see A Boulesbaa, The U.N. Convention on Torture and the Prospects for Enforcement (The Hague, Martinus Nijhoff, 1999) at 175–234. 12 D Sloss, “The Domestication of International Human Rights: Non-Self-Executing Declarations and Human Rights Treaties”, (1999) 24 Yale JIL 129 at 205 n. 362. 11

542 A Byrnes However, other provisions of the treaty do not at first sight seem especially apt to apply extraterritorially. For example, the obligation under article 10 to educate and inform all those involved in the custody, interrogation or treatment of individuals subject to detention appears on its face to apply only to the State’s own officials, or to private sector actors delegated these tasks by the State, as well as others, such as doctors, who are involved in such activities on the territory or under the jurisdiction of the State party.13 The requirement of article 11, that a State party systematically review law and practice relating to detention is even clearer, explicitly stating that it applies to detention “in any territory under its jurisdiction”. Similarly, the provisions of the CAT imposing obligations to investigate alleged cases of torture also have a territorial focus. Article 12 requires a State party to ensure that there is a “prompt and impartial investigation, wherever there is reasonable ground to believe that an act of torture has been committed in any territory under its jurisdiction”, while article 13 obliges a State party to ensure that there is a prompt and impartial investigation of any allegation by a person that “he has been subjected to torture in any territory under its jurisdiction”. The impression that the CAT’s operation was intended to be primarily territorial, except in relation to criminal matters (where extraterritorial application is explicit and central to the purpose of the treaty), is reinforced by the language of the general obligations in articles 2 and 16, which deal with torture and other forms of ill-treatment respectively. Each of these provisions requires a State party to take effective steps to prevent torture or other ill-treatment “in any territory under its jurisdiction”.14

4 THE SCOPE OF ARTICLE 14

Extraterritorial Applicability of Article 14: First Impressions Against this background, we come to article 14, which is intended to ensure that a victim of torture obtains redress for the injury suffered. It provides: “(1) Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture , his dependants shall be entitled to compensation. (2) Nothing in this article shall affect any right of the victim or other persons to compensation which may exist under national law.” 13 It could presumably be read as applying to training provided by one State to officials of another, including where that training is conducted abroad. 14 See generally Boulesbaa, supra n. 11, at 74–6.

Civil Remedies for Torture Committed Abroad 543 The critical issue is whether the CAT imposes the same obligation in relation to foreign torture as it does in relation to domestic torture. This question requires analysis of article 14 of the CAT, against the background of its other provisions and its overall goals. The central issue is whether the reference to “torture” in article 14 refers only to torture for which the forum State itself is responsible, or whether it also includes torture for which another State is responsible, and which took place outside the jurisdiction of the forum State.15 The placement of article 14, together with other articles which are mainly territorially limited, suggests that it should also be understood as limited to requiring remedies only for torture committed under the jurisdiction of the State party (and for which it is responsible).16 To oblige a State party to make its legal system—including, possibly, legal aid, national compensation funds, criminal injuries compensation funds, and rehabilitation facilities—available to all comers is presumably not an obligation to which States parties would have lightly agreed, and one might have expected an explicit statement to that effect if the drafters had intended to impose such an obligation. Indeed, the presence of the article 14(2) savings clause would seem to suggest that, at most, the drafters did not wish to preclude States from adopting a universal approach to redress such as that found in the United States’ Torture Victim Protection Act (TVPA).17 15 A further question is whether the CAT, supra n. 3, requires that a domestic legal order recognize a civil wrong in the terms of article 1—to contain a nominate tort of torture. In the criminal context, the Committee against Torture has consistently maintained that the CAT requires that a State introduce a criminal offence of torture defined as in article 1 as part of its national law, thus recognising a nominate crime of torture. However, it has not as yet expressed the same view in the context of civil remedies, though it has been concerned to ensure substantive coverage: see, e.g., the Committee’s concluding comments on Namibia, U.N. Doc. A/52/44, paras 235 and 240 (1997). A second issue is whether the CAT requires that a civil action in the courts be available or whether alternative avenues of redress would satisfy the obligation. These issues arise in relation to the article 14 obligations of a State party for torture within a State’s own territorial jurisdiction and are not confined to the question of extraterritorial torture. 16 See Murphy, supra n. 7, at 27. 17 See Gery, supra n. 7, at text accompanying nn. 96–9: this commentator has argued that the enactment of legislation such as the TVPA “contradicts” and is “inconsistent” with obligations under the CAT, since the CAT adopted a universalist stance in relation to criminal matters but “rejected” the permissibility of the assertion of universal civil jurisdiction, and further that supporters of universal civil jurisdiction cannot rely on the principle of universal jurisdiction under general international law. However, this position is not sustainable. Even the Reagan and Bush administrations, which opposed the passage of the Act and argued that the TVPA was inconsistent with the approach of the CAT, did not argue that the exercise of extraterritorial jurisdiction over torture was impermissible under international law. The Executive Branch expressed the view that the CAT did not require the provision of a remedy for extraterritorial torture, not that it forbade it, and that the unilateral passage of the TVPA, embodying universal jurisdiction, was not the most effective way of combating torture and could lead to problems of foreign relations. See Schwarz, supra n. 7, at text accompanying nn. 66–82; Gery, supra n. 7, at text accompanying nn. 74–138. The argument that the exercise of civil jurisdiction over extraterritorial torture is not permissible under international law also fails to take into account the well-recognised jurisdiction of many States to adjudicate private law actions involving civil wrongs under established rules of private international law. See also American Law Institute, Restatement of the Law of Foreign Relations (3d), §404, and Torture Victim Protection Act, Hearing and Markup before the Committee on Foreign Affairs and its Subcommittee on Human Rights and International Organisations of the House of Representatives, 100th Cong., 2d Sess, at 53 nn. 44 and 55 (1988) (prepared statement of

544 A Byrnes The reports of States parties to the Committee against Torture do not provide support for a broad reading of the obligation under article 14. Few States have referred, in their reports, to the legal or other entitlements of persons who may have suffered torture abroad and who have then come to the State party. Those States which have touched on the issue have made it clear that they view article 14 as requiring no more than the provision of redress for domestic torture, and that the provision of redress or rehabilitation to victims of extraterritorial torture remains a matter within their discretion—though it may be a matter of obligation that arises under treaties other than the CAT.18 It must also be said that little support for an expansive interpretation of article 14 can be gained—at least to date—from the practice of the Committee against Torture. While the Committee has been assiduous in its insistence on States parties’ obligations in the criminal law context, and in their adherence to article 3 in relation to torture which has occurred or which may occur outside the territory of the State, it has not considered or raised the question of the obligations of a State party in relation to torture outside its territory generally, or with particular reference to article 14.19 It has, on the other hand, stressed the importance of States’ having in place effective criminal sanctions and civil law remedies for torture for which that State or its officials may be responsible.20 In the Association of the Bar of the City of New York, Committee on International Human Rights, 23 Mar. 1988). 18 See, e.g., the Second periodic report of New Zealand, U.N. Doc. CAT/C/29/Add.4 (1997) paras 35–40 (discussing situation of asylum-seekers who may have been subjected to torture); Second periodic report of Germany, U.N. Doc. CAT/C/29/Add.2 (1997) para. 39 (stating that rehabilitation services provided to victims of torture who come to Germany as refugees go “[b]eyond its duties under article 14 of the Convention”); and Initial report of the United States of America U.N. Doc. CAT/C/28/Add.5 para. 268 (2000), (restating consistent Executive position that article 14 requires a State party to provide remedies only for torture committed in its territory). 19 For example, the export by States parties of weapons or instruments that could be used or misused for the purpose of torture is arguably one issue that could have been taken up by the Committee in terms of complicity in torture, but there has been little or no attention devoted to this matter. See generally the discussion in M Gibney, K Tomasevski, and J Vedsted-Hansen, “Transnational State Responsibility for Violations of Human Rights”, (1999) 12 Harvard HRJ 267 at 272, 288 and K Tomasevski, “Foreign Policy and Torture”, in Bertil Dunér (ed.), An End to Torture: Strategies for its Eradication (London and New York, Zed Books, 1998) 183 at 184–5. 20 A similar territorial focus, so far as the provision of remedies is concerned, is reflected in the various efforts to deal with the question of reparation for gross human rights violations and impunity for offenders. Neither in the Van Boven nor the Joinet studies prepared for the United Nations SubCommission on the Prevention of Discrimination and Protection of Minorities on these issues is it suggested that a State party is obliged to provide civil remedies for violations that took place outside its territory, and for which it is not responsible. Theo van Boven, Study concerning the right to restitution, compensation and rehabilitation for victims of gross violations of human rights and fundamental freedoms, U.N. Doc. E/CN.4/Sub.2/1993; Basic Principles and Guidelines on the Right to Reparation for Victims of [Gross] Violations of Human Rights and International Humanitarian Law, U.N. Doc. E/CN.4/1997/104, Annex [known as the Van Boven Principles]; and Question of the Impunity of perpetrators of human rights violations (civil and political), Final report prepared by Mr. Joinet pursuant to Sub-Commission decision 1996/199, U.N. Doc. E/CN.4/Sub.2/1997/20. Nor does the Human Rights Committee in its General comment 20(44) on article 7 of the International Covenant on Civil and Political Rights (prohibition of torture) suggest that it is anything other than violations by that State in respect of which there is an obligation to provide a remedy: U.N. Doc. HRI/GEN/1/Rev.3, at 31 (1997).

Civil Remedies for Torture Committed Abroad 545 those cases where States parties have provided information about the special rehabilitation services they provide to victims of torture who have come from abroad, the Committee has praised such efforts, but has not suggested that the States concerned were obligated by article 14 to provide such services.21

Assessments of Arguments for an Expansive Interpretation of Article 14 Despite the primarily territorial focus of article 14 and related provisions, there are a number of arguments that can be advanced to support a broader interpretation. Firstly, the general tenor of the CAT is to address the scourge of torture, to bring perpetrators to justice and to provide reparation for victims. Accordingly, a broad interpretation of the scope of the treaty is appropriate, especially where the rights of those who have been subjected to torture are in issue. Such purposive construction of human rights treaties is by now a wellaccepted central canon of treaty interpretation. But, purpose cannot be the only referent in interpretation and must be balanced against interpretive evidence pointing in the other direction. Second, it might initially appear that the drafting history of article 14 provides some support for the purposive argument that the article is not confined to torture committed in territory under the jurisdiction of the forum State. The initial Swedish draft of the article submitted to the United Nations Commission on Human Rights contained no express limitation on its territorial scope,22 and the discussion at the 1980 session of the Working Group of the Commission that considered the issue did not address the matter.23 However, at the 1981 session of the Working Group, the Group adopted a Netherlands proposal to include the phrase “committed in any territory under it jurisdiction” after the word “torture”.24 The phrase remained in the draft during the 1982 meetings of the Working Group.25 21 See, e.g., Concluding comments on third periodic report of Denmark, U.N. Doc. A/52/44, para. 177 (1997) (welcoming the provision by the Danish government of subsidies to private organisations involved with the rehabilitation of torture victims). 22 Article 12 of the original Swedish draft: Burgers & Danelius, supra n. 3, at 68, 205; U.N. Doc. E/CN.4/1285 (1978). 23 Burgers & Danelius, supra n. 3, at 68–9; Report of the informal open-ended working group on a Draft Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, (UN Doc. E/CN.4/L.1367), paras 74–81, reproduced in Commission on Human Rights, Report on the Thirty-Sixth Session (4 February–14 March 1980), UN ESCOR 1980, Supp No. 3, at 64–6, UN Doc. E/1980/13 (E/CN.4/408) (1980). 24 Burgers & Danelius, supra n. 3, at 74; Report of the informal open-ended working group on a Draft Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, (UN Doc. E/CN.4/L.1576), paras 40–4, reproduced in Commission on Human Rights, Report on the Thirty-Seventh Session (2 February—13 March 1980), UN ESCOR 1981, Supp No. 5, at 61–2, UN Doc. E/1981/25 (E/CN.4/1475) (1981). 25 Report of the informal open-ended working group on a Draft Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, (UN Doc. E/CN.4/L.1576), paras 40–4, reproduced in Commission on Human Rights, Report on the Thirty-Eighth Session, Addendum, Reports of the informal open-ended working groups of the Commission, at 10, para. 41, UN Doc. E/1982/12/Add.1 (E/CN.4/1982/30/Add.1) (1982).

546 A Byrnes Yet, by the time of adoption of the final version of the draft Convention in 1982, the phrase had disappeared from the text. Unfortunately, neither the travaux préparatoires nor the commentary by Burgers and Danelius on the CAT provide any insight into why this phrase was removed, and it is accordingly difficult to assess whether its disappearance was inadvertent or whether it has some significance. On the one hand, it could be argued that its removal, even if undocumented by the travaux préparatoires, must have been intended to make clear that the revised version was not territorially limited—it cannot be lightly assumed that a crucially important phrase is dropped for no reason. On the other hand, it could be contended that the territorial limitation of the provision must have been seen as being so obvious that it did not need to be spelled out.26 What seems clear enough, however, is that the Working Group expressly decided in 1981 to include the territorial limitation in its draft, but no similar express decision appears to have been taken to revert to the original Swedish unqualified proposal. To some, this might constitute strong evidence of a mistaken omission. This latter view can be seen in the analysis of the provision which accompanied President Reagan’s submission of the CAT to the United States Senate in 1988:27 “The negotiating history of the Convention indicates that Article 14 requires a State Party to provide a private right of action for damages only for acts of torture committed in its territory, not for acts of torture occurring abroad. Article 14 was in fact adopted with express reference to ‘the victim of an act of torture committed in any territory under its jurisdiction’. The italicized wording appear to have been deleted by mistake. This interpretation is confirmed by the absence of discussion of the issue, since the creation of a ‘universal’ right to sue would have been as controversial as was the creation of ‘universal jurisdiction’, if not more so.”28

When it ratified the CAT, the United States entered an understanding to that effect, a statement which has received no objection from other States parties.29

26 David P Stewart, Assistant Legal Advisor, US Department of State, testified in 1990 that, during the negotiations on the CAT, a number of States “considered the issue of states establishing civil jurisdiction over acts that take place abroad and rejected it”. Torture Victim Protection Act of 1989, Hearing before the Subcommittee on Immigration and Refugee Affairs of the Senate Committee on the Judiciary, 101st Congress, 2d Sess., at 31 (1990) [hereinafter Senate TVPA Hearings]. 27 “Summary and Analysis of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment”, in Message from the President of the United States transmitting the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, 20 May 1988, 100th Congress, 2nd Session, Treaty Doc. 100–20, at 13 (1988). See also Senate TVPA Hearings, supra n. 26, at 13 (1990) (prepared statement of John O. McGinnis, Deputy Assistant Attorney General, Office of Legal Counsel, US Department of Justice). 28 Some proponents of the TVPA maintained that the CAT obligated States parties to exercise universal civil jurisdiction over torture. However, the better position would appear to be that, while a State is permitted to exercise such jurisdiction, it is not obliged to. 29 “[I]t is the understanding of the United States that article 14 requires a State Party to provide a private right of action for damages only for acts of torture committed in territory under the jurisdiction of that State Party.” Para II (3), 136 Congressional Record S17,492 (1990), reprinted in (1991)12 HRLJ 276.

Civil Remedies for Torture Committed Abroad 547 Third, some support for a broader reading of the CAT may be derived from the practice of the Committee on the Elimination of Racial Discrimination (the CERD Committee) under the International Convention on the Elimination of All Forms of Racial Discrimination (the CERD Convention).30 Article 3 of the CERD Convention provides that States parties, in addition to condemning racial segregation and apartheid, undertake to “prevent, prohibit and eradicate all practices of this nature in territories under their jurisdiction”. In 1972 the CERD Committee adopted General recommendation III,31 in which it stated that “measures adopted on the national level to give effect to the provisions of the Convention are interrelated with measures taken on the international level to encourage respect everywhere for the principles of the Convention”, and that it welcomed the inclusion in State party reports under the CERD of information concerning States parties’ relations with the apartheid regime in South Africa. The Committee also declared that “all policies, practices or relations which have the effect of supporting, sustaining or encouraging racist regimes are . . . inconsistent with the specific commitment of States parties to condemn racial discrimination and apartheid in accordance with article 3 of the Convention”.32 This was a clear assertion by the Committee that States parties had an obligation not to take actions which would contribute to the violation of the norms embodied in the CERD in the territory of another State by that State. However, the approach of the CERD Committee was controversial, and a number of governments rejected the Committee’s interpretation of the CAT—although the overwhelming majority of members of the General Assembly seemed to endorse it.33 Perhaps significantly, the CERD Committee does not appear to have expressed such a view in the context of the most closely analogous provision of the CERD Convention, article 6. This article provides that a State party to the CERD Convention is obligated to provide remedies for racial discrimination in territory under its jurisdiction. Fourth, contextual interpretation does add something to the argument in favour of reading article 14 broadly, to include extraterritorial acts of torture for which other States bear state responsibility. Article 31(1) of the Vienna Convention on the Law of Treaties states that a provision should be interpreted according to the ordinary meaning of a text, read in its context and in light of its object and purpose. It has already been noted that a purposive approach may provide support for a wider reading of article 14. As for context, the CAT treaty 30 International Convention on the Elimination of All Forms of Racial Discrimination, open for signature 7 Mar. 1966, 660 UNTS 195 (entered into force 4 Jan. 1969). 31 General recommendation III (sixth session, 1972), CERD, U.N. Doc. A/8718, reproduced in U.N. Doc. HRI/GEN/1/Rev.3, at 101 (1997). 32 Decision 2 (XI), U.N. Doc. A/10018 (1975) 33 T Buergenthal, “Implementing the UN Racial Convention”, (1977) 12 Texas ILJ 187 at 194–5; N Lerner, The U.N. International Convention on the Elimination of All Forms of Racial Discrimination (Alphen aan den Rijn, Sijthoff & Noordhoff, 1980) at 110–11; K J Partsch, “The Committee on the Elimination of Racial Discrimination”, in P Alston (ed.), The United Nations and Human Rights: A Critical Appraisal (Oxford, Clarendon Press, 1992) 339 at 359.

548 A Byrnes text, on its face and taken as a whole, suggests that the omission of a territorial qualifier in article 14 must have some significance in light of the express inclusion of such a qualifier in other provisions. Article 14 is an important provision in the scheme of the treaty. Therefore, one might reasonably presume drafters were paying close attention to its wording and were cognizant of the import of a final vote on a provision that on its face was not territorially limited. The omission might seem all the more significant if one recalls that the CERD Convention was available as part of the background normative context for the CAT drafters. The CERD Convention’s redress provision, article 6, explicitly limits its application to conduct within States Parties’ territorial jurisdictions, and was adopted, and quite widely, well before the CAT negotiating process began. Yet, all that being said, the most plausible theory for the omission of the territorial qualifier from article 14 of the CAT remains that of inadvertence, given the lack of evidence in the travaux of an affirmative decision to remove the words “committed in any territory under its jurisdiction”. The United States’ claim that the words had to have been dropped by mistake cannot be lightly dismissed, and must squarely be addressed as a major counter-argument by those wishing instead to emphasize text, purpose and normative context in support of a universalist reading of article 14. Finally, it must also be recognized that treaties are living instruments and that over time they can over time be interpreted in ways which the original drafters did not anticipate or intend. In the field of human rights, there have been many cases where a court or monitoring body has developed expansive interpretations that would not necessarily have met with acceptance by the drafters at the time the treaty was adopted, but which States parties may be prepared to accept many years later as reflecting changed circumstances and expectations. This is part of the dynamic of applying international law. It can frequently be driven by a monitoring body (such as the Committee against Torture) which is committed to generous interpretations of its convention in an effort to realise the treaty’s goals—especially once it has established its standing among the States parties to the treaty. In its initial decade of operation, the Committee has, perhaps not surprisingly, taken a firm but traditional approach to the interpretation of the CAT, which has seen an understandable focus on the extraterritorial criminal elements and national remedies for national torture has provided the Committee with more than enough work. But there are signs that the Committee may be prepared to be more innovative both in its procedures and substance, and not to be constrained by overly formalistic readings of the CAT.34 It is, of course, hard to 34 See, e.g., Elmi v. Australia, Communication No. 120/1998, decision of 14 May 1999, U.N. Doc. CAT/C/22/D/120/1998 (1999), in which the Committee held that it was a violation of article 3 of the Convention to return a person to Somalia, since there was a likelihood that he would be subjected to torture at the hands of non-governmental actors. The Committee rejected the argument that, since Somalia had no central government and the CAT defined torture as actions inflicted by, at the instigation of, or with the acquiescence of, a public official, the Convention had no application. It held that since the opposing factions were exercising certain governmental functions, that was sufficient for them to be considered “public officials” within article 1.

Civil Remedies for Torture Committed Abroad 549 predict whether that trend will continue, and whether it will lead over time to the Committee eventually adopting the view that, notwithstanding the original intentions of the drafters, the full implementation of States’ obligations requires them to exercise extraterritorial civil jurisdiction. What one might envisage in the coming years is that the Committee may begin to encourage States to consider treating article 14, read with the treaty as a whole, as providing for a permissive basis for jurisdiction under international law.

5 CONCLUSION

In conclusion, it is difficult to argue unequivocally that article 14 of the CAT must be interpreted as requiring States parties to provide the same civil right to redress for torture which occurs outside its jurisdiction as it is obliged to provide for torture which is alleged to have occurred within its territorial and other jurisdiction.35 If there were such an obligation, then it would be necessary to address the issue of whether immunity could be claimed by State defendants if actions were brought against them in a foreign State. While immunity in criminal cases is crumbling, even in relation to former heads of state, there seems to be much less willingness to sweep away State immunity in civil cases—even the House of Lords in the Pinochet case firmly reasserted immunity in civil cases where a suit was in effect a suit against a foreign State.36 Yet, even if there is no firm obligation in the CAT which obliges a State to make civil remedies available for extraterritorial torture, there may nonetheless be ways in which the CAT may have an impact on actions which can be brought under existing law. The norms of the CAT might, for example, be taken into account, when deciding whether the law to be applied to a foreign tort should be the lex loci delicti. Or, if that law permits torture outlawed by the CAT, it should be ignored, and the law of the forum or of another jurisdiction should be applied.37 When the court of the forum is considering whether torture is lawful under the law of a jurisdiction where ratified treaties form part of domestic law, the forum court may arguably take the CAT’s provisions (including article 14) into account when deciding whether any requirement of liability under the lex loci delicti is satisfied. Finally, the fact that a person may have been or would be unable to avail himself or herself of the right to redress under article 14 in the 35 See also R Garnett, “The Defence of State Immunity for Acts of Torture”, (1997) 18 Australian YBIL 97 at 109–10. 36 The argument that the CAT necessarily did away with immunity in relation to criminal procedures which was accepted by some judges in Pinochet does not apply with the same force in relation to civil remedies, since the CAT does not clearly oblige States parties to provide remedies for foreign torture. R v. Bow Street Metropolitan Stipendiary Magistrate, ex parte Pinochet Ugarte (No 3) [1999] 2 All ER 97 at 167 (Lord Hutton) and 175, 179 (Lord Millett). 37 For example, under the flexible exception to the traditional double-actionability rule that applies in many common law jurisdictions (see Red Sea Insurance Co. Ltd. v. Bouygues SA [1995] 1 AC 190 (PC)) or an exception in the interests of justice where the lex loci delicti is otherwise normally applied (see Tolofson v. Jensen [1994] 3 SCR 1022 (SCC)).

550 A Byrnes lex loci delicti could be taken into account in determining whether an action should be stayed on the basis of the doctrine of forum non conveniens, on the ground that the plaintiff cannot obtain justice in the natural forum.38 However, if the goal is to enhance the exercise of a universal civil jurisdiction over torture, it remains perfectly possible to invoke existing tort categories, such as assault and battery, in tandem with choice of law rules for tort—including the exceptions built into these rules.39 To the extent this is correct, it seems, at least in common law jurisdictions, that the principal barriers lie not so much in obtaining jurisdiction over an alleged torturer (at least when (s)he enters the jurisdiction), but in overcoming other barriers, in particular the doctrine of forum non conveniens which may persuade a court not to hear such a case, and the question of State immunity (since torture in many cases will involve the acts of officials). While the CAT may be able to contribute somewhat to the expansion of the jurisdictional and choice of law rules of some States, its major contribution may yet be the erosion of the defence of State immunity in cases of torture falling within its definition.40

38 See, e.g., Spiliada Maritime Corporation v. Cansulex Ltd. [1987] AC 460 at 478 (HL) (Lord Goff). 39 See J Orange, “Torture, Tort Choice of Law and Tolofson”, G Virgo, “Characterisation, Choice of Law and Human Rights”, and T Hyland, “International Human Rights Law and the Tort of Torture: What Possibility for Canada?”, chapters 11, 12 and 15 in this volume. 40 On this, see the contribution of the present Chairperson of the Committee Against Torture supra: P Burns (and S. McBurney), “Impunity and the United Nations Convention against Torture: A Shadow Play Without An Ending?”, chapter 10 in this volume.

21

Doing the Right Thing? Foreign Tort Law and Human Rights JAN KLABBERS 1

1 INTRODUCTION H E R E I S , at first sight, something surreal about suing people in foreign courts, in particular when the tort in question has no tangible connection to the forum state. Thus, the chance that the Italian waiter who spilled coffee on the English tourist in a café in Paris, France, will ever be answerable in tort proceedings before a US court seems rather remote. Similarly, a Dutch-produced television set exploding in Ghana will rarely lead to litigation in, say, Canada. Not only do we feel that such actions would be impractical, quite apart from any concerns about causes of action, jurisdiction, or forum non conveniens, we even think them slightly frivolous examples of forum-shopping—not unlike the case in which the winner of a Colombian lottery prize sued in the US over the tardiness with which the prize was awarded.2 Moreover, we have a lingering, if not articulated, sentiment that local issues are best dealt with locally, for the obvious reason: that local courts, or other decision-makers for that matter, will have their fingers closest to the pulse of local communities and are therefore best placed to do justice.3 Yet when it comes to human rights cases, our intuitions take on a radically different character. With most human rights cases,4 we tend to feel that using foreign tort law is one of the greater inventions of the last few decades.5 If

T

1 Professor of International Law, University of Helsinki. This chapter has benefited considerably from the comments of participants in the 12th Helsinki Summer Seminar on International Law and from discussions with Anja Lindroos, Jarna Petman and Jutta Zalud. The usual disclaimer applies. 2 Zapata v. Quinn, 707 F.2d 691 (2d Cir.1983), mentioned in G A Christenson, “Customary International Human Rights Law in Domestic Court Decisions”, (1995/96) 25 Georgia JICL 225 at 235, n. 42. 3 Another expression of the same idea is the principle of subsidiarity: compare article 5 EC, holding in part that “the Community shall take action, in accordance with the principle of subsidiarity, only if and insofar as the objectives of the proposed action cannot be sufficiently achieved by the member States...”. A consolidated version of the EC Treaty is reproduced in (1998) 37 ILM 56. 4 Not all: Zapata, supra n. 2, unsuccessfully invoked a human right to property. 5 Indicative is the plea to use tort law as a complement or substitute to international criminal law enforcement. See J F Murphy, “Civil Liability for the Commission of International Crimes as an Alternative to Criminal Prosecution”, (1999) 12 Harvard Human Rights Journal 1.

554 J Klabbers anything, we wish that we had thought of it earlier, for the practice seems to provide a useful answer to two of our dearest sentiments. One of those sentiments is that of international lawyers, and, by extension, human rights lawyers collectively. Specifically, the familiar argument that international law is not really law, for lack of judicial enforcement, can be felicitously countered if we can claim that aspects of international law are habitually addressed by domestic courts. Thus, transnational public law litigation does wonders for our collective self-confidence.6 The other sentiment is the high moral ground occupied by human rights generally.7 It is difficult to argue with or about human rights without either being labelled or feeling obnoxious or reactionary, or being labelled the kind of communitarian willing to subordinate the individual to the greater good of the group. Surely, that is the shortest way to totalitarianism, it will be pointed out. In addition, any critique of human rights litigation, or human rights more generally, runs the risk of being seen to make a mockery of the plight of the victims.8 Human rights are, to put it in terms which appear only slightly exaggerated, the gospel of the late twentieth and the early twenty-first centuries—the last straw to clutch in a world otherwise filled with uncertainties and existential Angst. They embody, as one astute observer has put it, a second reformation.9 Perhaps as a result, transnational public law litigation has remained an undertheorised topic. Not only are critical studies hard to come by,10 but its advocates, following the well-established tradition of “diluted pragmatism”,11 6 This phrase has been borrowed from Harold Koh: H H Koh, “Transnational Public Law Litigation”, (1991) 100 Yale LJ 2347. I will henceforth use this phrase to refer to the practice of using foreign tort law in human rights cases. 7 Illustrated, perhaps, by the contention that customary international law does not allow the exercise of jurisdiction over non-commercial torts committed abroad, partly because of differing local public policy standards, but that human rights may “benefit from a shared conceptual universe of legality”. M Byers, Custom, Power and the Power of Rules (Cambridge, Cambridge University Press, 1999) 71, 83. 8 The general attitude is well captured by Jack Donnelly in discussing international enforcement: “. . . [A]ny victim (or potential victim) who is helped is a victory for international action, wherever that person resides.” J Donnelly, “State Sovereignty and International Intervention: The Case of Human Rights”, in G M Lyons and M Mastanduno (eds), Beyond Westphalia? State Sovereignty and International Intervention (Baltimore, Johns Hopkins Press, 1995) 115, at 128. 9 Z Bauman, In Search of Politics (Cambridge, Polity, 1999), esp. at 157. 10 A possible example, concentrating on private law, is C Focarelli, “The Right of Aliens Not to be Subject to So-called ‘Excessive’ Civil Jurisdiction” in B Conforti and F Francioni (eds), Enforcing International Human Rights in Domestic Courts (The Hague, Kluwer, 1997), 441. Rather straightforward is Hassan’s complaint about intervention: F Hassan, “A Conflict of Philosophies: The Filártiga Jurisprudence”, (1983) 32 ICLQ 250. A most vigorous critique is J-M Simon, “The Alien Tort Claims Act: Justice or Show Trials?”, (1993) 11 Boston University International Law Journal 1. It may not be irrelevant that Simon was the photographer-author of a book that, according to the editor of this volume, is considered by many to have played an important role in making known the horrible human rights conditions in Guatemala in the mid-1980s. See J-M Simon, Guatemala: Eternal Spring, Eternal Tyranny (New York/London, W.W. Norton, 1987). Simon’s critique of transnational human rights litigation in the US is most assuredly from someone with impeccable human rights credentials. 11 Ryan coins this term in reference to US legal developments in the 1920s and 1930s. See A Ryan, John Dewey and the High Tide of American Liberalism (New York, W.W. Norton, 1995) at 89.

Foreign Tort Law and Human Rights 555 have rarely put things in perspective12—except perhaps to have historical analysis take the place of normative argument.13 Accordingly, this chapter will be a first brief attempt to investigate the legitimacy of transnational public law litigation. It follows that several issues will be omitted. Thus, I will refrain from questioning whether local judges are sufficiently versed in international law to warrant suggestions that litigation is important in view of local courts’ role in enunciation of international legal norms.14 For similar reasons, I will not address a host of practicalities: whether or not witnesses can or should be transported; problems of translating pertinent documents, and evidence-gathering in general; the extent to which judges rely on affidavits introduced on behalf of the claimants; having to find and pay for a lawyer abroad; and even the question of the often unrecoverable costs of proceedings, including the issue of whether the local taxpayer should ultimately foot the bill so as to allow for proceedings between foreigners over something that took place abroad.15 Instead, I will focus on two issues related to the propriety of transnational public law litigation. The first involves the systemic level: does transnational public law litigation amount to unjustified intervention, both of judges into democratic politics and of foreign states into the affairs of other societies? Secondly, there is the circumstance that we tend to think that human rights litigation abroad provides a great service to humankind. What, if anything, does that tell us about ourselves and the societies we live in? Clearly, the analysis on both levels depends to a large extent on one’s conception of human rights. Therefore, I shall begin by providing a sketchy conceptualisation, loosely based on a mixture of Arendtian republicanism and insights deriving from critical legal studies and critical social theory. Thus, I distance myself from both the dominant liberal tradition, which generally—and instinctively—applauds transnational public law litigation, and from certain kinds of communitarianism. Here, however, it is important to acknowledge at the outset that my approach has some things in common with the latter. Straightforward communitarian critiques of transnational public law litigation would emphasise that there is nothing universal about human rights, or, at a minimum, much less than is commonly assumed by mainstream humanrights liberals. This is especially seen as the case when a communitarian considers the need to balance public-good rationales with the protection of individual rights when interpreting and applying abstract human rights standards to concrete social contexts. As adjudication by foreign courts involves balancing the 12 Instead, their aim is to ensure “that the rule of law becomes a reality in the international human rights arena”. See B Stephens, “Litigating Customary International Human Rights Norms”, (1995/96) 25 Georgia JICL 191. 13 A particularly brilliant example is A-M Burley, “The Alien Tort Statute and the Judiciary Act of 1789: a Badge of Honor”, (1989) 83 AJIL 461. 14 As suggested by Koh, supra n. 6, at 2368. 15 Useful discussions of some of those issues include Simon, supra n. 10, and R G Steinhardt, “Fulfilling the Promise of Filartiga: Litigating Human Rights Claims Against the Estate of Ferdinand Marcos”, (1995) 20 Yale JIL 65.

556 J Klabbers individual right with the position of the local community at issue, including its social and cultural sensitivities, these courts ought not to get involved. While I accept elements of such a communitarian argument, I depart from the basic conception of human rights most communitarians tend to share with liberals. Both accept at the very least that human rights are to be conceived of as intrinsic, worthy of being protected as rights in their own right, whether they are universal or particular to social and historical context. Instead, I will sketch out a more instrumental conception of human rights primarily as guarantors and facilitators of politics. In this view, human rights ultimately amount, as Hannah Arendt could have put it, to rights to have rights. By the expression “rights to have rights”, I wish to tap into Arendt’s view that the overriding condition for any enjoyment of human rights is to have a nationality or, in more accurate terms, membership in a political community.16 At one level this, of course, speaks to a certain primacy that must be given to self-determination of political communities, and freedom from foreign decision-making about choices that should instead be made from within that community.17 However, “nationality” is used more metaphorically than literally: it is not membership in states as such that is at issue, but instead membership, as already noted, in polities—in bounded political communities in which membership is capable of being active in the sense of being at least potentially participatory.18 In this sense, the notion of a “right to have rights” may not fully capture the entirety of the conception of human rights I wish to outline in the next section; we may well wish to speak also of rights to participate in deciding which rights we do have. My conception of human rights is, thus, communitarian in a certain sense only, namely in terms of a theory of the appropriate allocation of institutional power to decide. Whether it is communitarian in a substantively anti-liberal sense is a different matter: to argue against domestic courts or foreign states, and thus doubly so against foreign courts, being the legitimate actors to decide the content and scope of human rights says very little about which approach to human rights I would advocate as a member of a political community.

2 A CONCEPTION OF HUMAN RIGHTS

While human rights should serve to protect human dignity, they need not do so directly. Instead, they should primarily be seen as providing protection in indirect fashion, that is, by facilitating the conduct of politics or, even more 16 See, e.g., H Arendt, The Origins of Totalitarianism (San Diego, Harvest, no year, first published 1951) 296–7. 17 It is thus important to make it clear that what is being defended is not “sovereignty” of states as much as the political community for whom sovereignty is institutionalised. 18 For an argument that the right to political participation is a kind of right which speaks against an institutionalisation of decision-making power in the judiciary by way of the constitutionalisation of rights (in the UK context), see J Waldron, “A Right-Based Critique of Constitutional Rights”, (1993) 13 Oxford Jl Legal Studies 18.

Foreign Tort Law and Human Rights 557 fundamentally, by helping create a body politic and thereby making politics possible in the first place.19 Without human rights, it is difficult to think of conducting any form of politics whatsoever. Meaningful political debate not only presupposes protection of such freedoms as those of expression, religion and assembly but also presupposes freedom from arbitrary discrimination or torture—at least for those participating in the debate.20 That being said, it is immediately useful to recall the comments with which I ended the last section, and to note that the view that protection of those substantive (and—yes—classically liberal) values necessary to guarantee political participation does not entail that enforcement of those substantive values should come from outside the political process itself, from courts let alone from foreign courts. Politics, in the sense of meaningful and unrestrained debate about our common future, unrestrained by any concerns for one’s immediate survival, should be understood as the life-blood of community, not least when communities are increasingly characterised by plurality and diversity.21 And it is precisely by safeguarding politics that human rights serve to protect human dignity. For, as Arendt taught us, politics is the main safeguard we have against evil.22 Indeed, it can be no other way. Given man’s proclivity for changing an existing set of values for a new one, mid-twentieth-century Germany being the prime example, any moral connotation attached to human rights is bound to remain unpersuasive. However paradoxical—indeed frightening—it may seem to invoke an ideal of politics as the only defence to pathologically perverse politics such as those of the Nazis, it is simply an illusion to hope to protect politics from itself by creating an institutional legal realm that purports to be based on timeless or essential moral truths.23 Instead of contrasting human rights to politics, human rights are better seen as facilitators of politics, enabling us to engage in free discussion and debate. Nothing more, but certainly nothing less either. If this view of human rights as safeguarding politics is plausible, then a few important consequences follow. Firstly, on such a view, the very label “human rights” is something of a misnomer. Popular thought notwithstanding, human rights do not so much aspire to safeguard human beings qua individuals as seek to make possible their collective existence. Politics, after all, is only conceivable in a collective setting, and follows from the very plurality of the human condition.24 Human rights relate to human beings as they live together, not as each lives separately. 19 Similarly, S Benhabib, “Judgment and the Moral Foundations of Politics in Arendt’s Thought”, (1988) 16 Political Theory 29. 20 Compare H Arendt, On Revolution (London, Penguin, 1990, first published 1963) at 108. 21 Particularly helpful on this point is M Canovan, Hannah Arendt: A Reinterpretation of Her Political Thought (Cambridge, Cambridge University Press, 1992). 22 H Arendt, supra n. 16. Useful here is J McGowan, Hannah Arendt: An Introduction (Minneapolis, University of Minnesota Press, 1998). 23 On this point, see E M Meade, “The Commodification of Values” in L May and J Kohn (eds), Hannah Arendt Twenty Years Later (Cambridge , MA, MIT Press, 1996) 107. 24 See generally H Arendt, The Human Condition (Chicago, University of Chicago Press, 1958).

558 J Klabbers A second corollary is that, to put it in terms of a familiar if awkward dichotomy, human rights find their proper place not in the private sphere, but in the public sphere or, more accurately perhaps, in what Bauman calls the intermediary space between public and private. In Bauman’s view, the public and private spheres meet at an intermediary agora which can guarantee autonomy both for a society and for its members, but which is under constant attack from two directions.25 Traditionally, authoritarian tendencies channelled through state power have threatened the agora, but, nowadays, this space is increasingly under fire from private actors with growing, while yet elusive, power: companies, speculators and the like, rather than accountable public servants.26 Understanding such exercises of private power as analogous to public power is one thing; it is another thing entirely to allow human rights to be conceptualised as the normative basis for legal regulation of harm-causing behaviour amongst private persons unlinked to the exercise either of public power, or of power closely associated with it. After all, the circumstance that the political space may be threatened by the private sector by no means implies that, on such a basis, private relations ought to be invaded by public power and reconceptualised in terms of public law. Thus, it is somewhat odd to think of the prohibition of murder, even mass murder, as one elaboration of the human right to life. Also, no matter how horrendous spousal abuse is, it somehow appears counterintuitive to think of it in terms of torture.27 As I have sought to articulate, human rights typically deal with individuals as part of larger communities. By contrast, private law, and tort law in particular, brackets any social relation that individuals may have, instead focusing on the purely bilateral relation between victim and tortfeasor.28 The hallmark of human rights is precisely that they cannot be reduced to such pure, asocial bilateral relations. Thus, if torture paradigmatically involves official acts, such as those committed by government agents or, at least, under colour of law, then it becomes awkward to completely ignore this official aspect and instead focus on the individual relation between torturer and victim. To do so would be tantamount to denying the official element of torture, and thereby to render it indistinguishable from assault or battery or other forms of inflicting pain.29 25 See Bauman, supra n. 9: Bauman’s conception seems thoroughly Arendtian: her polis and his agora appear to embody the same type of thought. The difference is perhaps that Bauman would readily include social and economic relations in his conception of the public sphere, thus prompting him to reintroduce a sphere in which unencumbered politics can take place, with reference to ancient Greece. 26 Bauman, supra n. 9, esp. at 86–100. To be sure, he does not suggest that human rights can safeguard the agora. Instead, in a more communitarian vein he hints at the possibly detrimental impact of a rights-based society (ibid., 157–8). 27 For a powerful plea to that effect, in part, see C A MacKinnon, “Crimes of War, Crimes of Peace”, in S Shute and S Hurley (eds), On Human Rights (New York, Basic Books, 1993) 83. 28 See, generally, R W Wright, “Right, Justice, and Tort Law”, in D G Owen (ed.), Philosophical Foundations of Tort Law (Oxford, Oxford University Press, 1995) 159. 29 And render jurisdictional issues more complicated, for, while torture may perhaps invite universal jurisdiction, assault and battery probably do not.

Foreign Tort Law and Human Rights 559 A third corollary of such a view of human rights is not only that any particular content that may be given to them is most likely to be influenced by local considerations, but also that it cannot be any other way in a world marked by diversity of cultures and communities.30 While the abstract outlines of the main human rights may owe much to an appeal to universal humanity,31 the precise contours of a right within a given community are inevitably the product of that particular community’s historically-situated understandings. The question of whether a group of neo-Nazis will be allowed to march through a neighbourhood of Holocaust survivors cannot be given a universal answer. It depends on the value placed on freedom of speech, as well as on the sensitivities surrounding the Holocaust; the weighing of these contending factors may, and should, lead to radically different results from one community to the next.32 Hence, to the extent that human rights can be said to be universal, they are so only in an abstract sense; any concretisation is bound to be coloured by local sentiments, opinions, and sensitivities.33 And, to the extent that considerations of morality find their way back into discourse at the local level, those considerations—or, at the very least, their application—will necessarily vary in accordance with localised concerns and associated understandings. A final corollary is that this political–social conception of human rights has implications for the legitimacy of various possible means of institutional protection. For, if politics itself is the ultimate desired good, this end is arguably undermined by means of enforcement that lie outside politics itself, notably judicial enforcement. And the undermining effect is exacerbated if the nonpolitical means of enforcement stem from outside the relevant community whose politics are at stake.

3 THE EMBARRASSMENT OF INTERVENTION

Whatever the motives underlying a plaintiff’s complaints, any judgment in a transnational tort case somehow amounts to a statement about the propriety of 30 A powerful critique of universalism is A MacIntyre, After Virtue 2nd ed. (London, Duckworth, 1985). 31 So, e.g., in one of the few explorations of Arendt’s human rights views that I am aware of, J C Isaac, “A New Guarantee on Earth: Human Dignity and the Politics of Human Rights”, reproduced in his Democracy in Dark Times (Ithaca, Cornell University Press, 1998) 74. 32 The example stems from J H H Weiler, “Fundamental Rights and Fundamental Boundaries: On the Conflict of Standards and Values in the Protection of Human Rights in the European Legal Space”, reproduced in his The Constitution of Europe (Cambridge, Cambridge University Press, 1999) 102, at 105. 33 A prime example has traditionally been Israel’s insistence that security threats justify forceful interrogation techniques. While this attitude was struck down by the Israeli Supreme Court in September 1999, the Court nevertheless urged parliament to legislate on the propriety of forceful interrogation techniques, thus presuming a local content to the universal prohibition of torture. “La Cour suprême israélienne déclare illégaux les interrogatoires musclés du Shin Beth”, Le Monde, 8 September 1999, at 3. For a discussion of whether Israeli courts should accept foreign courts imposing liability on Shin Bet interrogators, see A Reichman and T Khana, “Israel and the Recognition of Torture: Domestic and International Aspects”, chapter 24 in this volume.

560 J Klabbers a foreign political system; its rules, mores, practices and standards. In short, an inevitable, and possibly often hoped for, effect of transnational public law litigation is the rendering of a verdict on what happens elsewhere. As such, it is an exercise in extraterritorial jurisdiction: courts in one jurisdiction sit in judgment on the propriety and legality of behaviour in another jurisdiction. Moreover, they sit in judgment not just on everyday activities, but on activities at the heart of the other political community. In such circumstances, the legitimacy of litigation is strongly dependent on the answers to at least two questions. The first question is: under what conditions may jurisdiction be rightfully claimed by a court over behaviour taking place between foreigners abroad? The second relates more directly to the conception of human rights sketched above: what is the connection between human rights and politics? The question of jurisdiction raises, if nothing else, an awkward conceptual problem. Usually, in the literature around the United States’ Alien Tort Claims Act and Torture Victims Protection Act, jurisdiction to litigate against foreign human rights violators is claimed on the basis of the universality principle.34 Some violations, so the argument runs, are so atrocious that every state may claim jurisdiction over the perpetrators. After all, those who engage in such activities are to be considered, as the Filártiga court put it, as hostis humani generis—enemies of all humankind.35 Indeed, a strong case could be made that, in its initial form, the very idea of universal jurisdiction had limited application, applying only to those who were to be such enemies of all humankind. It is no surprise that the examples of crimes most often mentioned as giving rise to universal jurisdiction are piracy and attacks on ambassadors. Those who interfere with diplomacy interfere with a right accruing to all states, in the proper maintenance of which they all share an interest. Piracy is characterised precisely by the pirate’s attempts to operate within and exploit the “interstices” of the inter-state system, thereby denying the system itself, and thus turning the pirate into a universal outlaw.36 Human rights violators, by contrast, do not typically operate in the interstices of the state system, or act so as to antagonise all states without exception. Rather, they act with the support of at least one state: the one that employs them and that puts them in a position from which they can commit torture or other human rights violations. Conceptually, then, the torturer differs from the classic enemies of mankind in the crucial. To subject the torturer to universal jurisdiction is to invoke this jurisdiction to acts that do not engage the direct

34 Alien Tort Claims Act of 1789, [ATCA] 28 USC §1350, and the Torture Victim Protection Act of 1991, Pub. L . No 102–256, 106 Stat. 73, 28 USC §1350 n (1992) [hereinafter “TVPA”]. 35 Filártiga v. Peña-Irala, (1980) 19 ILM 966. 36 The argument derives from H Arendt, Eichmann in Jerusalem (New York, Penguin, 1994, first published 1963), esp. 261–2. See also H Spruyt, The Sovereign State and Its Competitors (Princeton, Princeton University Press, 1994) esp. at 169.

Foreign Tort Law and Human Rights 561 interest of all states in the defence of the interstate system itself. The doctrine of universal jurisdiction must somehow come to terms with this difference.37 The second point is that transnational public law litigation must come to terms with is the crucial connection between human rights and politics. This connection implies not only that human rights are best seen as facilitators and guarantors of politics, as discussed in Section 2 above. It has another side as well: human rights violations usually take place under colour of government authority. Indeed, this remains their defining characteristic.38 What is more, they usually take place in the pursuit of some higher goal or ambition. The typical human rights violator, if such a person exists to begin with, is not to be compared to the typical criminal or tortfeasor. The typical human rights violation occurs, so it seems, because someone thinks that, by impeding freedom of assembly, freedom of expression, or even by committing torture, a larger goal is somehow being served. In short, the human rights violator acts upon the belief that he will, ultimately, receive the blessing of history. In other words, the human rights violator is politically motivated.39 International law has traditionally carved out a niche for the unencumbered conduct of politics.40 The doctrine of sovereign immunity guaranteed that states could conduct activities for political reasons without having to answer to anyone before a domestic court of law. Any disputes, instead, were to be dealt with on the interstate level.41 The law on sovereign or state immunity relates, most observers agree, to official acts, but no longer to commercial acts of governments.42 And the idea behind this state of affairs is precisely that safeguarding politics within sovereign societies is an important enough goal to warrant the creation of a separate niche for politics to take place in relative isolation from mundane economic matters. Thus, by sacrificing the economic conduct of states, and therewith facilitating litigation against foreign governments in these matters, immunity law managed simultaneously to safeguard a political space. 37 Perhaps for this reason Arendt lacks faith in the universality principle as the basis of Israel’s case against Eichmann. Instead, she proposes something of a “spiritual territory” principle. See Arendt, supra n. 36 at 262–3. 38 The applicability of human rights between private actors still faces serious theoretical problems. Compare A Clapham, “The ‘Drittwirkung’ of the Convention”, in R St J MacDonald, F Matscher and H Petzold (eds), The European System for the Protection of Human Rights (Dordrecht, Kluwer, 1993) 163. 39 Which may help answer Steiner’s famous query why even the highly cultured may commit atrocities. See G Steiner, Language and Silence 2nd ed. (London, Faber & Faber, 1985). 40 This argument is somewhat elaborated in J Klabbers, “The General, the Lords, and the Possible End of State Immunity”, (1999) 68 Nordic JIL 85. 41 This alone should ensure that the comparison drawn between transnational commercial and public law litigation is unpersuasive: one cannot simply compare contract with torture as, does Koh, supra n. 6, at 2365–6. In addition, the argument can plausibly be made that parties to a contract knew what they were getting themselves into, at least with regard to private law litigation. See H Fox, “State Responsibility and Tort Proceedings Against a Foreign State in Municipal Courts”, (1989) 20 NYIL 3. 42 On the genesis of the restrictive immunity doctrine, see Byers, supra n. 7, esp. 110–14. And see R Wai, “The Commercial Activity Exception to Sovereign Immunity and the Boundaries of Contemporary International Legalism”, chapter 8 of this volume.

562 J Klabbers Clearly, the violation of human rights constitutes an abuse of protected sovereign space in the name of protecting politics. Yet, that has traditionally been considered the price that may have to be paid for keeping a separate niche for politics. As long as political debate can still take place, there is hope for humankind’s common future. Where political debate is no longer possible, neither is a common future other than one characterised by the constant negation of any rights. Precisely because human rights go to the heart of a political community, foreign courts run the heightened risk of misreading the political map of the originating state or, where they read the map correctly, they may be torn between attempts to do justice to the individual plaintiff and the desire to avoid causing major embarrassment in the relations between their own states and the relevant foreign states. It is noteworthy that, hitherto, the proceedings brought before US courts have generally, although not invariably, dealt with either regimes that had already been ousted or with individuals who were no longer occupying any position within such regimes.43 In such circumstances, the cases as brought have rarely presented courts with major foreign policy dilemmas. Yet, such dilemmas are more than conceivable.44 A first is where a claimant would bring a case against an agent of a major power. Consider, for example, if a Chinese dissident were to initiate proceedings against incumbent leaders of the People’s Republic of China, or if those who were once imprisoned for IRA activities but later cleared decided to start proceedings against Margaret Thatcher or other British government members for having been imprisoned without a fair trial. Surely, in such cases, any finding of human rights violations will cause major foreign policy embarrassment.45 One way out for courts of a given state might be to invoke the Act of State or political questions doctrines, but that would quickly raise the charge of arbitrary justice, if the application of either doctrine is contingent on the level of foreign policy offence taken by foreign states and the degree of clout those states have in international relations, so as to cause the executive of the local court to notify the court that its relations with the foreign state would be embarrassed should the court proceed with the case. On what principled basis could one allow a claim against relatively low-ranking Ethiopian or Paraguayan officials, or even former government members of small states, but not against a former head of government of a major power, let alone an incumbent? Considerations of head of state immunity aside, such a distinction seems hardly justifiable. 43 Regarding ousted regimes, this applies to the several cases brought against Marcos (see Trajano v. Marcos and another, in 103 ILR 521; and Hilao and others v. Estate of Marcos, in 104 ILR 119), to Xuncax and others v. Gramajo (104 ILR 165, involving Guatemala’s former Defence Minister), and arguably also to Mushikiwabo v. Barayagwiza (107 ILR 457), where the Rwandan government waived immunity. Regarding loss of position in ongoing regimes, defendants in Filártiga v. Peña-Irala (supra n. 35) and Abebe-Jira v. Negewo (see 107 ILR 447) had already lost their positions. 44 Excluding different types of embarrassing issues (e.g., on recognition and statehood) such as arose in Tel-Oren v. Libyan Arab Republic (1984), 77 ILR 192. 45 Think only of how Pinochet’s possible extradition has troubled relations between Chile and the UK and Chile and Spain.

Foreign Tort Law and Human Rights 563 The second type of foreign policy dilemma would consist of the situation where the valid outcome of judicial proceedings in state A can be construed as a human rights violation by state B—in other words where the violation is actually the result of regular court proceedings. Examples would be the courtordered punishment of caning for relatively small offences in Singapore or the death penalty sentence discussed elsewhere in this volume.46 In such situations, any findings of human rights violations amount to a negation of the judicial system of the originating state. And that would imply that the courts in state A would start to function, in effect, as appellate courts for state B. Yet, in a world where states often hesitate to allow even the International Court of Justice to function as an appellate court, surely the same line of thinking applies a fortiori to the courts of other states.47 Hence, transnational public law litigation appears to be justifiable only if we can successfully invoke a special normative status for human rights. But, even if it were based on a plausible conception of human rights as emanations of morality, the availability of this approach at least presupposes being able to point to some form of agreement on which human rights could or should invite litigation abroad, and agreement on what the precise contents of those rights are. Presumably, it is with such concerns in mind that many theorists engage in compiling lists of “basic” or “fundamental” or “core” human rights. Yet, given the inconclusive nature of such exercises, not to mention their dubious legitimacy in terms of less-than-universal representation of competing views, the result is inevitably the transposition of one’s own political values to other communities. Unless those values can be shown to be morally superior, which appears at minimum to be debatable, transnational public law litigation itself remains morally doubtful.48

4 MODERNITY ’ S RUDE AWAKENING

A relentless insistence on the judicial enforcement of human rights, whether by domestic or by international tribunals, displays two distinct attitudes. One of those is the attitude to recast life into sets of bilateral rights and obligations. In this “rights” conception, everything bad that happens to us entails the violation of a right and must, somehow, be the result of someone else’s actions. The other, therewith, becomes answerable not just before God or his own conscience, but also, and most importantly, before a court. 46 See B Wells and M Burnett, “Cultural Challenges: Injunctions in Australian Courts and the Right to Demand the Death Penalty under Saudi Arabian Law”, chapter 23 in this volume. 47 The argument was made, e.g., by counsel for the United States before the ICJ in the 1998 Case concerning the Vienna Convention on Consular Relations (Paraguay v. USA), available at http://www.icj-cij.org/docket. 48 Some merely assert the superiority of human rights without more. For an extreme example, see F R Tesón, A Philosophy of International Law (Boulder, Colo., Westview Press, 1998).

564 J Klabbers The standard communitarian critique of this type of thinking is not only that it is based on a hopelessly inadequate vision of humans as unencumbered and rational in pursuit of self, but also that it ends up undermining all chances we have of forming an authentic community.49 It leads to atomisation, to a world of all against all. Therewith, it eventually undermines its own project for, in a world of all against all, any rights will be rendered meaningless.50 A related point, and one which is more important for present purposes, involves the above conceptualisation of human rights.51 A vision of human rights in terms of enforceability by courts of law, and an insistence that enforcement actually take place, illustrates that modernity has finally shed its last ambitions.52 The idea of individual human rights which must at all costs be vindicated, even if only for symbolism’s sake (compensation being a rarity), displays a retreat of human beings into their own little worlds, their own private cocoons. By insisting on judicial enforcement in individual cases, we have given up on our chances of creating and maintaining a common future. As long as our private realms find protection, the public realm, or Bauman’s intermediate agora, no longer attracts our interests, our aspirations, or our hopes. We have, in other words, given up on politics, and therefore on ourselves. Insisting on judicialised individual rights, tragically, results in the defeat of our best, perhaps only, way to combat evil: through the hard-won decisions and evolving consensus of political processes. Ironically, this is the ultimate idea behind the liberal conception of human rights. Human rights are supposed not just to be immune from politics and provide guarantees against politics gone wild, but are thought also to be beyond politics: they mark the point where the end of politics has been reached and we can progress to the rule of law. And that is no coincidence: all great ideologies (and Liberalism is proving no exception) carry an implicit promise of the end of politics.53 The one problem, of course, is that there is no apolitical space beyond politics that is not authoritarian; it is illusory to expect law to take the place of politics, if only because law itself (including human rights law) is intensely political.54 More fundamentally, as long as we cannot agree on what constitutes the good life (and such agreement is not to be expected in the immediate future), we cannot plausibly impose a preference for any type of society. And agreement on what constitutes the bad life, which some believe is occasionally within our 49 See M Sandel, “The Procedural Republic and the Unencumbered Self”, (1984) 12 Political Theory 81. 50 For a powerful argument, see C Taylor, “Atomism”, reproduced in S Avineri and A de-Shalit (eds), Communitarianism and Individualism (Oxford, Oxford University Press, 1992) 29. 51 See Section 2, supra. 52 See also Klabbers, supra n. 40. 53 An elegant brief conceptual study of the end of politics is B Tromp, Het Einde van de Politiek? (Schoonhoven, Dubio, 1990). 54 See M Koskenniemi, “The Effect of Rights on Political Culture”, in P Alston (ed.), The EU and Human Rights (Oxford, Oxford University Press, 1999) 99.

Foreign Tort Law and Human Rights 565 grasp,55 is too meagre a basis on which to found our common future.56 The only alternative then, or so it seems, is to muddle through, while keeping our options open. The only alternative, in other words, is to treasure, facilitate and safeguard politics. Here the role of human rights cannot be overstated—but human rights embedded as claims and demands within political discourse itself.

5 CONCLUSION

Asking courts to settle our political differences for us is rarely a good idea, and asking foreign courts to do so is even worse. Neither force nor authority can take the place of meaningful political debate, and every attempt to solve political disputes judicially is almost automatically suspect. Where courts, moreover, use a legal institution such as tort law for purposes for which it is not suited;where the universality principle requires reconceptualisation, where at bottom there is the debatable assumption of the universality of human rights; where many of the arguments in favour of using courts are mistaken; where the larger social ramifications ought to concern us all. Where all of the foregoing are the case, the use of courts to settle our political debates for us becomes a practice of even more doubtful validity than it already is in normal circumstances. And one cannot simply assimilate the phenomena of commerce and torture by asking polemically “if contract, then why not tort?”, as Harold Koh does.57 That is not, of course, to belittle the importance of human rights. Without human rights, meaningful political debate is difficult to conceive. Meaningful political debate is the only thing we have to help safeguard a common future. Thus, recognition of human rights is vital for our living together, and governments as well as non-governmental actors ought to be reminded time and again that human rights are not to be trampled upon. There are methods available for doing so, from reporting procedures before international supervisory organs, to press campaigns, to protest action of all sorts and measures, to formal political debate and action. Indeed, using such channels, rather than turning to foreign judges, may itself be both evidence of the vitality of political community and reinforcement of that vitality in a way that makes it less likely, in the future, that serious human rights violations will occur in the first place.58

55 Compare A Linklater, The Transformation of Political Community (Cambridge, Polity, 1998). See also Z Bauman, Postmodernity and Its Discontents (Cambridge, Polity, 1997). 56 In more detail, J Klabbers, “The Scope of International Law: Erga Omnes Obligations and the Turn to Morality”, in M Tupamäki (ed.), Liber Amicorum Bengt Broms (Helsinki, ILA, 1999) 149. 57 Supra n. 6. Other arguments are spurious: it is curious to note that advocates of transnational public law litigation refer to compensation as an argument in favour of such litigation, while acknowledging that full compensation is extremely rare. 58 Compare Linklater, supra n. 55, for whom alternative actions are indeed vital signs of political life.

566 J Klabbers Litigation cannot solve our political problems for us. Indeed, it starts from the mistaken presumption that political problems can ever be “solved” to begin with. Instead, we may have to get accustomed to the idea of having to live with ambivalence, and organise our political behaviour accordingly.59 And, if that is correct, then any attempt to reduce ambivalence by asking courts to enforce the unenforceable is at best misguided and at worst a downright dangerous and selfdefeating exercise in legal expansionism.

59

See generally Z Bauman, Modernity and Ambivalence (Cambridge, Polity, 1991).

22

Just Amnesty and Private International Law JENNIFER LLEWELLYN 1

1 INTRODUCTION

of the twentieth century was a time of transition for many countries previously ruled by repressive regimes. Out of these repressive pasts often marked by violence and wide-scale abuses of human rights, such countries have made and are making the transition to democratic governance. These transitions have in many cases resulted from delicate negotiations. Few, if any, countries have had the luxury of a victor able simply to begin afresh unfettered by the power politics of the past. Thus, the process of transition has required compromise and even cooperation with members of outgoing repressive regimes. The need to secure a stable environment in which to build a democratic future has made amnesties an oft-used tool in transitional contexts.2 The choice to grant amnesty for past abuses has been subject to criticism at the international level. Despite these criticisms, amnesties remain a familiar feature within transitional contexts.3

T

HE LATTER HALF

1 LL.M. Candidate Harvard Law School. The author wishes to thank Ronald Slye, Blake Brown and Hallett Llewellyn for their helpful comments and support. 2 For example, in the last twenty years amnesty laws have been passed in Chile (1978), Brazil (1979), Colombia (1982), Uruguay (1986), Argentina (1983, 1986, 1987, 1989), Guatemala (1986), Honduras (1987), El Salvador (1987, 1993), South Africa (1990, 1992, 1993), Peru (1995) and Spain (1977). Ronald Slye notes that “[a]mnesties of one form or another were used to limit the accountability of individuals responsible for gross violations of human rights in every major political transition in the twentieth century.” R Slye, “Justice and the South African Amnesty” in W Verwoerd and C Villa-Vicencio (eds), Looking Back/Thinking Forward: Reflections on the South African Truth and Reconciliation Commission (forthcoming, Juta Press). 3 The success of the international human rights movement, beginning with Nuremberg, in establishing the principle of universal jurisdiction for the gravest violations of human rights serves as the basis for much of this criticism. For example, see Human Rights Watch’s policy statement on accountability for gross human rights abuses: Human Rights Watch, Special Issue: Accountability for Past Human Rights Abuses, no. 4 (New York, Human Rights Watch, Dec. 1989) at 1. Amnesty International has criticized amnesty laws as a factor contributing to impunity. They have argued that “amnesty laws which have the effect of preventing the emergence of the truth and subsequent accountability before the law, should not be acceptable, whether effected by those responsible for the violations or by successor governments.” Amnesty International, Oral Submission by Amnesty International before United Nations Commission on Human Rights, Sub-Committee on Prevention of Discrimination and Protection of Minorities (Amnesty International, International Secretariat, August 1991) at 4. In addition, every international tribunal which has dealt with the issue of amnesty has held that the granting of amnesty for gross violations of human rights constitutes a violation of international

568 J Llewellyn At the root of these criticisms of amnesties and their role in transitional contexts is a claim that amnesties thwart the interests of justice. Critics assert the provision of amnesty makes it impossible for justice to be done by bringing those responsible to account. Curiously, this perspective on amnesty is, for the most part, shared by those who defend, justify and even advocate the use of amnesties in transitional contexts. Many argue that amnesties are prudentially justified—for the sake of peace, political stability and an end to violence—but few if any suggest that amnesties are just. Arguments for and against amnesties seem to share the same underlying assumption: that amnesties are inherently unjust. Opponents charge, and defenders seem to concede, that amnesty represents a trade-off or compromise with justice. This, then, is taken as the common starting point for discussions and debates on the subject of amnesty. Critics use this failure to meet the demands of justice as a basis to condemn the use of amnesty provisions. Those who defend the use of amnesty argue that justice does not encompass the whole of the moral universe, and point to a range of other important social values that might be served by the use of amnesties. Underscoring this common assumption that amnesty is inherently antithetical to the pursuit of justice is a particular conception of justice. The granting of an amnesty is only unjust by definition if justice is understood in terms of retribution. Retributive justice requires that punishment be meted out to the guilty. When judged against this standard of justice, amnesty is obviously unjust as it precludes punishment of the guilty by its very definition. This chapter seeks to challenge these common, basic assumptions in the debate over amnesty. It takes a new look at amnesty and its potential for serving the interests of justice. It examines the possibility and promise of just amnesties. Doing so entails a challenge to the foundational assumptions about justice that underlie traditional approaches to amnesty. This chapter takes as its starting point an understanding of justice as restorative not retributive. Restorative justice is fundamentally concerned with the restoration of equality in relationships—with restoring relationships to a plane of social equality. Since this restoration cannot be achieved through traditional resort to punishment, which has the effect of isolating individuals, and in many cases renders relationship of any kind impossible, amnesty and the lack of punishment it entails, are not necessarily antithetical to the pursuit of justice understood restoratively. What will be required instead of punishment to see justice done under a restorative understanding becomes a contextual question. One must ask the question: what will restore this relationship in this community between these individuals, groups or communities? Amnesty may be part of the answer, in some contexts. human rights law. Such decisions have been made primarily in the Inter-American system, owing to the prevalence of amnesties in South American territories. See for example: Consuelo v. Argentina, Report No. 28/92, Inter-Am. Comm.H.R., OEA/Ser.L/V/II.83 Doc. 14 at 41 (1993); Mendoza v. Uruguay, Report No. 29/92, Inter-Am. Comm.H.R., OEA/SER.L/V/II.83 Doc. 14 at 154 (1993); Garay Hermosilla v. Chile, Report No. 36/96, Inter-Am. Comm.H.R., OEA/Ser.L/V/II.95 Doc. 7 rev. at 156 (1997). For a detailed discussion of the treatment of amnesty in these cases see Slye, supra n. 2.

Just Amnesty and Private International Law 569 In fact, it may prove fundamental, rather than antithetical, to the pursuit of justice in transitional contexts. If this is the case, and just amnesty is possible, the important question for transitional contexts becomes: can just amnesties be accommodated within current international law? Despite the substantial critical attention paid to amnesties in the international community, much of it under the rubric of impunity, surprisingly little attention has been paid to the status of amnesties under international law.4 Specifically, the issue of the extent to which domestic amnesties bind or ought to bind foreign jurisdictions has yet to be seriously considered.5 With the International Criminal Court (ICC) on the horizon and the proliferation of transnational tort litigation, it is increasingly impossible for countries undergoing transitions to remain oblivious to prevailing international opinion in making choices about how to deal with their pasts. In an ever-globalising world, amnesty is meaningless if it is effective only within the borders of the granting country. Amnesties, then, cannot do the work transitional contexts require if they do not garner international respect. The arrest in the United Kingdom of the former Chilean head of state, General Augusto Pinochet, serves as an example of how the issue of amnesty could become central on the international stage. Pinochet was arrested by British authorities on 16 October 1998, pursuant to a Spanish arrest warrant. 4 Impunity means exemption or freedom from punishment. According to El Hadji Guissé, one of two UN Special Rapporteurs on Impunity, impunity means “the absence or insufficiency of penalties and/or compensation for voluntary or involuntary violations of the rights and freedoms of individuals or groups of individuals”. Understood in this way, he explains, “impunity is not necessarily a lack of penalties, but may also be a question of penalties which are inadequate given the seriousness of the violations they cover, as regards both the punishment itself and compensation for the resulting injury to the victims.” El Hadji Guissé, Special Rapporteur on Impunity (economic, social and cultural rights), Interim Report on Opposition to the Impunity of Perpetrators of Human Rights Violations UN Doc. E/CN.4/Sub.2/1995/19 at para. 16. Louis Joinet, the other Special Rapporteur on Impunity (civil and political rights), further explains that the term covers “all the measures and practices whereby, on the one hand, states fail in their obligation to investigate, try and sentence those responsible for violations of human rights and, on the other hand, impede the enjoyment by victims and their families of the right to know the truth and have their rights restored”. According to Joinet, the scope of impunity is restricted to “violations of a serious and massive nature or constituting a systemic practice”. Progress Report on the Question of the Impunity of Perpetrators of Violations of Human Rights (civil and political rights) UN Doc. E/CN.4/Sub.2/1995/18 at paras 4 and 5. According to this definition, then, impunity is not limited to a failure to prosecute. In terms of ensuring and restoring the rights of victims, a state’s failure to make civil remedies available might contribute to impunity. Joinet explicitly acknowledges this understanding in his consideration of the definition of impunity in a subsequent report. Impunity, he explains, “means the impossibility, de jure or de facto, of bringing the perpetrators of human rights violations to account—whether in criminal, civil, administrative or disciplinary proceedings—since they are not subject to any inquiry that might lead to them being accused, arrested, tried and, if found guilty, convicted”. Louis Joinet, Special Rapporteur on Impurity (civil and political rights). Final Report—The Administration of Justice and the Human Rights of Detainees: Question of the impunity of perpetrators of violations of human rights (civil and political rights) UN Doc. E/CN.4/Sub.2/1996/18 at section 2a. 5 United States courts, for example, have yet to face the question of the status of a foreign amnesty granted for violations of human rights. As Slye explains, however, in the immigration context US courts have resisted the assertion of foreign amnesties as a means of obliterating a foreign conviction: Slye, supra n. 2. Instead, courts have undertaken an evaluation of the underlying act for which amnesty was granted: see Marino v. INS, 537 F.2d 686 (2nd Cir. 1976) at 691.

570 J Llewellyn Subsequent to his arrest, the Spanish government made a request for his extradition. The various charges against Pinochet, including murder, torture and forced disappearances, arose from his actions after he became head of state not long after a military coup d’état in 1973. Pinochet remained in power until March 1990 when he agreed to step down to allow for democratic elections. In exchange for this peaceful transition, the military exacted an agreement that the incoming government would leave unchanged the constitutional structure put in place by the military regime.6 After the elections Pinochet remained head of the army and enjoyed the status of “senator for life”. These factors combined to create a climate in which prosecutions for prior human rights abuses were made difficult because of the threat they might pose to an already-fragile transition. The maintenance of power ensured de facto impunity for the former regime and the new government’s ability to exercise control over the military was constrained by the inherited constitutional structure.7 This meant that any action against the former regime would be met with a military response that would threaten the democratic transition. The former regime also enjoyed de jure impunity in the form of a 1978 amnesty law. In 1978 the Pinochet regime granted itself amnesty with respect to acts committed during the period of martial law following the coup.8 The amnesty covered all murders, mayhem, batteries, unlawful detention, kidnapping, torture and disappearances committed between 11 September 1973 and 10 March 1978.9 This amnesty law has been challenged on two occasions and found constitutional.10 While bound by the amnesty law, the civilian government did manage some scrutiny of militaryera human rights abuses within the judicial system. In response to the court’s 6 The Pinochet regime was ultimately ousted by a 1988 plebiscite. In order to compete in the plebiscite, the political parties had to accept the 1980 Constitution and promise that they would only change it according to its own rules. The Constitution contained broad, well-structured legal provisions aimed at preventing any radical change. All the political parties with the exception of the Communist party agreed to the terms in order to participate in the plebiscite. For a full discussion of the Chilean transition, see J Correa, “Dealing with Past Human Rights Violations: The Chilean Case After Dictatorship”, (1992) 67 Notre Dame Law Review 1457. 7 The jurisdiction of military courts over any criminal matters involving military personnel also contributed to the de facto impunity enjoyed by the former regime. The Pinochet government expanded the scope of military court jurisdiction to include “proceedings for common crimes committed by military personnel during time of war, being in the field, in an act of military service or occasioned by such service, in the barracks, camps, bivouacs, forts, military works, stores, offices, premises, foundries, workshops, and other military or police establishments or premises of the armed forces”. Article 5(3) Law No. 18,342. Such broad provisions made it easier to challenge the jurisdiction of the civilian courts and to allow the military to sit in judgment of its own actions. See Americas Watch, Human Rights and the “Politics of Agreements”: Chile during President Aylwin’s First Year (Human Rights Watch, 1991) at 40. 8 Legislative decree 2,191 (April 1978). 9 The only exception was made under pressure from the United States government for the murder of the former Foreign Minister Orlando Letelier who was killed while in Washington D.C. 10 In both recorded cases where the amnesty law was challenged, the Chilean Supreme Court found it to be both constitutional and consistent with international law. See: Insunza Bascuñán Case, Recurso de Inaplicabilidad, Corte Suprema de Chile, Revista de Derecho y Jurisprudencia y Gaceta de los Tribunales, pt. 2, §4, at 64 (May–Aug. 1990); Romo Mena Case, Corte Suprema de Chile (26 Oct. 1995).

Just Amnesty and Private International Law 571 judgments, President Aylwin instructed that the courts must determine whether amnesty applies in any given case, and in doing so they must undertake a full investigation of the facts of a case brought before them.11 As a result, the courts have undertaken investigations in hundreds of cases.12 This approach, combined with the work of the National Commission on Truth and Reconciliation, established in 1990 to investigate serious violations of human rights committed in recent years, has resulted in in-depth investigations into much of the abuse that occurred under the prior regime.13 The amnesty law provides immunity from criminal prosecution for “all persons who, as principals or accessories, have committed criminal offences during the period of state of siege, between 11 September 1973 and 10 March 1978. . . .”14 The amnesty act ensures impunity in the criminal context, but does not provide immunity against civil liability.15 The former regime may, however, enjoy de facto immunity in the civil realm owing to the unwillingness of the Chilean Supreme Court to pursue serious investigation into such cases.16 Ultimately, Pinochet did not found his defence 11 S. Brett, “Chile–When Tyrants Tremble: The Pinochet Case ” (Human Rights Watch, 1999), (available at: http://www.hrw.org/hrw/reports/1999/chile/). 12 Human Rights Watch World Reports : Chile (2000), (avaliable at http://www.hrw.org/ wrzk/americas.02.htm). 13 Interestingly, despite the amnesty, there are currently eleven criminal suits which have been filed against Pinochet in Chile and another five suits to which the Chilean courts have already decided the amnesty does not apply (R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (Amnesty International and others intervening) [1998] 4 All ER 897 at 900). The National Commission on Truth and Reconciliation was established by decree of the Chilean Government. Its mandate was to establish as complete a picture of the abuses of the past, gather evidence concerning such abuses, make recommendations for reparation measures and the prevention of future human rights abuses and finally to make a report of its work. See: Report of the Chilean National Commission on Truth and Reconciliation, trans. P E Berryman (Notre Dame, Ind., University of Notre Dame Press, 1993); D Weissbrodt and P W Fraser, “Book Review: Report of the Chilean National Commission on Truth and Reconciliation”, (1992) 14 Human Rights Quarterly 601; J Zalaquett, “Balancing Ethical Imperatives and Political Constraints: The Dilemma of New Democracies Confronting Past Human Rights Violations”, (1992) 43 Hastings Law Journal 1432. 14 Article 1, Legislative Decree 2,191 (April 1978). Article 2 provides amnesty to all those previously convicted of such crimes. 15 The Chilean amnesty would thus not have been an issue had the case against Pinochet been brought as a transnational tort action. This is not to say, however, that our investigation of amnesties is irrelevant in the tort context as many amnesties (e.g.: the recent South African amnesty) do offer protection from both criminal and civil liability. See discussion later in this section. 16 Correa, supra n. 6 at 1464. Correa notes that one of the problems facing the new government in terms of dealing with the abuses of the past regime was the ideological sympathies of the judiciary—“The Supreme Court applied the broadest interpretation possible to the amnesty law, making it difficult even to investigate the truth.” Shortly after the release of the report of the Chilean Truth and Reconciliation Commission, Americas Watch was optimistic about the possibility for civil cases.

“One immediate effect of the report was to bolster the credibility of legal cases before the courts. The Commission forwarded information on some 230 victims to the civilian courts with requests that existing cases be reopened on the basis of new data or, in some instances, that cases be initiated for the first time. . . . At this writing it is unclear what the legal future of those cases may be. However, another important result of the Commission’s work—public criticism of the Supreme Court—may end by improving somewhat the chances for investigation.” cont.

572 J Llewellyn against the Spanish extradition request on the amnesty granted in 1978. Since many of the crimes alleged against Pinochet were committed after 1978, and thus outside the amnestied period, the General looked elsewhere for protection. In addition to the de jure impunity claimed under the amnesty law for acts committed between 1973 and 1978, Pinochet also claimed sovereign immunity as head of state of Chile for those acts committed before the elections in 1990. This was ultimately the focus of the House of Lords decision in the matter.17 The House of Lords, in their consideration of the Pinochet extradition case, did not focus on the status of the Chilean amnesty.18 The amnesty was not at issue, Americas Watch, Human Rights and the “Politics of Agreements”: Chile during President Aylwin’s First Year (Human Rights Watch, 1991) at 30. Recent reports from Human Rights Watch suggest that there is cause for some optimism with respect to investigations. The report notes a marked change in the approach of the Supreme Court. “By mid-1999, however, there were signs of a significant change of attitude among senior judges, due in large part to a 1997 reform of the structure of the Supreme Court, and the appointment of a new slate of justices in 1998. In September 1998, the Second Chamber of the Supreme Court, now comprised entirely of democratic appointees, for the first time invoked Common Article 3 of the Geneva Convention of 1949 to reverse the closure of a ‘disappearance’ case . . . In its July 1999 ruling, the Supreme Court unanimously endorsed [Judge] Guzmán’s [justice appointed to investigate a series of criminal complaints lodged against Pinochet] decision to exempt from the amnesty law cases in which the fact of death could not be legally certified and in which the victim, abducted, therefore must be presumed to be still missing. . . . This decision implied that the amnesty law was inapplicable to other ‘disappearances’, and it opened the door to hundreds of new appeals against the closure of these cases. It meant that only in cases where the death could be certified to have happened between 11 September 1973 and 10 March 1978 would the author of the crime be exempt from prosecution.” The report warns however, that “[e]xperience dictates caution in interpreting the lasting significance of a Supreme Court verdict . . . [s]ince, under Chilean law, jurisprudence has no binding effect on future cases . . .” Human Rights Watch, World Reports: Chile (1999), (available at: www.hrw.org/reports/1999/chile/patrick.htm). See also A Faiola, “Doggedly following Pinochet’s trail: tenacious Chilean judge builds up evidence against former dictator,” (10–16 February 2000) Vol. 16, No. 7 The Guardian Weekly at 33. 17 A panel of five law lords rendered judgment in the Pinochet matter on 25 November 1998: R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (Amnesty International and others intervening), [1998] 4 All ER 897 [hereinafter Pinochet No. 1]. That decision was subsequently set aside by the House of Lords on 17 December 1998 after it was discovered that one of the Lords sitting on the case had an affiliation with one of the intervening organizations that appeared before the court. See: R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No 2) [1999] 1 All ER 577 [hereinafter Pinochet No. 2]. The House of Lords issued its judgment in the rehearing of the case on 24 March 1999. R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (Amnesty International and others intervening) (No 3) [1999] 2 All ER 97 [hereinafter Pinochet No. 3]. Finally the decision on extradition was made by Bartle J. of the Bow Street Magistrates’ Court on 8 October 1999. See: The Government of Spain v. Augusto Pinochet Ugarte, Bow Street Magistrates’ Court [Unreported judgment of 8 October 1999] [hereinafter Pinochet No. 4]. 18 The issue of amnesty was dealt with by some of the Lords in the Pinochet No. 1 judgment in connection with the head-of-state immunity question. For example, in addressing the issue of whether or not immunity could apply to acts which amounted to violations of international law, Lord Lloyd compared this proposition with the treatment of amnesty under international law. He concluded that, “[i]t has not been argued that these amnesties are as such contrary to international law by reason of the failure to prosecute individual perpetrators. Notwithstanding the wide terms of the Torture Convention and the Taking of Hostages Convention, state practice does not at

Just Amnesty and Private International Law 573 because the Court was asked to deal with the question of whether Pinochet was “entitled to immunity as a former Head of State from arrest and extradition proceedings in the United Kingdom in respect of acts allegedly committed while he was Head of State”.19 The Court was not asked to determine whether Pinochet should ultimately be extradited (under British law this is a question for the Secretary of State), or tried by an international court or domestic Chilean courts, or found guilty of the alleged crimes. Had they been engaged in these inquiries they undoubtedly would have had to address the amnesty issue directly.20 However, the Pinochet fact situation does serve as a useful example of how the status of domestic amnesties might become an issue under international law. In particular, the status of domestic amnesties becomes crucial within the private international law sphere, notably in the evaluation of claimed transnational tort liability. Should domestic amnesties be respected by foreign jurisdictions when those amnesties provide for freedom from civil liability? As noted above, while the Chile amnesty was limited in scope to protection from criminal prosecution, other recent amnesties have no such limitation. The South African amnesty, for example, provides amnesty from both civil and criminal prosecution. While the South African Truth and Reconciliation Commission took the Chile Commission as a model, there are significant differences between the two. Most notably, the South African Commission was responsible for granting amnesty as part of its work. Amnesty under the South African model was offered in exchange for truth. Amnesty from both civil and criminal proceedings with respect to political acts was constitutionally guaranteed. However, the amnesty provisions developed as part of the Commission process required individuals to apply with regard to specific acts. Blanket present support an obligation to extradite or prosecute in all cases.” Pinochet No. 1, supra n. 17 at 929–30. We will return to examine the status of amnesty under international law. For now, it is important simply to note that the implication of Lord Lloyd’s position is that the Chilean amnesty would be relevant to any attempt to prosecute Pinochet. This is made plain by his later comments on the issue of non-justiciability. For an English court to investigate and pronounce on the validity of the amnesty in Chile would be to assert jurisdiction over the internal affairs of that state at the very time when the Supreme Court of Chile is itself performing the same task. In my view this is a case in which, even if there were no valid claim to sovereign immunity, as I think there is, we should exercise judicial restraint by declining jurisdiction. Ibid. at 934–5. 19 Per Lord Slynn, Pinochet No. 1, supra n. 17 at 900. This immunity issue was central to the second judgment on the merits, Pinochet No. 3, in which the Lords addressed, prior to the immunity issue, the issue of the requirements for an extradition crime under British law. On the necessity of considering this extradition issue, not addressed in Pinochet No. 1, see Lord Browne-Wilkinson, Pinochet No. 3, supra n. 17, at 101. 20 Lord Slynn noted that the fact that amnesty had been granted may be relevant to the Secretary of State’s decision on extradition. Further, he maintained: “If the question for your Lordships on the appeal were whether these allegations should be investigated by a Criminal Court in Chile or by an international tribunal, the answer, subject to the terms of any amnesty, would surely be yes.” Pinochet No. 1, supra n. 17, at 900 (emphasis added). See also comments of Lord Lloyd in Pinochet No. 3, supra n. 17.

574 J Llewellyn amnesty either for groups of people or for all acts perpetrated by a single individual was not available. Instead, the South African model attempted to provide accountability in amnesty.21 There are a number of examples imaginable within the South African context where the issue of the status of amnesty under both public and private international law might arise. One recent example is the amnesty granted to those responsible for bombing the African National Congress (ANC) headquarters in London, England, during the apartheid era. Those responsible have been granted amnesty under South African law. Would the amnesty apply if a victim, or the family of a victim, decide to sue those responsible in UK courts, under UK law, for the loss or damage suffered as a result of the bombing? Would public international law require the South African amnesty to be respected by the UK courts and, if so, to what extent should domestic judges give effect to the norms of public international law when deciding a private international law issue? The South African situation also offers a number of other examples where individuals disappeared or were tortured while in exile outside the country. The justnoted contexts would all require a court to ask whether a foreign court should create an exception for the application of its own tort law to tortious acts occurring in the court’s own territory. Courts could imaginably and understandably take the view that they are under no duty to recognise a foreign state’s granting of an absolution from civil liability that purports to operate extraterritorially, thus shielding tortfeasors for acts committed outside the jurisdiction of the amnestying state. These courts might be more likely to give effect to the foreign amnesty where that foreign law purports to operate extraterritorially only with respect to torts involving a foreign state’s own nationals as both perpetrator and victim, especially if it is clear that the amnesty and its operation do not violate public international law. But these questions, while real, are somewhat esoteric when contrasted with the paradigm in which a domestic court may be asked to look to public 21 The South African model will be used as an example throughout the chapter. On truth commissions generally see: P B Hayner, “Fifteen Truth Commissions—1974 to 1994: A Comparative Study”, (1994) 16 Human Rights Quarterly 597; D Bronkhorst, Truth and Reconciliation: Obstacles and Opportunities for Human Rights (Amsterdam, Amnesty International Dutch Section, 1995); L Huyse, “A Devil’s Choice—Dilemmas of Backward-Looking Justice”, (1994) 2 European Journal of Crime, Criminal Law and Criminal Justice 120; L Huyse, “Justice after Transition: On the Choices Successor Elites Make in Dealing with the Past”, (1995) 20 Law and Social Inquiry 51. M Minow, Between Vengeance and Forgiveness: Facing History after Genocide and Mass Violence (Boston, Beacon Press, 1998). On the South African Truth and Reconciliation Commission see generally: K Asmal, L Asmal and RS Roberts, Reconciliation Through Truth: A Reckoning of Apartheid’s Criminal Governance (Cape Town, David Philip Publishers, 1996); J Sarkin, “The Trials and Tribulation of South Africa’s Truth and Reconciliation Commission”, (1996) 12 South African Journal on Human Rights 617; W Verwoerd, “Continuing the Discussion: Reflections from Within the Truth and Reconciliation Commission”, (1996) 8 Current Writing: Text and Reception in Southern Africa 66; J Llewellyn, “Justice for South Africa: Restorative Justice and the South African Truth and Reconciliation Commission” in C Koggel, (ed.), Moral Issues in Global Perspectives (Peterborough, Ontario, Broadview Press, 1999) 96 [hereinafter Llewellyn]; J Llewellyn and R Howse, “The South African Truth and Reconciliation Commission—An institutional model for dealing with conflicts of the past”, (1999) 49 UTLJ 355 [hereinafter Llewellyn and Howse (1999)].

Just Amnesty and Private International Law 575 international law in order to know whether it may or must recognise a foreign state’s amnesty as determinative of the existence or non-existence of civil liability. An example of such a situation is where torture occurred in the territory of the state which has adopted the amnesty, (for example South Africa), and one or more of the victims seek to avoid a finding of non-liability in South Africa by asking a foreign court, (for example the UK,) to adjudicate the civil claim and refuse to recognise the amnesty. Such refusal could be made either by avoiding the application of South African law entirely, or by applying South African law except for the amnesty law. What if the UK or other Commonwealth countries were to follow the United States’ lead and enact transnational tort legislation similar to the Alien Tort Claims Act (ATCA) and the Torture Victims Protection Act (TVPA)?22 Would the South African amnesty protect those responsible for such acts of torture from liability? A recent newspaper editorial contemplated this question, comparing the situation in Chile with that of South Africa:23 “. . . Archbishop Tutu [Chairperson of the South African Truth and Reconciliation Commission] told a story of four men who had murdered young people in a small town. They appeared before his commission in that same town in a crowded hall before the very people whose relatives had been lost. They admitted their guilt. They expressed their remorse. They asked for forgiveness. It was a hot night. The hall had been filled with anger and passion. After some moments of silence, the crowd broke into applause and the guilty men wept. . . . Will they [United Kingdom and Spain] do the same thing if the South African murderers in the hall presided over by Archbishop Tutu come to London on business or Spain for a holiday? Will the new World Court reject solutions chosen by places such as South Africa, and insist on a parallel system of justice that exposes everyone from Inkatha Chief Mangosutho Buthelezi to former president F.W. de Klerk and Winnie Mandela to prosecution if they travel beyond the borders of their reconstituted state?”24 22 Alien Tort Claims Act of 1789, [hereinafter “ATCA”] 28 USC §1350, and the Torture Victim Protection Act of 1991, Pub. L . No 102-256, 106 Stat. 73, 28 USC §1350 n (1992) [hereinafter “TVPA”]. 23 William Thorsell, “Whose justice is it?” The [Toronto] Globe and Mail (9 January 1999). 24 The “new World Court” to which Thorsell refers is the International Criminal Court (ICC), as established by the Rome Treaty of 1998 (UN Doc. A/CONF. 183/9, 17 July 1998). There are potential problems with the ICC, as envisaged in the Rome Treaty, that are worth at least a brief mention. As it stands, the ICC cannot trump domestic jurisdiction. It can only take jurisdiction if a country is unwilling or unable to prosecute, or if the ICC determines that they are not making a good faith effort to prosecute. It is unclear from the language of the statute whether or not amnesties would even be a consideration in this respect. However, such a decision would rest with the ICC itself, as it has final interpretive authority over the statute. Article 17 of the Rome Statute of the International Criminal Court specifies the terms under which a case will be admissible for consideration by the Court:

“ . . . [T]he Court shall determine that a case is inadmissible where: (a) The case is being investigated or prosecuted by a State which has jurisdiction over it, unless the State is unwilling or unable genuinely to carry out the investigation or prosecution; (b) The case has been investigated by a State which has jurisdiction over it and the State has

576 J Llewellyn The question of the status of domestic amnesties in foreign jurisdictions warrants close examination, in light of the role such amnesties have played and continue to play in transitional contexts around the world. Given the importance of amnesties in these contexts, the question must concern more than simply the current legal status of amnesties under international law. Rather, the issue of whether amnesties should be respected also requires an analysis of the moral, political and legal issues comprising the normative status of an amnesty. An examination of the treatment of amnesty under existing international law needs, then, to be informed by and critiqued from a normative perspective. The focus of inquiry in this chapter is this: on what basis should amnesties be respected, if indeed they should be respected at all? Further, I am concerned, for the purposes of this chapter, with amnesty granted with respect to gross violations of human rights—that is the granting of amnesty for violations of jus cogens norms. While there are some who advance the argument that amnesty can never be justified under international law, for the most part the objection to amnesty comes with respect to serious violations of human rights, such as war crimes, genocide, crimes against humanity, and torture. I will argue that it is precisely with respect to violations of jus cogens norms that amnesty and the normative standards by which it should be evaluated are most relevant.

2 DEALING WITH AMNESTY UNDER INTERNATIONAL LAW

There are three options available for the treatment of amnesties under international law: (1) treat all amnesties as violations of international law and therefore invalid; (2) treat all amnesties as deserving of recognition as lawful;25 and decided not to prosecute the person concerned, unless the decision resulted from the unwillingness or inability of the State genuinely to prosecute; . . . 2. In order to determine unwillingness in a particular case, the Court shall consider, having regard to the principles of due process recognised by international law, whether one or more of the following exist, as applicable: (a) The proceedings were or are being undertaken or the national decision was made for the purpose of shielding the person concerned from criminal responsibility for crime within the jurisdiction of the Court referred to in article 5; . . . (b) The proceedings were not or are not being conducted independently or impartially, and they were or are being conducted in a manner which, in the circumstances, is inconsistent with an intent to bring the person concerned to justice.” For a discussion of the practical concerns raised by these jurisdictional restraints, see: J Llewellyn and S Raponi, “The Protection of Human Rights Through International Criminal Law: A Conversation with Madam Justice Louise Arbour, Chief Prosecutor for the International Criminal Tribunal for the Former Yugoslavia and Rwanda”, (1999) 57 UT Fac L Rev 83. 25 This favourable position with respect to amnesties standing in contrast to the first “antiamnesty” position is, of course, too simplistic. A position which allows all amnesties might do so in one of two ways. It might defend amnesty using international law. This position would argue that international law permits amnesties, thus offering a positive defence of amnesty under international law. This is not, however, the only way one might argue respect for amnesties. Another position might view amnesties as deserving of respect by foreign legal systems by virtue of the fact that they are acts of sovereign states within their own territory. Such acts, according to this position, might demand respect from foreign states even if they violate international law.

Just Amnesty and Private International Law 577 (3) draw distinctions between types of amnesties and respect some.26 The current state of international law vis-à-vis amnesties is at best open-ended and at worst confused and unclear. Advocates for the first two options claim the support of international law for each of their opposed views. Both approaches, however, suffer similar shortcomings. They fail to take account of the important legal and normative differences between various models of amnesties. As a result, they obscure important issues at stake in the debate over amnesty. Advocates of both options are often guilty of overstating, and at times misrepresenting, international law in service of their case. The first option—the “anti-amnesty” view—equates all amnesties with impunity.27 It paints all amnesties with the broad brush of impunity. All amnesties then, on this view, thwart the efforts of justice. Amnesty, critics argue, is tantamount to amnesia and thus cannot serve as a sound foundation on which to base a democratic future.28 What these critics of amnesty overlook is the possibility that amnesties might be a part of a mechanism aimed at reconciliation, such that there may be models of truth-telling involving amnesty which meet the demands for accountability, investigation and redress to victims, and generally serve the interests of justice.29 Amnesties, then, need not always be blanket trade-offs with justice. To take the position that no amnesty should be respected beyond the borders of the granting country is to ignore the

26 This third option is obviously insufficient as it requires articulation of the standards by which one might respect “some” amnesties and not others. This is the work on which the remainder of the chapter will focus. 27 See discussion supra n. 4. 28 The implicit assumption here is that a successful transition requires one to deal with the past as opposed to simply sweeping it under the carpet. The present chapter assumes that history has proven Santayana’s dictum correct—those who ignore the lessons of history have found their destiny is to repeat it. The only way out of this cycle is to know and learn from one’s past mistakes. On the need to deal with the past see: C Braude, “The Archbishop, the Private Detective and the Angel of History: The production of South African Public Memory and the Truth Commission”, (1996) 8 Current Writing: Text and Reception in Southern Africa 39; J Méndez, “Accountability for Past Abuses”, (1997) 19 Human Rights Quarterly 255; R Botman and R Petersen, To Remember and to Heal: Theological and Psychological Reflections on Truth and Reconciliation (Cape Town, Human And Rousseau, 1996); Alex Boraine and Janet Levy (eds.), The Healing of the Nation? (Cape Town, Justice in Transition, 1995); A Boraine, J Levy and R Scheffer, (eds.), Dealing with the Past: Truth and Reconciliation in South Africa (Cape Town, Institute for Democracy in South Africa, 1997); J Herz, “Conclusion” in J Herz, (ed.), From Dictatorship to Democracy: Coping with Legacies of Authoritarian and Totalitarianism (London, Greenwood Press, 1982) at 275; G O’Donnell and P Schmitter, Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies (Baltimore, Johns Hopkins University Press, 1986); T Rosenburg, “Overcoming the Legacies of Dictatorship”, (1995) 74 Foreign Affairs 134. 29 Again, the model of the South African Truth and Reconciliation Commission comes to mind here. As mentioned above (supra n. 21) the South African model brings amnesty into the process of investigation and truth-telling. Through the exchange of amnesty for truth (recall amnesty here is not a blanket guarantee, rather, in order to qualify, individuals must apply with respect to specific events) the process offers accountability in amnesty. For a more thorough examination of this model see: Truth and Reconciliation Commission of South Africa Report (Cape Town, Juta Press, 1998); Llewellyn and Howse (1999), supra n. 21; M Parlevliet, “Considering Truth: Dealing with a Legacy of Human Rights Violations”, (1998) 16 Netherlands Quarterly of Human Rights 141.

578 J Llewellyn realities facing countries undergoing transitions. Such a limitation would render amnesties, as a tool of transition, both ineffective and meaningless.30 The second option—that favouring recognition for all amnesties—makes the opposite assumptions from the first. Advocates of this option tend to argue for total deference to the domestic choices of sovereign states.31 They assume amnesties are a necessary trade-off, for the sake of peaceful transitions. However, this option ignores the fact that amnesties do not always serve the interests of transition, peace and reconciliation. It does not differentiate between amnesties enacted pursuant to a plan of national reconciliation and democratic transition, and those granted by repressive regimes to their own members in an effort to ensure impunity for past abuses. As a result, the second option, by insulating amnesties from scrutiny by the international community, runs the risk of allowing those who would violate human rights to do so without fear of legal reprisal. As is often the case, the path out of the dichotomy lies somewhere between the two opposing options. A deferential, hands-off approach is insufficient because it may be tantamount to endorsing, or at least turning a blind eye to, further abuse. Nor is it a satisfactory approach to outlaw amnesties by refusing to give them recognition, as this would deny countries a valuable tool in dealing with the challenges of transition. It is unrealistic to expect countries in fragile states of transition to undertake mass-scale prosecutions. And yet, our intuitions tell us that “something” must be done to call to account those responsible for such abuse. The challenge, then, is to find a way to open the door for amnesties used in service of reconciliation and peaceful transition, and at the same time protect against opening the floodgates of impunity masquerading as amnesty. Accomplishing this requires the development of a normative standard

30 It might be argued that individuals would still enjoy amnesty so long as they remained in their own country. However, such amnesty is rendered much less meaningful when one considers the possibility of transnational tort litigation conducted in absentia. Further, the Eichmann scenario (whereby an accused might be brought by force from one country to another for the purposes of prosecution) is not beyond the realm of possibility, especially in cases of crimes against humanity where states take seriously their duty (and thus right) to prosecute. 31 As noted earlier, this position involves two separate issues with respect to recognition. Since our main focus is on the third option, and devising the standards by which international law might distinguish between valid and invalid amnesties, we will forgo a detailed analysis of the second option. It may be that there are, in fact, two distinct alternatives included under the “pro-amnesty” option. On the one hand is the position that amnesty is lawful—that is, permitted by international law as acceptable acts within a sovereign state. The other position concerns jurisdiction and recognition. Amnesties on this second formulation might not be lawful per se under international law and yet other states would have no jurisdiction to police that unlawfulness, either through transnational tort litigation or criminal prosecutions. Thus, the question of recognition might be cast either negatively or positively. In the negative, foreign states might be forced to recognise an amnesty because they lack the jurisdiction to do otherwise. A positive case for amnesty would go further and suggest that foreign states must recognise amnesties because they are consistent with international law. The standard as developed in the remainder of the chapter with respect to which amnesties should be recognised under option three will address the positive aspect of recognition.

Just Amnesty and Private International Law 579 by which to evaluate different amnesty provisions with a view to determining which deserve international endorsement and recognition.32

3 A NORMATIVE STANDARD FOR AMNESTY

Perhaps the best way to begin to articulate such a standard is to inquire into what it is with regard to amnesties to which people react negatively. What is it about amnesty provisions that people find so objectionable? What is it that they hope to achieve through criminal law prosecutions or civil law claims? A close examination of the arguments for criminal punishment and civil liability reveal them to be proxies for justice. The moral intuition people are trying to serve when they claim justice must be done is that an elusive “something” must be done in the face of wrongdoing.33 But what exactly is it that we think ought to be done? In other words, what do we mean by justice?34 As I suggested at the beginning of this chapter, I will assume that justice means establishing or re-establishing a social equality between victim and perpetrator. Social equality implies an equilibrium in societal relationships such that each party enjoys fully realised rights to dignity, equal concern and respect. Thus, justice is concerned with restoring relationships to this ideal.35 Justice on 32 As I indicated at the outset of this chapter, I am suggesting that a theory of justice must serve as this normative standard. Specifically, a restorative conception of justice will provide a basis for determining which amnesties warrant respect internationally and which do not. 33 Although amnesties are most commonly understood as exempting individuals from criminal prosecution, as pointed out earlier, many amnesties also preclude civil liability. The objection to amnesties in the civil context mirror those in the criminal context. What people find objectionable about amnesties from civil liability is the denial of justice. People perceive the right to bring a civil suit for damages as the right to justice—as fulfilling the demand that something must be done in the face of wrongdoing. See for example the arguments made by several families of prominent victims of the Apartheid regime in their unsuccessful constitutional challenge to the South African amnesty laws. The applicants objected both to the failure to prosecute and to the denial of their rights to seek civil redress through the courts: Azanian People’s Organization (AZAPO) and Others v. President of the Republic of South Africa and Others 1996 (8) BCLR 1015 (CC). 34 For a more in-depth treatment of these issues and further explanation of restorative justice theory see generally: J Llewellyn and R Howse, Restorative Justice—A Conceptual Framework (Ottawa, Law Commission of Canada, 1998) [hereinafter Llewellyn and Howse (1998)]. 35 Note the difference between this and the claim made by those advocating restitution—that the objective is to return things to the way they were before the wrong occurred. Even if this were conceptually possible, and I suggest this is questionable in many cases, within the approach to justice I am advocating it would not be desirable. Restoring relationships, then, does not mean restoring to the status quo ante, or the way things were before. Rather, the objective is to restore relationships to the ideal of equality in relationship. Despite this explanation some critics express difficulty with the use of the word “restore” in this context. “Restore” is apt here if one understands that the human self is inherently relational. The concept of restorative justice I advance is rooted in feminist relational theory. See generally: J Nedelsky, “Reconceiving Rights As Relationship”, (1993) 1 Rev of Constitutional Studies at 13; C Koggel Perspectives on Equality: Constructing a Relational Theory (Maryland, Rowman and Littlefield Publishers, Inc., 1998). If we are relational selves—fully human in and through relationships with others—we are inherently relational in nature. Thus, we always exist in a web of relationships. The term “restoration” recognises that relationships are preexisting, and thus right relationships are not created out of nothing. Instead the objective is to

580 J Llewellyn this account is restorative by nature. If this is the correct picture of justice, we must re-examine the claim that prosecution and punishment are required, or indeed even sufficient, to meet the demands of justice. We must ask: how is it that prosecution and punishment restore relationships? Cast in this light, the problem with equating prosecution and punishment with justice becomes obvious. If justice is about restoring relationships, how can the removal of one party from the social relationship through isolating punishment help to achieve equality in relationship? In fact, the paradigm of prosecution and punishment would seem to preclude relationship altogether (by removing one party from the relationship) let alone equality in relationship. Civil trials can be equally problematic from the perspective of achieving restoration. In many instances, civil liability is used as a means of punishing— particularly in cases where a criminal conviction is hard or impossible to obtain. Even in cases where compensation alone is the objective, there is no guarantee that this will serve the interests of restoration. Firstly, the process in a civil suit is adversarial, requiring two opposed parties (and perhaps joined third parties), and thus excluding many parties with an interest in the situation that does not go directly to their own legal right to compensation or to their own legal liability. Secondly, the means through which private law tries to address the harm do not serve the needs of restoration. The concept of corrective justice is at work in private law.36 Corrective justice assumes that doing justice requires giving back to one party what was taken by another. Thus, on this conception, doing justice between parties requires the commodification of harms done by one to the other. In this respect, corrective justice and the solutions it offers cannot achieve restoration of relationship—because it is assumed that all harm can be rectified through monetary rewards. Restorative justice recognises that restoring relationships requires a contextual approach to determine what is required to address the harm in a given case and is not limited to financial compensation. Thirdly, private law approaches, such as punitive damage awards in tort law, can have the same effect as punishment in the criminal law system. They can serve to isolate one party from the other and thus preclude the restoration of relationship.37 Justice then requires reintegration, rather than isola“restore” relationships to the ideal in which human selves flourish. To speak of restoring relationships acknowledges that this ideal is derived from the very nature of the self. The term “rehabilitate” familiar from the criminal correctional context, is founded on a similar assumption. Few would claim that every prisoner was once a decent, law-abiding citizen and all that is required is a return to that earlier state. Rather the term implies, I would argue correctly, that the aim is to rehabilitate wrongdoers to the ideal state of law-abiding behaviour possible for most human beings. I say “most” here to allow for the possibility that some individuals may not be capable of being rehabilitated, just as some individuals may not be capable of taking part in the restoration of relationships. Such cases are however rare, generally the result of some pathology. For a more detailed discussion of these issues, see Llewellyn and Howse (1998), supra n. 34. 36 E J Weinrib, The Idea of Private Law, (Cambridge, Mass., Harvard University Press, 1995). 37 For more on the relationship between restorative and corrective justice see: Llewellyn and Howse (1998), supra n. 34.

Just Amnesty and Private International Law 581 tion.38 To the extent that amnesty provisions are aimed at reintegration and not isolation, they might be crucial to the pursuit of justice and not, as the traditional view would have it, antithetical to the very notion of it. Does this move toward a restorative understanding of justice mean that any and all amnesties are acceptable, or indeed worthy of respect? Are we left unable to distinguish between different amnesty models? We are assuming, from our discussion of the second option in the previous section, that respecting all amnesties is an undesirable decision. It would entail not only respect for amnesties in service of peaceful transition and national reconciliation, but also for those amnesties which serve primarily to protect abusers by ensuring they can continue to act with impunity. Restorative justice would not permit amnesty on any terms. On the contrary, a restorative approach to justice contains the legitimate parameters for amnesty within its very notion of justice. Justice as restoration of relationships can only be served by amnesties aimed at restoration. What this means in practical terms will depend on the social context in need of restoration. Recall, restorative justice asks the question: what is required to restore these relationships, in this context, between or amongst these people, groups or communities? The contextual nature of restorative justice means it is impossible to offer a “one size fits all” blueprint for restorative justice practice and likewise for any amnesty that might be part of restoration. However, this contextual nature does not leave us unable to distinguish restorative practices from those that are not. Nor does it mean we cannot differentiate between amnesties consistent with restorative justice and those that are not. Underlying the various practical forms restorative justice may take are various common commitments: to restoration over retribution, reintegration over 38 The suggestion here is not that justice will never place a burden on the wrongdoer to make up for what she has done. On the contrary, for reintegration to be possible, the harm done must be repaired. According to a restorative justice approach, however, such repair cannot be accomplished simply by isolating the wrongdoer or by imposing some other resolution. Instead, what will be required to repair the relationship will depend on the particular context and must be the result of dialogue between wrongdoer, victim and the community. What is required may actually turn out to be much more onerous than traditional modes of punishment or compensation. Whereas punishment and imposed compensation are passive, restorative justice requires the wrongdoer to play an active role in determining how to make reparation and in meeting these demands. From the perspective of restorative justice then, it is clear that the private law system, such as tort law, has an advantage over the public law system, such as criminal. The private law system opens the possibility for dialogue by including both the wrongdoer and the victim in the process, whereas the victim is not a direct party in most criminal law systems. Further, the remedies in private law do not include the ultimate isolating mechanisms of imprisonment, let alone capital punishment. Yet, although the remedies do not tend to be isolating, neither are they generally determined by restorative objectives. Even in cases where one party is not physically removed from the relationship, the processes and remedies of civil law systems can lead to isolation from relationships, insofar as they impose a solution on the wrongdoer. This precludes the process of true dialogue between wrongdoer and victim whereby a restorative resolution is to be found. As a result of this failure of traditional approaches, for the most part the remedies provided are not context-specific and often do not address the actual needs of the victim. Finally, for a resolution to be restorative the parties must agree to participate in the remedy and such participation must be active. In current systems, the process is a passive one for the wrongdoer, as the outcome is imposed on her—thus precluding her active participation in addressing the harm.

582 J Llewellyn isolation; to understanding the community as an integral part in the creation and solution of social conflict, with a concomitant acknowledgement that the focus is always broader than the individual or the immediately interpersonal; to looking at the implications for the future of a wrong and of proposed means of redress; and to bringing together all those with a stake in the development of that future.39 These commitments then provide a normative standard for amnesty. For an amnesty to be consistent with justice understood restoratively, it must demonstrably embrace these principled commitments. Those amnesties that do meet the demands for justice should thus enjoy the support of the international community. Amnesty provisions that fail on these terms are likely to result in impunity, thwarting the efforts of justice. Thus, they are not deserving of respect, and indeed ought to be condemned by the international community. Given this understanding of the theoretical requirements for just amnesties, what practical guidance does this offer us in distinguishing between amnesties? We must bear in mind that in order to remain sensitive to the specificity of each transitional context, it is not possible or desirable to lay down a blueprint to which all amnesties must conform. It is helpful and possible, however, to provide general guidelines for how this theory of just amnesty might work itself out in practice. What amnesty models might be rejected on their face according to these standards and what are the basic features of a just amnesty? In other words, if amnesties are to be judged by this restorative standard, what picture of amnesty emerges? What does this standard preclude? The answers to these questions will tell us which amnesties ought to be respected by the international community as just. Perhaps most obviously, the restorative justice standard forces one to distinguish between blanket and individualised amnesty provisions. Blanket amnesties generally specify a set of acts and a time period to which they apply. Sometimes the provisions will name a particular group or groups as the beneficiaries.40 A blanket amnesty applies automatically to all those covered by the provision once enacted or decreed.41 Generally, no investigation or admission is required to take advantage of the provision.42 Blanket amnesties obviously fail 39

Llewellyn and Howse (1998), supra n. 34, at 41–2. Most often application is restricted to the military or state actors, although this is not always the case. For example, the first Indemnity Act in South Africa, passed in order to facilitate the negotiations between the various political parties, applied only to members of the resistance movement. See Indemnity Act 1990 (Act No. 35 of 1990) and the Indemnity Amendment Act 1992 (Act No. 124 of 1992). 41 Whether decreed as secondary legislation or enacted as primary legislation, such amnesties are rarely the product of a democratic process. 42 There are of course exceptions to this rule. Particularly as new democratic governments try to deal with (or negotiate around the effects of) amnesties granted by a previous regime to its officials, new exceptions or rules of application have been developed. For example, as noted at the beginning of this chapter, Chile’s President Aylwin has developed a creative approach to the Pinochet regime’s 1978 amnesty, whereby courts are authorised to investigate each allegation before determining whether or not the amnesty applies. In this way, the Chilean government has tried to lower the wall of silence erected by the 1978 amnesty about the abuses during that the pre-1978 period. 40

Just Amnesty and Private International Law 583 to meet restorative standards. Such provisions preclude restoration of relationships by concealing the perpetrator’s identity. Rather than facilitating a process whereby all those with a stake in the situation can come together to discuss an appropriate resolution, blanket amnesties shroud wrongdoing in a cloak of silence. In this way, blanket amnesties are no better than prosecution and punishment in that they isolate the perpetrator from relationships with victims and society as a whole and thus preclude dialogue and action aimed at restoring equality to those relationships. Furthermore, blanket amnesty provisions deprive the victim and the community of one of the basic requirements for restoration—knowledge of the past. If the harm to the relationship and its parties resulting from the wrongdoing is to be healed, the truth about what happened must be known. Blanket amnesties do not require individuals to make application for amnesty, and thus there is no way to know precisely what acts, done by whom, have been amnestied.43 Blanket amnesties pose another significant problem for restoration, a basic requirement for which is that the wrongdoer admit responsibility for her actions.44 It is only when there is an acceptance of responsibility that the process of repairing the harm can begin. Under a blanket model, because individuals do not submit applications for amnesty with respect to their particular acts, there is no requirement that they admit responsibility for their actions. The most obvious alternative to blanket amnesties is individualised amnesties.45 Individualised amnesty refers to provisions requiring individuals to apply for amnesty with respect to specific conduct.46 Under this model, the onus

43 While it is possible to set up commissions of inquiry or “truth commissions” after a blanket amnesty, there is no impetus for those responsible to come forward and surrender information once such a provision is in place. Thus, the prospects of discerning the truth of what happened are seriously injured by blanket amnesties. 44 It is important to recognize that this admission of responsibility is not the same as requiring an admission of guilt. The contextual nature of restorative justice allows the perpetrator to offer an explanation of what happened instead of requiring the all-or-nothing choice of guilty or not guilty as in the criminal justice system. For example, in a restorative system a perpetrator might admit that she harmed the other person but claim that she did not intend to do so, or truly felt it was a necessary evil to do so, or even sincerely believed the harm was not morally wrong if ordered by authorities, and so on. This allows the perpetrator to take at least a form of responsibility for her actions and opens the door to repairing the harm done. Restorative justice promotes understanding and explanation of the event rather than simply labelling it. This raises the question of the extent to which an expression of regret or contrition, perhaps in the form of an apology, is required for restoration. This is a difficult issue since requiring an apology negates the genuine qualities necessary for a meaningful apology. This fact was recognized by the South African Truth and Reconciliation Commission when they chose not to make apology mandatory for amnesty. This is not to say that some expression of regret for the harm caused by the wrongdoing is not important to restoration. However, such an expression may be the result of, rather than the starting point for, a restorative process. Through the process, the wrongdoer is given a chance to hear from those affected by the wrongdoing and to understand the harms caused to others. This understanding is necessary before any contrition or regret can be expected. See Llewellyn and Howse (1998), supra n. 34, at 55. 45 We have already seen this model of amnesty provision in the discussions of the South African Truth and Reconciliation Commission, supra nn. 21 and 28. 46 The word “conduct” embraces both acts and omissions.

584 J Llewellyn rests on the individuals seeking amnesty to prove they meet the eligibility requirements for amnesty rather than resting the burden to prove noneligibility on those wanting to prosecute, thus forcing the latter to prove the amnesty provision does not apply. Individuals who qualify will receive amnesty from prosecution with respect to the particular crimes for which they made application. Under an individualised amnesty structure, individuals will not be immune from prosecution for acts not included in their amnesty application. This model of amnesty builds accountability into its very structure. Individuals must admit responsibility for particular acts in order to receive amnesty.47 Thus, this model meets the basic preconditions for just amnesty. The individualised approach requires the provision of information about what happened as a precondition for amnesty. In fact, the exchange of amnesty for truth sits at the very core of the individualised amnesty process, paving the way for restoration by making known the identity of the wrongdoers through the application process.48 Of course, simply making amnesties individualised is not enough to satisfy the demands of restorative justice. As is clear from the above discussion, the design of the process through which such amnesties are granted is key to determining whether an amnesty is just. Before dealing further with the issue of process, it is interesting to note that, under this individualised model, amnesty looks more like a pardon than an amnesty as it is traditionally understood. Those who argue pardons are preferable to amnesties base their case upon the fact that pardons occur after an assumption of responsibility.49 Given that individualised amnesties share this 47 This approach does create some concern with respect to self-incrimination—namely that individuals who come forward to reveal details of their wrongdoing and who are subsequently denied amnesty will in effect have provided evidence against themselves for future prosecution. In order to avoid such a scenario, it might be necessary to preclude the use of information derived from amnesty applications in other court proceedings. However, using revelations about an individual contained in another person’s amnesty application would not amount to self-incrimination, and may serve as an important source of information to prosecutors. Thus, one of the incentives to apply for amnesty is fear of incrimination by other amnesty seekers. 48 In any individualised amnesty process, this information will be known at least by those administering the process. Logically, in order to grant an individual amnesty one must know the individuals’ identity—both the identity of the perpetrators and, as far as possible, the victims. However, it is possible that such information might not be released to the public, and thus the victims might still be unaware of who was responsible for harming them. A restorative justice standard would then obviously have an additional requirement that the amnesty process be public, at least at some point, and its records accessible. 49 The claim is made not only that pardons are morally superior but also that they might be acceptable under international law, as they do not entail a violation of the duty to prosecute. For example, the Declaration on the Protection of All Persons from Enforced Disappearance, G.A. Res. 47/133 (18 Dec. 1992), UN Doc. A/RES/47/133, (Feb. 12, 1993), prohibits amnesty but nevertheless preserves a state’s right to pardon (see article 18(b)). N Roht-Arriaza offers a clear articulation of this distinction between amnesty and pardon:

“Although amnesty implies no investigation or acknowledgement of responsibility, pardon comes after both and merely constitutes a suspension of punishment. Moreover, amnesty implies a categorical decision, whereas pardon requires individualized consideration. Many of the functions of criminal prosecution—judicial inquiry into the truth, acknowledgement of responsibilities, public delegitimation—may be carried out without anyone actually going to jail. Indeed, it is

Just Amnesty and Private International Law 585 feature of accepted responsibility and case-by-case determinations, it seems they must also share the status of morally and legally acceptable practice.50 The individualised model of amnesty thus moves us some way forwards in our efforts to identify the basic characteristics of just amnesty provisions. However, simply guaranteeing an individualised model is, as pointed out above, insufficient. Without regard to the process through which individualised amnesties are granted, there is no guarantee that amnesties will meet the restorative standards set out above. For example, individualised amnesties could be granted through a confidential process.51 In order for amnesties to meet the demands of restorative justice, they must be granted through a public process that brings together all those concerned.52 Or, an amnesty process may fail by paying no or little regard to the needs of victims and the community for compensation or, in the alternative, the appropriate form of redress. Restorative justice, as it is concerned with restoring relationships, must attend to repairing the harm resulting from the wrongdoing. This means linking amnesty to noteworthy how uniformly the victims of past abuses indicate they are uninterested in seeing their former tormentors imprisoned, so long as their pardon is based on an acceptance of responsibility rather than its denial. While a promise of pardon may not satisfy some reluctant perpetrators, it may be enough for others.” N Roht-Arriaza, (ed.), Impunity and Human Rights in International Law and Practice (New York, Oxford University Press, 1995) chapter 21 at 300. 50 The analogy to pardons does raise an important point regarding individualised amnesties. The promise of accountability in individualized amnesties is contingent upon genuine consideration of each case. Just as legitimate pardons require unbiased and impartial consideration of the facts of each case prior to pardon, so too individualised amnesties will amount to little more than blanket amnesty for some if individuals are permitted to judge their own case. Of course it is not likely that an individual would literally be judge in their own case. However, in so far as one party to the conflict is in a position to draft and enact an amnesty provision on their own, the effect may be the same. If one side to the conflict is able to select the criteria and design the process by which amnesty is to be granted the result may be a privileging of certain applications over others. Such privilege might even amount to an amnesty guarantee for some. 51 Some might justify this in order to avoid the concern over self-incrimination: recall discussion supra n. 47. However, other mechanisms might be used to avoid the use of confessions in other proceedings where amnesty is refused. See, for example, rules of evidence developed regarding illegallyobtained confessions. 52 Recall that justice as restoration of social relationships reveals the importance of involving the relevant community in the work towards restoration. The relationship between wrongdoer and victim exists within a web of interconnecting relationships. Understanding these relationships between the individuals and their communities and between communities is crucial to restoring the relationship between wrongdoer and victim. Restoring this relationship then will require attention to the broader needs for restoration within the society. It is key to restoration that the appropriate communities/groups are involved in the process. Who is the community for the purpose of restoration? This is a contextual question. The answer must be found through an investigation of the connections/interconnections between victim and wrongdoer and those connected to them. This contextuality does not mean we know nothing about what is necessary for restoration. We know that the context in which the wrongdoing occurs, the society and communities to which each party is connected will be implicated in the harm and thus must be part of the process of restoration. This has implications for where restorative processes must be located. If restoration involves the community as much as the individual parties, the process must be undertaken domestically. Restoration is not something which can be achieved between the parties by removing them from the context in which the wrongdoing occurred. We will return to this issue when we consider the role of foreign courts under the act of state doctrine.

586 J Llewellyn reparation.53 Just amnesties then will not only be individualised but will constitute part of a larger process aimed at dealing with the past and working towards reconciliation. It is only within this kind of process that amnesties could be connected to reparations. What is entailed by linking amnesty with reparations is a more open question. Is it enough for amnesties to be granted by the same body responsible for providing reparations to victims or is a more immediate connection required? It is clear from our examination of restorative justice that it requires more—it requires, that the wrongdoer participate in repairing the harm. It seems acceptable within the bounds of the concept of amnesty to require at least that wrongdoers participate in reparation insofar as this involves appearing at a public hearing, facing their victims and accepting responsibility for their actions. This public acknowledgement would provide symbolic reparation. The question of material reparations is less straightforward. It raises the difficult issue of how far the concept of amnesty can be stretched and still be considered amnesty. Does requiring participation in reparation programmes or mandatory direct payment of compensation still qualify as amnesty? Certainly, organisations responsible for abuses could be required to pay reparations (i.e. the government or resistance movements) but what of individual perpetrators? At the very least, the possibility that individual amnesty applicants be required to contribute to reparations ought to be left open.54 However, all that being said, so long as the process of granting amnesties is connected to a larger project of reconciliation that ensures reparations to victims, it seems that amnesty can be consistent with restorative standards. Testing amnesties against restorative justice standards also suggests the way in which amnesty provisions ought to be drafted and enacted. So-called selfamnesties, those granted by one side to the conflict (generally the side in power) to its own members, are more likely to be of the blanket variety. Even if they are not blanket amnesties, self-amnesties fail by restorative standards because they do not take all the parties involved into account in designing a contextual response to wrongdoing.55 Restorative standards provide us with a picture of just amnesties. Such a version of amnesty would facilitate restoration instead of furthering the destruction of relationships amongst perpetrators, victims and the community. It can accomplish this by being: individualised, thereby requiring perpetrators to 53 Reparation includes both symbolic and material measures. Symbolic measures may produce an overlap between the right to know and justice in so far as such measures might include redress in the form of public acknowledgement of what happened. Other symbolic reparations range from commemorative activities or monuments, reburials, the erection of tombstones at previously unmarked graves, to the simple recognition of individuals as victims. Material reparations include compensatory and rehabilitative measures. For a discussion of reparations, see The Truth and Reconciliation Commission of South Africa Report, supra n. 29 at vol. 5, ch. 5. 54 That is to say, it ought not to be precluded by the process. It is important to note that this may also provide an opportunity for perpetrators to show repentance for their actions. 55 The 1978 Chile amnesty is an example of such a “self-amnesty”. See also discussion supra n. 49.

Just Amnesty and Private International Law 587 provide information and accept responsibility; part of a broader process of reconciliation; connected to reparations; and designed so as to benefit all parties to a conflict fairly.

4 AMNESTY , INTERNATIONAL LAW AND TRANSNATIONAL TORTS LITIGATION

Duties and Powers to Provide Avenues of Redress for Torture Under International Human Rights Treaty Law We now have a picture of what a just amnesty might look like within a theory of restorative justice. It has, hopefully, been shown that we can use a restorative standard to distinguish such amnesties from those that would bring injustice in the form of impunity. The question is: what use can we make of this standard? Is there room for this approach to amnesty under current international law? How might this analysis be applied to the case of a Pinochet, or Tutu’s young men amnestied by the Truth and Reconciliation Commission, if they were being prosecuted or held civilly liable before a foreign court in the face of having been amnestied in the country where their wrongdoing occurred? As mentioned earlier, there are two distinct situations in which the amnesty issue might arise. The first is litigation brought under public international law. This is the Pinochet-like case. A foreign court generally asserts jurisdiction over a case under its own criminal laws based on the universal jurisdiction provided for by a multilateral human rights treaty. The second situation in which amnesty might be at issue is in transnational tort litigation under the rules of private international law. This type of litigation remains a novel action in most jurisdictions. However, since the Filártiga case in the United States, it has the potential to become a more familiar means of dealing with wrongdoing on the international stage.56 It is this second situation which will be explored in the remainder of the chapter, as the focus of this collection is on redress of human rights abuses, notably torture, by way of transnational tort actions. Transnational tort litigation opens the possibility for victims of human rights violations to file suit in a domestic court of a foreign jurisdiction against those responsible. Thus, litigation may occur in the United States for a case involving two citizens of another country concerning abuse which occurred in that foreign country. The question is what status ought to be accorded to an amnesty granted in the country where the violation occurred.57 How could one make an argument that such an amnesty is just, under private international law? 56

Filártiga v. Peña-Irala, 630 F.2d 876 (2nd Cir. 1980). It is important to note that this may not be an issue with respect to those amnesties which only protect against criminal prosecution. However, even in cases where the amnesty seems clearly restricted to criminal prosecutions only, one might want to look closely at the intended results of the amnesty provision. In countries where civil claims for damages relating to a criminal act can only be made following a criminal conviction, or even in some cases alongside a criminal case, then an amnesty from criminal prosecution amounts to amnesty against civil actions as well. In this 57

588 J Llewellyn The first hurdle faced by advocates of amnesty in the context of private international law comes from treaty provisions. The problem for amnesty in the private law context comes not in the form of a duty to prosecute, as it would in a public law context, but rather with the duty to provide a remedy. The argument might be made that the right to effective remedy prohibits amnesty, at least with respect to civil liability. Amnesty on this account violates victims’ rights to seek effective remedy. There is implicit in this argument an idea of what constitutes an effective remedy. Effective remedies are of course to be secured by avenues provided by the proper authorities. In most legal systems, those proper authorities are the courts. This understanding of the right to a remedy is explicit in the International Covenant on Civil and Political Rights (the ICCPR). Article 2(3) reads: “Each State Party to the present Covenant undertakes: (a) To ensure that any person whose rights or freedoms as herein recognised are violated shall have an effective remedy, notwithstanding that the violation has been committed by persons acting in an official capacity; (b) To ensure that any person claiming such a remedy shall have his right thereto determined by competent judicial, administrative or legislative authorities, or by any other competent authority provided for by the legal system of the State, and to develop the possibilities of judicial remedy.”58 (emphasis added)

The ICCPR text does not support the claim that the right to a remedy requires the right to bring a civil suit. The granting of an amnesty is not then a violation of the right to remedy simply on the basis that it prevents civil liability. The right to remedy, it seems, might well require an appeal to some official authority to ensure its provision—but this need not be in the form of enabling civil actions. That being said, the right to bring both a civil action for torture and a transnational action is sometimes grounded in article 14 of the Convention against Torture (the Convention) which provides: “Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has enforceable right to fair and adequate compensation, including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation.”59 (emphasis added)

The argument might be advanced that in order to fulfill its responsibility under the Convention, a state party must facilitate victims of torture obtaining redress through their legal system in the form of civil suits. However, as with the situation, where the possibility of transnational litigation may not have been recognised, the drafters assumed the amnesty provision would work within its own judicial system. Caution is warranted in simply taking the amnesty at face value and proceeding with civil litigation as if the amnesty were a non-issue. 58 International Covenant on Civil and Political Rights, 19 December 1966, 999 U.N.T.S. 171 at art. 2(3). 59 Convention against Torture or Other Cruel, Inhuman or Degrading Treatment or Punishment, G.A. Res. 39/46, 39 U.N. GAOR Supp. (No. 51) at 197, U.N. Doc. A/39/51 (1984) at art. 14.

Just Amnesty and Private International Law 589 ICCPR, this is not the only possible interpretation of article 14. The wording of the provision is such that it could be interpreted to be consistent with the approach to a just amnesty described here. The Convention does not specify the way in which victims must obtain redress. So long as the victim has an enforceable right to compensation, there seems to be no reason why this could not be fulfilled within a system using just amnesty provisions. Article 14 does not require states to discharge their responsibility through enabling civil actions in domestic private law let alone by way of transnational (private international law) tort claims. Amnesty then, with respect to criminal and civil liability, need not constitute a violation of the Convention’s right to redress and compensation or the ICCPR’s right to an effective remedy. So long as the amnesty is part of a process through which compensation is provided, it will not be inconsistent with the right to a remedy. While it is not necessary for states to fulfil their obligations under the Convention to provide a remedy through the use of transnational tort legislation or common law processes, such litigation is nevertheless being utilised with greater frequency. It may thus be that international law permits such civil remedies. It is important then to consider how one would make the case for just amnesty in the context of transnational tort litigation assuming, arguendo, it is permissible. There are two main issues in any private international law suit: jurisdiction and choice of law.60 The jurisdiction issue logically precedes the choice of law question. A court must determine whether it can properly hear a case and, even if it can, whether it should hear the case before electing what law to apply when deciding it. The claim that a just amnesty ought to be respected must then be made at both these stages.

Invocation of Foreign Amnesty at a Jurisdictional or Preliminary Motion Stage The first question of jurisdiction—whether a court can hear a case—does not turn on the existence or non-existence of a foreign amnesty. All that is required is that the requisite degree of connection is present between the dispute, parties and forum.61 In any event, courts do not usually have the liberty to decide the 60 Other courts will face a third stage question of whether to enforce the judgment. This enforcement issue is relevant to the first two stages, as the issue of whether or not the judgment is likely to be enforced abroad might affect the jurisdictional inquiry for example at the forum non conveniens stage of analysis. Also, in deciding whether to take jurisdiction and also what law to apply, the forum court should, if it wishes to avoid engaging in double standards, ask hypothetically whether it would enforce its own judgment were the facts transposed and were it therefore the court such that a successful plaintiff is seeking to enforce a foreign judgment identical to that which the forum court is contemplating reaching. If the forum court would not be prepared to enforce in these circumstances, what justification could it have for expecting a foreign court to do so? 61 See A McConville, “Taking Jurisdiction in Transnational Human Rights Tort Litigation: Universality Jurisdiction’s Relationship to Ex Juris Service, Forum Non Conveniens and the Presumption of Territoriality”, chapter 6 of this volume.

590 J Llewellyn question of basic jurisdiction by direct recourse, or even direct reference, to international law principles on jurisdiction. Instead, they generally have to work with legislated rules of jurisdiction, whether they form part of the court’s rules of civil procedure or whether, as in the United States, they also take the form of specific statutes enabling causes of action for transnational human rights torts to be founded.62 The case is not the same however for the second question at the jurisdictional stage—should the court hear this case? The considerations at this stage include such factors as whether there is another more appropriate forum for this case (forum non conveniens) and whether the subject matter of the case is such that this court ought not to hear it (act of state and non-justiciability doctrines).63 The latter doctrines are more relevant than the doctrine of forum non conveniens as the basis for a defendant seeking to make the case for respecting just amnesties in transnational tort cases.64 The act of state doctrine may be traced back to the 1897 decision of the US Supreme Court in Underhill v. Hernandez.65 Fuller C.J. articulated the principle 62

ATCA and TVPA, supra n. 22. Note that those who view the act of state doctrine as a conflict of laws rule see it as relevant only at the choice of law stage of the case because it deals with an issue of applicable law, namely whether to apply foreign law or presume its validity, out of deference to the territorial sovereignty of the foreign state. That being said, however, it can function as a kind of jurisdictional doctrine. If a defendant raises the act of state doctrine at the outset, one would argue that it must be applied at the merits (choice of law and substantive law) stage of analysis and, thus, there is no point in continuing with the case at all. On this approach, the act of state doctrine is raised in effect to argue that no violation of law can be made out on the facts as pleaded, and, therefore, the case must be dismissed for want of a cause of action. As it would be futile to take jurisdiction only to subsequently apply the act of state doctrine where it operates not simply to determine one issue amongst several but to bar the action as a whole, the case should be dismissed at the outset. In this way, the doctrine operates like a jurisdictional doctrine. Others maintain the act of state doctrine operates jurisdictionally, because they understand it as having a non-justiciability strand according to which the purpose of deferring to a foreign state’s official acts is to avoid hampering the ability of the forum court’s own state speaking with one voice in international relations: for discussion of the two strands of the act of state doctrine within US jurisprudence, see A-M Burley. “Law among Liberal States: Liberal Internationalism and the Act of State Doctrine” (1992) 92 Colum. L. Rev. 1907. For the distinction between the act of state doctrine and the non-justiciability doctrine regarding certain kinds of international legal issues, in Anglo-Canadian law, see M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of ‘Act of State’ in Anglo-Canadian Law”, chapter 13 of this volume. 64 The forum non conveniens stage might allow room for the political/moral arguments about the appropriateness of the forum where the harm occurred for trial. These arguments would be similar to those contemplated under the non-justiciability head discussed later in this section. However, this discussion at the forum non conveniens stage is highly dependent on the context (specifically with regard to the state of the legal system in the place of the tort). See A McConville, supra n. 61. 65 168 US 250 (1897). The facts of this case are worthy of note, as they serve as a historical mirror of modern human rights claims. Hernandez was in command of a revolutionary army in Venezuela when an engagement took place with government forces which resulted in the defeat of the government. The revolutionaries subsequently occupied Bolivar. Underhill was living in Bolivar where he carried on a machinery repair business. He had also constructed the city’s waterworks system under a contract with the government. Underhill applied for a passport to leave the city, but this was refused by Hernandez in the hopes of coercing Underhill to continue to operate the waterworks system and his machinery business for the benefit of the revolutionary forces. Underhill eventually obtained a passport and brought suit in the United States against Hernandez for the tort of unlawful confinement. Since the United States recognized Hernandez’s revolutionary government as the legitimate government of Venezuela, his acts were found to be those of the state of Venezuela, and thus not properly the subject of adjudication in courts of another state. 63

Just Amnesty and Private International Law 591 that “[e]very sovereign State is bound to respect the independence of every other sovereign State, and the courts of one country will not sit in judgement on the acts of the government of another done within its own territory.”66 The act of state doctrine was thus, in its inception in the United States, based on the idea of comity between sovereign nations. Comity acts like a golden rule for the family of nations, instructing sovereign states to do unto others as they would have done unto themselves. This explanation for the act of state doctrine leaves the door open for the exclusion of certain acts from the ambit of the doctrine as not deserving respect because they were not ones which any state could will to be respected, even if they were their own. Most obviously, these are acts that contravene international law—or, at least, contravene fundamental norms of international law. In Banco Nacional de Cuba v. Sabbatino, the US Supreme Court refused to apply international law even to determine the validity of an act of a foreign sovereign state within that state’s own territory, where the Court felt that there was no settled international law against which these acts’ legality could be judged.67 Here, putative rules relating to the expropriation of aliens’ property were at issue. The Court, in effect, concluded that questions concerning the acts of other sovereign States were to be treated as political questions deserving of judicial restraint and more appropriately within the scope of the Executive’s power, at least in the absence of clear and settled international legal standards which the Court could apply. One of the baseline rationales for such judicial restraint was that the choice of whether or not to respect an act of another sovereign State might affect relations with that State and thus such decisions were inherently political. Given this baseline premise, the Court thus found that in the absence of a strong international consensus on a rule of customary international law, separation of powers and judicial incompetence weighed heavily in favour of judicial restraint. As Harold Koh explains, in invoking this rationale, the Court “went far beyond the comity rationale that had guided its previous Act of State decisions, now emphasizing separation of powers and judicial incompetence as the main reasons why American courts should not adjudicate cases under international law”.68 The Sabbatino approach to the act of state doctrine in effect equated it with the principle of non-justiciability.69 This trend in the approach to the act of state doctrine continued until the 1980, when a proliferation of transnational commercial litigation occurred.70 The effective litigation of transnational commercial cases, according to Koh, raised the persistent question: “If contracts why not torture? If American courts could subject the commercial conduct of a foreign sovereign to legal scrutiny 66

Ibid., at 252. 376 US 398 (1964). H Koh, “Transnational Public Law Litigation”, (1991) 100 Yale LJ 2347 at 2348. 69 This remains the approach taken by some courts that continue to view the act of state question as combined with the issue of justiciability. 70 Commercial suits brought by individuals or corporate entities against foreign governments, generally in the area of contracts, were being successfully litigated in US Courts. 67 68

592 J Llewellyn without offending comity, why should comity immunise the same sovereign from judicial examination of its egregious public conduct?”71 At virtually the same time as Koh was posing this question, the US Supreme Court moved to narrow the act of state doctrine in its decision in W.S. Kirkpatrick Co. v. Environmental Tectonics Corp.,72 in which the Court unanimously declared that the act of state doctrine was not, as many American courts and commentators had been interpreting it to be, a general rule of non-justiciability of issues which might affect diplomatic relations.73 Thus, the door is open once again for the exclusion of certain acts from protection under the act of state doctrine. In light of the preceding brief account of its trajectory, it seems clear how the act of state doctrine might be of assistance in making the case for respecting just amnesties. The argument can be made quite simply that, by any reading of the nature of the act of state doctrine, the enactment of an amnesty is the act of a sovereign state undertaken as an exercise of territorial jurisdiction and as such ought to be respected.74 Viewed in this way, courts would be forced to refuse jurisdiction in a case where amnesty has been granted under the foreign state’s positive law. But what of the argument that the act of state doctrine must be limited to those acts which are properly those of a sovereign, and thus only those that do not violate international law, ought to be respected as part of the idea of comity?75 Does this preclude amnesty from being treated as an act of state? To the contrary, this restriction allows exactly the kind of analysis of amnesties advocated in a just amnesty approach. While the act of state doctrine understood in non-justiciability terms poses problems for distinguishing between just amnesties and other amnesty provisions, a Kirkpatrick-like narrowing of the act of state doctrine to an applicable law issue enables a court to entertain the precise issues raised by a restorative analysis of amnesties. The court might answer the question of whether a given amnesty is the kind of amnesty that they ought to respect by assessing whether the amnesty violates principles of international law. One would then be able to advance arguments about the ability of just amnesties to serve the objectives of international law. The act of state doctrine actually holds out the possibility that just amnesties could be found to be legitimate acts of a sovereign state within its own territory, and thus worthy of respect. Assessing legitimacy becomes a process of weeding out those amnesties inconsistent with the requirements for justice under international law. 71

Koh, supra n. 68 at 2349. 110 S. Ct. 701 (1990). 73 The Court found that the act of state doctrine was actually a federal choice of law rule applicable when the validity of a sovereign act of a foreign state was a necessary element in deciding the scope of rights and obligations of the parties. See Koh, supra n. 68 at n. 106. 74 Underhill v. Hernandez, supra n. 65. Note that the Court in Underhill found that it was irrelevant whether the government was legitimate or recognised as such. “Nor can the principle be confined to lawful or recognised governments . . .” per Fuller C.J. at 252. 75 See Bühler, supra n. 63, for a discussion of the issue of limiting the duty to recognize foreign acts to acts that do not breach fundamental norms of international law. 72

Just Amnesty and Private International Law 593 The narrowing of the act of state doctrine has occurred as a matter of common law in the United States. It has not, however, meant the complete end of a non-justiciability approach to jurisdiction, even in the US. A notable example of this is the invocation of a distinct “comity of nations” doctrine by several US courts citing passages from older cases such as Underhill v. Hernandez and avoiding entirely any mention of the act of state doctrine.76 A non-justiciability approach to act of state, including when it comes labelled as “comity of nations”, might fail to make the very distinctions sought in a just amnesty approach. On a non-justiciability approach, a court must decide whether this is the kind of matter it can legitimately decide. The problem with this approach is that the “matter” at issue is likely to be amnesties as a category. Thus, it is either always appropriate or always inappropriate for courts to interfere in this area. There is little room for the investigation suggested by a just amnesty approach—namely that, where amnesties are legitimate according to a justiceinformed interpretation of international law, foreign courts ought not to interfere. The temptation then, on a non-justiciability approach, is to decide either that foreign courts can rule on the status of all amnesties, or on none at all, and to thereby miss the opportunity to explore a third option.77 All of the above does, however, raise important issues with respect to the desirability of foreign courts’ involvement in making judgments about the justness of amnesties at all. Is there not a reason to say that foreign courts should refrain from passing judgment on such contextual legal questions where international law itself can only supply standards when read in tandem with legal and political theory? It is thus important to inquire whether the theory of just amnesty has anything to say to the appropriateness of domestic courts judging foreign amnesties. It might be said that, as a pragmatic matter, foreign courts are considering amnesty and thus it is crucial they hear just amnesty arguments in the process of their deliberations. But the simple fact that they are dealing with the issue of amnesty does not answer the question of whether the decision about whether an amnesty is just or not should be left in the hands of foreign courts. Even if we accept the normative analysis of just versus non-just amnesties, this does not resolve the institutional issue of which body should make such an evaluation. The non-justiciability approach to act of state at least has the merit of bringing these issues to the fore. Recall that in order for an amnesty to be just, it must be restorative. Just amnesties, then, must be part of a process of restoration aimed at restoring 76 A good example of this is the judgment of Rakoff J. in Aguinda v. Texaco, 175 FRD 50 (1997). While the federal Court of Appeals took issue with Rakoff J.’s application of the “comity of nations” doctrine as a separate jurisdictional bar, it did not question that the doctrine was available: see Jota v. Texaco, 157 F.3d 153, (2d Cir. 1998). 77 It does not appear impossible, however, to entertain this third option even under a nonjusticiability approach to act of state. One might answer the justiciability question positively—yes, courts are competent to deal with amnesties—and utilize a just amnesty approach in coming to their decision with respect to a given amnesty. But the danger remains that, without clear international law on amnesties, courts will not even get to the “yes” answer.

594 J Llewellyn societal relationships and what restoration requires is a contextual determination. Coming to an agreement about the needs for restoration and how to achieve it will ideally require the participation of all of the parties affected by the situation. Thus, a plan for restoration must be the result of dialogue within the community affected by the wrongdoing. Such dialogue will include, but not be limited to victim, perpetrator and affected communities. Our earlier examination of restorative justice reminds us that these relationships are embedded in, informed by and shaped by the network of relationships of which they are a part. In turn, these relationships affect the broader web of relationships that constitute the community to which they belong. Thus, any attempt to restore the relationships harmed by wrongdoing will require the participation of the broader community. The extent to which amnesty serves the interests of restoration can only be determined by attending to the specific needs for restoration of the relationships at issue—the needs of these perpetrators, these victims, these groups and the affected communities. The contextual nature of such a determination would seem to raise serious questions about the ability of foreign courts to judge an amnesty just or not. Foreign courts are just that—foreign. As such they stand removed from the community in which the amnesty was enacted and to which it is to be applied. Without intimate knowledge of the community and its needs for restoration, how could a foreign court be expected to undertake an evaluation of the restorative potential of an amnesty? This concern fits well with the normative concerns underlying the non-justiciability approach to the act of state doctrine. Quite simply, one would argue that foreign courts lack both the interpretive competence and the legitimacy to pass judgment on the justness of amnesties in other national contexts. However, this approach, while it might avoid the problem of foreign courts imposing their own understanding of the needs for restoration in a particular context, creates some pragmatic problems for the case in favour of just amnesties. Most obviously, the problem is: if not foreign courts, or some body outside the context in which the amnesty is granted, then who is to judge the justness of amnesties? If we disqualify all but those within the context from evaluating the justness of an amnesty, thus removing the possibility of external judgment, then we run the obvious risk that every country will find their own amnesty just. This is the risk of relativism. It is precisely to avoid this problem that we look to developing normative standards for amnesty in the first place. The ability of foreign courts to make a choice regarding whether to respect an amnesty or not then serves an important function in ensuring that amnesties are just. It creates an incentive for domestic governments contemplating the use of amnesty to design and implement law which meets the standards for just amnesty, so as to ensure foreign respect. Judgment by foreign courts may also serve to prompt a country to work towards restoration in light of that fact that, in cases where an amnesty is unjust, thus not meeting needs for restoration, perpetrators are likely to keep themselves and their assets in their home country for

Just Amnesty and Private International Law 595 fear of prosecution or civil liability in foreign jurisdictions. Thus, there is an incentive, so long as they remain in the community in which they committed their wrongful acts, to work towards restoration and the achievement of a just amnesty. A just amnesty approach highlights the challenges inherent in using foreign courts to evaluate amnesties. As a purely practical matter, in the absence of some international body able to monitor states’ evaluations of the justness of their own amnesties, it is likely that foreign courts will continue to find themselves in this role.78 As was alluded to above, this may prove advantageous to the project of encouraging just amnesties. However, along with such benefits, the inherent dangers remain. The challenge is to strike a balance between deference to the decisions made within communities concerning their needs for restoration and protection against unjust amnesties perhaps better understood as legalized impunity.

Amnesties and Tort Choice of Law Should the above arguments prove unpersuasive, and a court choose to take jurisdiction despite the fact of a just amnesty, the other possibility for a defendant is to argue the case for just amnesties at the choice-of-law stage of analysis. At this stage, a court must decide which law to apply in deciding a case. Typically, in transnational tort cases, this involves a choice between two legal systems—the lex fori (law of the forum) and lex loci delicti (law of the place of the tort). It is interesting that one of the foundational British Commonwealth common law cases on tort choice of law concerned the application of an indemnity law passed in the place of the tort. Phillips v. Eyre arose out of events surrounding the suppression of a rebellion in the then British colony of Jamaica.79 Eyre, then governor, declared a state of martial law and imprisoned many of the rebels in his efforts to end the rebellion. Upon his return to England at the end of his tenure as governor, a suit for false imprisonment and other torts was brought against him by Phillips, one of the rebels. Eyre contended that an 78 Plans for an international civil court do not appear to be anywhere on the horizon, however, although the necessity for one was suggested at least as early as the 1960s: see T Franck, The Structure of Impartiality: Examining the Riddle of One Law in a Fragmented World (New York, Macmillan, 1968). With respect to criminal matters (or, at least, criminal matters that form a subset of crimes under international law), we might vest some hope in the proposed international criminal court. This body might be in a position to evaluate amnesties against the restorative standards we have articulated. But, looking at the Rome Statute with this objective in mind, one has cause for concern. Recall the conditions under which a case will be admissible to the Court (supra n. 24). In situations where a country has failed to make a good faith effort to prosecute, the Court will take jurisdiction and prosecute the case. Is there any room in the Statute’s formulation for the Court to find an amnesty acceptable as a good faith effort to do justice, or is the Court so caught up in the fallacy of equating justice with prosecution that anything else will be viewed as insufficient? If the latter is the case, it would seem the International Criminal Court will disappoint as a means of distinguishing among amnesty provisions. 79 Phillips v. Eyre (1870) LR 6 QB 1 (Ex. Ct.).

596 J Llewellyn indemnity law passed by the British colonial authorities in Jamaica following the rebellion protected him against any liability for his action during the rebellion. With Jamaica as the physical locus of the tort, the court was faced with the choice of whether to apply Jamaican law, with the result that Eyre could not be held liable for his acts, or whether to substitute the law of England, as the forum with adjudicative jurisdiction over the parties. In making their decision, the court enunciated the tort choice-of-law rule which would remain in force in Canada and the United Kingdom for the next century. In a now-famous passage, Willes J. offered the following choice-of-law rule: “As a general rule, in order to found a suit in England for a wrong alleged to have been committed abroad, two conditions must be fulfilled. First, the wrong must be of such a character that it would have been actionable if committed in England . . . Secondly, the act must not have been justifiable by the law of the place where it was done.”80

This rule came to be known as the double actionability rule because it required a wrong to be actionable under both the law of the place of the tort and the law of the forum. According to this rule then, a suit can be brought with respect to wrongs committed elsewhere only if the wrong was not justifiable in the place where it occurred. With respect to the case of Phillips v. Eyre, this meant that the indemnity law had the effect of making Eyre’s acts lawful and thereby absolving him of any civil liability for the tort of false imprisonment. The Phillips v. Eyre non-justifiability condition was subsequently interpreted in Machado v. Fontes as distinct from the actionability condition articulated in the first branch of the test.81 The Court found all that was required to find a defendant’s acts “not justifiable” was a breach of the foreign law—any law, civil or criminal. Thus, the act need not be civilly actionable, either at all or on the same grounds, in both the place of the tort and under the law of the forum. This interpretation of double actionability would not rule out the Phillips v. Eyre scenario, where an indemnity act precludes liability by justifying otherwise unjustifiable acts. The only effect this formulation might have is in cases where amnesty is provided against criminal but not civil liability. Where such amnesties exist, an action might be founded on the grounds that the case would be actionable under the civil law of the place of the tort. However, acts amnestied by provisions covering both criminal and civil liability would provide protection under this interpretation of double actionability. Under this tort choice-of-law rule, then, all amnesties would have to be respected, provided they cover criminal and civil liability. Under this rule, transnational tort actions could not be brought against individuals who enjoy amnesty under the lex loci delicti. The problem with this rule from a just amnesty perspective is obvious from our earlier discussions. It fails to draw any distinction between just amnesties and other provisions. In other words, this rule is over-inclusive with respect to amnesties. It leaves no room for courts to 80 81

Phillips v. Eyre, supra n. 79. [1897] 2 QB 231 (CA).

Just Amnesty and Private International Law 597 disregard the lex loci delicti and apply the lex fori in cases where not doing so would result in grave injustices. This justice concern was, in fact, what prompted latter developments in some jurisdictions in the tort choice-of-law rule away from double actionability. In the United Kingdom, this move occurred in the 1969 decision of the House of Lords in Boys v. Chaplin.82 The House of Lords per Lord Wilberforce restated the basic rule in English law with respect to foreign torts as “requiring actionability as a tort according to English law, subject to the condition that civil liability in respect of the relevant claim exists as between the actual parties under the law of the foreign country where the act was done”. However, the Lords, recognizing the need for flexibility in the interests of justice, qualified this rule, finding that “[t]he general rule must apply unless clear and satisfying grounds are shown why it should be departed from and what solution, derived from what other rule, should be preferred.” Accordingly, the House of Lords found firstly that Machado v. Fontes should be overruled and civil actionability must exist in the lex loci delicti in order for a suit to be brought in England. Secondly, they found that the double actionability rule is open to exceptions requiring only that “satisfactory” grounds be shown for departure. How would just amnesties fare under this reformulation of the tort choice-oflaw rule? At first glance it looks as if the decision opens the door to the antiamnesty view considered earlier. If one takes the view that all amnesties amount to impunity and are therefore unjust, the very fact of an amnesty might be cited as cause to depart from the general rule. However, a closer analysis reveals it is possible to make the case for just amnesty, even given the broad exception contemplated in Boys v. Chaplin. Given the connection of the lex loci delicti to the parties, and the context in which the wrong occurred in typical transnational tort cases concerning human rights violation, it might be difficult to reason that “clear and satisfying grounds” exit for another law to be applied. In addition, a restorative perspective reveals that doing justice between parties requires the involvement of the community in which the wrong occurred. Justice understood as restoration of relationships can only be done when the context in which those relationships exist is involved in restoration. This requirement argues against departing from the lex loci delicti, particularly in cases where the choice has been made in favour of amnesty as part of a process of restoration. In such cases, then, the threshold for “clear and satisfying grounds” should be extremely high. The ruling in Boys v. Chaplin was replaced in 1995 by the enactment by the British Parliament of the Private International Law Act.83 Section 10 of that Act abolishes the double actionability rule and the exception to it as decided in Boys v. Chaplin. In its place, the Act institutes a lex loci delicti rule with a slightly narrower exception provision from that offered by the House of Lords. The Act allows for departure from the general rule only where it is “substantially more 82 83

[1969] 2 All E R 1085 (H.L.). The Private International Law (Miscellaneous Provisions) Act (UK) 1995, c. 42.

598 J Llewellyn appropriate” to apply the lex fori. Such a determination is to be made according to a list of significant factors, including the connection between the country and the wrong. This rule, with its narrower exception, would seem to make the case for just amnesties in the United Kingdom even easier than it was under the Boys v. Chaplin rule. In Canada, the tort choice-of-law issue was not addressed until the rule was set out in the 1994 decision of the Supreme Court in Tolofson v. Jensen.84 According to Tolofson, the lex loci delicti rule is to determine the choice of law in torts cases. This is not, however, an absolute rule in the Canadian context either. Mr. Justice La Forest, for the court, carved out a very narrow exception to the lex loci delicti rule for cases involving a jurisdiction outside Canada, (i.e. jurisdictional relationships that are truly international as opposed to interprovincial): “. . . [B]ecause a rigid rule on the international level could give rise to injustice, in certain circumstances, I am not averse to retaining a discretion in the court to apply our own law to deal with such circumstances.”85 The discussion of La Forest J. earlier in the same judgment gives some clue as to the circumstances he might have in mind. “On the international plane, the relevant underlying reality is the territorial limits of law under the international legal order. The underlying postulate of public international law is that generally each state has jurisdiction to make and apply law within its territorial limits. Absent a breach of some overriding norm, other states as a matter of ‘comity’ will ordinarily respect such actions and are hesitant to interfere with what another state chooses to do with those limits.”86 (emphasis added)

La Forest J. is, however, clear that such exceptions must be carefully defined and that he can imagine very few cases to which these exceptions would apply. By using the term “overriding norm”, he seems to have had in mind here the kind of violations of jus cogens norms at stake in the debate over the concept of amnesty.87 Departure from the lex loci delicti is, then, warranted when it amounts to a violation of fundamental international law. Understood in this way, the Tolofson exception strikes the perfect balance from the perspective of just amnesty. According to Tolofson, the lex loci delicti is to be applied unless doing so would violate jus cogens norms. Just amnesties, then, will survive this exception given that, as argued earlier, they meet the demands by international law that justice be done in the form of investigation, accountability and redress 84

[1994] 3 SCR 1022 [hereinafter Tolofson]. The extent to which this exception is restricted to international situations (as opposed to interprovincial cases) is the subject of significant debate. However, this need not concern us for our current purposes. 86 Tolofson, supra n. 84 at 1047. 87 Article 64 of the Vienna Convention on the Law of Treaties defines jus cogens norms as peremptory norms of international law which are recognized as having an overriding quality by the international community as a whole. Should such a norm emerge, according to the Convention, any treaty which is in conflict with that norm becomes void and terminates. It is interesting to note that La Forest J. was the only judge on the Supreme Court of Canada at the time Tolofson was decided who came to the bench after an academic career in international law. 85

Just Amnesty and Private International Law 599 for victims. Amnesties that fail on these terms will fall under the Tolofson exception and be set aside in favour of the lex fori.88 The Canadian choice-of-law rule, then, seems on its face to require courts to respect just amnesty laws adopted in the place where the tort occurred. However, this raises the issue of determining of the place of the tort. One way for advocates of transnational tort litigation to get around the application of amnesties provisions, as necessitated by the lex loci delicti rule, is to locate the tort elsewhere. Particularly with respect to actions for torture, it can be argued that the tort of torture continues so long as the harmful effects are still felt by the victim.89 Thus, the tort is said to have “occurred” in the place where the victim resides and continues to suffer the effects of the wrongdoing. If one is able to claim that the tort or a significant part of the tort occurred in a jurisdiction other than the one offering the amnesty, then the whole issue of amnesty can be avoided. There are clearly problems with such an argument. Firstly, it may produce inconsistent results and end up treating some victims unfairly. Under this rule, a victim who manages to deal with her experience of torture in such a way as to put it behind her, at least with respect to continuing harms, would not be able to bring a claim under the law of the country in which she resides—assuming she has moved from the country in which the act of torture occurred. On the other hand, a victim who lives with the ongoing effects of torture in a more immediate way would benefit from the law of the new country in which she lives. Even if one leaves aside the difficulty in testing the extent to which the victim still suffers, it is clear that this rule would have discriminatory effect on those who manage to deal with their experiences more quickly. Secondly, if we return to the idea of restorative justice explained earlier, it is clear that this argument misconstrues the full nature of the harm involved in violations of human rights. While it is true that the victim might suffer ongoing physical harm from the act, the most significant harm is that done to the victim’s relationship with others. A restorative understanding of justice makes clear that the resulting harm from wrongdoing is, or at minimum includes, harm to social relationships. As such, all parties to the wrongdoing experience this harm—victim, 88 Recall that a just amnesty approach does not differentiate with respect to the act for which amnesty is granted. On this approach amnesty is just, and thus permissible according to its role in restoration. As restoration is the goal of justice, there is no group of acts, i.e.: violations of jus cogens norms, that is not the subject of justice and thus may not be the subject of a just amnesty. Whether it is in the interests of restoration or not, it is the standard against which amnesties are to be judged, not according to the status of the violation for which they are granted. Thus, a just amnesty approach to the Tolofson exception would limit the exception to cases where the amnesty at issue was unjust, rendering the subject of the amnesty irrelevant. 89 In his 8 October 1999 extradition decision in the Pinochet case, Bartle J. goes even further than this, to suggest that “the effect on the families of those who disappeared can amount to mental torture.” The idea that torture might include the ongoing psychological effects on family members of persons who have been “disappeared” has potentially far-reaching implications. Such family members could be present in any jurisdiction in the world, beyond the jurisdiction in which the disappearance took place. Pinochet No. 4, supra n. 17.

600 J Llewellyn perpetrator and community—albeit to different extents. Understood in this light, it is clear that a significant part of the ongoing harm is rooted in the place where the torture occurred. Thus, it remains appropriate for this to be regarded as the primary location of the tort for the purposes of a choice of law-analysis.

5 CONCLUSION

The case for the possibility of a just form of amnesty can be made under current private international law. The international community is able to, and should, draw a distinction between different amnesty provisions on the grounds of justice. The mere absence of either criminal law prosecution or civil liability should no longer stand as the mark of injustice crying out for remedy from the international community. Instead, one must attend to the context at issue and ask whether the interests of justice are being served. The concern is appropriately transformed from a focus on prosecution and civil litigation to a focus on restoration. Just amnesty, then, stands ready to play a fundamental role in achieving such restoration—in achieving justice.

23

Cultural Challenges: Injunctions in Australian Courts and the Right to Demand the Death Penalty under Saudi Arabian Law BELINDA WELLS and MICHAEL BURNETT 1

“A transnational action that defines general international standards by taking into account the context, legal norms, and cultural norms of the other state would be more in accordance with the principle of comity.”2

1 INTRODUCTION

litigation frequently raises issues of international comity, and often of inconsistent cultural norms. In what we will call the Saudi nurses case, the conflict of cultural values challenged assumptions about the objective nature of rights interpretation, and heightened the importance of issues such as jurisdiction and justiciability. The case provides a vivid illustration of the interaction of competing forces in such litigation. First, there is clearly a tension between the principle of the universality of international human rights and the need to appreciate the cultural context of rights. Secondly, the case provides an example of one of the central conundrums for domestic courts: should a court decide to uphold international human rights standards where the effect of its decision will be to interfere in foreign legal proceedings? The case began in December 1996 in Saudi Arabia. Two British nurses, Deborah Parry and Lucille McLauchlan, were accused by Saudi authorities of murdering an Australian nurse, Yvonne Gilford, at the foreign nurses’ compound at the King Fahd Military Hospital in Damman, Saudi Arabia. The criminal

T

RANSNATIONAL HUMAN RIGHTS

1 Respectively, Lecturer in Constitutional Law and Human Rights at the Flinders University of South Australia and Director of Enforcement (Queensland) for the Australian Securities and Investments Commission. 2 S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, in C Scott (ed,), Torture as Tort: Comparative Perspectives on the Development of Transnational Human Rights Litigation (Oxford, Hart Publishing, 2001), chapter 14 in this volume, at 380.

602 B Wells & M Burnett investigation process and the ensuing judicial hearings (which led to McLauchlan being convicted as an accessory to murder and sentenced to eight years imprisonment and five hundred lashes, and a likely conviction and death sentence for Parry3) did not involve a number of the procedural protections that would be regarded as fundamentally necessary by the British or Australian legal systems. The procedures adopted by the Saudi Arabian court would more clearly have been regarded as an area of exclusively Saudi concern (from the perspective of Australian law although not necessarily international human rights law) were it not for a particularly distinctive feature of Saudi law which empowered the murder victim’s relatives—Australian citizens—to demand the death penalty. Under the Shari’a, the Islamic religious law applicable to Muslims and nonMuslims alike in Saudi Arabia, relatives have the right to insist upon quesas: that is, to demand that the accused, if convicted, be beheaded. If the relatives unanimously demand quesas, then neither the Court nor the King can interfere with that sentence.4 At this point, there was a cultural intersection. Australian citizens, whilst in Australia and subject to its laws and culture, were authorised to perform the sentencing aspect of a judicial process that had been conducted in accordance with the Shari’a. The Saudi judicial process, and the criminal investigation process which preceded it, had not been consistent with Anglo-Australian practice in a number of respects. Most strikingly, the part of the process for which the Australian citizens were made ultimately responsible—the implementation of the death penalty—would amount to an offence under Australian law if carried out in Australia.5 The British nurses, whilst detained in Saudi Arabia, initiated proceedings in the Supreme Court of South Australia seeking declarations and injunctions to prevent the Australian relatives from taking action which they argued would result in the violation of their human rights. The case (which subsequently settled) very clearly raised questions of individual—private—responsibility for infringements of human rights norms. The manner by which a settlement came about is important to the themes of this chapter, and will be described later. In this chapter we consider the Saudi nurses case from the perspective of lawyers seeking to prevent the implementation of the death penalty.6 We discuss 3 The conviction and sentence of Parry were never formally announced. However, there are two reasons why a conclusion can be drawn as to her conviction and sentence. First, as McLauchlan was convicted as an accessory it is logical to assume that Parry had been convicted as a principal offender. Secondly, it is arguable that the trial in Saudi Arabia had already determined the guilt of the accused at the initial hearing and that the remainder of the hearings dealt only with the imposition of the sentence. 4 See A al-Alfi, “Punishment in Islamic Criminal Law”, in M Bassiouni (ed. and trans.), The Islamic Criminal Justice System (New York, Oceana Publications, 1982) at 227. 5 Section 4 of the Statutes Amendment (Capital Punishment Abolition) Act, 1976 (SA) (“the Abolition Act”) provides, inter alia, that no sentence of death shall be imposed on any person. The implementation of the death penalty is therefore unlawful in South Australia. 6 The second-mentioned author of the present chapter was the Australian lawyer who represented Parry and McLauchlan in the proceedings before the Australian courts.

Cultural Challenges 603 the way in which applications by British citizens for equitable remedies—declarations and injunctions—were used to bring the issues “on-shore”, within the consideration of an Australian court. Clearly, the granting of these remedies, even if only a declaration, but especially if an injunction, raised issues of jurisdiction and international comity. Any remedy granted in these circumstances, although directed against Australian citizens (and issued to them in Australia itself) rather than against the Saudi Arabian authorities, might understandably be regarded by Saudi Arabia as an interference in its judicial process and an affront to international comity. The discussion in this chapter focuses on the application of Australia’s Crimes (Torture) Act 1988 (Cth) within the Saudi Arabian context as an example of one of the central tensions in transnational human rights litigation: the universality or cultural relativity of international human rights norms. The Act results from a direct decision by the Australian government to partially implement an international human rights treaty ratified by it—the UN Convention against Torture (CAT).7 If the Saudi nurses case had proceeded to trial (rather than ending with a settlement), an Australian court could have been faced with a major question: in the application of international human rights standards such as the rights to be free from torture and inhuman or degrading punishment, can and should the content of the rights be determined in a way that maintains the underlying morality of the rights but which is simultaneously sensitive to the different cultural context in which the potential violation may occur?8 In any event, should non-Muslims be subjected to punishments prescribed by the Shari’a,9 or be permitted to participate in the Shari’a sentencing process? We will focus on what could have been achieved had an Australian court been required to rule on the issues raised by the Saudi nurses case. However, even in the case of the right to be free from torture (which is as universally supported by states’ pronouncements as any other right and which is conferred by Australian law in a statute which does have clear extraterritorial application), there are major difficulties with the scope of the concept of torture, especially in light of the cultural context (a death sentence to be performed in Saudi Arabia in accordance with its religious law). And there are doubts as to whether, in the amended injunction proceedings (described below in the next section), an Australian court would have been prepared to exercise jurisdiction, 7 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, 39 U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027 [hereinafter “CAT”]. 8 See A A An-Na’im, “Toward a Cross-Cultural Approach to Defining International Standards of Human Rights: The Meaning of Cruel, Inhuman or Degrading Treatment or Punishment”, in A A An-Na’im (ed.), Human Rights in Cross-Cultural Perspectives: A Quest for Consensus (Philadelphia, University of Pennsylvania Press, 1992). See also the discussion of An-Na’im’s work in S Raponi, supra n. 2, at 373. 9 In the view of An-Na’im, they should not be, since such punishments are “essentially religious in nature”: A A An-Na’im, supra n. 8 at 34.

604 B Wells & M Burnett indeed whether it would even have regarded some of the questions before it as justiciable.

2 THE FACTUAL BACKGROUND

For about eight days in December 1996, the British nurses were detained by Saudi authorities and questioned after being accused of Yvonne Gilford’s murder. During this period, they were not given access to lawyers or consular officials. Whilst detained, each nurse signed two confessions. The second confession in each case was fundamentally different from the first. The nurses later stated that both confessions had been made under physical and sexual duress, and that they had signed the second confessions only after being given assurances that they would then be allowed to return to Britain. The trial was conducted by the High Shari’a Court in Al Khobar, Saudi Arabia. The Court sat as a closed court. The prosecution did not attend any of the eight short hearings which comprised the trial, nor call witnesses, since under Islamic law a confession or two eye-witnesses is sufficient proof of murder. The British nurses did not have an opportunity to cross-examine witnesses nor to present evidence on their own behalf to attempt to show that they had not committed any offence, that another party may have done so, or that the confessions were obtained under duress. The prosecution did not at any stage inform the nurses of the nature of the case against them or of any evidence against them other than the confessions. The nurses’ lawyers were not granted access to the scene of the crime or permitted to investigate or conduct examinations of that scene. The British nurses were tried under and in accordance with the Shari’a. In Saudi Arabia, Islamic religious law is applicable to Muslims and non-Muslims alike. As noted in the Introduction, under the Shari’a the relatives of a murder victim have the right to insist upon quesas (also kisas or qisas): that is, to demand that the accused, if found guilty of murder, be put to death. In the Saudi nurses case, Saudi Arabian lawyers acting for the murder victim’s brother (an Australian living in Australia) initially informed the High Shari’a Court that the brother demanded the imposition of the death penalty in that event. In August 1997, at the conclusion of the hearings, and without having informed the nurses of its decision or reasons for decision, the High Shari’a Court referred the case for review to the Court of Cassation. Apart from the opportunity to make written submissions, neither the nurses nor their lawyers were able to participate in the appeal. During the course of the appeal (and after the parties had reached an agreement to settle the proceedings in Australia) one of the British nurses, McLauchlan, was brought before the High Shari’a Court and advised that she had been convicted of being an accessory to murder and sentenced to eight years imprisonment and five hundred lashes. The quesas death penalty procedure was therefore not ultimately relevant to her sentence,

Cultural Challenges 605 since the penalty was imposed by the State and not by the murder victim’s relatives. Although it was widely reported that the second nurse, Parry, had been found guilty of the murder and was subject to the death penalty quesas procedure, the verdict and sentence for her were never formally announced.10 The murder victim’s brother advised the Court in Saudi Arabia that he no longer sought quesas (as a settlement had been achieved). Prior to the High Shari’a Court completing its hearings (and at a time when both nurses were subject to the imposition of the death penalty), the British nurses instituted proceedings in South Australia, where the brother lived, seeking declarations and certain interlocutory injunctions. Injunctions were initially sought to restrain the brother from advising the Court in Saudi Arabia that it was the unanimous view of the heirs that the death penalty should be inflicted if the accused were found guilty of intentional murder. In addition, a mandatory injunction was sought requiring the brother to advise the court in Saudi Arabia that the mother’s mental competence was to be determined by the South Australian Supreme Court. Declarations as to the mother’s competence and the brother’s right to say that the mother agreed to seek quesas were also sought. As will be discussed in detail below, the Supreme Court granted the interim injuctions sought.11 The plaintiffs then amended the legal proceedings to seek an additional injunction to prevent the brother from communicating a demand for the imposition of the death penalty to the Saudi Arabian Court. The injunction was sought on the basis that Australian law prevented the brother from demanding the death penalty. This question was never resolved by the courts, as the brother entered into a settlement agreement with the British nurses four days before the trial was due to begin. He agreed that upon the payment of an agreed sum of compensation he, as an heir, would waive his right to demand the death penalty.12 The private right having been waived, the Saudi Arabian State retained its public right to impose a sentence of imprisonment on the British nurses.13 The nurses remained imprisoned until an act of clemency by the King of Saudi Arabia led to their release from prison and return to Britain in mid-1998.

10

See supra n. 3. See McLauchlan and Another v. Gilford (1997) 69 SASR 269 [“McLauchlan”], and the discussion infra text at n. 50. 12 The settlement sum was Aus$1.7m. The brother gave Aus$1m to the Women’s and Children’s Hosptial in Adelaide. Most of the remainder of the settlement sum went to the payment of the brother’s lawyers in Australia and Saudi Arabia. 13 A al-Alfi, supra n. 4 at 227. HE Dr G Al-Gosaibi, Ambassador to the United Kingdom, as quoted in S Nimr, “Al-Gosaibi Speaks to Asharq Al-Awsat about Blair’s Discussions in Saudi Arabia”, Asharq Al-Awsat No. 7083 at 4 (20 April 1998); Press Release from the Office of Ambassador Ghazi Algosaibi dated 19 May 1998, as set out in The Advertiser newspaper (Adelaide, South Australia), 21 May 1998. Most of the factual information set out in the text above, as well as much of the narrative in the rest of this chapter, was obtained from private sources. 11

606 B Wells & M Burnett

3 ISLAMIC LAW ( THE SHARI ’ A )

As has already been stated, Saudi Arabia is governed according to Islamic law (generally referred to as “the Shari’a”).14 The Basic System of Saudi Arabia issued in 1992 reaffirms the State’s commitment to the implementation of Shari’a, and clearly indicates the fusion of religion and state in Saudi Arabia.15 There are four sources of the Shari’a. The most important of these is the Koran (or Qu’ran). The Koran contains about five hundred verses which set out legal rules—each in the form of either an order (‘amr) or a prohibition (nahi).16 These legal rules govern all aspects of public and private behaviour,17 including many matters that would not normally be regulated by the state in secular legal systems such as diet, sexual conduct and aspects of child-rearing.18 Accordingly, application is often personal.19 The other sources of the Shari’a are the Sunna (the practice of the Prophet Mohammed or his disciples),20 Ijma’ (the consensus of scholars of a particular age in relation to the application of a particular rule), 21 and Qiyas (analogical reasoning). As to the latter source, if no rule is found in the Koran or Sunna, then the jurist finds the rule in an “inductive process governed by rules of logic”.22

14 King Abd al-Aziz Ibn Sa’ud unified the Saudi Arabian legal system in 1926, based on the Hanbali School of Islamic law: see SH Amin, Middle Eastern Legal Systems (Glasgow, Royston Limited, 1985) at 314; H J Liebesny, The Law of the Near and Middle East (Albany, State University of New York Press, 1975) at 107. 15 See, for example, Articles 7, 8, 23 and 48 of the Basic System, as set out in “Kingdom of Saudi Arabia” in G H Flanz (ed.), Constitutions of the World (New York, Oceana Publications Inc, 1995) at 45–6, 48 and 51. For a discussion of the status and background to the Basic System (or “Basic Law”) in Saudi Arabia, see A E Mayer, Islam and Human Rights (2nd edn, Boulder, Colo., Westview Press, 1995), at 24, 34–5; R Aba-Namay, “The Recent Constitutional Reforms in Saudi Arabia” (1993) 42 ICLQ 295. 16 See T Kamel, “The Principle of Legality and its Application in Islamic Criminal Justice”, in M Bassiouni, supra n. 4 at 152; M Lippman, S McConville and M Yerushalmi, Islamic Criminal Law and Procedure: An Introduction (Westport, Conn., Greenwood Press, 1988) [“M Lippman et al.”] at 23. 17 See T Kamel, supra n. 16. 18 C G Weeramantry, Islamic Jurisprudence: an International Perspective (London, Macmillan, 1988), p 1. See also H J Liebesny, supra n. 14 at 4. 19 M Bassiouni, “Sources of Islamic Law and the Protection of Human rights in the Islamic Criminal Justice System”, in M Bassiouni, supra n. 4 at 15. 20 See M Lippman et al., supra n. 16 at 30–1; H J Liebesny, supra n. 14 at 13–18; and F Malekian, The Concept of Islamic International Criminal Law (London, Graham & Trotman/Martinus Nijhoff, 1994) at 31–2. 21 See C G Weeramantry, supra n. 18 at 39; F Malekian, supra n. 20 at 32; H J Liebesny, supra n. 14 at 16; and M Lippman et al., supra n. 16 at 31. 22 H J Liebesny, supra n. 14 at 18.

Cultural Challenges 607

Criminal Process The Shari’a provides clearly defined and rigid rules of evidence and procedure to apply during criminal investigations and trials.23 In the case of murder, the required proof is two eye-witnesses or a confession.24 There is no crossexamination or testing of a witness’s credibility or the weighing of one party’s evidence against the other.25 According to the Hanbali School of Islamic law applicable in Saudi Arabia, counsel may not be retained to assist defendants.26 The trial of the British nurses was a departure from this rule: for the first time accused persons charged with murder were permitted counsel. In some respects, Saudi Arabian practice may not be consistent with the dictates of the Shari’a. For example, under the Shari’a a confession may be withdrawn at any time prior to execution.27 In the Saudi nurses case, however, the British nurses were not permitted to withdraw their confessions. And they argued that their confessions were not given freely and voluntarily in the first place, as is required by the Shari’a.28 In the Basic System, the Saudi Arabian state has made a commitment to “protect the rights of the people” but only “in line with the Islamic Shari’a”.29 The Shari’a expressly prohibits torture and other cruel and inhumane treatment.30 In addition, the Universal Islamic Declaration of Human Rights, which Saudi Arabia supported,31 provides that no one shall be subjected to “torture in mind or body, or degraded”.32 However, a general prohibition is one thing and giving concrete interpretive content to that prohibition is another: no Islamic court 23 “[T]he rigid system of procedure and evidence makes fact-finding an almost automatic process and advocacy wholly superfluous:” N J Coulson, Conflicts and Tensions in Islamic Jurisprudence (Chicago, University of Chicago Press, 1969) at 61, 63. 24 M Lippman et al., supra n. 16 at 43. 25 N J Coulson, supra n. 23, at 63; M Lippman et al., supra n. 16 at 66–73. 26 See M Lippman et al., supra n. 16 at 65; N J Coulson, supra n. 23 at 61. 27 O al-Saleh, “The Right of the Individual to Personal Security in Islam”, in M Bassiouni, supra n. 4 at 73. 28 Ibid; M Lippman et al., supra n. 16 at 63. In addition, see J Young, “Torture and Inhumane Punishment of United States Citizens in Saudi Arabia and the United States Government’s Failure to Act” (1993) Hastings Int’l & Comp. Law Rev 663 at 673–9. Young refers to the hearings before the Sub-committee on Europe and the Middle East of the US House of Representatives Committee on Foreign Affairs, 101st Congress, 2nd Session (1990), in support of her contentions of torture and inhumane punishment. 29 The Basic System of Saudi Arabia, 1992 as set out in G H Flanz (ed.), supra n. 15, article 26. For an excellent discussion of cultural relativism and Islamic approaches to human rights, see A E Mayer, supra n. 15 esp at 6–18, 179–83; see also n. 28 supra. For a conservative Islamic perspective, see S Tabandeh, A Muslim Commentary on the Universal Declaration of Human Rights (London, Golding and Company, 1970); see also F Malekian, supra n. 20. 30 O al-Saleh, supra n. 27 at 72–3; M Lippman et al., supra n. 16 at 63. 31 A E Mayer, supra n. 15 at 22. 32 Universal Islamic Declaration of Human Rights, Islamic Council, 1981, article VII, as set out in F Malekian, supra n. 20 at 187. See also discussion of Saudi Arabia’s support for the Cairo Declaration on Human Rights in Islam of 1993 and its attitude to the concept of universal human rights norms, in A E Mayer, supra n. 15 at 23–4, 173–81, 183.

608 B Wells & M Burnett would regard a punishment prescribed by the Shari’a as “torture”, “cruel and inhumane”, or “degrading”.33

The Right to Seek Quesas It will be recalled that the Koran confers on the relatives of murder victims a right to seek quesas.34 “Quesas” strictly means “equality,” notably equality of outcome as between the victim’s and perpetrator’s families. It was introduced to prevent the pre-Islamic practice in which the relatives or tribe of a murder victim would exact revenge on the tribe of the murderer.35 The Koran refers to “a life for a life, an eye for an eye, . . . a tooth for a tooth”36—one of the few phrases in the Koran with which westerners are familiar. When applied to murder, quesas means that the relatives (or “heirs”) of the murder victim may, if unanimously agreed, demand that the death penalty be carried out against the convicted person. If the relatives unanimously exercise this so-called “private” right, then neither the Court nor the King can interfere in the sentence.37 The sentence must be carried out. The Koran does not specify a particular method of execution. In Saudi Arabia the death penalty is carried out by public beheading. Quesas crimes—crimes against the person and the body—are said to be private wrongs that are prosecuted by the heirs of the victim, rather than by the state.38 The right of quesas therefore has a strong element of retributive “justice for the victim”,39 although in Saudi Arabia it also includes a significant element of general deterrence since the penalty is carried out by way of a public beheading. However, the relatives of the murder victim also have other alternatives to 33 The different perspectives underlying Western interpretations of these standards, on the one hand, and Qur’anic punishments, on the other, are discussed in A A An-Na’im, supra n. 8. According to M Lippman et al., the punishments prescribed by the Shari’a for some crimes include stoning to death for adultery, death by beheading, amputations of hand(s) and/or feet, live burial and being thrown from a high building: M Lippman et al., supra n. 16 at 42–5. In relation to practice in Saudi Arabia, see M Lippman et al., supra n. 16 at 91–2, nn. 34 and 42. 34 The Koran states as follows: Verse 2:178: “Oh believers, the law of qisas, exact retaliation, is prescribed for you in cases of murder: a free person for a free person, a slave for a slave, and a woman for a woman. But if any remission is made by the brother [kin] of the slain, then grant any reasonable demand, and compensate him with handsome gratitude. This is a concession and a mercy from your Lord. Whoever exceeds this limit shall receive great penalty.” Verse 5:45: “We have ordained [in the Torah] that a life [should be taken] for a life, an eye for an eye, a nose for a nose, an ear for an ear, a tooth for a tooth, and wounds [are to be punished] by qisas, exact retaliation But if anyone remits exact retaliation by way of charity, that will be an act of atonement for himself. Whoever fails to judge in accordance with what God has revealed shall be of the unfair ones.” This translation appears in A An-Na’im, Toward an Islamic Reformation (Syracuse, Syracuse University Press, 1990) at 115–16. 35 See M Lippman et al., supra n. 16 at 2, 85 which refers to the pre-Islamic “system of private vengeance [which] often set in motion ever-widening cycles of retaliation between tribes”. 36 The Koran, verse 5:45. 37 See A al-Alfi, supra n. 4. 38 M Lippman et al., supra n. 16 at 41. 39 Ibid. at 84.

Cultural Challenges 609 demanding quesas. They may instead elect to receive diya (“blood money”—the amount of which is fixed40), or to pardon the accused.41 The Koran clearly indicates a preference that the heirs exercise their rights to receive diya or to give forgiveness, rather than exercise their right to quesas.42 And the Sunna states that a person who forgives the offender is “elevated by one more step” and his or her sins are “reduced by one step”.43 It is said that this preference for diya or forgiveness “illustrates a bond of continuity between the temporal law and religion since the forgiver will be rewarded in heaven, which for the Muslim is a greater reward than any other”.44 There are therefore powerful religious reasons for a Muslim not to exercise his or her right to quesas. These religious reasons do not apply to, and thus do not constrain, a non-Muslim Australian resident. As such, Australian law should not, in our view, permit an Australian citizen to adopt an aspect of Islamic law and demand the death penalty in circumstances in which the rationale and countervailing restraints of the Shari’a restricting the exercise of the power have no application. This, it seems to us, is consistent with the point that An-Na’im makes about the application of Islamic punishments to non-Muslims. In his view such punishments should not apply to non-Muslims “because they are essentially religious in nature”.45

4 BRINGING THE RIGHTS

“ ON - SHORE ”:

THE INTERLOCUTORY PROCEEDINGS

The Linked Issues of Mental Competence of an Heir and the Tort of Injurious Falsehood: The Question of the Jurisdiction of the Australian Court Whilst the two British nurses were held in custody in Saudi Arabia , accused of having committed or assisted in a murder, most of the events affecting their fate were taking place in that country as part of the investigative and judicial process. However, the people who had ultimate control over the fate of the nurses—namely the murder victim’s “heirs” (her brother and mother)—lived in South Australia. 40 “According to a legal opinion of the Mufti of Saudi Arabia of 1955, the blood money for the intentional or quasi-intentional killing of a male Muslim is one hundred camels”: H J Liebesny, supra n. 14 at 232. “Diyya payment for a woman is half the payment for a Muslim man. For a Jew, Christian or other nonbeliever the payment is a third of that for a Muslim”: M Lippman et al., supra n. 16 at 85. H J Liebesny points out that the concepts of retaliation and blood money have historically been “widespread in the laws of many peoples”: supra n. 14 at 232. 41 A al-Alfi, supra n. 4 at 127; M Lippman et al., supra n. 16 at 84. And, as is evident from the Saudi nurses case itself, the relative(s) of the murder victim may agree to receive compensation other than Court-ordered blood money in return for waiving their right to quesas. 42 M Bassiouni, “Quesas Crimes”, in M Bassiouni, supra n. 4 at 205. See also the passages from the Koran set out at n. 34 supra. 43 M Lippman et al., supra n. 16 at 30. 44 M Bassiouni, supra n. 4 at 205. 45 A A An-Na’im, supra n. 8 at 34.

610 B Wells & M Burnett The High Shari’a Court, during one of its hearings, raised the question of the medical condition of the mother. The British nurses’ lawyers in Saudi Arabia considered that if the mother of the victim was mentally incompetent, then the nurses might have an argument that under Shari’a law there could not be the required unanimity of heirs due to the fact that two heirs were alive but one was not able to give the requisite quesas demand. Lawyers in South Australia were engaged to find out whether the mother was mentally competent to express an informed view to the Saudi Arabian Court. Since initial investigations indicated that she was not, legal proceedings were commenced in South Australia against the brother and mother. The British nurses sought a declaration that the mother was mentally incompetent, and interim injunctions to preserve the situation in the Saudi Arabian legal proceedings. They sought injunctions to prevent the brother from advising the Saudi Arabian court that it was the unanimous view of the heirs that the death penalty should be imposed if the accused were found guilty of intentional murder, and to require the brother to advise the Saudi Court that the mother’s mental competence was to be determined by the Supreme Court in South Australia. In addition, the British nurses relied upon the tort of injurious falsehood. They argued that this tort had been committed in South Australia (rather than in Saudi Arabia) when the brother had informed his Saudi Arabian lawyers that it was the unanimous view of both heirs that the death penalty should be implemented. The proceedings were heard with great urgency, since the High Shari’a Court in Al Khobar was due to reconvene and complete its deliberations in two days’ time. The jurisdiction of the South Australian Supreme Court to hear the proceedings was clearly put in issue by the defendants. However, King AJ held that he did have jurisdiction both to grant the declaratory relief sought and to adjudicate upon the cause of action brought for malicious or injurious falsehood. As to the jurisdiction over the application for declaratory relief (a statement of the mother’s mental competence), King AJ relied upon the fact that the mother (the first defendant) was within the jurisdiction of the Court, having been served within the jurisdiction and having entered an appearance, and emphasized that “[t]he truth of her mental state . . . [was] more readily ascertainable in proceedings in . . . [the Supreme] Court than they would be in Saudi Arabia or elsewhere”.46 As to jurisdiction to entertain the cause of action for malicious or injurious falsehood, King AJ considered that the location of the tort in Australia was not obvious (it was “an arguable proposition”), but held that in any event the tort was actionable in the Supreme Court since the defendants had been served within the jurisdiction and the tort would “not (be) justifiable according to the law of Saudi Arabia”. He took the view that Saudi law could be presumed for present purposes to be the same as the law of South Australia, no proof of its content having been pleaded before the court at that

46

McLauchlan, supra n. 11 at 271.

Cultural Challenges 611 stage.47 This presumption of foreign law being identical to local law is a common rule of evidence in private international law cases in jurisdictions where foreign law must be proven as a matter of fact. The defendants had also argued that the proceedings constituted “an abuse of the process of the court” since they were brought “not for the purpose of obtaining any genuine remedy” in the Supreme Court, “but rather as a means of interfering . . . with the proceedings in the court in Saudi Arabia”.48 King AJ acknowledged that such a purpose could be characterised as an abuse of process in some cases, but regarded it as “quite legitimate” for the plaintiffs in this particular case “to invoke the jurisdiction of . . . [the Supreme] Court in order to obtain relief here which . . . [would] avert damage to . . . [them] elsewhere”.49

The Remedies: Declarations and Injunctions In the Saudi nurses case, the British nurses were attempting to obtain a declaration as to the incompetence of the murder victim’s mother so that they could then obtain final injunctions to enforce that declaration.50 In South Australia, and in other jurisdictions, a court may grant a declaration whenever a plaintiffs’ legal rights are or will be affected, regardless of whether any other substantive relief is sought.51 In the Saudi nurses case King AJ held that the statement alleged to have been made to the Saudi Arabian Court as to the unanimous view of the heirs, would, if the statement were untrue, infringe the plaintiff’s rights to ensure that all statements made in South Australia and provided to the Saudi Arabian court were true.52 King AJ was thus indicating that he would be prepared to grant a declaration if the mother’s incompetence was established on the evidence. In the meantime, King AJ granted a prohibitory injunction restraining the defendants “from advising . . . the Saudi Arabian Court that it is the unanimous view of the heirs of . . . [the murder victim] that the death penalty should be inflicted on the plaintiffs in the event that they are found guilty of . . . intentional murder.”53 In addition, he granted a mandatory injunction requiring the brother 47 Ibid. King AJ was here applying the tort double actionability choice of law rule for tort which Australia has retained even as other Commonwealth jurisdictions like Canada and the UK have moved to a presumptive lex loci delicti rule: see J Orange, “Torture, Tort Choice of Law and Tolofson”, G Virgo, “Characterisation, Choice of Law and Human Rights” and S Raponi, “Grounding a Cause of Action for Torture in Transnational Law”, chapters 11, 12 and 14 of this volume. 48 Supra n. 11 at 272. 49 Ibid. 50 A legal or equitable right, once declared, would justify the granting of injunctive relief: see G Dal Pont and D Chalmers, Equity and Trusts in Australia and New Zealand (Sydney, LBC Information Services, 1996) at 574. 51 See Supreme Court Act 1935 (SA), s 31; JN Taylor Holdings Ltd v. Bond (1993) 59 SASR 432 at 435–7 per King CJ. 52 McLauchlan, supra n. 11 at 271. 53 McLauchlan, supra n. 11 at 273.

612 B Wells & M Burnett to advise the Saudi Arabian Court that the Supreme Court of South Australia would determine the mental competence of the mother. It is important to bear in mind that these were interlocutory injunctions designed to preserve the rights of the plaintiffs prior to the adjudication of the claims on the merits. The prohibitory injunction was accordingly issued to preserve the status quo pending trial and not as any indication that the judge had reached a final view on either the mental state of the mother or the truth of the alleged statements already made on behalf of the brother before the Saudi court.

5 THE CULTURAL CHALLENGE : POTENTIAL ARGUMENTS

Despite the grant of interlocutory injunctions by King AJ, the prospects of a successful outcome to the case soon diminished. The High Shari’a Court at its next hearing indicated that it would be prepared to act upon the demand of the brother alone, notwithstanding the lack of mental competence of the other heir. Therefore, soon after the grant of the South Australian interim injunctions, the British nurses’ lawyers amended their claim in the South Australian Supreme Court. As amended, the legal proceedings in South Australia now squarely raised the fundamental issue of the legality of the brother’s actions. In clear terms that could have been viewed as a direct challenge to Saudi Arabian law and culture; the Australian court was now asked to consider whether it is contrary to Australian law for an Australian citizen to demand the death penalty. The Court’s attention was no longer confined to the limited issues of the unanimity of the heirs and the mother’s mental competence (issues which did not directly challenge the right to make a quesas demand if both heirs were indeed competent). The plaintiffs now sought declarations that as a matter of Australian law an Australian citizen could not demand the death penalty. In addition, they sought both a permanent prohibitory injunction preventing the brother from continuing to demand the death penalty and a mandatory injunction requiring the brother to advise the Saudi Arabian Court that he no longer sought the death penalty. The arguments that were advanced on behalf of the British nurses in the Saudi nurses case exposed an unusual area of uncertainty in Australian law. Although Australia has abolished the death penalty, it is unclear whether an Australian citizen who demands rather than implements the death penalty is acting contrary to Australian law.54 By definition, of course, any such demand addressed 54 The death penalty was last imposed in Australia in 1967: M Finnane, Punishment in Australian Society (Melbourne, Oxford University Press, 1997) at 126. It was finally abolished as a penal option in 1985: see LBC, Laws of Australia , vol. 21, Human Rights, “21.7 Life, Health and Welfare” (as at 31 May 1998), [9], n. 3. Australia was one of the first countries to ratify The Second Optional Protocol to the International Covenant on Civil and Political Rights, Aiming at the Abolition of the Death Penalty (“the Second Optional Protocol”), which places an obligation on States to abolish the death penalty: see GA Res 44/128, UN Doc A/44/824, UN Doc A/44/PV 82, p 11; LBC, Laws of Australia, ibid., n. 1; W A Schabas, The Abolition of the Death Penalty in International Law (2nd edn, Cambridge, Cambridge University Press, 1997) at 175–6.

Cultural Challenges 613 to the Australian state would not be unlawful as there would be no possibility of the state acting on the demand. Thus, the question really was whether such a demand made to another state (which, per hypothesi, that other state would act on) was unlawful. As such, the issue fused a substantive issue (is such a demand unlawful?) with a jurisdictional issue (is the demand intended to cause consequences abroad and, if so, does Australian law govern such extraterritorial consequences?). In seeking to establish the illegality of the brother’s actions, the British nurses had various possible arguments at their disposal. They could have requested the Australian court to grant an injunction to restrain breaches of common law principles constraining the exercise of judicial power. They could also have sought to restrain breaches of legal rights set out in the 1976 Abolition Act, the 1354 Statute of Edward, the 1689 Bill of Rights and the 1914 Australian Crimes Act.55 These statutes prohibit, respectively, the imposition of the death penalty, the imposition of the death penalty in the absence of “due process of law”, the infliction of “cruel and unusual punishment”, and aiding or abetting the commission of an offence (including the offence of torture under the Crimes (Torture) Act 1988). The second and third-mentioned statutes are old English statutes which, for reasons particular to Australia that need not detain us here, were received into Australian law and continue in force. Only some of these arguments were in fact relied upon in the amended legal proceedings. However, despite the initial victory in obtaining an interim declaration and injunctions, we are not at all certain that the plaintiffs would have prevailed on the merits of the case. Even aside from issues of the applicability of the statutes,56 and choice of law questions,57 there would have been major difficulties in persuading an Australian Court to apply these statutes and common law principles to the Saudi nurses case. With the exception of the Crimes (Torture) Act, none of the statutes or principles expressly stated that they applied extraterritorially, that is, to activities occurring outside South Australian territory. If, as seems quite likely, an Australian court decided that the brother’s actions in demanding the death penalty occurred in Saudi Arabia rather than in South Australia, it would have been necessary to establish that the common law and statutes that the plaintiffs relied upon were intended to apply extraterritorially. Even if the court were prepared to characterise some or all of the values espoused in the laws as “universal” in nature,58 this would have been a difficult

55 The Abolition Act, supra n. 5; 28 Edw. III, c.3 (1354); 1 William III and Mary, session 2, c.2 (1689); Crimes Act 1914 (Cth). 56 For example, in order for the Abolition Act to apply to the facts under consideration, it would be necessary to establish that the demand for the death penalty by the murder victim’s relative amounted to the imposition of a death sentence: see the Abolition Act, s 4. 57 See the more detailed discussion of private international law issues in B Wells and M Burnett, “When Cultures Collide: An Australian Citizen’s Power to Demand the Death Penalty under Islamic Law” (2000) 22(1) Syd LR 5. 58 Ibid.

614 B Wells & M Burnett hurdle to overcome given, inter alia, the general presumption of statutory interpretation that Australian laws are not intended to apply extraterritorially.

The Crimes (Torture) Act In the case of the Crimes (Torture) Act 1988 (Cth), however, there was no such problem with respect to extraterritorial application. Reflecting a late twentiethcentury perspective on human rights, the Crimes (Torture) Act expressly states that it is extraterritorial in operation and applicable to non-nationals. It imposes responsibility in the form of criminal liability on individuals rather than on states.59 The difficulties to be faced in basing arguments on the Crimes (Torture) Act are of a different nature from those involving the formal scope of application. Here, one must consider the interrelationship between Islamic cultural traditions and the concept of torture, as well as the failure of the Act to create an extraterritorial crime of cruel, inhuman or degrading punishment (thereby limiting its extraterritorial application to “torture”). The potential for cultural conflict over the interpretation of torture is obvious when one considers the nature of the arguments that the British nurses could have chosen (but did not60) to rely upon in the amended legal proceedings. The argument was, first, that if the Saudi Arabian public officials implemented the death penalty by means of a public beheading, they would be committing an act of torture which is made an offence under Australian law by the Crimes (Torture) Act.61 Secondly, so the argument runs, the murder victim’s relative, in demanding the imposition of the death penalty, would be aiding the commission of this offence and would therefore himself be committing an offence since the 1914 Crimes Act makes it an offence to aid or abet, or urge or encourage the commission of an offence.62 The Crimes (Torture) Act partially implements the Convention against Torture (CAT) which Australia subsequently ratified in 1989.63 The Act creates an extraterritorial criminal offence: it provides that any person who commits an act of torture outside Australian territory is guilty of an offence under 59 Nonetheless the Crimes (Torture) Act is, like the statutes mentioned earlier (which leave all issues of application unclear), directed at constraining public action—the actions of “a public official or other person acting in an official capacity”, rather than private action. In this respect, neither international nor domestic law reflects contemporary recognition of the need to constrain private action in the interest of protecting human rights. See, generally, A Clapham, Human Rights in the Private Sphere (Oxford, Clarendon Press, 1993); R McCorquodale, “First Correct the Wrongs on Human Rights”, The Australian, 12 October 1998. 60 The Crimes (Torture) Act 1988 (Cth) was not relied upon by the British nurses in the legal proceedings as their lawyers considered that this argument would cause unavoidable delay to the hearing of the proceedings. 61 Crimes (Torture) Act, s. 6. 62 Crimes Act 1914 (Cth), ss 5 and 7A. 63 Australia signed the CAT, supra n. 7, on 10 December 1985 and ratified it on 8 August 1989: LBC, Laws of Australia, vol 1, “1.7 International Law” (as at 1 September 1995) [35].

Cultural Challenges 615 Australian law.64 It confers jurisdiction to prosecute such an act of torture on the basis of the alleged offender’s Australian nationality or on the broader basis of his or her presence in Australian territory.65 The Act does not, however, create civil liability for torture committed outside Australian territory.66 The facts of the Saudi nurses case raise a central issue as regards the concept of torture: could the implementation of the death penalty ever constitute “torture” for the purpose of the Crimes (Torture) Act and/or the CAT? The Crimes (Torture) Act has adopted the same definition of torture as that in the Convention—except perhaps in one important respect. The possible difference lies in the scope of the exception to the definition. The CAT excludes all lawful sanctions from the definition of torture.67 The Crimes (Torture) Act, however, states that an “act of torture” does not include an act which arises “only from, inherent in or incidental to, lawful sanctions that are not inconsistent with the Articles of ” another human rights treaty, the International Covenant on Civil and Political Rights (the ICCPR).68 It is unclear whether an Australian court would regard this additional phrase in the Crimes (Torture) Act as a clarification, or as an extension, of the ambit of the CAT.69 In any event, the effect of the exception in the Crimes (Torture) Act is that lawful sanctions (such as the imposition of the death penalty after conviction for an offence) will only be excluded from the scope of the Act if they are consistent with rights conferred by the ICCPR. One such right is the right to life set out in article 6 of the ICCPR. Article 6(2) states that “[i]n countries which have not abolished the death penalty, sentence of death may be imposed only for the most 64

Crimes (Torture) Act (1988) (Cth), s. 6(1). Ibid., s. 7. (thereby omitting two of the four bases for jurisdiction mentioned in the CAT). 66 Article 14 of the CAT, which imposes on each State party an obligation to “ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation”, does not state that it applies only to acts of torture committed within the jurisdiction. In relation to the US, see the Torture Convention Implementing Legislation, 1994, 18 USCA, ss 2340–2340B and the Torture Victim Protection Act 106 Stat 73 (1992), 28 USCA s, 1350: H J Steiner and P Alston, Notes, International Human Rights in Context: Law, Politics, Morals (Oxford, Clarendon Press, 1996) at 801–6. See also the Alien Tort Claims Act, 28 USC s. 1350. By contrast, in Australia’s report on its implementation of the CAT, article 14 of the CAT is said to be implemented through the existence of State criminal injuries compensation legislation (and other like statutory provisions) which are unlikely to be regarded as applicable to acts of torture committed outside the jurisdiction: Australia, Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment: First Report by Australia (unpublished, 1991) (“First Report”) at 42–3. 67 CAT, supra n. 7, article 1. 68 Crimes (Torture) Act, s. 3 (1) (emphasis added). 69 In their discussion of the “lawful sanction” exception in the CAT, Burgers and Danelius indicate that views differ on the scope of the exception: J H Burgers and H Danelius, The United Nations Convention against Torture (Dordrecht, Martinus Nijhoff Publishers, 1988) at 121–2. If the exception contained in the Crimes (Torture) Act was narrower than that contained in the CAT (so that the definition of torture in the Act was wider), there would be doubts as to the validity of this aspect of the Act, since the extension might not be regarded as an “external affair” for the purpose of s. 51(xxix) of the Australian Constitution. However, in its first report to the UN Committee against Torture on the implementation of the CAT, Australia has stated that the Act “defines torture as it is defined in the Convention”: Australia, First Report, supra n. 66 at 27. 65

616 B Wells & M Burnett serious crimes in accordance with law in force at the time of the commission of the crime and not contrary to the provisions of the present Covenant.” (emphasis added). The British nurses, in seeking to rely on the Crimes (Torture) Act, would argue that, since the criminal trial in Saudi Arabia did not comply with a number of the procedural guarantees in article 14 of the ICCPR, the imposition of the death penalty did not fall within the “lawful sanctions” exception in the Act. It could thus constitute an “act of torture” for the purposes of the Act. The UN Human Rights Committee and the European Court of Human Rights have not made any statements which indicate that they regard the use of the death penalty as such as “torture.” That being said, they have agreed that in some instances the imposition of a lawful sanction of this nature amounts to “cruel, inhuman or degrading” punishment contrary to article 7 of the ICCPR or article 3 of the European Convention on Human Rights. As already indicated above, this latter possibility would not assist the British nurses in their case because the Crimes (Torture) Act does not extend to conduct that does not rise to the level of torture. However, the line between torture and cruel, inhuman or degrading conduct is a matter of degree and drawing that line inevitably forces one to focus on the tension between universal content and accepted local norms and practices. According to the UN Human Rights Committee and the European Court, there are various factors to be considered in determining whether the death penalty constitutes cruel, inhuman or degrading punishment, including “the manner in which . . . (the death penalty) is imposed or executed”.70 In Kindler v. Canada, the UN Committee said that “[i]n determining whether, in a particular case, the imposition of capital punishment could constitute a violation of article 7, the Committee will have regard to the relevant personal factors regarding the author, the specific conditions of detention on death row, and whether the proposed method of execution is particularly abhorrent.”71 In a subsequent case, Ng v. Canada, the Committee stated that capital punishment will violate article 7 unless the execution of the sentence is “carried out in such a way as to cause the least possible physical and mental suffering”.72 It decided here that execution by gas asphyxiation does not meet this requirement,73 and that it 70 Soering v. United Kingdom and Germany (1989) 11 EHRR 439, Series A, Vol. 161 (European Court of Human Rights, 7 July 1989) (“Soering”), paragraph 100. 71 Kindler v. Canada (No. 470/1991), UN Doc. CCPR/C/48/D/470/1991 (1993) (UN Human Rights Committee, 11 November 1993), paragraph 15.3. See also Soering, ibid., where the age and mental state of the defendant were considered to be important factors: see W A Schabas, supra n. 54 at 234–5. 72 Chitat Ng v. Canada (No 469/1991), UN Doc. CCPR/C/49/D/469/1991 (1994) (UN Human Rights Committee, 7 January 1994) (“Ng”), paragraph 16.2. See also the passage to the same effect in the Human Rights Committee’s General Comment 20(44), UN Doc. CCPR/C/21/Rev/1/Add 3, para. 6, as set out in W A Schabas, supra n. 54 at 136. 73 The Committee accepted that asphyxiation by cyanide gas may take more than ten minutes: Ng, ibid., paragraph 16.3. As Schabas remarks, this comment suggests that the Committee “adopts the criterion of instantaneity”: W A Schabas, supra n. 54 at 137. Two of the dissenters stated that they would view death by stoning as contrary to article 7: see W A Schabas, ibid. at 138.

Cultural Challenges 617 therefore constitutes cruel and inhuman treatment contrary to article 7 of the ICCPR. This approach, of course, takes the view that particular methods of implementing the death penalty are inhumane, and therefore abhorrent. It assumes a particular cultural attitude towards the death penalty, an attitude that is far from universally shared. The condemnation of torture at the abstract level, and indeed its qualification as a jus cogens norms, is universal.74 The same might be said of the prohibition of cruel, inhuman or degrading punishment. But the same cannot be said of either the death penalty per se, or of particular methods of implementation.75 Critics of assimilating the death penalty with cruel, inhuman or degrading punishment, let alone with torture, would argue that it is one thing to encourage all nations to ratify the Second Optional Protocol to the ICCPR Aiming at the Abolition of the Death Penalty.76 It is quite another to insist that the prohibitions against torture and against cruel, inhuman or degrading punishment in the CAT and in the ICCPR— treaties ratified by various countries who still permit the death penalty—be interpreted as allowing certain methods of execution, but not others. The diverse attitudes of states (or, rather, their societies) towards sentencing practices was presumably a major reason for ensuring that “lawful sanctions” were not regarded as “torture” under the CAT.77 Some would argue that they should not be regarded as “cruel, inhuman or degrading” either. Abdullahi An-Na’im, for example, argues that “the interpretation and practical application” of the prohibition against cruel, inhuman, or degrading treatment or punishment “in the context of a particular society should be determined 74 With particular respect to torture, this condemnation is deepened by the fact that the prohibition on torture is a rule of jus cogens: see J H Burgers and H Danelius, supra n. 69. 75 This is reflected in the level of ratification of the Second Optional Protocol (as at 1 January 1996, twenty-nine ratifications), when compared to that of the CAT (as at 31 December 1995, ninety-three ratifications): W A Schabas, supra n. 54 at 142, 176; cf. at 221. For a description of various attitudes to the death penalty: see ibid. at 296–7, 306–8. According to Schabas, most Islamic States have retained the right to use the death penalty. Saudi Arabia was one of the Islamic countries who spoke against, and voted against, the adoption of the Second Optional Protocol to the ICCPR Aiming at the Abolition of the Death Penalty: W A Schabas, ibid. at 175. (For an argument that it is politics rather than religion which prevents many Islamic States from abolishing the death penalty, see H Esmaeili, “Capital Punishment in Islamic Countries” (Sept. 1997) 6(3) Hum Rts Defender 7 at 9. The United States has retained the death penalty. Accordingly, it entered a reservation to article 6(5) of the ICCPR: see W A Schabas, ibid. at 314. It is interesting to note that article 2(1) of the European Convention on Human Rights states that the death penalty may only be imposed by a court: see article 2(1); W A Schabas, ibid. at 228. 76 The Second Optional Protocol, supra n. 54 As of 1 January 1996, twenty-nine parties had ratified the Second Optional Protocol, only six more than the twenty-three which had ratified Protocol No. 6 to the Convention for the Protection of Human Rights and Fundamental Freedoms Concerning the Abolition of the Death Penalty (in force: 1 March 1985): W A Schabas, supra n. 54 at 176, 221. However, only three parties had ratified the Additional Protocol to the American Convention on Human Rights to Abolish the Death Penalty (in force: 6 October 1993): W A Schabas, ibid. at 347. 77 Burgers and Danelius suggest that during the preparation of the CAT opinion was divided on whether “lawful sanctions” should constitute an absolute exception to the prohibition against torture: J H Burgers and H Danelius, supra n. 69 at 121–2.

618 B Wells & M Burnett by the moral standards of that society”.78 He concludes that “[n]either internal Islamic reinterpretation nor cross-cultural dialogue” is likely ever to result in the total abolition of Islamic punishments, but such mechanisms may act to restrict their implementation—for example, to cases in which strict procedural requirements have been satisfied.79 Conversely, it seems likely that the UN Human Rights Committee would regard certain conduct as “cruel, inhuman or degrading punishment” or in some cases as “torture” (given that the ICCPR does not have the “lawful sanctions” fetter that CAT contains) whenever certain objective and subjective critieria are satisfied, regardless of the cultural context. This is an approach which seems particularly persuasive when the potential victims are non-Muslim and do not participate in the culture of which the punishment is a part. The two poles of universal content and cultural particularity could merge to some extent in that cultural appropriateness—which is lacking when a person does not share the cultural (including religious) premises that generate the national “norms”— could be posited as a minimal universal criterion for acceptable application of a particular form of punishment. An Australian court, in applying the Crimes (Torture) Act, would be likely to follow the lead of the UN Human Rights Committee. In determining whether a public beheading amounts to an “act of torture” under the Act, the court would consider both the potential physical and mental suffering involved. A public beheading, if performed by a skilful executioner with appropriate instruments, might satisfy the Human Rights Committee’s test of “least possible physical suffering”.80 But the Human Rights Committee’s test also focuses on the degree of mental anguish suffered by the defendant. This is a more subjective question. What is more, the mental anguish element will often mean that it is not the act as such that violates the right but the effects on the person to be killed in the period leading up to the planned execution. At least one of the British nurses had factual grounds for arguing that the prospect of a public beheading was causing her such extreme mental anguish as to amount to an “act of torture” under the Act.81 78

A A An-Na’im, supra n. 8 at 37. Ibid. at 36. 80 Ng, supra n. 72 at paragraph 16.2. Query whether there comes a point when the execution of a death sentence in the absence of the procedural requirements of article 14 of the ICCPR does amount to torture under the Crimes (Torture) Act because of the method chosen to implement the sentence. Would it, for example, be regarded as potential torture if a person was sentenced to death under the laws of an Islamic State which still allowed the victim’s relatives not only to pronounce the sentence, but also to carry it out (invariably crudely and inexpertly)? According to Sandra MacKey, the victim’s family have until relatively recently retained (and sometimes exercised) this right in Saudi Arabia: S MacKey, The Saudis: Inside the Desert Kingdom (Boston, Houghton Mifflin, 1987), as cited in M Lippman et al., supra n. 16 at 91–2, n. 42. 81 At the time that the British nurses’ proceedings were to be heard by the South Australian Supreme Court, one of the nurses had already suffered severe depression because of the prospect of undergoing a public beheading, and had consequently been hospitalized and placed on strong medication to control her depression. 79

Cultural Challenges 619 Nonetheless, it is reasonably likely that an Australian court would decide that the nurses’ suffering was not of a sufficient “intensity” to constitute torture.82 The court might instead be prepared to hold only that public beheading is “cruel, inhuman or degrading”. However, as already noted, such a finding would not assist the British nurses, since there is no Australian legislation which prohibits the extraterritorial imposition of cruel, inhuman or degrading punishment. There would therefore be no extraterritorial criminal offence (committed in Saudi Arabia) to which the “aiding and abetting” provisions of the 1914 Crimes Act could attach. It is worth making one further point in relation to the CAT. If the Saudi nurses case had more clearly involved an act of torture, the British nurses might have considered accusing Australia of breaching its obligations under the CAT. It requires State parties to ensure that any act by a person “which constitutes complicity . . . in torture” shall be punishable as a criminal offence.83 And it imposes an obligation on States parties to prosecute (or extradite) persons alleged to have committed such an offence.84 The nurses could have argued that the Australian government was therefore obliged to initiate a prosecution to prevent demands for the death penalty by the murder victim’s relative. They might have considered seeking an injunction to force Australia to initiate such proceedings against the victim’s brother, calculating that this could prove an indirect way to put pressure on the brother to retract any quesas demand. However, it is unlikely that an Australian court would be prepared to adjudicate upon such an argument. As Australian law on reception of international treaty obligations currently stands, a court would hold that the question of the Australian government’s compliance with the CAT (or with any other international treaty obligation) is non-justiciable by an Australian court.85 As for putting a parallel argument that the government was illegally failing to prosecute a crime prohibited under an Australian statute (the Crimes (Torture) Act), this would not advance the matter further given the traditional deference accorded by the courts to prosecutorial discretion. All that being said, however, there is perhaps a formally valid argument that could knit these two rules of non-involvement of the courts into a judicial power to order that a torture prosecution be launched. A court could note that torture has been criminalised in 82 See Ireland v. United Kingdom (1978) 17 Int’l Legal Materials 680 at 702 (Eur Ct of Human Rights); see also J H Burgers and H Danelius, supra n. 69 at 114–19. 83 CAT, supra n. 6, article 4. 84 Ibid., article 7 (1). 85 “A breach . . . (of Australia’s international obligations) . . . is not a matter justiciable at the suit of a private citizen: Tasmanian Wilderness Society Inc v. Fraser (1982) 153 CLR 270 at 274 per Mason J; Re Ditford, Ex parte Deputy Commissioner of Taxation (NSW) (1988) 83 ALR 265 at 286 per Gummow J; Koowarta v. Bjelke-Peterson (1982) 153 CLR 168 per Gibbs CJ. See also R Balkin, “International Law and Domestic Law”, in S Blay, R Piotrowicz and BM Tsamenyi (eds), Public International Law: An Australian Perspective (Melbourne, Oxford University Press, 1997) at 127–8; and G Lindell, “Judicial Review of International Affairs”, in B R Opeskin and D R Rothwell (eds), International Law and Australian Federalism (Melbourne, Melbourne University Press, 1997), ch. 6, at 161–5, 167, 186–8.

620 B Wells & M Burnett Australia in implementation of a treaty obligation that, in the treaty, comes accompanied by a duty to prosecute (or extradite). By reference to the treaty norm, the statute itself could be interpreted as requiring prosecution and not just permitting it, a limited but important exception to the prosecutorial discretion rule—an exception justified by the principle of statutory interpretation that a statute should be interpreted in conformity with relevant international obligations if at all possible.86

6 POTENTIAL BARRIERS

Thus far we have focused on the successful application for interlocutory injunctions in the Saudi nurses case, and on the possible substantive grounds for obtaining a declaration and injunctions in the amended proceedings. We now turn to consider some of the potential barriers to success in the amended proceedings, barriers which tend to be present in transnational human rights litigation generally. We focus first on the difficulties faced in obtaining one of the main remedies sought: injunctions.

Injunctions to Restrain Breaches of Public Rights The various rules of civil procedure applicable in Australian courts confer a wide jurisdiction on courts to grant declarations and injunctions in appropriate cases.87 However, this wide jurisdiction is by and large premised on the assumption that an injunction lies to enforce private rights, the legal rights of one individual against another individual. In the amended proceedings, however, a prohibitory injunction was sought to prevent an Australian citizen from acting contrary to Australian law, that is, contrary to the common law and statutory prohibitions mentioned earlier. These prohibitions—for example, prohibitions on the imposition of the death penalty, on the infliction of “cruel and unusual punishment”, and on aiding or 86 In Baker v. Canada (Minister of Citizenship and Immigration), Supreme Court of Canada, File No.: 25823, 9 July 1999, the Supreme Court of Canada held that even where a statutory provision appears to contain an unfettered discretion, it can be interpreted as imposing duties to act in accordance with certain principles and values, including the values reflected in international human rights law, particularly those contained in treaties ratified by Canada. A similiar approach was taken by the High Court of Australia in the case of Minister of State for Immigration and Ethnic Affairs v. Teoh (1995) 183 CLR 273. In both Australia and Canada, there is a (rebuttable) presumption of statutory interpretation that statutes are intended not to violate established rules of international law: see D C Pearce and R S Geddes, Statutory Interpretation in Australia 4th ed., (Sydney, Butterworths, 1996), paragraph 5.10; R Sullivan, Dreiger on the Construction of Statutes 3rd ed., (Toronto, Butterworths, 1994) at 330. This is particularly so where the statutory provision purports to implement a provision in a treaty. 87 See, for example, Supreme Court Act, 1935 (SA), s. 29, 30. A final injunction will be granted in aid of a declared right, which need not be a proprietary right: see G Dal Pont and D Chalmers, supra n. 50 at 574.

Cultural Challenges 621 abetting the commission of an offence—can only be characterised as conferring public (rather than private) rights. An Australian court might well have concluded in these circumstances that the British nurses were seeking to enforce against an individual the sort of legal rights that are ordinarily only enforceable against the State. On this view, the nurses would be regarded as not seeking an injunction to enforce private rights. However, even if a court held that the case did not involve the enforcement of private rights, the British nurses could nonetheless have argued that they were legitimately seeking an injunction to enforce public rights and to restrain illegal acts.88 It seems to us that, on this basis, an Australian court would in some circumstances be prepared to grant an injunction against an individual in order to enforce the Abolition Act and the 1914 Crimes Act.89 The general rule here, set out in the leading case of Gouriet, is that it is the role of the Attorney-General, and not that of private individuals, to enforce and protect public rights.90 As a result, the Attorney-General may seek an injunction (on his or her own initiative, or as a relator action on behalf of an individual) to protect a public right.91 However, there are several exceptions to the exclusivity of the AttorneyGeneral’s right. In particular, it seems that an individual may seek an injunction to restrain the commission of a crime or to enforce a public right where the individual has “a special interest in the subject matter of the action”.92 The British nurses would of course have had no difficulty in establishing a special interest in preventing the potential imposition of the death penalty on them. They would therefore have been in a good position to seek an injunction.

“International Comity:” Would the Injunctions Interfere with Saudi Legal Proceedings? Even aside from the general difficulties of enforcing public rights, the British nurses may have faced arguments that international comity precludes the grant of injunctions in these circumstances. International comity “in the legal sense” 88 See H G Hanbury and J E Martin, Modern Equity 15th ed., (London, Sweet & Maxwell Ltd, 1997) at 740–3, 797–8. 89 On this basis an injunction could also be obtained to enforce the common law “judicial power” principles and the Imperial statutes mentioned earlier: see supra text at n. 55. 90 Gouriet v. Union of Post Office Workers [1978] AC 435 (“Gouriet”). 91 See the description of a relator action given by Dixon AJ in A-G v. T S Gill & Son Pty Ltd [1926] VLR 414 at 416, cited with approval in Attorney-General (on the Relation of Daniels and Others) v. Huber (1971) 2 SASR 142 per Bray CJ (diss.) at 170–1; H G Hanbury and J E Martin, supra n. 88 at 797. 92 Wentworth v. Woollahra Municipal Council (1982) 149 CLR 672. See also Australian Conservation Foundation Inc v. Commonwealth (1980) 146 CLR 493 at 526 per Gibbs J; G Lindell, “Judicial Review of International Affairs”, in B R Opeskin and D R Rothwell (eds), supra n. 85 at168 and n. 138; J D Heydon and P L Loughlan, Cases and Materials on Equity and Trusts 5th ed., (Sydney, Butterworths, 1997), at 968–77; and H G Hanbury and J E Martin, supra n. 88 at 741, n. 55 and 797, n. 96 and the cases cited therein.

622 B Wells & M Burnett has been described by the United States Supreme Court (in a passage approved by the Australian High Court) as “neither a matter of absolute obligation, on the one hand, nor of mere courtesy and goodwill, upon the other. But it is the recognition which one nation allows within its territory to the legislative, executive or judicial acts of another nation, having due regard both to international duty and convenience, and to the rights of its own citizens or of other persons who are under the protection of its laws.”93

Clearly the interest of preserving comity would have been an influential factor in deciding the Saudi nurses case. Is it likely that an Australian court would give precedence to the rights of individuals over the principles of sovereign noninterference embedded in the notion of international comity? Perhaps it is likely. Analogies provided by two other areas of the law indicate that, despite references to comity, common law courts are prepared to grant injunctions which impact upon legal proceedings in foreign jurisdictions. The courts have enforced or preserved the legal rights of parties before them by the grant of anti-suit injunctions, and by the grant of Mareva injunctions which restrain the disposition of property in a jurisdiction other than that in which proceedings have been (or will be) initiated. Of these two forms of injunction, we will discuss that which is the most closely analogous: the anti-suit injunction. The anti-suit injunction has some very similar features to the injunction that the British nurses were seeking in order to restrain conduct by the Australian relative of the murdered nurse. Such an injunction is sought to prevent the initiation or continuation of foreign proceedings which are related to local proceedings. The injunction operates in personam—that is, against one of the parties who is subject to the local court’s jurisdiction—and not against the foreign court. However, in CSR v. Cigna the High Court recognised that the grant of an anti-suit injunction nonetheless “interferes with the processes of the foreign court and may well be perceived as a breach of comity by that court”.94 It therefore emphasised that the power to grant these injunctions should be exercised “with caution” and, indeed, declined in that case to issue such an injunction.95 On the other hand, in CSR v. Cigna the main motivation for seeking an anti-suit injunction was to avoid court proceedings in the US under a federal antitrust statute, the Sherman Act, which provided for damages three times the normal amount as the remedy. Clearly, anti-suit injunctions granted in such commercial contexts are not on all fours with the sorts of human rights interests at stake in the Saudi nurses case.

93 Hilton v. Guyot 159 US 113 at 163–4 (1895), approved in CSR Limited v. Cigna Insurance Australia (1997) 189 CLR 345 (“CSR v. Cigna”) at 436 per the majority. The CSR v. Cigna case is closely analogous to the situation at hand as it involved the issue of whether Australian courts should issue an anti-suit injunction to order a party not to pursue proceedings in another country’s courts. 94 CSR v. Cigna, supra n. 93, at 436 per the majority. 95 Ibid.

Cultural Challenges 623 The interests of comity have no doubt restricted the grounds upon which the High Court is prepared to grant anti-suit injunctions. Nonetheless, the court remains willing to grant these injunctions in a variety of circumstances based upon the demands of “the administration of justice” or in order to restrain “unconscionable conduct or the unconscientious exercise of legal rights”.96 As in the case of an anti-suit injunction, the injunction that would be sought by the British nurses would operate in personam. The injunction would be sought against the relative (who is within the local court’s jurisdiction, although not strictly speaking a party to the foreign proceedings) to prevent him from performing a particular procedural step in legal proceedings on foot in Saudi Arabia. On the basis of the Australian anti-suit injunction case law, there is no reason why international comity should prevent the grant of the injunction to enforce the nurses’ legal rights.

Jurisdiction There is, however, another strand to the argument that an Australian court should avoid granting a remedy which would have the effect, if not the intention, of interfering with the judicial process in Saudi Arabia. A court may also consider the issue in the context of its decision of whether to exercise jurisdiction. An Australian court would not have any difficulty in deciding to exercise initial jurisdiction. In common law countries, a court will assume jurisdiction over a matter and a party so long as the initiating proceeding has been served on the party according to the rules of civil procedure of the court.97 This much was evident in King AJ’s approach in the initial hearing in the Saudi nurses case.98 However, the defendant may then put arguments based upon the forum non conveniens doctrine. This doctrine allows a defendant to argue that the forum chosen by the plaintiff is an “inappropriate” one to hear the matter. If the defendant succeeds in establishing this, then the court will stay or dismiss the proceedings before it. The precise requirements for establishing forum non conveniens are not consistent across the common law world. In the United Kingdom, for example, the defendant need only establish that there is a “more appropriate forum” in another country.99 In Australia, however, defendants have a difficult onus: they 96

Ibid. See P Nygh, “The Common Law Approach”, in C McLachlan and P Nygh, Transnational Tort Litigation: Jurisdictional Principles (Oxford, Clarendon Press, 1996). 98 McLauchlan, supra n. 11 at 271. 99 See Spiliada Maritime Corp v. Cansulex Ltd [1987] AC 460 at 477. According to Spiliada, the plaintiff then has the opportunity to show that it would nonetheless be substantially unjust not to allow the case to continue in the forum which the plaintiff has selected. The US position is set out in Piper Aircraft Co v. Reyno 454 US 235 (1981) at 241, as cited in Voth v. Manildra Mills Pty Ltd (1990) 171 CLR 538 (“Voth”) at 561. For a comparison between the Australian, English and US approaches, see P Prince, “Bhopal, Bouganville and Ok Tedi: Why Australia’s Forum Non Conveniens Approach is Better”, (1998) 47 Int & Comp LQ 573 at 574–5. 97

624 B Wells & M Burnett must demonstrate that the Australian court chosen by the plaintiff is a “clearly inappropriate forum”.100 In order to establish this, the defendant must satisfy the court that the proceedings are “oppressive, vexatious or an abuse of process”, and that they should therefore be stayed in order to avoid injustice between the parties.101 Here again we see the juxtaposition of international and domestic law, with concepts from international law influencing the attitude of domestic courts. While English courts have adopted an approach which is less protective of their own jurisdiction because of their “respect for judicial comity between nations”,102 the Australian approach, although described by some as parochial103 and chauvinistic,104 can be justified on pragmatic and even on international comity grounds.105 Most significantly, the Australian approach is structurally disposed to take seriously a plaintiff’s rights claims including the plaintiff’s view that s/he is most likely to achieve justice by bringing proceedings in Australia. That is, it is a view of jurisdiction more geared towards achieving justice between individuals than a view of jurisdiction that is preoccupied with justice between states. It is therefore regarded by some commentators as having the benefit of allowing poor third-world plaintiffs to access against multinationals the remedies and high levels of compensation that would be denied them in their own courts.106 We consider it unlikely that an Australian court would accept an argument of forum non conveniens in the circumstances presented by the Saudi nurses case. The British nurses had effected service in South Australia on a person who was a resident and citizen of South Australia. This was not a case in which a defendant was being asked to defend himself before a foreign court. Secondly, the British nurses were asking the court to apply the law of the forum—South Australian law.107 And obviously (and perhaps dispositively in the context of forum non conveniens doctrine) the specific relief that the British nurses were seeking—the declaration and injunctions—were not available in the foreign forum (Saudi Arabia).108 In order to rebut any argument that the application of the forum non conveniens doctrine should include comity considerations, the British nurses could have emphasised that they were seeking to prevent the actions of an Australian 100 Voth, supra n. 99, at 564 per Mason CJ, Deane, Dawson and Gaudron JJ (“Mason CJ et al.”) (Brennan J concurring as to the nature of the test). 101 Ibid. at 554 per Mason CJ et al. 102 P Prince, supra n. 99 at 576. 103 M Pryles, “The Struggle for Internationalism in Transnational Litigation” (1987) 61 ALJ 434 at 435, as cited in P Prince, supra n. 99 at 573. 104 A Briggs, “Wider Still and Wider: The Bounds of Australian Exorbitant Jurisdiction”, (1989) 2 Lloyd’s Maritime and Commercial LQ 216 at 222, as cited in P Prince, supra n. 99 at 573. 105 See Voth, supra n. 99 at 558–9. 106 See, for example, the discussion of the Bhopal litigation and the more recent Ok Tedi litigation in P Prince, supra n. 99 at 577, 580–1. 107 See Voth, supra n. 99 at 566 per Mason CJ et al. 108 See Voth, supra n. 99 at 558 per Mason CJ et al.

Cultural Challenges 625 citizen in Australia which interfered with their rights under Australian law; they were not seeking to attack the Saudi Arabian justice system. The human rights on which they relied—rights to life and to freedom from torture and cruel punishment—are well-established in international instruments such as the ICCPR. Such international legal pedigree should influence an Australian judge’s interpretation of Australian law (including forum non conveniens law) in line with the above-mentioned presumption that domestic law should be presumed to accord with international law. This suggests that a court should be particularly cautious in denying jurisdiction on a forum non conveniens basis in such a case.109

State Immunity and Act of State Finally, any argument based on the Crimes (Torture) Act in conjunction with the Crimes Act could be met with defences based on the doctrines of state immunity and “act of state”. This is because any decision by a court to prohibit the murder victim’s relatives from requesting the death penalty would be dependent upon the court first determining that Saudi public officials (that is, representatives of the Saudi Arabian State) had committed an act of torture. Would an Australian court consider such a question as non-justiciable by an Australian court? This is one of the most fascinating and contentious issues arising from the Saudi nurses case. There has long been a doctrine of public international law that foreign States are immune from the jurisdiction of the courts of other states.110 In Australia, the Foreign State Immunities Act (FSIA) codifies this principle and sets out various exceptions which include actions in tort involving “the death of, or personal injury to, a person” but only where the act or omission takes place in Australia.111 The Act does not, however, confer specific exceptions to immunity in cases involving acts of torture or crimes against humanity. International law also confers immunity upon public officials who are acting in an official capacity within actual (and probably also “apparent”) State authority.112 Immunity on this basis was recognized, for example, in two decisions of the Ninth Circuit 109 This is perhaps what Justice Michael Kirby (now of the Australian High Court) was suggesting when he wrote, extra-judicially, that an “internationalist approach”—an approach which pays regard, where possible, to international human rights law—should be applied in cases involving “the respect of the laws of other fora and the principles of forum non conveniens”: M Kirby, “The Role of the Judge in Advancing Human Rights by Reference to International Human Rights Law” (1988) 62 ALJ 514 at 529. 110 See, for example, I Brownlie, Principles of Public International Law 4th ed., (Oxford, Clarendon Press, 1990) at 322–37; J Brohmer, State Immunity and the Violation of Human Rights (The Hague, Martinus Nijhoff Publishers, 1997), ch. 1. 111 Foreign States Immunities Act, 1985 (Cth) (“FSIA”), ss 9, 13; I Shearer, “Jurisdiction”, in S Blay, R Piotrowicz and BM Tsamenyi (eds), supra n. 85 at 186–7. See J Brohmer, supra n. 110 at 96–8. 112 See I Brownlie, supra n. 110 at 450.

626 B Wells & M Burnett Court of Appeal in the US, but the immunity may be overcome when a domestic court is prepared to characterise the relevant conduct—for example, torture—as involving acts which are not official acts within the scope of the official’s authority.113 Since Saudi Arabia would not itself be a party to proceedings before the Australian courts, it could not be contended that the FSIA and international law prevent an Australian court from exercising jurisdiction in relation to a claim that Saudi Arabian public officials (acting in an official capacity and within their authority) would commit acts of torture by carrying out a public beheading on the British nurses. Such a contention would defeat the whole purpose of the Crimes (Torture) Act and the CAT, which is to prevent officially sanctioned torture—that is, torture committed or sanctioned by the State. The only possible immunity claim would arise if the murder victim’s brother argued that he was acting in an official capacity and thus, for this purpose, was acting as part of the Saudi Arabian state. We consider it unlikely that the Australian courts would accept such an argument.114 However, we shall assume, arguendo, the applicability of the Australian FSIA to the brother qua state agent. Even on that assumption, his invocation of State immunity should still fail. The Crimes (Torture) Act criminalises acts of torture which occur overseas where they are carried out by persons who are “public official(s)” or “acting in an official capacity”, or are “acting at the instigation, or with the consent or acquiescence” of such a person.115 And it provides that Australian authorities may prosecute such acts in Australian courts (or extradite non-Australian citizens) when the alleged perpetrator is an Australian citizen or present in Australia.116 As a result, it could be argued that the Crimes (Torture) Act overrides or operates as an exception to certain parts of the State immunity principle. This was very much the approach taken by the House of Lords in the recent Pinochet decision, where a majority of the Court decided that liability for international crimes such as torture could not be avoided by reliance on another aspect of state immunity.117 Consequently there are now strong prospects of success with such an argument. 113 Chuidian v. Philippine National Bank, 912 F 2d 1095 (9th Cir 1990), and In re Estate of Ferdinand Marcos, 25 F 3d 1467 (9th Cir 1994), as discussed in H J Steiner and P Alston, supra n. 66 at 793. 114 The Australian FSIA only extends foreign state immunity to a foreign State, “a province, state, self-governing territory or other political subdivision . . . of a foreign State”, “the head of a foreign State, or of a political subdivision of a foreign State, in his or her public capacity” and “the executive government or part of the executive government of a foreign State or of a political subdivision of a foreign State or subdivision:” ss 3 and 9. 115 Crimes (Torture) Act, 1988 (Cth), s. 6. 116 Ibid., s. 7. 117 In R v. Bow Street Metropolitan Stipendiary Magistrate; Ex parte Pinochet Ugarte (No. 3) (1999) 2 WLR 827 (“Pinochet (No. 3)”), a majority of the House of Lords decided that the express words of the CAT displace the “former head of state immunity” principle set out in statute and in customary international law: at 847–8 per Browne-Wilkinson LJ, at 899, per Hutton LJ, at 903–4, per Saville LJ, at 913–14, per Millett LJ and at 924 per Phillips LJ; cf. at 854–63, 865 per Goff LJ, and at 881–5 per Hope LJ. Several of the judges held that previous authorities on state immunity in

Cultural Challenges 627 As we concluded above, it is unlikely that the state immunity doctrine would be raised at all, since neither Saudi Arabia nor its officials would be before Australian courts. But another doctrine closely related to sovereign immunity would be more relevant: the “act of state” doctrine. According to this common law doctrine (one of private international law), Australian courts may generally not inquire into “the legality or validity of public acts performed by a foreign recognised state within the latter’s own territory”.118 On this basis, it might be argued that an Australian court was precluded from adjudicating upon the legality of an act—a public beheading—which was clearly authorized by Saudi Arabian law. In the US, some courts have held that acts of torture will be shielded by the “act of state” doctrine when they reflect government policy (for example, if they have been ratified by government or have been authorised by legislation).119 However, the indications from the Pinochet case are that a majority of the House of Lords would hold that the common law principle was, at the very least, not intended to apply in cases in which statutory provisions (such as those in the Crimes (Torture) Act) evince a clear intention that certain issues involving state action are to be justiciable.120 As a result, Australian civil proceedings were not directly applicable to the question of immunity from criminal prosecution: see at 890–2, 901 per Hutton LJ, at 914 per Millett LJ and at 916 per Phillips LJ. The suggestion seemed to be that immunity may be more available in the civil context. If this view was adopted in future, the potential for transnational torture claims in United Kingdom courts could be seriously affected. 118 J Karkar, “Nationalisation in the Conflict of Laws” (unpublished thesis submitted for the LLM degree at the University of Melbourne, 1976) at 241, as quoted in E I Sykes and M C Pryles, Australian Private International Law 3rd ed., (Sydney, The Law Book Company Limited, 1991) at 297. Australian courts have accepted and applied the doctrine: see B Wells and M Burnett, supra n. 57 at n. 123. 119 See Forti v. Suarez-Mason, 672 F Supp 1531 (1987) (US District Court, Northern District, California) per Jensen J, as quoted in H J Steiner and P Alston, supra n. 66 at 799. But compare the opinion expressed in the Senate Report on the Torture Victim Protection Act (S Rep 102–249, Committee of the Judiciary, 102nd Cong, 1st Sess, 1991): “[S]ince the ‘act of state’ doctrine applies only to ‘public’ acts, and no state commits torture as a matter of public policy, this doctrine cannot shield former officials from liability under this legislation”: H J Steiner and P Alston, supra n. 66 at 804. This may be an overly sanguine view when one considers that a government may officially adopt a policy to treat persons in a certain way but not regard the methods used as cruel, inhuman or degrading much less as torture, while courts in another state may have a different interpretation of whether these methods cross the normative thresholds in question. One example would be Israeli interrogation techniques in national security cases: see Reichman and Khana, “Israel and the Recognition of Torture: Domestic and International Aspects”, chapter 24 of this volume. Another example is, of course, precisely the question at hand: Saudi Arabia has adopted an official policy of beheading but, in so doing, has not (in its view) adopted an official policy of inhuman treatment or torture. 120 This was the approach taken by a majority of the House of Lords in the original decision (subsequently annulled) in the Pinochet case: R v. Bow Street Metropolitan Stipendiary Magistrate; Ex parte Pinochet Ugarte (1998) 3 WLR 1456 per Lord Nicholls (with whom Lord Hoffmann agreed) and per Lord Steyn. And it would seem to be supported by the majority of the reconstituted House of Lords in Pinochet (No. 3), supra n. 117, who regarded the express words of the CAT as overriding the “former head of state immunity” principle set out in statute and in customary international law: see esp. at 904 per Saville LJ and at 906 per Millett LJ. See also R Higgins, Problems and Process: International Law and How We Use It (Oxford, Clarendon Press, 1994), . 217–18 at nn. 42–4.

628 B Wells & M Burnett courts should now be less likely to regard the “act of state” doctrine as shielding from their consideration those extraterritorial acts of torture which are the intended target of the Crimes (Torture) Act.

7 SETTLEMENT

Notwithstanding the barriers which may be placed in the way of transnational human rights litigation, the Saudi nurses case demonstrates that instituting legal proceedings and, in particular, seeking equitable remedies, may make a difference to the outcome of a case even if the case does not proceed to judgment on the merits. While the British nurses were detained in Saudi Arabia, the British press applied considerable pressure for their release. The British government was working towards such an end. Valuable trade contracts between Britain and Saudi Arabia were potentially at risk. However, initially, there was no useful communication between the murder victim’s brother and the British nurses’ lawyers. This dialogue only began as a result of the commencement of legal proceedings in the South Australian Supreme Court. The litigation created a new framework which crystallized the legal issues at stake. A compromise solution became possible. Ultimately, the brother agreed to accept compensation in lieu of quesas, and donated most of the money to a public hospital.

8 CONCLUSION

In Australian law, international comity is usually regarded as a principle which constrains Australian courts in adjudicating on matters which interfere with proceedings in foreign courts, and one which encourages our courts to recognize foreign laws and to decline jurisdiction where appropriate. It is less clear, however, how international comity as a principle is to work in the context of attempts to enforce international human rights in transnational litigation. Inevitably, there is a tension which domestic courts must explore. And clearly, for many from non-Western cultures, comity is also intrinsically involved in the process of interpreting and applying human rights norms in a way that is culturally sensitive. Abdullahi An-Na’im considers it possible and worthwhile for the proponents of international human rights to engage in dialogue with Muslims about the imposition of Islamic penalties. He believes that there are potentially areas of common ground.121 The Saudi nurses case provides an example of a case in 121 For example, both Western criminal justice systems and “Islamic religious texts emphasise extreme caution in inflicting any criminal punishment”: if there is any doubt about the guilt of the accused, the punishment should not be imposed: A A An-Na’im, supra n. 8 at 36.

Cultural Challenges 629 which there was potential for an Australian court to deal with different cultural attitudes to rights through the application of a variety of legal principles. In fact, however, there was little or no cross-cultural dialogue. The Saudi Arabian court regarded a non-Muslim Australian as qualified to enforce a right under Islamic law, despite the fact that he did not share the religion or culture from which the right sprang. And the Australian court in which the British nurses initiated legal proceedings did not have an opportunity to take a stance on the cultural challenge before it, since the case settled before trial. If the case had proceeded to trial, an Australian court might have clarified its position on many of the potential barriers involved in transnational human rights litigation generally—jurisdiction and extraterritoriality, international comity more generally, and the interpretation of universal human rights in different cultural contexts. Nonetheless, the decision to initiate transnational human rights litigation produced its own dialogue—a dialogue with the murder victim’s relative. Interlocutory injunctions had the effect of changing the dynamics of the conflict. The objective of the litigation was achieved: the imposition of the death penalty was avoided. Ironically, or perhaps fittingly, it was avoided in a way which saw the victim’s brother come to a decision that combined Islam’s preference for forgiveness (consistent with diya, or analogous compensatory payments) with a principle of charity that may well have owed much to the brother’s own moral convictions.

24

Israel and the Recognition of Torture: Domestic and International Aspects1 AMNON REICHMAN 2 and TSVI KAHANA 3

1 INTRODUCTION H I S C H A P T E R A D D R E S S E S a recent decision by the Israeli Supreme Court, in which it was held that the use of physical force by the General Security Service (GSS) in the course of its interrogations was illegal under Israeli law.4 The Court ruled that, in the absence of explicit enabling legislation, the GSS investigators’ general interrogation powers do not encompass the power to resort to these methods. At the same time, the Court asserted that in “ticking bomb” situations (as defined by the Court),5 the defence of necessity was available to an investigator who employed such methods, provided that no other means were available to save human lives and that the pressure resorted to was no more than was necessary to extract the information needed to diffuse the bomb. As a consequence of this ruling, future applications for injunctions

T

1 This chapter originated in papers presented by the authors in the constitutional roundtable at the Faculty of Law of the University of Toronto. The roundtable session focused on the limits of defensive democracy in the aftermath of the new Israeli Supreme Court ruling. The authors wish to thank the participants of the roundtable for their comments and remarks and Idan Erez for his editorial assistance. The responsibility over the content of this article lies, of course, with the authors. Section 2 was drafted by co-author Kahana; sections 3 to 5 were drafted by co-author Reichman and the arguments contained therein are further elaborated in A Reichman, “When We Sit To Judge We Are Being Judged: The Israeli GSS case, Ex Parte Pinochet and Domestic/Global Deliberation”; forthcoming in 9 Cardozo J. Int’l & Comp. L. (2001). 2 LL.B (Hebrew University); LL.M (University of California at Berkeley); S.J.D (University of Toronto). 3 LL.B (Tel Aviv University); LL.M (Tel Aviv University); S.J.D (University of Toronto). 4 H.C. 5100/94 Public Committee against Torture in Israel et al. v. The State of Israel and the General Security Service [1999] [as yet unpublished] [hereinafter GSS case], para. 40. An English translation of the judgment can be found at: http://www.court.gov.il/mishpat/html/en/system /index.html. Our impression is that the translation, while a valuable contribution to the world of comparative law, raises some difficulties. Perhaps the most glaring is found in paragraph 17, where the Court, in the official Hebrew version, acknowledges that there have been occasions where it chose not to issue an injunction against the use of force by GSS investigators in interrogations. The English version omits this sentence, giving the impression that at no time in the past has the Court, even if by omission, sanctioned the use of force. 5 See in paras. 15, 33 and 34.

632 A Reichman & T Kahana against the use of force by GSS investigators will succeed in Israeli Courts in all circumstances. However, should an individual GSS investigator inflict pain and suffering (presumably before such an injunction, so as not to violate a Court order), the investigator will be sheltered from criminal liability under the defence of necessity if the conditions appropriate to the defence were present. The Court stated that, should the Knesset (the Israeli Legislature) wish to change this legal picture, it would have to address the issue through explicit legislation. The GSS judgment is clearly an administrative law ruling. This emerges from the presentation of the question before the Court, in the analysis of its answer, and in the phrasing of its conclusion. The question in the case was presented in terms of the “legality”6 of “directives”7 and “practices”,8 rather than on the “constitutionality” of “legislation”. The Court’s analysis focused on the power of a governmental organ rather than on constitutional rights and their limits. The conclusion explicitly stated that if and when the legislature decides to act on the matter and to enact relevant legislation, then the legislation will have to be within the limits of Israel’s Basic Law: Human Dignity and Liberty (BLH).9 It is only at that point that this issue would become a constitutional issue. However, it is possible to read the ruling as signalling that such an enactment would be constitutionally futile because any legislative attempt to legitimise the use of physical force during GSS interrogations would fail constitutional muster. Sections 2 and 3 of this chapter focus on the domestic arena. After describing the decision, the analysis examines the Court’s suggestion that the legislature could, after appropriate consideration, enact legislation explicitly to authorize GSS interrogators to use physical force. The respective sections suggest two competing readings of this judicial advice. According to the first, the reading favoured by co-author Kahana, there are sufficient indications in the decision that even legislative acts explicitly authorising the use of physical force will be unconstitutional and will be struck down by the Court. The second interpretation, that favoured by co-author Reichman, contends that such enabling legislation is possible as long as it is carefully tailored to the matter of preventing terrorist attacks. It should be noted that the authors do not express any position regarding the desirability of such legislation, and restrict their analysis to the legal holdings of the GSS case. For purposes of setting up the terms of ongoing post-GSS interpretive debate in Israel, the authors assume both of these two competing interpretations are reasonable readings and that the legal process in Israel has yet to clarify which is the preferred reading. It may be that the evolving nature of Israel’s judicial 6

GSS case, supra n. 4, at para. 1. Ibid., at the introductory paragraph of the case (not numbered). 8 Ibid. 9 Hok Yesod: Kevod HaAdam VeHeiruto (Basic Law: Human Dignity and Liberty), Sefer HaHukim (S.H.) 150 (1992); GSS case, supra n. 4, at para. 39. 7

Israel and the Recognition of Torture 633 scrutiny of human rights violations in interrogations should be factored into the jurisdictional and choice of law analysis of foreign courts should one or more of them be faced with a tort action brought abroad against GSS officials for torture. Similarly, Israeli courts may well have to address which is the better reading of the GSS case should a foreign court award damages to a plaintiff and should that plaintiff then seek to have that award enforced in Israel. Thus, this chapter will move from the domestic arena to the international arena in Sections 4 and 5. Section 4 reflects on how the GSS judgment may be received by foreign and international institutions in light of recent developments in public international law, specifically international criminal law. Section 5 then considers the possible effect of diminished state immunity and expanded universal jurisdiction in domestic courts on the emerging global constitution and, in turn, on questions of private international law.

2 READING ONE : THE ISSUE HAS IN EFFECT BEEN DECIDED

The first plausible reading of the case that the authors suggest is that legislation that will make it legal for the GSS to use force during interrogations will be struck down since, by the Court’s reasoning in GSS, it will necessarily stand in contradiction to the BLH.10 According to the BLH, constitutional scrutiny involves a two-stage process: the rights stage and the limits stage.11 In the rights stage, the Court must ask whether the impugned measures infringe on a protected right. If such infringement is found, the Court must then decide whether the infringement is valid according to the limitation clause in section 8. That clause saves those infringements that are contained “in a statute befitting the values of the State of Israel, aimed at a proper purpose, and to an extent no greater than required, or by virtue of explicit authorisation in such a statute”. In the United Mizrahi Bank decision, the Supreme Court ruled that an act is considered to infringe on a right “to an extent no greater than required” if that act satisfies three requirements. First, the act must be rationally connected to the purpose of the legislation. Second, the act must pursue its goal while minimally impairing the right. Third, the negative consequences of the act must be proportionate to the purpose that the act achieves.12

10 The BLH does not include a supremacy clause according to which contradicting legislation is to be struck down. The Supreme Court of Israel, however, decided that despite this, the BLH is supreme. See C.A. 6821/93 United Mizrahi Bank Ltd., et al. v. Migdal Cooperative Village, et. al., 49(4) PD 221 (1995). 11 In this respect, the BLH follows the model created by section 1 of the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11. 12 See United Mizrahi Bank, supra n. 10, at 436–7. The Court thus adopted the test developed by the Supreme Court of Canada in R v. Oakes, [1986]1 S.C.R. 103 at 138.

634 A Reichman & T Kahana Suppose now that the Knesset enacts a statute that authorizes GSS interrogators to use force during interrogations (a hypothetical Use of Force Act).13 This Act would undoubtedly be found to undermine the rights to dignity and to liberty of those being interrogated. Both of these rights are protected in sections 2, 4 and 5 of the BLH and, symbolically, both also appear in its very title. In the next stage of inquiry, the Court would look to section 8’s limitation clause. Section 8’s first requirement is that the infringement of the protected rights be contained within a statute. For the purposes of this analysis, the Use of Force Act is presumed to be a statute. The second and third requirements are that this statute befits “the values of the State of Israel” and that it be passed “for a proper purpose”. Again, it is assumed that fighting terrorism indeed befits the values of the State of Israel and that this represents a proper purpose. On these assumptions, the Use of Force Act would therefore likely survive the first three requirements of the limitation clause. From the language of the Court’s decision in GSS, however, the Use of Force Act will fail to meet the fourth requirement of the limitation clause. Recall that this condition stipulates that the act infringe on the interrogation subjects’ rights only “to an extent no greater than required”. When the Court examined the impugned physical means of interrogation in the GSS case from an administrative law perspective, it asked whether these means fell within the powers of the GSS’s interrogators. The central principle of Israeli administrative law is the principle of reasonableness. Correspondingly, the Court developed what it called “the rules of ‘reasonable interrogation’ ”.14 While discussing the principle of reasonableness, however, the Court explicitly discussed the issue of proportionality. The Court stated: “[A] democratic society, desirous of liberty seeks to fight crime and to that end is prepared to accept that an interrogation may infringe upon the human dignity and liberty of a suspect provided it is done for a proper purpose and that the harm does not exceed that which is necessary.”15

The Court’s conclusion is anchored in administrative law and not in constitutional law. That is, the Court refers to the powers of the GSS interrogators and 13 A private member bill that authorises the GSS to use force during interrogations was indeed tabled in the Israeli legislatures and was supported by 47 (out of 120) Knesset members. See G. Allon, “47 MKs back “special tactics” for Shin Bet”, Haaretz (24 October 1999). The Israeli Justice Minister did not support this bill (ibid.) Instead, the Israeli Prime Minister appointed a team in the Ministry of Justice to examine the legal regime regarding GSS interrogation following the GSS decision, and to suggest possible responses to the decision. The team’s final report was not open to the public, but the media reported that the team failed to arrive at unanimous recommendations and was split in its approach to the matter. One approach supported legislation that would enable the use of “special means” in cases where “the agency is convinced that detainee holds information likely to prevent a clear and present danger posed to the state”; a second approach was to allow the GSS “to use certain methods only in special and exceptional circumstances and following strict legal guidelines”; a third approach opposed any legislation legitimising the use of force in GSS interrogation. See G. Alon, “Justice Ministry panel split on legalising Shin Bet ‘torture’ ”, Haaretz (10 December 1999). 14 GSS case, supra n. 4 at paras. 21–3. 15 Ibid. at para. 22.

Israel and the Recognition of Torture 635 not to the constitutionality of any legislation. The Court’s language, however, is obviously based on the text of the BLH. The Court mentioned “human dignity and liberty”, which echoes both the name of this constitutional statute, as well as the specific rights for human dignity and liberty identified in sections 2, 3 and 5. In addition, the Court made reference to the “proper purpose” of the legislation, and to the infringements not “exceeding that which is necessary”; these are terms that appear in the section 8 limitation clause. Similarly, when the Court moved from discussing the general standard for legality to discussing whether the specific physical means of interrogation met this standard, it followed the constitutional proportionality test set out in the United Mizrahi Bank case, notably the standards of rational connection and minimal impairment. Thus, when the Court disqualified the “frog crouch” because it “does not serve any purpose inherent to an investigation”,16 it was applying the rational connection test. When the Court found shaking to be illegal because it “surpasses that which is necessary”,17 the Court was addressing the standard of minimal impairment.18 When the Court held that normal handcuffing techniques sufficiently ensure the interrogator’s security without requiring recourse to handcuffing the suspect’s hands behind his back or to restraining the suspect with small handcuffs, the Court is utilizing the concept of alternative means within a minimal impairment inquiry. When the Court found deliberate sleep deprivation to “harm the rights and dignity of the suspect in a manner surpassing that which is required”,19 the Court is applying the concept of necessity, again within the minimal impairment inquiry. In other words, although working under the administrative law’s disguise of “reasonableness”, the Court was in reality mounting an inquiry into rational connection and minimal impairment. The Court, then, effectively ruled that the impugned physical techniques either were not rationally connected to the purpose of the interrogation or they were not the least drastic means with which to achieve their goal. It should be noted that the Court did not arrive at this conclusion due to the government’s failure to submit evidence regarding these means; rather, the Court ruled positively that these means did not withstand a scrutiny according to a reasonableness standard that incorporates constitutional norms. It follows from this that a statute legalising the use of these or similar means in interrogations will be presumptively unconstitutional. Nevertheless, the Court’s address to the legislature implies that such legislation is possible. The Court said that, if it will “nonetheless be decided that it is appropriate for Israel, in light of its security difficulties to sanction physical means in interrogations (and the scope of these means which deviate from the ordinary investigation rules), this is an issue that must be decided by the 16 17 18 19

GSS case, supra n. 4 at para. 25. Ibid. at para. 24. Ibid. at para. 26 Ibid. at para. 31.

636 A Reichman & T Kahana legislative branch which represents the people.”20 The Court made note of the fact that “various considerations must be weighed”,21 and that it should be the legislature’s objective to ensure that such legislation falls within the protection of the limitation clause. In other words, the Court pointed out the topics for discussion, identified the factors to be weighed, and allowed for the possibility that physical means of interrogation could escape a finding of unconstitutionality under the limitation clause. Indeed, the Court wrote that it would not “take any stand on this matter at this time”,22 thereby implying that it would be willing to consider the question at a future date. Can the Court’s advice be reconciled with the Court’s reasoning? How could the impugned means of interrogation ever be adjudged to be rationally connected to their goal and minimally impairing of any impugned right simply by virtue of being adopted by the legislature? One word contained within the Court’s discussion can arguably resolve this difficulty: this is the word “ordinary”. The Court implied that the physical means of interrogation were “unreasonable” in terms of “ordinary investigation rules”. The legislature might enact these means by deciding that ordinary investigation rules are simply not sufficient for Israel. Even though physical means of interrogation do not survive the administrative law standard of “reasonableness” for ordinary interrogation rules, they still might survive the external constitutional law standard of proper purpose and minimal impairment should the legislature ground the statute in extraordinary necessity. This dualism lines up with the idea of internal and external standards of scrutiny which is a familiar one in the rights-protection world. For example, it can be found in the Canadian framework of internal and external limitation mechanisms provided by sections 7 and 1 of the Canadian Charter of Rights and Freedoms.23 Section 7 provides that the rights to life, liberty and security of the person can be deprived “in accordance with principles of fundamental justice”, and section 1 provides that all Charter rights are subject to “reasonable limits”.24 Reading these two sections together suggests that it is possible, at least theoretically, to have some limits on the right to life, liberty, and security that might not be in accordance with the principles of fundamental justice but that would nevertheless still be considered “reasonable limits”. Such limits would not survive section 7’s internal limitation clause, but they would survive section 1’s external limitation clause.25 A similar structure can inform analysis of the situation at hand. 20

GSS case, supra n. 4 at para. 39. Ibid. 22 Ibid. (emphasis added). 23 Supra, n. 11. 24 The full text states: “The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.” 25 In a very recent case, Canada’s Federal Court of Appeal issued a judgment that followed this exact pattern of a violation of section 7 but a justification under section 1: see Suresh v. Canada, [1999] F.C.J. No. 865. Significantly, the case involved deportation of a person to a country where there was a substantial risk he would face torture. 21

Israel and the Recognition of Torture 637 But the analogy to section 7 of the Canadian Charter illustrates why this structure cannot explain the contradiction between the Court’s granting of permission to the legislature to consider constitutionally permissible physical means of interrogation and the Court’s finding that these means were not reasonable according to administrative law. The reason why section 1 can still protect a measure that section 7’s own limitation clause does not protect is that the formulas prescribed by these clauses are not identical; specifically, a measure could fail to be “in accordance with principles of fundamental justice” (section 7), but could still be “demonstrably justified in a free and democratic society” (section 1). Imagine now, that at some point a Court rules that a certain measure is found not to have been done “in accordance with principles of fundamental justice” because this measure was not “demonstrably justified in a free and democratic society”. In such a case, it would be impossible to hold that this measure could be saved by section 1, since it was in violating the substance of section 1 that it violated section 7. This is exactly the structure of the Court’s analysis here. As we have indicated previously, the very reason why the Court found the impugned means of interrogation to be “unreasonable” in terms of administrative law was that they were not rationally connected to the goal of the interrogation or that they were not the least drastic means available. Once these were declared to be the reasons, it became impossible to hold that such means could ever be saved by section 8, the Israeli limitation clause. These means had already been judged to have failed the clause’s requirements.26 Legally speaking, this reading of the case suggests that there is a tension between the Court’s address to the legislature and the Court’s normative finding. Politically speaking, it suggests that the Court put itself in a no-win situation that would arise should the legislation that it suggested the legislature consider in order to legitimise physical means of interrogation ever be challenged in the Courts. In such a situation, the Court would likely be harshly criticized regardless of its ruling. If it upheld the legislation, it would be admitting that it had allowed the legislature, simply by virtue of its being the legislature, to violate the limitation clause. This is the worst form of deference. If, on the 26 In order to avoid confusion by the reader who is not familiar with the Israeli context, it should be emphasised that the second part of section 8’s limitation does not empower the Knesset to override the BLH. Section 8 prescribes that rights infringements are valid only if they are contained “in a statute befitting the values of the State of Israel, aimed at a proper purpose, and to an extent no greater than required, or by virtue of explicit authorisation in such a statute.” In other words, an explicit authorisation by the legislature to impinge on rights must conform to this substantive test. The purpose of the second part of section 8 (“or by virtue of explicit authorisation in such a statute”) is not to provide for an exemption to the standards of justification prescribed in the first part of section 8, but to provide that a justified limit on a right can be also be exercised by executive regulation, and not only by direct provision in statutes. Comparing again with the Canadian limitation clause in Section 1 of the Charter (supra n. 11), what the Israeli provision stipulates is that limits on rights do not have to be written in a statute but merely “prescribed” by a statute. The idea behind both the Canadian and the Israeli authorisation for the non-legislative rights infringement is that, in the modern state, statutes cannot cover all areas of legislation and there is a need for administrative regulation. But, once limitation by regulation is expressly permitted, the idea is obviously not to then exempt this regulation from the burden of the limitation clause’s standards of justification.

638 A Reichman & T Kahana other hand, it struck down the legislation, it would be accused by the legislature of sending it to work on legislation that it knew or ought to have known was doomed to be struck down.

3 READING TWO : ALTHOUGH THE THRESHOLD IS HIGH , THE DOOR IS STILL OPEN

According to the reading advocated below, the Israeli parliament has not been precluded by the judicial analysis in the GSS case from empowering the GSS or its investigators to apply force, when the recourse to force is the only effective measure available to prevent loss of human lives in a terrorist attack. This reading attempts to reconcile the tension between the Court’s analysis and its explicit statement that the legislature may seek to empower the GSS to apply force. The legal starting point for the Court’s analysis is the lack of any specific enabling legislation under which the GSS operates. As a result, the GSS, as an organisation, is not authorised to exercise state power or to deprive individuals of their liberty, a deprivation inherent in any non-voluntary interrogation. The Court found that the only legal basis for the exercise of state power by specific GSS investigators to interrogate a person can be found in an ordinance, dating from the British mandate, which authorises “a police officer or . . . any other officers generally or specially authorised by the [Justice Minister] to hold inquiries into the commission of offences”.27 The Justice Ministers throughout Israel’s history have indeed named specific GSS officers to conduct interrogations with respect to crimes committed under the Penal Code as well as under specific legislation put in place to combat terrorism. Thus, “[b]y virtue of this authorisation, [named] GSS investigators are tantamount to police officers in the eyes of the law”28 for the purposes of conducting interrogations. In other words, GSS investigators are not empowered to combat terrorism with arrows not in the quiver available to police forces in their investigations of any other crime.29 Does the general authorisation to conduct investigations, shared by the police and the named GSS investigators, include the legal power to use physical pressure methods or otherwise to inflict pain? The administrative law answer is in the negative, since according to administrative law principles any violation of human rights, such as the one incurred by harmed suspects, must be based upon 27 Article 2(1) of the Criminal Procedure Statute [Testimony] as amended in 1944, as cited in GSS case, supra n. 4, at para. 20. 28 Article 2(1) of the Criminal Procedure Statute, supra n. 27. 29 “The power to interrogate given to the GSS investigator by law is the same interrogation power that the law bestows upon the ordinary police force investigator. It appears that the restrictions applicable to the police investigations are equally applicable to GSS investigations. There is no statutory instruction endowing a GSS investigator with special interrogating powers that are either different or more serious than those given to the police investigator.” GSS case, supra n. 4, at para. 32.

Israel and the Recognition of Torture 639 an explicit authorisation; otherwise, the enabling statute is interpreted so as not to allow such state action.30 Since the statute that empowers the Justice Minister to authorise GSS officers to conduct interrogations does not mention the use of force or other means that inflict physical pain or cause degrading psychological harm, such use is interpreted as being unauthorised, and is therefore ultra vires. The state argued, however, that some methods that inflict physical or emotional harm are nonetheless ancillary to the conduct of ordinary investigations.31 Accordingly, no specific authorisation by law was required since, in order to carry out its powers under the statute to interrogate, interrogators had to be able, legally, to resort to methods that are inherent to the conduct of interrogation as such. More specifically, the state argued that the tying of a suspect’s hand to a chair is justified for the protection of the investigator; that the placing of a sack to cover the suspect’s head is justified so as to prevent eye contact and communication between the suspects; that the playing of powerfully loud music is justified so as to prevent the suspects from verbally communicating with each other; and that the deprivation of sleep is derived from “the needs of the interrogation”, such as the timetable of the investigators and the recognized need for a thorough and uninterrupted investigation. It is with respect to these justifications (or legitimate purposes) that the Court found no rational connection, a failure to meet the least restrictive means requirement and disproportionality.32 The Court did not engage other pairs of 30 This presumption dates back to the landmark decision of Kol Ha’am (H.C. 73/53 “Kol Ha’am” Company Limited v. Minister of the Interior, 7 P.D. 871), and was reinforced by later jurisprudence, such as H.C. 2918/93 The City of Kiryat Gatt v. The State of Israel and others, 37 (5) P.D. 832, referred to by the Court in the GSS case, supra n. 4, at para. 19. 31 It should be noted that the state did not try to justify all the methods used by GSS investigators as inherent to the conduct of an ordinary investigation. For example, the state did not argue that the physical shaking of a suspect is a method that every investigator (in the GSS or the police) is permitted to employ, nor that there is any justification one can put forward for using this method as part of an ordinary investigation. The Court, therefore, had no difficulties in finding that the use of such physical means “surpasses that which is necessary” [para. 24]; however, on the reading of the case advanced here, this finding does not necessarily curtail the use of this method—or others— should the state explicitly empower the GSS to use such methods in tandem with putting forward an alternative justification, such as the need to extract information from a suspect in a “ticking bomb” situation. 32 The Court found that while handcuffing the suspect’s hands is indeed included in the inherent powers of interrogation (and therefore can be resorted to by the GSS interrogators without specific authorisation in the empowering statute), the cuffing used by GSS “is unlike routine cuffing. The suspect is cuffed with his hands tied behind his back. One hand is placed inside the gaps between the chair’s seat and back support, while the other is tied behind him, against the chair’s back support. This is a distorted and unnatural position. The investigators’ safety does not require it.” (para. 26.) Moreover, the Court found no rational connection between the use of particularly small handcuffs and the safety of the interrogators (ibid.). Similarly, the Court accepted that “seating a man is inherent to the investigation”, but stated that “this is not the case when the chair upon which he is seated is a very low one, tilted forward facing the ground, and when he is sitting in this position for long hours. This sort of seating is not encompassed by the general power to interrogate” (para. 27). The Court accepted that means designed to prevent contact and communication between suspects are indeed inherent to the conduct of an interrogation. But since the specific method used by the GSS included covering the suspect’s head with a long, opaque sack that caused the suspect to suffocate, the Court found those methods not rationally connected to the furtherance of the legitimate

640 A Reichman & T Kahana means-justifications that go beyond the ambit of ordinary rules of investigation into ordinary crimes.33 Such examination was not necessary, since the Court found that the GSS investigators have no statutory authority permitting them to exercise any measures beyond those that constitute an ordinary investigation (such as asking a suspect questions, handcuffing a suspect in order to prevent harm to the investigator and keeping the door to the suspect’s cell closed to avoid an escape or eye-contact with other suspects). In particular, the Court did not examine whether the desire to prevent bodily injuries and the loss of human lives by terrorist attacks may justify recourse to the use of force. According to this reading of the case, the Court found that these measures and this justification are not an inherent element of an investigation. Since the official state power invested in the police and the GSS investigators did not cover these means as designed to combat terrorism and save human lives, their use requires specific authorisation in law. The state offered one possible authorisation—the necessity defence—which was dealt with by the Court separately, and will be addressed below. At this point, suffice it to note that, since it was not necessary to its decision, the Court, in its analysis of the “reasonableness” of the methods used by GSS investigators, did not decide whether applying force to pressure a suspect to reveal information crucial to foil a terrorist attack is rationally connected to the prevention of the loss of human lives. Similarly, the Court did not rule on the availability of other less restrictive means capable of achieving this purpose, and it did not engage in a proportionality analysis between the harm inflicted by applying these means and the harm caused by frustrating such legislative purpose. Nor did the Court consider evidence on these points. Thus, these issues remain open; the state and the human rights organisation may be required in the future to bring forth evidence to investigative purpose (para. 28). Given that the suspect is covered for long hours, the Court stressed that other means—less restrictive—are available, such as an eye cover (ibid.). The Court was also prepared to assume that precluding the suspect from hearing other suspects or “voices and sounds that, if heard by the suspect, risk impeding the interrogations’ success” (para. 29) is an inherent part of an interrogation, and hence does not have to be explicitly authorized by the empowering statute. That being said, the Court found that being exposed to powerfully loud music for a long period of time is unreasonable and therefore ultra vires. In the same line of reasoning, the Court found that “a reasonable investigation is likely to cause discomfort; it may result in insufficient sleep”, but “sleep deprivation for a prolonged period, or sleep deprivation at night when this is not necessary to the investigation time-wise, may be deemed a use of an investigation method which surpasses the least restrictive means” (para. 23; see also para. 31) and hence unauthorised in law. The Court added that combining all of these methods is particularly problematic, given the cumulative harm inflicted; the use of each of these methods, and especially their use in combination, is therefore outside the current scope of the general powers of GSS investigators. 33 According to the state, these methods could be used whether the purpose of the interrogation was to uncover past wrongdoing or to prevent future harm, and could be used regardless of the nature, magnitude and imminence of any such future harm. By casting these methods as inherent to an interrogation as such, the state implied that these means could be used by any police-person conducting any kind of interrogation for any kind of offence. This position, as the Court pointed out, contrasts with the rules governing ordinary interrogations, which reject the use of force (GSS case, supra n. 4, at para. 23).

Israel and the Recognition of Torture 641 support their positions, and the Court will have to determine, should the Knesset legislate, the normative and evidentiary aspects of such legislation. The alternative legal ground for the use of extraordinary investigative methods (i.e., the use of force or the infliction of physical and emotional harm) put forward by the state was the necessity defence, found in the Penal Code. This ground, according to the state, justifies the use of force by GSS investigators in “ticking bomb” situations, defined by the Court as follows: “A given suspect is arrested by the GSS. He holds information respecting the location of a bomb that was set and will imminently explode. There is no way to diffuse the bomb without this information. If the information is obtained, however, the bomb may be diffused. If the bomb is not diffused, scores will be killed and maimed.”34

According to the state, because the GSS is burdened with safeguarding the public against terrorist attacks, and because no other alternative is available in “ticking bomb” situations, the Penal Code section establishing the defence of necessity should be read as authorising GSS investigators to resort to shaking a suspect (or applying similar methods) in order to extract the information from him and thus save human lives. The Court disagreed. While accepting that the defence of necessity could, in these extreme circumstances, be available to GSS investigators as individuals in ex post facto criminal investigations,35 the Court found that this defence cannot serve as an a priori authorisation for government to use these methods.36 The Court reasoned that the administrative power is based on establishing explicit, general, forward-looking criteria, whereas the necessity defence is engaged as an ad hoc reaction to an event: “It is the result of an improvisation given the unpredictable character of the events.”37 Since the statutory authority relied upon by the state—the necessity defence—failed to empower the executive to act in a manner violating human rights, and since rules of ordinary investigation did not include the authorisation to inflict pain and suffering, the state lacked an explicit authorisation in law to use force in interrogation aimed at extracting crucial information in “ticking bomb” situations. Consequently, an application to a court of law, on administrative law grounds, for an injunction against such use of force by the GSS will succeed. Read as such, the decision suggests that the legislature is indeed not free to empower the GSS to use the methods above as inherent to any interrogation qua interrogation. Any attempt to enact into law the methods above as ancillary to an “ordinary” interrogation will likely fail the proportionality test imposed by 34 GSS case, supra n. 4, at para. 33. It should be noted that the “ticking bomb” situation is broader than an actual ticking bomb. Under the necessity defence, the Court does not require the actual timing mechanism of a particular bomb to have been activated. The term “ticking bomb” is in essence a metaphor that stands for a terrorist attack that has been set in motion so that harm to human life is imminent unless the security forces intervene so as to foil the attack: GSS case, supra n. 4, at para. 34. 35 Ibid., at paras. 33–4. 36 Ibid., at para. 36. 37 Ibid.

642 A Reichman & T Kahana BLH in the same way that it failed the review under administrative law grounds. At the same time, this judgment does not foreclose the possibility that the legislature may empower the GSS to use force in “ticking bomb” situations, given the nature of the harm, its magnitude, its imminence and its character.38 Such legislation would overturn the GSS case to the extent that it would allow the state to demonstrate a “ticking bomb” situation so as to avoid an injunction. Should the legislature see fit to empower the GSS to use force or to inflict harm on suspects it interrogates, this legislation would be subjected to constitutional scrutiny under the BLH using legal formulae similar to those employed by the Court in GSS for its administrative law review. It seems, however, that the content poured into the different variables of the legal tests would be different. For example, the Court’s proportionality analysis under the BLH will be different from the administrative analysis conducted in the GSS case because the purpose of the legislation in question will be different. Whereas the legislation that currently empowers the GSS incorporates ordinary rules of investigation, that is, authorises GSS investigators to employ only those means integral to any investigation as such, a specific statute could empower the GSS to combat “ticking bomb” situations by specifically outlining the harm the legislation seeks to prevent and the methods the GSS can use to prevent such harm, including methods that go beyond those integral to ordinary investigations. In other words, while the Court in the GSS case did resort to the language of the proportionality analysis under the BLH, it did not place at the centre of its analysis the variables that would be relevant in a constitutional review of an enabling statute. Should the Knesset legislate, the Court will have to scrutinise very carefully whether such empowering legislation meets the constitutional requirements set out by the BLH. As the GSS case made clear, any such empowering legislation will amount to an infringement of liberty, and thus will prompt the Court to determine whether that infringement was prescribed by law, was part of a statute enacted for a proper purpose that befitted the values of the State of Israel, and whether the impugned statute infringed the right to an extent no greater than was required.39 As part of its analysis of the “prescribed by law” requirement, the Court will have to examine whether the law is explicit, clear and precise enough so as to confine and to guide executive discretion. Further, the state will have to show exactly how each method prescribed directly advances the gathering of information necessary for the defusing of the ticking bomb, real or more metaphorical. Further, the state will have to show the absence of any other less restrictive means of avoiding the imminent deaths. Arguably (and hopefully), this would not be an easy burden to meet. It might very well be the case that the legislature, having read the current judgment, will be unable to defend in court the authorisation of playing loud music, or the cuffing of a prisoner in 38 Acts of terror are aimed against the state as such, so as to terrorise its citizens as citizens, and thus may be seen as requiring means of collective self-defence. 39 Section 8 of BLH.

Israel and the Recognition of Torture 643 the “Shabach” position, because the causal link between such methods and the gathering of information necessary for the diffusion of the ticking might be too remote, or there might be less restrictive means; however, under the reading of the GSS case suggested here, this inquiry has not yet been conducted by the Court. In conclusion, this reading reconciles the Court’s use of constitutional terminology in administrative garb with its explicit ruling that the legislature may respond and enact enabling legislation. Under this reading, the door is open, but the threshold is high; any legislation will likely give rise to another round of dialogue between the Court and the legislature so as to ensure deliberation and the assumption of shared responsibility by the three branches of government.

4 THE INTERNATIONAL PERSPECTIVE : THE GSS CASE AND EX PARTE PINOCHET

Having analysed the possible readings of the GSS case domestically, it is worthwhile to place the case in the international context, notably vis-à-vis the matter of Senator Pinochet. The common denominator of these cases is clear: they both deal with aspects of the prohibition against the infliction of severe pain and suffering by state officials. The remainder of this chapter will deal with some aspects of the interaction between the domestic and the international arenas in light of Ex Parte Pinochet.40 It will not fully canvass the emerging global jus cogens (hence constitutional)41 norms as they relate to inflicting harm in interrogations. Nor will it delve into the intricacies of the three incarnations (to date) of the Pinochet case before the House of Lords.42 Accordingly, we will not attempt in this chapter to develop a theoretical framework in which to place the interaction between the GSS case and Ex Parte Pinochet.43 Rather, we will focus on a host of concrete legal questions, through which the possible ramifications of the GSS case can be assessed in the context of transnational adjudication. 40 Regina v. Bartle and the Commissioner of Police for the Metropolis and Others Ex Parte Pinochet (House of Lords, 24 March 1999), published as R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No 3), [1999] 2 All ER 97 (HL) [hereinafter Ex Parte Pinochet]. 41 On jus cogens, see O Schachter, International Law in Theory and Practice (Dordrecht, Boston, M. Nijhoff Publishers; sold and distributed in the U.S.A. and Canada by Kluwer Academic Publishers, 1991) at 30–1; L Henkin, R C Pugh, O Shachter and H Smith, International Law, Cases and Materials 3rd ed.,(St. Paul, Minn., West Publication, 1993) at 92–4; I Brownlie, Principles of Public International Law 4th ed., (Oxford, Clarendon Press, 1990) at 512–15. 42 Regina v. Bartle and the Commissioner of Police for the Metropolis and Others Ex Parte Pinochet (House of Lords, 25 November 1998) published as R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte, [1998] 4 All ER 897 (HL) (hereinafter Pinochet No 1); In Re Pinochet (15 January 1999) published as R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte, [1999] 1 All ER 577 (HL) (hereinafter Pinochet No 3); Ex Parte Pinochet, supra n. 40. 43 For a fuller exposition of the argument and a possible matrix see A Reichman, “When We Sit to Judge We Are Being Judged: The Israeli GSS case, Ex Parte Pinochet and Domestic/Global Deliberation” supra n. 1.

644 A Reichman & T Kahana Ex Parte Pinochet dealt with a request by the Spanish Government to extradite Senator Augusto Pinochet so that he could stand trial in Spain for offences that he allegedly committed—primarily in Chile—as Chile’s Head of State. The alleged offences included torture, hostage taking, conspiracy to take hostages and conspiracy to commit murder, all in a large-scale, systematic manner. According to British law, for the request to succeed the alleged acts had to be criminal in the United Kingdom as well as in Spain. Since the acts were committed in Chile, the United Kingdom would have had to assume extraterritorial jurisdiction over the acts. This had been done, with respect to torture, by section 134(1) of the Criminal Justice Act 1988,44 which incorporated the UN Convention against Torture, and Other Cruel, Inhuman or Degrading Treatment or Punishment (the Convention).45 It should be noted that article 1 of the Convention leaves little doubt as to its application to interrogations.46 Senator Pinochet advanced an immunity claim, enjoyed in international law by Heads of State, including former Heads of State, for acts conducted in their official capacity as such. This defence was rejected by the majority of their Lordships, for different reasons. The relevance to the Israeli case is clear: conceivably, Israeli officials—the Prime Minister, members of the GSS, and even Supreme Court justices who failed to prohibit the infliction of severe pain or suffering—could be subject, at least after leaving state service, to criminal liability in Britain and/or to extradition requests by third countries for torture committed after the enactment of the prohibition in the Criminal Justice Act 1988. It is pointless to speculate about whether the Israeli Court has acted so as to preempt such future liability, yet it seems reasonable to assume that, although Ex Parte Pinochet was not referred to specifically in the GSS case, the Court was well aware of its possible legal consequences. In any event, the legal result of the Israeli case minimises the possible friction between the British and Israeli legal regimes, given that the use of force in interrogation would cease to enjoy a priori judicial sanctioning. In the aftermath of the GSS case and Ex Parte Pinochet, there remains the question of the necessity defence. First, on that point there is a discrepancy between the position of the Israeli Court and the Convention, and thus foreign 44 Section 134(1) of the Criminal Justice Act 1988 provides: “A public official or person acting in an official capacity, whatever his nationality, commits the offence of torture if in the United Kingdom or elsewhere he intentionally inflicts severe pain or suffering on another in the performance or purported performance of his official duties.” Section 134(3) of the Act provides that it is immaterial whether the pain or suffering is physical or mental and whether it is caused by an act or an omission. 45 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted 10 Dec. 1984, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51 (1985) (entered into force 26 June 1987), reprinted in (1984) 23 I.L.M. 1027. 46 Article 1 of the Convention defines torture, for the purposes of the Convention, as “any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind . . .”.

Israel and the Recognition of Torture 645 courts will have to decide what legal weight the Israeli position should be accorded in criminal proceedings. Second, individual culpability under Israeli law is relevant to any foreign civil cases brought against GSS officials. Most notably, should necessity be recognised abroad as a defence to criminal liability, it may well speak to the justifiability of parallel recognition of such a defence in private law—for instance, by way of a comity doctrine or the act of state doctrine.47 Here, it will be relevant to determine whether, or to what extent, Israeli law shields GSS officials from civil liability and not only from Penal Code culpability.48 In the absence of such specific immunisation, a foreign court might well conclude that a necessity defence in Israeli criminal law cannot be given legal effect abroad in a civil law context until Israel itself has unambiguously legislated as such, or until the matter is so concluded by the Israeli Supreme Court. It would be understandable if a foreign judge reasoned that it is not his or her job to do the Israeli Knesset’s work for it, or to otherwise preempt the Israeli legal process. The question of transnational criminal law responsibility must first be addressed before returning to consider civil law analogies.49 Would an Israeli GSS officer incur criminal liability under British law for inflicting severe pain 47 See M Bühler, “The Emperor’s New Clothes: Defabricating the Myth of ‘Act of State’ in Anglo-Canadian Law”, chapter 13 of this volume. 48 The civil responsibility of individual GSS investigators or the state has not yet been fully resolved under Israeli law. In the aftermath of the GSS case a civil suit was filed by a Lebanese citizen, Mr. Dirani, who was allegedly tortured by GSS investigators after being kidnapped and brought to Israel. In another case, the state has agreed to settle and pay compensation to a Palestinian suspect who was tortured in a non-ticking-bomb situation, arguably in part in order to avoid judicial resolution of state responsibility. As a general matter, the issue is governed by Civil Wrongs Ordinance (New Version), 2 L.S.I. 5 (1972), and Civil Wrongs (Liability of the State) Law, 6 L.S.I. 147 (1951–52). Section 2 of Civil Wrongs (Liability of the State) establishes the general principle of state liability, which mirrors the liability of any other juridical personality. Section 3 exempts the state from tortious liability for any action committed under authority conferred by law, or under a bona fide mistaken belief of legal authority, but does not exempt state liability for a negligent action. Section 5 adds an exemption for any military action committed by the Israeli Defence Force. Section 7(b) extends the latter exemption to state agents or officials. However, the GSS and its investigators are not explicitly enumerated, and the court will have to decide whether that omission qualifies as a negative disposition of the matter. The Civil Wrongs Ordinance pertains to individual liability. Section 6 establishes a defence for any action or omission done pursuant to any statute or in accordance with its provisions. Section 7(a) establishes personal liability of state officials for any tortious liability, but exempts actions done pursuant to authority conferred by law, or a bona fide mistaken belief of legal authority. As with state liability, this exemption does not extend to negligence. Section 7(b) exempts state officials for actions done by other state officials, unless the principal explicitly authorised—ex ante or ex post—the tortious action of the agents. The emerging legal picture from the above sections suggests that, after the GSS case, it would be difficult for state officials, including GSS investigators, to avoid civil liability for the use of force in non-ticking bomb situations, since such use of force is unauthorised in law. The court will have to decide whether the necessity defence qualifies under “action pursuant to a statute or according to its provisions”, as stated in section 6 referred to above. With respect to action committed prior to the GSS case, the courts will have to decide whether GSS investigators could successfully claim a bona fide mistaken belief regarding their authority to use force in “ticking bomb” situations. In any event, the court will have to decide whether the tort of negligence can be used in order to claim physical, psychological and dignitary damages from the state and individual GSS investigators. 49 Transnational tort issues will be addressed infra in Section 5.

646 A Reichman & T Kahana and suffering while interrogating a suspect in Israel in a “ticking bomb” situation? This question could arise presently, since the necessity defence remains valid following the GSS case, and the Attorney General of Israel is therefore not likely to press charges where she is satisfied that the requirements of the defence are present. It may also arise where a GSS investigator is tried in Israel and is found not guilty because the necessity defence applied, although prosecution of the same case abroad might add a “double jeopardy” element. Alternatively, the question may arise in both criminal and civil law contexts should the Israeli Knesset go beyond ex post facto relief from responsibility and explicitly empower the GSS to employ specified violent methods in “ticking bomb” situations. As mentioned in the presentation of the second reading of GSS, such legislation may pass constitutional judicial review, or it may involve amendments to the BLH so as to immunise it from judicial review. The prohibition against torture, as defined in article 1 of the Convention, is absolute as a matter of international law. This was acknowledged by the Israeli Supreme Court50 and by the House of Lords.51 It is therefore not surprising that the Convention’s Committee against Torture has rejected Israel’s claim that exceptional circumstances are a justification under the treaty.52 Should the English Attorney General press charges, or seek extradition against a former GSS officer who inflicted pain in a “ticking bomb” situation, a British court will have to determine whether to adopt the finding of the Committee that the methods used by the GSS fall under the prohibition embodied in article 1 of the Convention—and therefore are absolute under article 2 of the Convention—or whether they fall under the prohibition embodied in article 16, dealing with cruel and inhumane and degrading treatment—and therefore may be subject to justifications and excuses as a defence.53 The Israeli Court chose not to engage 50

GSS case, supra n. 4, at para. 23. See e.g. Lord Browne-Wilkinson Ex Parte Pinochet (supra, n. 40), citing Prosecutor v. Furundzija, Tribunal for Former Yugoslavia, Case No. 17-95-17/1. 52 In para. 6 the Committee stressed that the prohibition against torture is absolute, as plainly expressed in article 2 (see Summary Records of the Public Part of the 297th Meeting of the United Nations Committee against Torture (Geneva, 9 May 1997): Israel, (CAT/C/SR.297/Add.1 (4 Sept 1997) ). 53 Supra n. 52, at para. 5. It should be noted, however, that the Committee is not a judicial organ, and its findings are therefore not legally binding, although they ought to form an important reference point in the interpretation of the Convention by a judicial organ. Moreover, Israel has reserved from article 20 of the Convention, which empowers the Committee to conduct investigations and issue findings. The Committee, apparently as part of its political efforts to promote the goals and purposes of the treaty under its guardianship, called upon Israel to withdraw this reservation. Thus, the Committee could be seen as part of the general enforcement mechanisms set in place by the Convention, so as to expose members that might be acting in violation of their treaty obligations. The Committee is not charged with other rules of international law, such as those prohibiting terrorism, and since it is not part of a structure of governance, as a judicial organ would be, it is not faced with a responsibility to the governed for its decisions. Under such a structure, a vigilant Committee can acknowledge the “terrible dilemma” (as the Committee put it) confronted by Israel in dealing with terrorist threats, but it cannot accept such a dilemma as a justification for deviating from the explicit rules of the Convention to which it owes its existence and the enforcement of which it must promote. A court, however, especially a national court, is in a different position. The Committee might have overreached somewhat, with respect to its finding regarding “using cold air 51

Israel and the Recognition of Torture 647 the matter at this stage, since it was not necessary to do so, but it is possible— given the Court’s sanctioning of the necessity defence—that, should the Court be confronted with the question, it will decide to classify some interrogation methods as something less than torture, and thus, in effect, permit their use in “ticking bomb” situations under the flexibility offered by article 16. After deciding whether the methods fall under the first subsection of article 16 of the Convention, the British court will have to determine to what degree the absolute rejection of defences and justifications embodied in article 2 is part of British law on point. This investigation is not unique to British law, since any legal system that incorporates universal jurisdiction to try infliction of severe pain and suffering by an official of a foreign state in a foreign territory will have to address whether the incorporation of the offence includes its absolute nature, or whether general defences are still applicable. In so doing, the court will have to decide whether the Israeli conclusion that necessity is indeed a defence is acceptable. Where there is no explicit statutory disposition of the matter, a foreign AngloAmerican court will have to decide whether article 2 of the Convention, namely the rejection of any defence to torture, is a matter of treaty law or customary jus cogens so as to have been incorporated into domestic law, and whether article 2 or the custom conflicts with other international law instruments (contractual or customary). In the British case, this investigation seems unnecessary, given that section 134(4) of the Criminal Justice Act 1988 explicitly negates the proposition that British law is as absolute as the Convention by accepting “acting under lawful authority, justification or excuse” as a defence. Moreover, even if, under British law, resorting to the use of force in “ticking bomb” situations would not constitute a lawful defence, section 134(5) requires that the British court should apply “the law of the place where it was inflicted” in its determination of whether a GSS officer in Israel is acting under lawful authority, justification, or excuse. It may be observed that the UK’s criminal law on torture has thus incorporated a choice of law approach to one of the issues going to culpability. The idea of applying foreign law is normally associated with private international law and not commonly associated with criminal law. The presence of section 134(5) will be a potentially powerful analogy in any tort case in which a defendant GSS official argues deference to, or application of, foreign (Israeli) law— either through a contended-for choice of law rule of lex loci delicti or by way of a version of the act of state doctrine.54 The GSS case makes it clear that the resort to violence outwith “ticking bomb” situations is unlawful and does not fall under the excuse of necessity. At to chill”. As the GSS case revealed, such a method apparently was not used by GSS investigators. Thus, the committee’s finding that this method was used, and that its use amounted to torture, might have hurt the Committee’s credibility, as was expressed in the Israeli reaction to the Committee’s recommendations (see supra n. 52). 54 On the current place of lex loci delicti in UK tort choice of law, see J Orange, “Torture, Tort Choice of Law and Tolofson”, chapter 11 of this volume. On the shape of the Anglo-Canadian act of state doctrine, see Bühler, supra n. 47.

648 A Reichman & T Kahana the same time, the Court in the GSS case reaffirmed its position that inflicting pain and suffering in “ticking bomb” situations could fall under the necessity excuse, provided that no other less restrictive means were available. Under this structure, an empowering Israeli statute might change little (assuming it will empower nothing but the application of force in “ticking bomb” situations). It is beyond the scope of this chapter to investigate British law further, including the possible effects of European Union law on this matter; however, it should be noted that the availability or unavailability of a lawful defence under British law (including its reference to “the law of the place”) might have implications regarding extradition requests, given the requirement of double criminality. In that context, it should be recalled that Israel, as the state with arguably the most obvious jurisdiction under article 5(1) of the Convention, might request the extradition of its officials (or ex-officials) to stand trial in Israel should they confront an extradition request by a third country while in Britain. Or, Israel might seek to intervene to oppose extradition by pointing to the combined effect of the double criminality requirement and the section 134(5) lex loci rule. As mentioned above, Senator Pinochet’s argument for former Head of State immunity was rejected by the Court. The Law Lords were all aware of the tension between assuming universal jurisdiction (without an explicit waiver from the state whose official, or former official, is put on trial) and a principle of sovereignty that draws vitality from the concept of self-determination: since no one state society is supreme over another, the courts of one sovereign cannot judge the official acts of another. In facing such tension, some of the Lords chose to see a waiver of immunity in the signing of the Convention.55 Others chose to analyse the scope of the immunity, finding that it did not extend to acts of torture since such acts cannot, as a matter of international law, be considered a part of the official function of any state official.56 It should be noted that before the Law Lords were allegations of systematic, state-sanctioned acts of infliction of severe pain and suffering, as part of a scheme designed to silence opposition and to thwart processes of selfgovernment. In the terminology advanced above, the Law Lords saw prima facie evidence of a system failure, a collapse of the capacity or will—or both—of the internal Chilean legal process to check executive abuse of power through the

55 The opinion of Lord Browne-Wilkinson ultimately rests on the Convention as establishing the missing link in constituting a fully effective international crime of torture, a crime which is incompatible with state immunity: Ex Parte Pinochet, supra n. 40, at 114f–15c. Therefore, his Lordship’s opinion rests, in part, on the fact that Chile has ratified the Convention, thereby agreeing to the creation of the international crime, the logic of which rejects state immunity: ibid., at 115d. The opinion of Lord Saville of Newdigate approaches the treaty as expressive enough to contain a waiver: ibid., at 169f. 56 Lord Millett was perhaps the clearest advocate of the proposition that state immunity cannot contradict jus cogens, (Ex Parte Pinochet, supra n. 40, at 179e–f) and that jus cogens against torture evolved as early as 1973 (ibid., at 178b–c). The majority in the first decision of Ex Parte Pinochet based its decision on jus cogens norms against torture: Pinochet No 1, supra n. 42.

Israel and the Recognition of Torture 649 prosecution and punishment of offenders.57 Under such circumstances, the Law Lords reached a conclusion that the international community, in this case through the agency of a British court acting in response to a Spanish magistrate, must step into the breach in view of the violations of the jus cogens against torture. On this dimension, the Israeli Supreme Court’s GSS case could stand for the proposition that the Israeli legal institutions, namely those governed by the legal process, have not abdicated their responsibility to maintain the rule of law. The Court explicitly stated that the use of force was not authorised under Israeli law, with the effect of distancing the state, and the legal process, from the condoning of such methods. As a result, the Israeli judicial system enjoins the executive from using such ultra vires methods. The Israeli Court expressly referred to the circumstances that can support the necessity defence, namely terror attacks against the Israeli civilian population, and narrowed the purpose that the infliction of pain and suffering by GSS interrogations must achieve before forming the subject of criminal prosecution ex post facto, namely, the saving of human lives. In declaring that the infliction of pain and suffering in all other circumstances is criminal, and likely to remain so given the presence of human rights constitutional legislation, the Israeli Court affirmed the availability of the 57 Lord Hope of Craighead advanced a position according to which only in cases of large-scale, systematic torture does former Head of State immunity not apply, since such conduct cannot be part of the public function of Heads of State. In his Lordship’s opinion, the alleged facts in Pinochet satisfied this requirement: “As a whole, the picture which is presented is of a conspiracy to commit widespread and systematic torture and murder in order to obtain control of the government and, having done so, to maintain control of government by those means for as long as might be necessary.” Lord Browne-Wilkinson disagreed on the law, and found that a single act of torture merits the lifting of the immunity. Yet Lord Browne-Wilkinson was careful to point out the following:

“. . . [T]he objective [of the Convention] was to ensure a general jurisdiction so that the torturer was not safe wherever he went. For example, in this case it is alleged that during the Pinochet regime torture was an official, although unacknowledged, weapon of government and that, when the regime was about to end, it passed legislation designed to afford an amnesty to those who had engaged in institutionalised torture. If these allegations are true, the fact that the local court had jurisdiction to deal with the international crime of torture was nothing to the point so long as the totalitarian regime remained in power: a totalitarian regime will not permit adjudication by its own courts on its own shortcomings. Hence the demand for some international machinery to repress state torture which is not dependent upon the local courts where the torture was committed.” (Ex Parte Pinochet, supra n. 40, at 109e). In other words, Lord Browne-Wilkinson noted the “system-failure” component of the case. In light of that context, Lord Browne-Wilkinson rejected Lord Goff of Chieveley’s dissenting position that states could be trusted to waive their immunity or prosecute the offenders themselves, given the political and moral pressure available in the international arena. For Lord Browne-Wilkinson such a structure will allow safe haven for torturers—Heads of States and all other officials alike—in cases where the legal system under which torture was committed failed to confront the problem. Lord Millett concurred that “[t]he evidence shows that other states were to be placed under an obligation to take action [under the Convention] precisely because the offending state could not be relied upon to do so”. (Ibid., at 179d). Lord Phillips of Worth Matravers noted that the nature of the jus cogens crimes, such as torture, is such “[t]hat they are likely to involve the concerted conduct of many and liable to involve the complicity of the officials of the state in which they occur, if not the state itself. In these circumstances, it is desirable that jurisdiction should exist to prosecute individuals for such conduct outside the territory in which such conduct occurs”. (Ibid., at 188j).

650 A Reichman & T Kahana internal legal process to curb the powers of the GSS—in future. Hence, the Israeli Court could be seen as addressing the British concern of system failure, so as to alleviate the need to pierce the veil of sovereignty inherent in the exercise of universal jurisdiction by a foreign court. It remains to be seen whether the Court’s reasoning—translated to English and put on the Internet as soon as the case was handed down—will have a forestalling effect on foreign courts. Or, will foreign jurisdictions in which the incorporation of the Convention leaves discretion regarding the availability of defences with the judiciary reject the necessity defence, or any other related defence such as collective self-defence, in favour of an absolute prohibition? Arguably, there could be jurisdictions that would proceed to prosecute former (or current) Israeli officials who were involved in inflicting pain and suffering in “ticking bomb” situations, as an absolute prohibition would demand. Such jurisdictions seemingly would even have to consider whether to view the Israeli judges themselves as accomplices, given the Court’s refusal to withdraw the necessity defence (accorded to such conduct by the Court in earlier cases). By continuing to endorse this defence, the Court could be construed as participating, even if by omission, in the state practice of inflicting pain and suffering on suspects who hold life-saving information in “ticking bomb” situations. The stance of foreign courts could depend, at least in part, on which of the two readings of the Court’s reasoning in GSS, advanced in sections 2 and 3, they find most persuasive.58 The Israeli Court did not have only international audiences in mind. In its reference to the rules of international laws that prohibit torture as well as cruel, inhuman and degrading treatment, the Court could be seen as reminding its local audiences—the GSS, the Parliament, the legal and academic communities and the general public—of the presence of the international and foreign legal machinery and of the concomitant possibility of legal interventions should Israel (the judiciary, the legislature or the executive) ignore the basic commitment of all nations to jus cogens norms of international law.59 In so doing, the GSS case, by referring to the Convention without explicitly applying it, made sure that the Convention would be part of the public debate, so as to ensure meaningful deliberation.

5 TORT ACTIONS IN FOREIGN COURTS AGAINST GSS OFFICIALS

We turn now to another dimension of the intersection of international law, foreign jurisdictions and Israeli law, that which centres around the availability of 58 Note also, in passing, that such jurisdictions will be subjected to the rule of reciprocity, under which their officials may be denied domestic defences if tried in a foreign court. This is provided such a denial of defences did not, in and of itself, amount to a violation of a pre-emptory norm. See Henkin et al., supra n. 41, at 578–9. 59 GSS case, supra n. 4, at para. 23.

Israel and the Recognition of Torture 651 civil actions against torts as well as the availability of immunity-like defences in such cases. As dicta in Pinochet suggests, under UK law sovereign immunity is accorded not only to the state but also to its officials. Such officials are thus protected from the exercise of jurisdiction by foreign courts with respect to their actions carried out while in public service after their service has ended. In 1978, the British Parliament enacted the State Immunity Act, modifying the immunity so as to allow litigation in some matters, primarily commercial, as an exception to the rule and at the same time removing the matter from the common law domain. In the realm of tortious harm to persons, the statute lifted foreign state immunity from personal injuries caused by act or omission of the foreign state in the United Kingdom, but not with respect to acts or omissions elsewhere. In his analysis of state immunity in Ex Parte Pinochet, Lord Hutton concluded that, under international law, the state is responsible for acts of torture carried out by its former heads of state (or other officials) “but could claim state immunity if sued for damages for such acts in a court in the United Kingdom”. As for Senator Pinochet, he “could also claim immunity if sued in civil proceedings for damages under the principle stated in Jaffe v. Miller”.60 Lord Hutton was satisfied that enjoyment by a state and its officials of state immunity in civil matters (“notwithstanding that the acts are performed in excess of [the officials’] proper function”) is not inconsistent with the lack of immunity for former state officials in criminal matters.61 In other words, according to Lord Hutton, the ruling of non-immunity in Ex Parte Pinochet did not affect immunity in civil proceedings. If so, Pinochet did not expose GSS investigators to new civil liability in Britain. For his part, Lord Millett agreed civil suits were still barred against officials for acts committed under the colour of their office. Finally, Lord Phillips of Worth Matravers acknowledged the “impressive, and 60 Jaffe v. Miller (1993), 13 O.R. (3d) 745 (C.A.). In this case, Mr. Jaffe sued Florida state officials for their role in his abduction from Toronto, for their role in his unlawful incarceration in Florida and for their role in initiating criminal charges against him so that he would settle a civil suit. The court dismissed the claim, finding immunity. The court stated that the allegedly illegal and malicious nature of the acts complained of did not in themselves move those acts outside the scope of the official duties of the defendants so as to deprive them of the protection afforded to functionaries of the state of Florida. See the discussion of this case in Bühler, supra n. 47. 61 In Lord Hutton’s opinion, in civil matters, the state, as the principal, is responsible for the actions of its agent, the official, whether or not the official was indeed vested with the actual authority to commit the acts. As long as the actions are carried out by state agents in the “ostensible performance of their official functions”, notwithstanding that the acts may be performed in excess of the officials’ proper function, the state is responsible. Under such a legal regime, a suit against the agent is de facto a suit against the state, which cannot be entertained in a national court without the agreement of the foreign sovereign state. In criminal matters, both the state and the official are separately responsible; hence the agent can be held criminally liable as an individual, without implicating the state, and without reaching legal and factual conclusions that ipso facto affect the responsibility of the state. (Ex Parte Pinochet, supra, n. 40, at 155j, 156f and 157j). Lord Millett noted that the distinction between the criminal and the civil lies in the “official” nature of the crime of torture: “The very official or governmental character of the acts which is necessary to found a claim of immunity ratione materiae, and which still operates as a bar to the civil jurisdiction of national courts, was now to be the essential element which made the acts an international crime.” The official character, it should be noted, is not dependent on actual authorisation by internal law to commit torture. (Ibid., at 175a).

652 A Reichman & T Kahana depressing” list of authorities for the proposition that even when the tortious act is also criminal, state immunity still stands.62 This is also the case when a state is not involved in the litigation, but when the litigation instead turns on the validity or invalidity of the public acts of a foreign state63 and a fortiori when an act was committed at the behest of the state64. These opinions, and the authorities relied upon therein,65 are difficult to reconcile with the possibility of entertaining a tort claim in Britain against Israel or its officials for torture. It seems that GSS officers act in their official capacity when they inflict pain and suffering in the course of an interrogation, even if in so doing they act without Israel’s “command or authorisation, or . . . in excess of their competence according to the internal law”66 of Israel. For a tort claim in Britain to succeed in the current state of affairs as understood by at least some of the Lords, it seems that Israel would have to waive its immunity. However, as Lord Phillips of Worth Matravers observed in the context of international criminal law, “this is an area where international law is on the move.”67 Therefore, it could very well be that, in the future, the relationship between state responsibility and individual responsibility in civil proceedings will mirror that in criminal ones. Arguably, if state-sponsored torture, no less than state-sponsored terrorism, contravenes jus cogens norms, such actions cannot qualify as official functions of state agents. Individual perpetrators could lose any immunity based on having acted on behalf of a foreign sovereign state—both in civil and in criminal proceedings. Such an approach would track the reasoning in Pinochet with respect to criminal law culpability and apply it to the realm of civil law accountability. Should this development occur, state immunity against civil claims of torture will be withdrawn, at least from former officials, regardless of the location where the torture was committed. 62 Saltany v. Reagan, 702 F. Supp. 319 (1988), (claims of assassination and terrorism); Siderman de Blake v. Republic of Argentina, 965 F.2d 699 (1992), (claim of torture); Princz v. Federal Republic of Germany 26 F.3d 1166 (D.C. Cir. 1994) (claim in respect of the Holocaust); Al-Adsani v. Government of Kuwait (1996, British Court of Appeal) 107 I.L.R. 536 (claim of torture); Sampson v. Federal Republic of Germany, 975 F. Supp. 1108 (N.D. Ill. 1997) (claim in respect of the Holocaust); Smith v. Libya, 886 F. Supp. 406 (E.D.N.Y., 1995), 101 F. 3d 239 (2d Cir. 1996) (claim in respect of Lockerbie bombing); Persinger v. Islamic Republic of Iran, 729 F.2d 835 (D.C. Cir. 1984), (claim in relation to hostage-taking at the U.S. Embassy). 63 Here, the doctrinal shield bears the name “act of state” doctrine. It is not state immunity as it can be pleaded by litigants with no connection, existing or former, to the foreign state. For classic statements of the doctrine in the US and the UK, respectively, see: Underhill v. Hernandez (1897) 168 U.S. 456; Buck v. Att. Gen., [1965] Ch. 760, 1 All ER 882. 64 See recently Kuwait Airways Corporation v. Iraqi Airways Company and Republic of Iraq, [1995] 2 Lloyd’s Rep. 317, where the House of Lords found Iraqi Airways immune from suit from Kuwait Airways for removing KAC from Kuwait during the Gulf War at the behest of the Iraqi Government. 65 Al-Adsani v. Government of Kuwait (supra n. 62); Siderman de Blake v. Republic of Argentina (supra n. 62); Jaffe v. Miller (supra n. 60); Marcos and Marcos v. Federal Department of Police (1989, Switzerland Federal Tribunal) 102 I.L.R. 198. 66 R Jennings and A Watt, (eds.), Oppenheim’s International Law, 9th ed. (Harlow, Essex, England, Longman, 1992), at 545, as cited by Lord Hutton, Pinochet, supra n. 40. 67 Ex Parte Pinochet, supra n. 40, at 188h.

Israel and the Recognition of Torture 653 In accordance with the theme advanced in this chapter, it is suggested that this development may be justified when the state under whose regime torture was committed fails to hold its officials, or itself, responsible for the wrongs. Such responsibility can include civil liability and resultant monetary compensation, public admission of responsibility towards the victims (the Truth and Reconciliation process in South Africa is an example in point) or various other means. However, should the state choose to ignore its civil responsibility in toto, it cannot cry foul if confronted with a future development in civil proceedings similar to that in international criminal law and akin to the development represented by Pinochet. In support of the just-stated hypothesis, we must consider the rationale behind the act of state doctrine, including possible supporting British authority on point. Part of the rationale behind the doctrine lies in the perceived need for judicial restraint in a court pronouncing on either the validity or the meaning of the foreign sovereign’s laws.68 However, after the validity of the state action and the content of the law in question have already been established via litigation in the foreign state (that is, by a judicial organ of the same nationality as the legislature or executive whose acts are challenged), this justification no longer applies. As is apparent from the GSS case, the Israeli Court has made clear that in non-“ticking bomb” situations, Israeli GSS officers act not only without any authorization in law, but in fact against the law. Therefore, there could be an argument that, in the absence of any enabling legislation requiring further interpretation, restraint vis-à-vis a foreign judicial system is no longer applicable when a non-Israeli court is confronted with the matter. The British court would not have to “pronounce upon the validity of a law of a foreign sovereign state within its own territory”, because the Israeli Court has already done so.69 Rather than conceiving the situation as one of confrontation between sovereigns, the initial position could be one of “full faith and credit”, in which each jurisdiction takes the laws of other jurisdictions seriously. Thus, it could be argued that when force is applied in non-“ticking bomb” situations, the GSS officers act not merely “in excess of their competence” (which might still allow them to claim immunity), but in violation of both Israeli and international criminal law. Under both Israeli and international law, they may accordingly be viewed as stepping outside their “official function” altogether, so as to lose the benefit of state immunity. In such circumstances, future developments in AngloAmerican jurisprudence, not necessarily dependent on amendments to the relevant state immunity statutes, could recognise civil liability even if the executive chose not to waive state immunity. Turning to the use of force covered by the necessity defence, it should be recalled that the Israeli Court found, as a matter of positive law, that GSS 68

See Lord Diplock in Buck v. Att. Gen., supra n. 63, at 770. Buck v. Att. Gen., supra n. 68. The GSS decision could be read as, in effect, pronouncing on the validity of secondary legislation, i.e. the directives issued by the Ministerial committee which authorized the use of force in “ticking bomb” situations. 69

654 A Reichman & T Kahana investigators were not authorised by statute to use force in any circumstances, including in “ticking bomb” situations. In view of this, the civil liability, both domestic and international, of GSS investigators could theoretically be triggered by conduct that oversteps the investigators’ administrative boundaries rather than only by the higher threshold of conduct overstepping criminal law boundaries. In other words, even if the necessity defence could shield GSS officers from criminal investigation, it is not clear, in the absence of explicit enabling legislation, that it could—or should—shield them from civil liability in Israel or abroad. Of course, for the reverse comity rationale to apply in full, there would have to be an Israeli case regarding civil liability of GSS officers, whether acting in “ticking bomb” situations or not. It is beyond the scope of this chapter to examine the possible effect of the GSS case on the civil liability of the state and/or GSS investigators in Israel for acts committed outwith the authority conferred upon GSS investigators by law. However, to the extent civil liability is shown to be possible in Israel, there will be reason for foreign courts to decline jurisdiction in deference to Israeli courts on grounds of forum non conveniens.70 Similarly, in analysing the justifiability (and efficacy) of foreign civil proceedings, derivative matters should not be overlooked. One such important matter is the future possibility of enforcement in Israel (or elsewhere) of monetary damages awarded by a foreign judgment against a GSS investigator when the GSS investigator has pleaded state immunity and the government of Israel has not explicitly waived its immunity in foreign civil proceedings. It could be that for public policy reasons Israeli courts would refuse to recognise and enforce foreign court orders.71 Where Israeli judges perceive the foreign court’s role as confrontational rather than cooperative (that is, one that focuses on the struggle between sovereigns in lieu of full faith and credit between judicial systems), they might see the foreign judgment as an affront to sovereignty—a still-central pillar of international law, public and private—that should not be given legal effect in Israel. Such considerations speak in favour of allowing Israeli courts to establish the law regarding civil liability before accepting that the matter can (also) be addressed in foreign jurisdictions. In other words, comity would suggest allowing the Israeli legal process an opportunity to address the matter, before addressing it in foreign jurisdictions. Comity considerations, however, should not be confused with full deference. In the GSS case, the Israeli Supreme Court could theoretically have taken the opposite position, one that deferred to the GSS in one way or another. Alternatively, the Israeli legislature could, theoretically, respond to the GSS 70 Forum non conveniens arguments will, also be affected by the possible need to match a specific investigator to a specific suspect, or the availability of cross-claims by victims of terrorist activities. 71 Enforcement of foreign judgments pertaining to civil matters is governed in Israel by the Foreign Judgments Enforcement Law, 1958, 12 L.S.I 82 (1957–8), which in section 8 lists contravention of public policy as an instance when the Court may declare a foreign judgment to be not enforceable in Israel.

Israel and the Recognition of Torture 655 judgment by fundamentally altering the legal landscape through an amendment to the BLH. Under either a counter-factual GSS result or the still-possible scenario of a problematic statutory response from the Knesset, the question of comity arises starkly: should foreign courts respect territorial acts by foreign states that themselves arguably constitute a severe and systematic violation of human rights? While the general principle may be that the content of foreign legislation is beyond the purview of common law domestic courts, there is House of Lords authority, albeit obiter, for the proposition that British courts would not recognise foreign law that is “unacceptably unfair, racial or barbaric”,72 or, in modern terminology, so repugnant to basic human rights protected by jus cogens that it cannot be recognised by members of the family of nations.73 In this spirit, the British House of Lords, in Oppenheimer v. Cattermole, indicated that it was not prepared to recognise a Nazi decree denying the German citizenship of Jews who left Germany.74 It goes without saying that comparing Israeli legislation to Nazi decrees is problematic, not only given the historical context, but also the different, if not opposite, legislative purpose behind the two legislative schemes (one designed to racially discriminate, the other designed to minimise the loss of human lives). However, from a human rights perspective, should the Israeli legislature seek to authorise the use of force with only a loose connection to the harm sought to be prevented, or should the authorisation amount to legalising torture, such legislation might not merit recognition by foreign jurisdictions for the purposes of the act of state doctrine. Such legislation is tantamount to a failure scenario, in which one sovereign state 72 Lord Denning M.R. in Attorney-General of New Zealand v. Ortiz and Others, [1984] A.C. 1 (HL) refers to Oppenheimer v. Cattermole, [1976] A.C. 249 (HL) for that proposition. 73 See opinions of Lord Salmon and Lord Chelsea in Oppenheimer, ibid. 74 Oppenheimer, ibid. It should be noted that the case itself involved a Jewish person, Mr. Oppenheimer, who received a stipend from Germany, and under a tax treaty was exempted from paying taxes on the remuneration if he held double citizenship. The legal fiction under which he was exempted by a lower court required that a Nazi decree, which stripped Jews (and only Jews) of German citizenship upon leaving Germany, be recognised as valid under German law and international law (so as to preclude the applicability of a 1913 German statute which would have stripped Oppenheimer of citizenship upon becoming a British citizen). At the same time, the fiction required that the Nazi decree be ignored under British law (which did not recognise the effect of foreign law pertaining to the citizenship of “enemy aliens” in times of war). The House of Lords rejected the fiction on the grounds that the German Basic Law, and not the Nazi decree, denied Mr. Oppenheimer his citizenship, by requiring presence in Germany for those who were denied citizenship by the Nazis (or a positive act to reaffirm German citizenship so as not to force German citizenship on those who were abroad and might not have wanted it). Thus, Mr. Oppenheimer ceased to be a citizen in 1949, after the hostilities were over, and the British doctrine regarding enemy aliens did not apply. In obiter, the Court proceeded to consider what the result would have been had the German Basic Law not had the aforementioned effect, and stated (Lord Chelsea and Lord Salmon, with Lord Hailsham of St. Marylebone and Lord Hodson concurring) that the Nazi decree would likely not be recognised in British courts, since Germany could deny German citizenship to a class of citizens it did not like, simply on racial grounds. In other words, the British Court would not give effect to a foreign statute that flagrantly violates basic human rights of citizens governed by that foreign statute. The somewhat ironic result is that the British Court, in the name of human rights, was prepared to ignore the position of a Jewish victim who still saw himself a German citizen despite the Nazi episode, and proceeded to tax him for compensation received from Germany for racial Nazi conduct.

656 A Reichman & T Kahana chooses to ignore in toto its reciprocal duties as a member of the family of nations, and is thus arguably estopped from demanding sovereignty-based respect and recognition within the framework of private international law. It remains to be seen, of course, whether a narrower statute, the purpose of which would be to enable the GSS to inflict some pain and suffering in “ticking bomb” situations and only as the least restrictive means, would enjoy recognition by foreign judiciaries.

6 CONCLUSION

The political reaction to the GSS decision in Israel was, as one could have expected, mixed. On the one hand, human rights organizations hailed the decision as an important step forward, towards normalcy.75 On the other hand, the GSS expressed concern that it could not adequately protect Israeli citizens against terrorist attacks, calling upon the government to present legislation to empower investigators to apply force in “ticking bomb” situations.76 After reviewing possible legislative options presented by a special committee headed by a senior deputy in the Attorney General’s office, and after several fierce debates, the Prime Minister reportedly decided to reserve judgment for the time being, and to examine the ability of the GSS to deal with terrorist threats under the structure laid out by the Court in the GSS case.77 The Attorney General reiterated his commitment not to prosecute investigators who apply force in “ticking bomb” situations.78 According to media reports, the desire to avoid international repercussions likely to follow legislation authorizing torture weighed heavily against explicitly empowering the GSS to use force in interrogations.79 It remains to be seen whether the GSS case will prove operable. Hopefully, GSS investigators will abide by the decision and refrain from using force, but still be able to foil terrorist attacks. Hopefully, the Court will abide by its own ruling and issue injunctions in real time against the use of force, including in a clear “ticking bomb” situation; the reader will recall that the Court’s reasoning viewed the use of force as ultra vires current Israeli law in all circumstances. Other legal challenges may lie ahead: a possible review of an Attorney General’s 75 Cf. the reaction of B’tselem, www.betselem.org, under the topic “torture”, and T Talmor, (ed.), B’tselem Quarterly, October 1999, at 1. 76 See supra, n. 13 and G Allon, “The Likud initiates Special Session to Debate Freeze on Legislation Allowing GSS to Torture” Ha’aretz, 18 February 2000. 77 Y Melman, “Shin Bet Drops Its Demand For ‘Torture’ Law” Ha’aretz, 17 February 2000. According to the report, a budget increase for more investigators and better electronic surveillance equipment helped secure the acquiescence of the GSS Chief. It should be noted, however, that the Prime Minister stated in response to a query in the Knesset that in principle he is in favour of providing an a priori authorisation for GSS investigators to apply force in “ticking bomb” situations. G Allon “PM Says Force Is Sometimes Needed”, Ha’aretz (15 March 2000). 78 Ibid. 79 Ibid.

Israel and the Recognition of Torture 657 decision not to prosecute GSS investigators who used force in “ticking bomb” situations; the civil liability of the State and GSS investigators for using force in “ticking bomb” situations; and a possible suit by victims of terrorist attacks (should the dire circumstances arise) against the state and the GSS for not doing enough to prevent terrorism. Similarly, it remains to be seen what position foreign courts will take regarding the GSS case should litigants seek to sue GSS officials abroad and especially should they invoke the GSS judgment as itself evidence of the practice of torture. In that context, the temporal, or prospective effect of the GSS ruling should be noted: torture alleged to have occurred prior to the ruling may be the most likely to be subject to suit, at least where a foreign court system’s law on time limitations permits.80 In contrast to the absolute ban on use of force found in the Committee against Torture’s interpretation of the Convention, the GSS case explicitly ruled that in “ticking bomb” situations GSS investigators will be sheltered from criminal liability through the defence of necessity, thus implying that the use of force in such circumstances does not carry the same moral blameworthiness as the use of force in other circumstances. Assume for present purposes that the courts extrapolate from the necessity defence in the Israeli Penal Code and find an analogous common law necessity defence in civil suits.81 This Israeli law modification of the ban on use of force might create situations where an Israeli official acts in a way that is legal in Israel, but would be illegal in other states had the same conduct occurred there. In such cases, as in many cases of private and public international law, domestic courts will have to translate complicated and delicate issues of individual rights and collective interests into bottom lines of criminal responsibility and punishment, and of tortious liability and remedy. For example, former Prime Ministers of Israel or former GSS heads may, in the aftermath of Pinochet, be indicted or sued abroad. GSS investigators who apply force in “ticking bomb” situations may enjoy the necessity defence in Israel, but might find themselves exposed in foreign jurisdictions. Consequently, Israeli courts might find themselves having to assess the enforceability of civil judgments rendered in the foreign state, including judgments that may well cite the reasoning in GSS as a basis for finding tort liability. Moving from transnational adjudication in municipal courts to international adjudication, or processes before international organs, Israel, or citizens of 80 Contrast the ten-year limitation period in the US Torture Victims Protection Act to the recent ruling by an Ontario court that, inter alia, time had run out on a tort action brought against Canada by the family of a young Somali who had been tortured to death in Somalia by Canadian soldiers: see Abukar Arone Rage and Dahabo Omar Samow by their Litigation Guardian Abdullahi Godah Barre v. The Attorney General of Canada (unreported, 6 July 1999, Ontario Superior Court of Justice, Cunningham J). 81 This extrapolation is by no means logically required given the different consequences of criminal responsibility (prison) and civil liability (monetary payment, indeed payment that would in all likelihood be underwritten by the state even when a GSS official is sued in his or her official capacity).

658 A Reichman & T Kahana Israel, may be brought to trial before the International Criminal Court (ICC) or other supranational bodies for the use of force in “ticking bomb” situations, should such means of interrogation continue past the date on which the ICC Statute enters into force. Future developments may also include revisiting the domestic legislative option in Israel. The government or the opposition may seek to empower the GSS to use force after all. However, the current decision, to let the GSS case stand unmodified, may be seen as reflecting growth. From a country which has traditionally argued that the constant threat of terrorist attacks justifies special means which are unacceptable in other democracies, Israel is maturing and joining the group of nations which takes the rule of law— both the domestic rule of law and the international rule of law—seriously.

25

An Uncivil Action: The Tort of Torture and Cosmopolitan Private Law MAYO MORAN 1

1 INTRODUCTION

curious quality to the American tort of torture cases.2 For close to two decades now, federal courts in the United States have been adjudicating responsibility for torture and other fundamental human rights abuses committed abroad. Thus, for example, courts have considered the tort liability of Ferdinand Marcos,3 the former President of Haiti,4 Radovan Karadzic,5 as well as of officials in various regimes including Paraguay,6 Argentina,7 Guatemala,8 the Ethiopian “Dergue”9 and many others. Yet, at first blush these cases are oddly unsatisfying. They should rank among the most interesting of all contemporary torts cases, but tort theorists don’t write about them. And perhaps reflecting a similar sense that these cases are “something else”, leading and comprehensive textbooks on the law of torts don’t even mention them.10 But this is only the beginning of the puzzle. Though arguably overwritten in some ways, the cases are also curiously undertheorised. Indeed, on initial examination, this hardly seems surprising: most of the cases are remarkably short and the reasoning notably thin. In some sense,

T

HERE IS A

1 Associate Professor, Faculty of Law, University of Toronto. This article outlines one dimension of a research project on norms in a complex world that I am undertaking with my colleague Karen Knop. It has benefited enormously from her input. I can am also grateful to Attila Attaner and Pauline Rosenbaum for their invaluable research assistance. 2 These cases arise from two enabling statutes in the United States, the Alien Tort Claims Act (ATCA) 28 U.S.C. ss.1350, and the Torture Victim Protection Act (TVPA), Pub.L.No. 102–256, 106 Stat.73 (1992) (codified at 28 U.S.C. ss. 1350 note (1994). See overview of the case law and issues litigated under ATCA and the TVPA in M Swan, “International Human Rights Tort Claims and the Experience of United States Courts: An Introduction to the US Case Law, Key Statutes and Doctrines”, chapter 3 in this volume. 3 Marcos Estate I 978 F.2d 493 (9th Cir.1992); Marcos Estate II, 25 F.3d at 1475 (9th Cir. 1994). 4 Lafontant v. Aristide, 844 F.Supp. 128 (E.D.N.Y. 1994). 5 Kadic v. Karadzic, 70 F.3d 232 (2nd Cir. 1995). 6 Filártiga v. Peña-Irala, 630 F.2d 876 (2d Cir.1980); 577 F.Supp. 860 (E.D.N.Y. 1984). 7 Forti v. Suarez-Mason, 672 F.Supp. 1531 (N.D. Cal. 1987). 8 Xuncax v. Gramajo, 886 F.Supp. 162 (D.Mass. 1995). 9 Abebe-Jira v. Negewo, 72 F.3d 844 (11th Cir. 1996). 10 For instance, see Prosser and Keeton on Torts, 5th ed., (St. Paul, Minn., West Publishing Co., 1984).

662 M Moran then, the fact of the cases seems more interesting than the cases themselves (perhaps accounting for the relative lack of interest on the part of private law theorists). But this inattentiveness, I shall argue, is a mistake. Despite their unpromising appearance, the torture cases repay the effort of serious study. Indeed, there are reasons for private law theorists as well as those concerned with public international law and human rights to pay attention to them. At the most obvious level, the torture cases arguably provide a glimpse of the challenges that will face the private law of the future. In this respect, these cases are but the cutting edge of a much broader phenomenon.11 And the torture cases hold other deeper, though related, lessons. But these lessons are somewhat buried—revealing themselves only in the tension between what courts say they are doing and what they actually do. Yet here, beneath the unpromising surface, lies perhaps the most important feature of these cases: these tensions reveal the extent to which the rapidly changing geography of the law is challenging the very foundations of the modern legal tradition. And one can also detect the emergence of an alternative picture of judgment in an increasingly interconnected world. Yet, while in some sense new, this picture is also oddly familiar. Like the old common law, the sources of law in this evolving legal field are multiple, diverse and often conflicting: the judicial task is thus constructive and integrative. Authority is on this view (as perhaps it always was) as much created as it is inherited, and a persuasive exercise of that authority comes only through constructive engagement of the local(s) and the distant. The significance of place, one might say, has shifted but not disappeared.

2 THE TORT OF TORTURE AND THE SHIFTING LEGAL LANDSCAPE

Law, like other major contemporary institutions, is in the midst of a dramatic shift. The borders of legal argument are undergoing rapid change as a whole variety of boundaries become increasingly porous. Theorists and practitioners alike struggle to make sense of these changes. Domestic courts in particular are confronted with unprecedented challenges. The recent extradition proceedings of General Augusto Pinochet focused international attention on this fact, but the Pinochet case is part of a much larger pattern. Recent years have also witnessed the proliferation of a whole range of legal actions—often private law in form—that challenge domestic judges with intricate webs of public and private law claims, as well as complex multijurisdictional sources and litigants. In this respect, the “tort of torture” exemplifies an emerging type of action that poses a fundamental challenge to our traditional understanding of the legal system. Perhaps the most dramatic and noted feature of the changing legal landscape is found in the contemporary zeal for comparativism. Constitutional law, for 11 Harold Koh describes a related phenomenon in “Transnational Public Law Litigation”, (1991) 100 Yale L.J. 2347.

Torture and Cosmopolitan Private Law 663 instance, is in many ways the cornerstone of this development. And this despite the fact that constitutional law has traditionally been thought of as the least appropriate candidate for fruitful analysis—an orthodoxy that contemporary judicial and academic practice is rapidly calling into question.12 But the increasing interpenetration of legal systems is by no means confined to the constitutional sphere. To some degree, private law has always involved an important and unavoidable comparative element. In Canada at least this is so central to our understanding of private law that we typically do not even think of the task of examining the jurisprudence of other jurisdictions as “comparative”. In fact, the sources of the common law have always been complex and of varied origin.13 But, even given this inherent hybridity, the common law itself is also in a period of unprecedented development. Commentators have noted that the “process of learning from each other” is gathering strength and will most certainly not be confined to common law countries.14 Indeed, this increasing interaction and even interpenetration of common law is such that it has revived the dream of “an International Common Law”.15 These developments in which courts (and other institutions) routinely look beyond their own jurisdictions have prompted Anne-Marie Slaughter to note the rise of what she calls the “real new world order”.16 She describes this new world order as a form of “transgovernmentalism”, characterised by “horizontal” communications between networks of courts, agencies and institutions in various places in the world. The legal world, as Slaughter and others suggest, is increasingly characterised by more interaction between jurisdictions and much more “looking around”. But the increasing interpenetration of legal norms and perhaps legal systems does not end here. The world of contemporary law may be characterized by increased recognition of the fluidity of the boundaries of the state and the relevance of “external” sources to “internal” norms, but there are also boundaries beyond those of the state at issue.

12 See for instance, Sujit Choudhry, “Globalization in Search of Justification: Toward a Theory of Comparative Constitutional Interpretation”, (1999) 74 Indiana L.J. 819 at 820–2 discussing Princz v. United States 117 S.Ct.2365 at 2377 and noting that Scalia J.’s majority opinion that comparative analysis is “inappropriate to the task of interpreting a constitution” is increasingly out of step with constitutional practice as well as constitutional theory. See also, Claire L’Heureux-Dubé, “The Importance of Dialogue: Globalisation and the International Impact of the Rehnquist Court” (1998) 34 Tulsa L.J. 15 and Mark Tushnet, “The Possibilities of Comparative Constitutional Law” (1999), 108 Yale L.J. 1225. 13 Rt. Hon. Lord Bingham L.C.J., “The Future of the Common Law” (1999) 18 Civil Justice Quarterly 203. 14 Ibid. at 143. 15 Lord Cooke of Thorndon, “The Dream of an International Common Law” in Cheryl Saunders (ed.), Courts of Final Jurisdiction: the Mason Court in Australia (Annandale, NSW: Federation Press, 1996) at 138. 16 “The Real New World Order” (Sept./Oct. 1997) 76:5 Foreign Affairs 183. See also Anne-Marie Slaughter, “A Typology of Transjudicial Communication” (1994) 29 U. Richmond L. Rev. 99 and “Governing the Global Economy Through Government Networks” in Michael Byers, (ed.), The Role of Law in International Politics (Oxford: Oxford University Press, 2000) 205.

664 M Moran In fact, this developing porousness of previously sacrosanct lines between jurisdictions is accompanied by more attentiveness to the problematic nature of other kinds of legal “boundaries”. This trend has manifested itself even within the heart of traditional private law, as courts increasingly move away from an either–or approach to the relationship between contract and tort, to cite but one example.17 And just as private law itself is increasingly not seen as composed of watertight compartments, so too is there a developing sense that private or civil law does not exist in its own hermetically sealed world but instead interacts in important ways with the larger legal order. In the United States, this was acknowledged in New York Times v. Sullivan when the Supreme Court of the United States held that the guarantees of freedom of speech applied to the common law of defamation.18 And in Canada, the courts have stressed that, although the Charter does not apply directly to the common law, the common law must nonetheless develop in a manner consistent with the Charter.19 And in jurisdictions without constitutionalised bills of rights, the interactions are if anything even more striking. Thus, for instance, we find the English Court of Appeal treating a wide array of precedents, including those of Australia, Canada, and the United States as relevant to the interpretation of the common law of libel.20 In Dietrich v. The Queen, the Australian High Court discussed the influence of the constitutional law of various jurisdictions, the European Convention on Human Rights and international obligations on Australian common law.21 Similarly, in an important recent book, Murray Hunt notes the growing importance of human rights law (particularly but not exclusively European) to the elaboration of English law, including administrative and common law.22 It is within this larger context of increasing interpenetration of legal norms and systems and the related demise of a separate-spheres approach to law that the “new” private law actions are found. Indeed, these actions constitute a kind of frontier because of the complications they introduce not only about jurisdiction but also about bodies of law, sources of legal rules and the relations between them. Although the developments which gave rise to this cosmopolitanization of private law can arguably be traced back to Nuremberg at least, it is in the last two decades 17 Key Canadian developments are Central Trust v. Rafuse (1986) 31 D.L.R. (4th) 481 (S.C.C.) holding that concurrent liability could arise in contract and in tort, and BG Checo International Ltd. v. B.C. Hydro and Power Authority (1993), 99 D.L.R. (4th) 577 (S.C.C.) holding that the right to sue in tort remains intact notwithstanding the presence of a contract unless the contract expressly excludes tort liability. See also, Stephen Waddams, “Can Private Law be Mapped” (Toronto, 1999, manuscript on file with the author). 18 (1963) 376 U.S. 254. 19 Canadian Charter of Rights and Freedoms, Part I of the Constitutional Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), c.11. [hereinafter Charter]. See RWDSU v. Dolphin Delivery, [1986] 2 S.C.R. 573; Dagenais v. CBC [1994] 3 S.C.R. 835; Hill v. Church of Scientology [1995] 2 S.C.R. 1130. 20 Derbyshire C.C. v. Times Newspapers [1992] 3 W.L.R. 28. See also Australian Capital T.V. P/L v. Commonwealth [No.2] (1992) 66 A.L.R. 695. 21 (1992), 67 A.L.R. 1. 22 Human Rights Law in English Courts (Oxford: Hart Publishing, 1997).

Torture and Cosmopolitan Private Law 665 of the second millennium that such litigation has begun to move into the heart of the traditional legal system—private law claims in domestic courts. The explosion of contemporary litigation began in the United States with the landmark decision in Filártiga v. Peña-Irala—an action brought in the US federal court for torture that occurred in Paraguay.23 And since that time, the civil action for torture has perhaps been the most prominent example of this new private law action that increasingly faces domestic courts. Indeed, American courts have adjudicated dozens of civil actions for torture committed in other jurisdictions.24 For this reason, the torture cases serve as a useful point of departure for examining how the changing geography of legal argument forces a fundamental reconsideration of the way that we think about law, the state and the relations between them. In fact, the tort of torture is only the most dramatic exemplar of the kinds of private law actions that increasingly confront domestic courts. Perhaps the most prominent other example of these new private law claims is the Holocaustrelated litigation.25 This rapidly growing category encompasses a wide array of suits, including claims that various companies profited form Nazi actions as well as actions against banks that hid assets stolen from Holocaust victims. Suits have also alleged that various major German companies including DaimlerChrysler, Volkswagen, BMW and Siemens profited from slave labour drafted from concentration camps. Thus, for instance, in March of 1998, Elsa Iwanowa filed a class action complaint in the District Court of New Jersey against Ford Motor Company for its use of forced labour.26 And in February 1999, Eva Kor filed a class action suit alleging that Bayer AG participated in the horrifying Mengele twin experiments.27 Litigation of this kind is not limited to Holocaustrelated wrongs. In fact, US veterans recently launched a class action against the Japanese government and various companies including Mitsubishi based on forced slave labour.28 Like the torture cases, these cases invoke complex questions of jurisdiction and of the appropriate sources of law as well as of the relations between various sources within and beyond a given jurisdiction. These new private law actions—exemplified by but not limited to the tort of torture—are thus characterised in part by how they complicate various “boundaries” long considered sacrosanct. In fact, as we shall see, the torture cases suggest that it is precisely this feature that makes unavoidable the worries about the increasing anachronism of our traditional separate-spheres model of law. The 23

Supra n. 6. For a survey of cases see Koh, supra, n. 11; Curtis Bradley and Jack Goldsmith, “The Current Illegitimacy of International Human Rights Litigation”, 66 Fordham L.Rev. 319 (1997), and “Customary International Law as Federal Common Law: A Critique of the Mordern Position”, 110 Harvard L.Rev. 815; Clyde H Crockett, “The Role of Federal Common Law in Alien Tort Statute Cases”, 14 B.C. Int’l & Comp.L.Rev. 29 (1991). 25 For an overview of these actions see Michael J Bazyler, “Litigating the Holocaust” (1999), 33 U. Richmond L.R. 601. 26 Iwanowa v. Ford Motor Company, and Ford Werke A.G. (1999) 67 F.Supp. 2d 424. 27 Kor v. Bayer A.G., Class Action Complaint and Jury Demand, February 17, 1999 (United States District Court, Southern District of Indiana, Terre Haute Division) Civil Action No. 2:99–36. 28 In re World War II Era Japanese Forced Labour Litig., (2000) 114 F. Supp. 2d 939. 24

666 M Moran tensions and ambiguities in the reasoning of those cases intimate that the traditional model actually obscures the extent to which our legal world is increasingly characterised by interactive and even hybrid norms—norms which therefore demand a different and more constructive understanding of the doctrine of sources and of the judicial role. But in order to see the extent of the challenge they pose to our traditional understanding of law, it is first necessary to remind ourselves of the structure of the traditional model.

4 SEPARATE SPHERES : CONFLICT , CHOICE AND THE TRADITIONAL MODEL

At least since the rise of the modern nation-state, the Western legal imagination has had its own distinctive geography, dominated by the concept of binding law.29 This notion, which not only animates traditional legal theory but also structures its debates with its “radical” counterpart,30 is characterised by an emphasis on formally articulated rules which oblige judges within a given territory. Clearly delineated boundaries are thus central to this understanding of law. A “radically formal” concept of binding law also dominates the conception of the judicial role as legitimately exercised only in the task of identifying and applying the binding legal rule. On the most extreme versions of this model, law is the expression of the Nation State which is “the natural forum for making local, substantive, precise law” as well as “the ideal instrument for seeking compliance with it”.31 The unsurprising concomitant of this view is the suspiciousness with which the traditional domestic legal system approaches foreign law.32 Similarly central to this model of law is a formalistic picture of the division of the domestic legal system into sharply distinct and mutually exclusive “bodies” of law.33 An understanding of authority that places such a premium on sharp boundaries can be maintained only through a system of decision-rules that establish how to choose among the possible candidates for binding law. Thus, this model of law also relies heavily on meta-rules for determining whose legal rules are plausible candidates for the binding law in any particular case, essentially conflict of laws. On this separate-spheres understanding of law, choice 29

H P Glenn, “Persuasive Authority” (1987), 32 McGill L.J. 262. See, M Moran, “Right Answers and Wrong Questions: Objectivity, Feminism and the Partiality of Legal Theory” (work in progress; manuscript on file with the author) 31 Glenn, supra n. 29 at 279. Perhaps for this reason, this period tended to prefer legislative reform and other methods of generalisation such as treatises, over the presumably more unmanageable and inherently particularistic common law. 32 Similarly, the commitment to a sharp delineation—in legal and other terms—between self and other is actually necessary to preserve the traditional understanding of comparative law, and perhaps even international law. Without this distinction, the comparative project is incomprehensible. Perhaps for this reason, Glenn argues that comparative law is a moribund discipline that only has meaning in a system committed to a strong version of binding law (H P Glenn, Public Lecture, Toronto, Ontario, October 1999). Thus, he notes the flourishing of comparative law in the United States. Presumably elsewhere, comparative law is just law. 33 Stephen Waddams, Lecture, Faculty of Law, University of Toronto, December 1999. See also Waddams, supra n. 17. 30

Torture and Cosmopolitan Private Law 667 among various contenders for the sources of law largely exhausts the judicial task, which is conceived of as identifying, then applying “binding” law. Indeed, the centrality of the traditional understanding to our thinking about law may make even this very simplistic description seem uncontroversial. But this understanding actually drives much of what comes to seem puzzling in the civil actions for torture—plagued as they are by an “embarrassment of riches”. So, for instance, domestic legal systems contain prohibitions against torture both in private law (assault and battery) and in criminal law (assault or more likely aggravated assault). Similarly, specific constitutional prohibitions against torture as well as more general clauses on “cruel and unusual punishment or treatment” are foundational to bills of rights.34 In addition, prohibitions against torture are among the least controversial of the various kinds of supranational norms and are found not only in international conventions but also in various other international instruments as well as in regional rights-protecting regimes.35 Undoubtedly this is simply a restatement of the obvious. But there is a reason why we may need to be reminded of the obvious here. Viewed against this backdrop, both the cases on torture and related academic discussions have a curious quality—a quality that can ultimately be traced back to the hold of the traditional model on legal thinking. One might expect that the arguments would draw on the rich array of resources offered by various fields of law within and across legal systems. Indeed, this would seem eminently desirable. After all, the very pervasiveness of the norms against torture may plausibly be thought to be testimony to their centrality and to the fundamental importance of the interest they protect—human dignity. Thus, the task of justification ought to be enormously aided by an integrative understanding of the various norms implicated, for instance, in a civil action for torture. Yet, at the most obvious level, this is exactly what the torture cases do not do. Instead, courts and commentators in these cases emphasise the importance of choosing which norms ought to apply—a question which is generally resolved in favour of supranational law. And, perhaps unsurprisingly given the general either-or disposition that is a corollary of the separate-spheres orientation of the binding law model, other sources and their normative possibilities are minimised or marginalised. 34 Thus, the Court of Appeals in Filártiga, supra n. 6 at 884 n. 12, points out that as of 1980 torture was prohibited in over fifty-five nations citing C Bassiouni and D Derby “ Appraisal of Torture in International Law and Practice: the Need for an International Convention for the Prevention and Suppression of Torture” (1977) 48 Revue International de Droit Penal 23. 35 Universal Declaration of Human Rights, GA Res. 217A (1948) art. 5; International Covenant on Civil and Political Rights (ICCPR), GA/Res. 2200 (XXI) reproduced in (1967) 6 I.L.M. 368, art. 7(1); Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment, UN Doc. A/Res/39/46 reproduced in (1984) 23 I.L.M. 1027; European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment, (1987) E.T.S. No. 126; Inter-American Convention to Prevent and Punish Torture, OEA/Ser.P, AG/doc 2023/85 rev. 1, 12 March 1986 reproduced in (1986) 25 I.L.M. 519; African Charter on Human and Peoples’ Rights, (1982) 21 I.L.M. 58.

668 M Moran This conflict-of-laws (or, conflicts) orientation of the traditional model when faced with a multiplicity of sources of norms helps to explain some of the more curious features of the civil actions for torture. For instance, without appreciating the backdrop this model provides, the preoccupation of the courts (and the commentators) with the choice of a source for the substantive norms is perplexing. After all, what most distinguishes the problem of torture is not the multiplicity of conflicting norms but rather the striking convergence of virtually identical ones.36 Nonetheless, even in this situation of minimal conflict among substantive norms, courts view the “choice” of a particular body of norms as fundamental and deeply controversial. And commentators generally agree. The idea, it seems, of watertight spheres of normativity is so profound that choice remains the predominant controversy even where any one of the “choices” would yield essentially the same substantive norms. The model’s commitment to the centrality of choice among presumably conflicting and mutually exclusive spheres also largely accounts for the otherwise baffling treatment of domestic law. Since international norms typically get chosen in these cases, the norms of the domestic legal order are minimized or ignored. And while this emphasis on the supranational sphere as the locus of normativity may be understandable, it is worth noting how indebted even the human rights rhetoric is to the binding law model. It is true that international lawyers, particularly those concerned with human rights, have increasingly begun to look to domestic courts to “solve” the long-standing “enforcement” problem of international law.37 But significantly, their reasons for looking to domestic courts are primarily strategic and not normative.38 This normatively thin picture of domestic courts is also implicit in the very language of “enforcement” used to describe the role that internationalists are prepared to confide to domestic courts. And international law’s term of art for domestic law—municipal law—echoes this view of the international sphere as the important locus of 36 This is not to suggest, of course, that torture is unproblematic in the contemporary world. As Amnesty International and other organisations that work against torture make only too clear, the practice of torture is, notwithstanding its universal condemnation, horrifyingly widespread. 37 See also Karen Knop, “Here and There: International Law in Domestic Courts” (forthcoming 2000 N.Y.U.L.Rev.) discussing the same phenomenon and citing Benedetto Conforti and Francesco Francioni, (eds.) Enforcing International Human Rights in Domestic Courts (The Hague: Martinus Nijhoff, 1997); Thomas M Franck and Gregory H Fox, International Law Decisions in National Courts (Irvington-on-Hudson, N.Y.: Transnational Publishers, 1996); Hunt, Using Human Rights Law in English Courts, supra, n. 22; Eyal Benvenisti, “Judicial Misgivings Regarding the Application of International Law: An Analysis of the Attitudes of National Courts” (1993), 4 E.J.I.L. 159. Various reports and studies are also under way including the work of the International Law Association’s Committee on International Law in National Courts (London: International Law Association, 1996 and 1998, for instance), and an ASIL-sponsored project on the application of international environmental law in domestic courts, preliminary results published in (1998) 7 Review of European Community and International Environmental Law. 38 See examples, ibid. Another illustration is found in A. Clapham Human Rights in the Private Sphere (Oxford: Oxford University Press, 1993), where Andrew Clapham describes the strength of the European Convention as its “ability to straddle national and international dimensions, synthesising the international standards with national enforcement procedures” (at 9).

Torture and Cosmopolitan Private Law 669 normativity. So it looks like domestic law provides the machinery while international law provides the normativity.39 But it seems clear that the traditional picture is undergoing a dramatic shift in its contours. Individuals and groups seeking to vindicate their rights—including fundamental dignity and equality rights—are increasingly looking to the resources of private law and raising claims that fuse private and public, domestic and international norms. And for domestic courts confronted with these complex claims, the traditional conflicts frame of reference seems inapt. So too is there is a growing sense—especially among advocates of international human rights—that the traditional language of monism and dualism is inadequate to the more complex process that is already well under way in our courts.40 Indeed, the civil action for torture reveals the extent to which judges faced with such claims struggle against the confines of the traditional understandings in order to justify their conclusions. And in their reasoning, the judges seem impelled to articulate the relationships between these disparate norms even as they feel pressed to apparently choose among them. 4 CONFLICT , AMBIGUITY AND THE TRADITIONAL PICTURE

In fact, the very reasoning of the courts in adjudicating the civil action for torture is illustrative of the strain that these new private law actions inevitably place on the traditional model of binding law. In the torture cases, this strain makes itself felt, for instance, in the tension between what courts say about the sources of law and what sources they actually invoke in their reasoning. The controversy about the appropriate sources of law for the tort of torture is apparent, even in the landmark decision of Filártiga itself. However, it is really with Tel-Oren that the question of the appropriate sources of law is sharply posed.41 And it is largely in response to Tel-Oren that the trial judge in Xuncax v. Gramajo42 feels compelled to exhaustively address the debate about the 39 This echoes the odd symbiosis between international law theory and traditional legal theory (which has overwhelmingly focused on domestic law). Thus, for instance, both tend to treat the dominant characteristic of national or domestic law as its coercive power and the dominant characteristic of international law as its very absence of that power. It is simply over the significance of those characteristics that they differ. Thus, since traditional legal theory tends to treat coercive power as the sine qua non of law, it is primarily concerned with the conditions under which the exercise of that power can be justified. On this view, since international law largely lacks this power it is, in an important sense, not really law and thus theoretically uninteresting. As noted in the text, international law also identifies the domestic sphere as the locus of coercive power, but treats this simply as a convenient solution to the enforcement problem in international law. 40 For instance, while Murray Hunt seems at times to invoke a fairly traditional monist stance, at other times he expresses frustration with the limitations of this approach—and with its unidirectionality in particular: Hunt, supra n. 22. As discussed, infra, n. 100 and accompanying text, a similar ambivalence is also apparent in the work of Harold Koh who at times seems to adopt a relatively traditional view that treats the normativity as coming from above, but who at other times suggests that the domestic and private side of the equation should play a more robust role. 41 Tel-Oren v. Libyan Arab Republic, 726 F.2d 774 (D.C.Cir. 1984). 42 Supra, n. 8.

670 M Moran appropriate sources of law. What these decisions ultimately reveal is the strain the traditional model faces as legal boundaries and even the foundational concept of sovereignty itself shift and become more fluid.43 As noted above, courts and commentators routinely stress that torture and like wrongs are international crimes and, as such, violations of international law (or, per ATCA, the “law of nations”). Thus, the Court of Appeals begins its decision in Filártiga by stressing how the fledgling thirteen-colonies America embraced the accepted norms of international law as a vital feature of its constitutional order.44 Indeed, this emphasis on the universal sphere is evident throughout the opinion. Thus, the Court stresses the “universal condemnation of torture in numerous international agreements”45 and attempts to distinguish between wrongs of mutual—versus those of “merely several”—concern.46 And the decision closes by noting the relationship between the Nuremberg ideal and international human rights law. It is therefore perhaps unsurprising that domestic law—including constitutional law—is treated as marginal, perhaps even normatively inferior. For instance, in the long discussion of the various international, and suprantional, prohibitions against torture, a mere sentence is devoted to the fact that similar prohibitions are also reflected in “modern municipal [i.e. national] law as well”.47 Similarly terse mention is made of the fact that torture is prohibited by the constitutions of over fifty-five nations.48 Indeed, in an excerpt, the Court quotes with approval a commentator who describes domestic matters as those which international law has “left for regulation by states”.49 Despite this insistence on the centrality of international law and the relative unimportance of domestic law, the message that the Court’s reasoning actually conveys about the sources of law is a complicated one. Thus, alongside the emphasis on international law, one finds repeated references to the domestic law of the most implicated jurisdictions, namely Paraguay and the United States. Interestingly, however, much of this discussion is either buried in footnotes or simply noted in passing, typically in a subordinate clause on the way to another point. Thus, for instance, the Court points out on several occasions, with most of the discussion occurring in footnotes, that Paraguayan law would afford a civil remedy.50 43

Neil MacCormick, “Beyond the Sovereign State” (1993), 56 Modern L. Rev. 1. Supra, n. 6. 45 Ibid. at 880. See also 881 (few issues on which opinion seems to be so united as the limitations on a state’s power to torture persons held in its custody”), 882 (“no dissent” from the view that the “right to be free from torture” is the bare minimum of human rights guarantees), 883–5 (outlining sources of international and supranational law prohibiting torture. 46 Filártiga, supra n. 6 at 888. Distinguishing torture from theft, which is several because prohibited everywhere and yet not mutual because not a violation of the law of nations. 47 Ibid. at 884. 48 Ibid. 49 Ibid. at 888 quoting Preuss, “Article 2, Paragraph 7 of the Charter of the United Nations and Matters of Domestic Jurisdiction,” Hague Receuil (Extract, 149) at 8, reprinted in H Briggs, (ed.) The Law of Nations: Cases, Documents and Notes (2nd Ed.) (New York: Appletor Century-Crofts, 1952) at 24. 50 See, for instance ibid. 879, n. 5 quoting the affidavit of Paraguayan counsel and 889, n. 25. 44

Torture and Cosmopolitan Private Law 671 Similarly, the Court notes that the acts in question would also violate not only Paraguayan civil law but also the constitution of Paraguay.51 Mention is also made—again in a footnote—of the fact that the conduct alleged here also violates the US Constitution and, if performed by a government official, would be an actionable deprivation of civil rights.52 And although much of the reasoning is concerned with subject-matter jurisdiction, the Court does insert a very abbreviated consideration of the choice of laws and suggests—again in a footnote—that when the District Court considers the case on the merits, it may well decide that “fairness requires it to apply Paraguayan law”.53 So although the Court of Appeals notionally leaves open the question of what substantive rules apply, the reasoning and rhetoric is such that the District Court’s subsequent preference for the norms of international law seems entirely unsurprising. Throughout this discussion of what substantive rules apply to actions for torture (the choice of law question), Nickerson J. echoes the Court of Appeals’ depiction of the spheres of normativity as mutually exclusive. He is thus constrained to choose either domestic or international law. And in part because of this separate-spheres view, the Court’s emphasis on the universal comes at the expense of denigrating the domestic. Thus, he stresses that this is not “an ordinary case of assault and battery”.54 Torture is sharply contrasted with the “ordinary” wrongs of domestic private law: “[T]orture is viewed with universal abhorrence; the prohibition of torture by international consensus and express international accords is clear and unambiguous.” The torturer is, in the famous words of the Court of Appeals, “the enemy of all mankind.”55 Thus, despite the latitude apparently given by the Court of Appeals, the substantive principles derive from international law.56 Yet, the insistence on the universal nature of the abhorrence of torture could actually have been fortified, not diminished, by clarifying the extent to which this kind of view is also enshrined in the domestic legal system, including in the Eighth Amendment.57 Indeed, it is arguable that, by using the ordinary conflict of laws language and in its characterisation of the choice of law question, the Court actually undermines to some degree the suggestion that there is universal abhorrence. But the complexities of jurisdiction in Filártiga do not end here. The limitations of the traditional model seem to account for another peculiarity of the District Court’s reasoning. Despite its insistence that the norms regarding torture derive exclusively from the international sphere, the actual reasoning of the 51

Filártiga, supra n. 6 at 885 (in a subordinate clause), and at 884, n. 14. 42 U.S.C. §1983 (Civil action for deprivation of rights secured by law or by the Constitution). Ibid. at 889, n. 25. 54 577 F.Supp. at 863. 55 Ibid. 56 Declaration on the Protection of All Persons from Being Subjected to Torture, General Assembly Resolution 3452, 30 U.N. GAOR Supp. (No. 34) 91, U.N. Doc. A/1034 (1975) adopted without dissent at the General Assembly. Note that the Court’s decision slightly predated the Convention against Torture (1984). 57 “Excessive bail shall not be required, nor excessive fines imposed, nor cruel or unusual punishment inflicted” (Amendment VIII, 1791). 52 53

672 M Moran Filártiga court suggests otherwise. Significantly, once the Court has established that the international norms are the only appropriate source of law, its consideration of Peña-Irala’s culpability makes virtually no mention of international law—not even the UN Declaration Against Torture.58 Thus, Section III of the decision begins by re-emphasising the importance of the international norm against torture, only to continue in the following way: “Clearly, the court should consider the interests of Paraguay to the extent they do not inhibit the appropriate enforcement of the applicable principle of international law or conflict with the public policy of the United States.”59 This seems so obvious that it is easy to overlook the tension with the earlier discussion. And, as the Court continues, the whole focus is on the domestic law of Paraguay not on international norms. The Court emphasises that “written Paraguayan law prohibits torture, that Paraguay is a signatory to the American Convention on Human Rights, and that Paraguayan law purports to allow extensive recovery for wrongful death, including punitive and moral damages.”60 Indeed, notwithstanding its earlier insistence that mere domestic law is inadequate to these situations, the importance of the convergence between domestic and international law is evident throughout this consideration of the substantive issues in the case. So the signal sent by the Court of Appeal about the possible relevance of Paraguayan law emerges as not only appropriate but crucial to ensuring the “fairness” of the decision. In fact, by the end of this section, the Court describes its analysis as the application of “the written law of Paraguay”.61 Similarly, the Court later gives American damages principles a background interpretive role in reliance on traditional conflict of laws rules pointing to the application of the lex fori on this issue.62 In this way, the Court’s actual reliance on a multiplicity of sources, many of which are in fact domestic to the relevant jurisdictions, complicates and even undermines the approach it says it is taking when it insists that the substantive principles that govern such actions are the exclusive province of international law. Notwithstanding these complexities in the actual sources relied on in Filártiga, the standard reading of the case was that §1350 gave jurisdiction to the Federal Court to determine whether the defendant’s act violated international law—there, the prohibition against torture.63 But the exact origin and shape of the cause of action in tort had become controversial. By 1993, a number of possibilities were on the table. Perhaps the dominant understanding was that §1350 granted both jurisdiction and a federal private cause of action where the actions

58 The Declaration was the predecessor to the UN Convention against Torture which had yet to be adopted at the time of Filártiga. 59 577 F.Supp.860 at 863–4. 60 Ibid. at 864. 61 Ibid. 62 Ibid. at 864–7. 63 See, for instance, Marcos Estate II, supra n. 3.

Torture and Cosmopolitan Private Law 673 complained of violate international law.64 But the Ninth Circuit had also concluded that plaintiffs may look to domestic law in a §1350 action.65 The most complex of the possible interpretations of the §1350 cause of action was found in the D.C. Circuit Court of Appeal decision in Tel-Oren.66 Edwards J., Bork J. and Robb J. all issued separate opinions dismissing the claim of the survivors and representatives of those murdered during an armed attack on a civilian bus in Israel. Bork J.’s concurrence insists that since §1350 operates only to give jurisdiction, a further inquiry is necessary to determine whether either international law or a federal statute provides an independent private cause of action with respect to the conduct in violation of the “law of nations”. This, combined with his very limited view of international law, would seriously limit the scope of §1350 and it has been criticised on that basis among others.67 However, the analysis that is more illuminating for our purposes here is that found in the concurrence of Edwards J. Because deriving concrete standards of liability from “amorphous” international law norms places “an awesome duty on federal courts,” Edwards J. outlines an alternative approach to §1350 which finds standards of liability in domestic tort law.68 He notes that this reading of Filártiga obviates the difficulty of discerning applicable standards of liability from international law.69 However, he goes on to point out the shortcomings in this approach, primary among them the fact that such an approach would require a conception of the “necessary degree of convergence between the domestic and international law”.70 Yet, despite Edwards J.’s intuition that domestic law may have an important contribution to make to these kinds of actions, his approach ultimately seems hampered by exactly the kind of watertight compartments analysis of Filártiga. And, since he only considers whether either international law or domestic law could found the cause of action, the possibility that the two could be read together in some helpful way is invisible as an option even in his thoughtful analysis. The traditional model thus limits the range of possibilities even here. Notwithstanding this, Edwards, J.’s suggestion that municipal law can also figure in §1350 actions retained enough promise to exert considerable influence over subsequent courts. Perhaps the most extensive discussion to date of the appropriate sources of the substantive rules in claims under the Alien Torts Claims Act is found in the decision in Xuncax.71 There, Woodlock D.J. canvasses the various approaches to §1350, which he divides roughly into the Filártiga approach and what he calls the “domestic law alternative approach”. 64

Ibid.; Forti v. Suarez-Mason, supra, n. 7; Paul v. Avril, 812 F.Supp. 207 (S.D.Fla 1993). Marcos Estate I, supra, n. 3. 66 Supra, n. 41. 67 Xuncax, supra n. 8 at 179–81. 68 Tel-Oren, supra n. 41 at 781. See also, Marcos Estate I, supra n. 3 at 503 (with reference to the jurisdiction of the Philippines). 69 Ibid. at 787, discussing Adra v. Clift, 195 F.Supp. 857 (D.Md. 1961). 70 Ibid. at 788. 71 Supra n. 8. 65

674 M Moran He notes Edwards J.’s suggestion in Tel-Oren that municipal law could be drawn on to provide the kind of relatively definite and concrete standards of liability lacking in international law norms. However, he rejects municipal tort law as a source for a §1350 cause of action. One important—and by now familiar—reason why municipal law is an inadequate source, according to Woodlock D.J., is that it cannot capture the nature of the wrong. As in Filártiga, the sense of mutually exclusive spheres of liability is predominant. So too is the picture of domestic law as intensely local in its concerns. Invoking municipal law would, according to Woodlock D.J., “reduce” the wrong to a “garden-variety municipal tort”.72 By contrast, international wrongs of the kinds addressed in §1350 implicate “all of mankind” in their condemnation.73 And the Court’s trivialisation of the concerns of domestic private law is reinforced through the insistent use of the term “municipal” law.74 This is, of course, international law’s term of art for domestic law, but the narrow connotations of the language are revealing. And, having thus sharply drawn the dichotomy between the grave harms of international law and “garden-variety” municipal torts, the conclusion that international law is the exclusively appropriate tool for redressing human rights violations seems inescapable. In this way, the separate-spheres approach drives the ultimate rejection of any role for domestic law on the ground that it is an inadequate placeholder for the values protected under §1350. Once again, what seems striking is the choice Woodlock D.J. felt compelled to make between municipal law and international law. And, in order to justify his decision to prefer international law, he relegated domestic law to a position of normative inferiority.75 And, indeed, when one examines the sources actually invoked in Xuncax, the parallels to what we saw in Filártiga are remarkable. The reasoning in Xuncax is similarly characterised by the tension between the protestations regarding the innate superiority of international law and the actual sources invoked in the substantive analysis in the case. Even in his discussion of the innate superiority of international law norms over domestic norms, Woodlock D.J. notes in a footnote that “I would reach substantially the same result here if I were to apply 72

Xuncax, supra n. 8 at 183 Ibid. In the accompanying footnote, Woodlock, D.J. questions, for instance, the appropriateness of using a municipal wrongful death statute to address summary executions or disappearances (n. 24). 74 Interestingly, this is his favoured description in those sections where he seems particularly at pains to emphasise the importance of the concerns of international law: Ibid. at 182–3. It may be worth noting that Black’s Law Dictionary, 6th ed., (St. Paul, Minn.: West Publishing Co., 1990) at 1018 distinguishes between the common and narrower connotation of municipal law—laws which pertain to towns, cities and villages and their local government—and the broader connotations of its international law use which refers to “that which pertains solely to the citizens and inhabitants of a state” as distinct from, inter alia, international law (at 918). 75 In this sense, his approach seems to oddly mirror the approach of Bork J. in Tel-Oren which arguably rests on a somewhat anachronistic and narrow view of international law. On either view, one set of legal norms is rejected as inadequate in order to support the applicability and validity of the other. Once again, both approaches seem indebted to the “watertight compartments” view that characterises the traditional model. 73

Torture and Cosmopolitan Private Law 675 Massachusetts tort law” or indeed, the equivalent private law in Guatemala.76 And an examination of the sources invoked in the case reveals an incredibly complex web of authorities, jurisdictions and relationships. Thus, for instance, Woodlock D.J. invokes and relies to some extent on federal law (in the form of both ATCA and the TVPA) and its various interpretations, the state law of Massachusetts and Kentucky, Guatemalan law, American constitutional law, various international laws norms, the European Convention on Human Rights and its interpretation by the European Court of Human Rights. This open-textured approach to authority is further complicated by the fact that much of the invocation of various statutes and other sources is done as part of a search for the most analogous statute.77 This tension with the traditional model also manifests itself in what otherwise seems a perplexing treatment of the relationship between constitutional and international law norms. The question is whether there is a binding international norm against cruel, inhuman or degrading treatment. An affirmative answer requires universal consensus not only on the binding status of the norm but also on its content.78 The plaintiffs in Xuncax argued that this requirement was satisfied when the US ratified the Convention against Torture with a specific reservation that made the meaning of the phrase coextensive with the interpretations of the same phrase in the Fifth, Eighth and Fourteenth Amendments to the US Constitution. But Woodlock D.J. suggests that, because the Senate reservation explicitly ties the content of the abstract standard not to international norms but to domestic constitutional law, it may actually undermine rather than affirm the international law status of the norm.79 He eventually dismisses this argument by stating that, where American constitutional law and international law overlap, the “domestic constitutional directive” counts as the national voice in what Woodlock D.J. describes as the consensus that makes up the rule of international law.80 But what seems so striking here is the force of the idea that a norm is either domestic or international, but not both. Although he ends up confirming that a relationship between a constitutional and an international law norm does not deprive the norm of international force, the assumption seems to be that there is an inherent incommensurability between 76

886 F.Supp. 182, n. 22. See, for instance, the discussion of Claims on Behalf of Third Parties 886 F.Supp. 189–93. Similarly on the limitations issue US federal courts are empowered to look to analogous state or federal law where a federal statute does not address the limitations question. In determining that the claims under the ATCA are not statute-barred, Woodlock D.J. found that the most analogous federal law, the TVPA, should govern the issue but in the course of this discussion, he also examined the law of Massachusetts and Guatemala: ibid. at 192–3. While this reliance on analogical reasoning is statutorily mandated, it is nonetheless worth noting that the examination of the underlying principles that it requires (and that the court engages in wholeheartedly here) sits somewhat uneasily with a straightforwardly binding conception of authority. I am grateful to Craig Scott for calling my attention to this point. 78 Xuncax at 186 quoting Forti II, 694 F. Supp. 707 at 712. 79 Xuncax, supra n. 8 at 187. 80 Ibid. at 187. 77

676 M Moran municipal and international law. Indeed, it is striking that the conflicts model even seems to have some force in the discussion of the relationship between municipal and international law where a constitutional prohibition almost perfectly tracks the wording of an international law norm.81

5 CONVERGENCE , AUTHORITY AND THE TRADITIONAL MODEL

Even this snapshot of the adjudication of the tort of torture suggests the troubling influence of the traditional picture. At best, this picture misses much of what is going on and, at worst, it obscures what may be important convergences and normative possibilities. The traditional model of binding law theoretically constricts the range of acceptable sources. But the complexity of the sources courts actually invoke in the torture cases reveals the extent to which this model is fraying around the edges. And the cases also illuminate a closely related phenomenon that troubles the traditional model. Concern with constraining judicial decision-making is an important part of the impetus for the model’s focus on binding authority. In part for this reason, the positivistic predispositions of the traditional model view legislation as the paradigmatically legitimate source of law. Emphasis on legislation as the paradigmatic legal rule also implies a certain narrow picture of judicial legitimacy—familiar at least since Dicey—in which the role of the judge is to identify and apply the legal rule. Too much judicial “creativity”—read as lawmaking—is thereby minimised. Thus, the binding law approach entails both a distinctive conception of the sources of law and a concomitant picture of judicial authority. But, just as the torture cases reveal the limitations in the model’s doctrine of sources, so too do they challenge the picture of authority. Ironically, at the very same time that there is more and more use of international and comparative law, much of the current zeal is unlikely to win the approval of traditionalists. Indeed, whether scholars or judges, they tend to deplore much of what is happening under the guise of international law, at least in part because recourse to international law increasingly ignores the salience of the concept of binding authority.82 Here too, though, the predispositions of the 81 There is, of course, a reason why Woodlock D.J. is addressing this worry, and that is that a norm cannot be said to be truly international if its confines depend on domestic interpretations. Again though, the description of the VIII Amendment as a “domestic constitutional directive” imples a picture of domestic authority as normatively thin. Here, it seems that the major contribution of this provision of the Constitution is that it helps make the rule international. 82 See, for instance, the Supreme Court of Canada decision in Baker v. Canada [1999] 2 S.C.R. 817 and the specific objection of Iacobucci J. (dissenting) on this point. In Baker, a majority of the Court found not only that the interpretation of a statutory provision should presume compliance with international law—an uncontroversial principle of statutory interpretation—but also that that presumption of compliance extends to Canada’s legal obligations under unincorporated treaties. On this aspect of Baker see C. Scott, “Canada’s International Human Rights Obligations and Disadvantaged Members of Society: Finally into the Spotlight” (1999), 10 Constitutional Forum 97 at 100–1; and Knop, supra n. 37.

Torture and Cosmopolitan Private Law 677 traditional model explain much of the controversy the torture cases have aroused about the scope of judicial authority. Strong versions of the binding law model have a pronounced affinity for sharply dualist conceptions of the relationship between the separate spheres of national and international law. The close relationship the traditional model mandates between the state and law thus fuels both a suspicion of foreign law and a hostility to international law not expressly executed by the legislative arm of state. Thus, at a conceptual level, the increased openness of a legal system to international law cuts against the binding law model. And this is particularly true where human rights are at issue.83 Such cases are typified not only by substantial overlay of various sets of norms but also by the fact that it is here that the Nuremberg ideal most undermines faith in the traditional model’s blind adherence to sovereignty. Consequently, the temptation to move towards a more persuasive approach to authority is also at its greatest. Here, it seems, the persistence of the traditional binding law model has a paradoxical effect. If binding law is paradigmatically the province of the nation state, and if international law on that model looks a bit like a poor relation struggling to keep up appearances, then it hardly seems surprising that one of the temptations that international law might hold out to judges who are instinctively aware of the inappropriateness of a binding law model is precisely some kind of “release” into the sphere of persuasive authority. Indeed, it might be just this feature that explains the paradoxical quality of the torture cases—affirming the salience of international law as the source of the cause of action and then all but ignoring it in favour of a wide range of norms drawn from a number of jurisdictions. Perhaps, then, what the “choice” of the international law gives the judges in these increasingly cosmopolitan cases is the freedom to move beyond the cramped confines of the traditional model with its narrow conception of authority and the judicial role. In fact, the Court in Xuncax affirms the centrality of international law at least in part precisely because it releases the judge from the inapt strictures of the traditional model and permits a more nuanced and wide-ranging approach to the authorities. Thus, one of Woodlock D.J.’s dominant rationales—reiterated and actually employed later in the judgment—for rejecting reliance on municipal law is that the invocation of international law norms actually leaves more discretion to the judge. Thus, he notes that “courts will be freer to incorporate the full range of diverse elements that should be drawn upon to resolve international legal issues than they would if bound to a straight-forward recurrence to extant domestic law.”84 Indeed, he argues that a cause of action grounded in international law would constrain a federal court less, in the event that the court wanted to look to “the municipal law of other interested countries for guidance, so long as such law is not inconsonant with international or domestic [US]

83 84

L’Heureux-Dubé, supra n. 12 at 24–5. 886 F. Supp. 182.

678 M Moran law”.85 So, oddly enough, although much of the force of his argument derives from his emphasis on the stature and binding nature of international law norms against torture, one of the major reasons that he gives for preferring reliance on such norms is precisely the interpretive latitude associated with them. Similarly, in his consideration of whether the municipal law of the place of the tort (lex loci delicti) should determine the choice of law, Woodlock D.J. distinguishes between the “desirability of leaving a court free to draw upon diverse sources of law where appropriate” and the difficulties associated with a finding that “United State courts are obliged” to apply foreign law in considerations of liability.86 Thus, it seems that an important part of the appeal of the international law approach is that it frees the judge by allowing recourse to a range of persuasive authorities, unlike “state” law with its more articulated conception of binding authority and its conflicts mechanism. Indeed, the view that an important part of the appeal of the international cause of action is the interpretive freedom it holds out is confirmed by what otherwise seems a baffling recent controversy. On its face, the current impassioned debate among American academics about the content of “federal common law” may seem of limited general interest. But beneath the almost mind-numbing intricacies of federal court structure and constitutional history, much of what is at stake in that debate concerns how the shift to international law in the torture cases inevitably involves a more persuasive conception of authority. In a series of recent articles, Bradley and Goldsmith attack the constitutional basis of Filártiga-style actions and the incorporation thereby of the “new” customary international law into federal common law. But the heart of their charge—and of the recent controversy—is that these cases represent a departure from the traditional model of binding law, which expresses the interplay between the domestic and the international through the distinction between self-executing and non-self-executing treaties.87 Thus Bradley and 85 Ibid. at 183. It is worth noting that at the same time that the Court rejects reliance on domestic tort law and finds that an advantage of international law is that it permits examination of other domestic laws, it insists that these other sources be consistent with US domestic law. The consistency condition may not seem surprising, except that the whole point of his argument is to establish that US domestic law is an inadequate source! In this sense it begins to look like the advantage of international law is precisely that it permits the introduction of a broad range of “persuasive” authorities that would perhaps be precluded by domestic law with its stronger notion of binding authority. 86 Xuncax, supra n. 8 at 183 (emphases in original). 87 The effects of this controversy and its connection to the binding law model can also be seen elsewhere, as the traditional relationship between international and domestic law seems to “lose its edge”. An example can be found in the recent Supreme Court of Canada decision in Baker, supra n. 82. Something like this controversy is also at work in the differences between the various positions in the two House of Lords decisions in Pinochet: R v. Bartle and the Commissioner of Police for the Metropolis and Others Ex Parte Pinochet [1999] 3 W.L.R. 1456 (Pinochet No.1) and R v. Bow Street Metropolitan Stipendiary Magistrate and Others, ex parte Pinochet Ugarte (No. 2) [1999] 2 W.L.R. 272 (Pinochet No. 2). In a recent article, Bradley and Goldsmith argue that the wider readings of Pinochet are inapplicable to the United States for various reasons: 97 Mich. L. Rev. 2129 (1999). See Hunt for an extensive and thoughtful discussion of the larger issues: supra, n. 22, especially Chapter 1.

Torture and Cosmopolitan Private Law 679 Goldsmith complain that courts in these actions cite multilateral treaties as a source of customary international law even where “the United States has not ratified the treaty or has declared the relevant provisions of the treaty to be nonself-executing.”88 In this sense, Filártiga-style actions actually enable judges to circumvent the limitations of the binding law model in its most technical moment by applying “the norms derived from the treaties as self-executing federal common law”.89 As Curtis and Bradley put it, Filártiga-style actions enable judges to “accomplish through the back door of CIL what the political branches have prohibited through the front door of treaties”.90 And the Curtis–Bradley criticism has itself inspired reactions that illustrate how centrally the torture cases implicate the integrity of the concept of binding authority.91 Indeed, Lawrence Lessig adverts to the relationship between this approach and the binding law model when he describes the Curtis–Bradley position as a “strictly positivistic view” according to which “the only law is domestic law and the only domestic law is statute or constitution based.”92 Similarly, Koh notes that, at bottom, what Curtis and Bradley object to is the “traditionally fluid, accretive, osmotic process of legal internalisation . . . as somehow threatening to state interests”.93 In this sense, implicitly both the critics and the supporters recognise the extent to which the Filártiga-style actions are invoking a different conception of authority, one that depends much less on the relatively sharp-edged “positivistic” notions of binding law and much more on relatively amorphous “persuasive” authority. This suggests that courts in the torture cases look to international law at least as much for interpretive latitude as for some kind of higher normativity. As discussed above, their preference for international normativity is justified partly by denigrating the status of domestic tort law—thus, it becomes the “garden variety” municipal tort. Yet the oddity noted earlier is that, although domestic law is theoretically rejected as a source of law, in fact domestic law norms from more than one legal system surface in the reasoning as extremely important 88 Curtis A Bradley and Jack L Goldsmith, “Customary International Law as Federal Common Law: A Critique of the Modern Position” 110 Harv. L. Rev. 815 at 858 (1997). For similar arguments see also Bradley and Goldsmith, “The Current Illegitimacy of International Human Rights Litigation”, 66 Fordham L. Rev. 319 (1997). The responses have been swift and include Harold Koh, “Is International Law Really State Law?” 111 Harv. L. Rev. 1824 (1999); and Lawrence Lessig, “Erie-Effects of Volume 110: An Essay on Context in Interpretive Theory”, 110 Harv. L. Rev. 1785 (1997). See also M O Chibundu, “Making Customary International Law Through Municipal Adjudication: A Structural Inquiry”, 39 Virginia J.I.L. 1069 (1999). 89 “Customary International Law as Federal Common Law”, supra n. 88 at 858. 90 “Current Illegitimacy”, supra n. 88 at 330–1. Recall Iacobucci J.’s back-door objection in Baker, supra n. 82. As commentators have pointed out, this is difficult to sustain as an interpretation of the Congressional stance, particularly given that, far from repealing the ATCA, Congress effectively extended its provisions to non-aliens in the Torture Victim Protection Act: The Honorable John M Walker, Jr., “Domestic Adjudication of International Human Rights Violations under the Alien Tort Statute”, 41 St. Louis U. L. J. 539 at 560 (1997). 91 See for instance, Koh, supra n. 88 at 1829 n. 24, 1841. 92 Lessig, supra n. 88 at 1810. 93 “Is International Law Really State Law?”, supra n. 88 at 1857.

680 M Moran sources of normativity. A clearer understanding of how the binding law model conceives of authority and of the relationship between domestic and international law thus helps to account for this tension in the reasoning of the torture cases. But, even given this, continued adherence to the binding law model comes at a high price. In order to justify the interpretive freedom that seems so crucial in cases where the relationships among norms are not most aptly understood on a conflicts model, the torture cases drastically underplay the normative significance of domestic and private law. This seems a serious mistake. In fact, the reasoning itself reveals that judges find substantial significance in the norms of domestic private and public law: thus they invoke the domestic law of the relevant jurisdictions to ensure fairness as well as to locate appropriate principles of responsibility. This is obvious testimony to the fact that, whatever their protestations to the contrary, judges view the norms of domestic law as important. But there are reasons beyond those of consistency to worry about the predominance of the separate-spheres view of domestic and international law that underlies the binding law model. Indeed, that model’s imperative for maintaining a sharp delineation between the normativity of domestic law and that of international law seems seriously misguided. It is simply inaccurate to describe domestic law and supranational law as unrelated and mutually-exclusive spheres of normativity.94 In fact, the relationship between domestic law of various sorts and supranational law has been, and continues to be, characterised by complicated interactions and borrowings. Thus, as we have seen, public law norms—including supranational ones—have played an increasingly important role in shaping the contours of private law. But, contrary to the rhetoric sometimes adopted— by commentators as well as in the torture cases themselves—to justify recourse to supranational law, domestic private law is not an empty vessel noteworthy only for its efficient machinery of enforcement. Indeed, a more satisfying picture of the interrelationships between domestic and supranational law suggests that there is much more two-way traffic than the traditional model would allow.95 It therefore seems important to note that domestic law, including private law, has always been an important contributor to the articulation of supranational norms. Indeed, the common law is often described as the “father [sic] of the very idea of the rule of law”.96 Similarly, Hunt,and others, note the important influ94 Of course, for the traditional model of law with its concern with pedigree and its understanding of law as the univocal imperatives of the nation state, it is quite natural to draw this sharp distinction because the concern with pedigree focuses the inquiry on sources not substance or method. 95 It is worth noting here that although they differ significantly in other ways, the “conservative” dualist sovereigntists and the “progressive” monist internationalists share a markedly unidirectional picture of the relationship between the national and the supranational. Thus, for traditional sovereigntists, the normativity all comes from the national while, for the internationalists, it comes from the international. As noted in the text, recent commentators like Koh and Hunt have evidenced discomfort with the strictures of this dichotomy. 96 Lord Goff of Chieveley, “The Future of the Common Law”, (1997) 46 I.C.L.Q. 745. It is worth noting that such statements themselves most certainly also reflect a conception of the genealogy of norms that is also both somewhat self-important and still markedly unidirectional. The point here,

Torture and Cosmopolitan Private Law 681 ence of the English common law on the development of the European Convention of Human Rights.97 At the constitutional level, the interactions are perhaps becoming more acknowledged. The influence of the American Bill of Rights both on the development of international human rights as well as other rights-protecting documents is well-known and well-documented.98 And in the recent spate of constitution building, the influence of the post-war models of Canada and Germany have had a profound impact.99 Thus, for the sake of accuracy alone, it seems worth questioning the unidirectional conception of norm development that characterises the traditional model. But there are also other reasons to attend to the normative interactions suppressed by the traditional model. In fact, notwithstanding the insistent dichotomising of this model, judges and scholars—particularly those involved in boundary-crossing problems like the torture cases—are beginning to note the hybridity and interpenetration of legal orders. Indeed, one could take it as further evidence of the depth of the challenge to the traditional model that its foundational orthodoxy is being called into question. Thus, for instance, although Koh sometimes opts for the standard interpretation that Filártiga-style claims derive their normativity from international law, elsewhere he suggests a more complex—and more satisfying—model. In such moments, he describes the development of truly transnational norms based “not solely on domestic or international law, but rather on a body of “transnational” law that blends the two”.100 And, uncovering the complex origins of the norms actually applied in cases like those involving torture may actually be pragmatically as well as conceptually important. In fact, recognition of the interactive nature of norm development also helps to formulate a response to the critics of Filártiga-style claims. Much of the criticism, inspired by the traditional model, depicts such claims as illegitimate because of their “foreignness”. But, interestingly, Koh finds that the most persuasive response here is the one that actually undermines the veracity of the traditional model as an understanding of the process of norm development. Without the ability to sharply delineate one set of norms from the other, the model begins to lose its bite. Koh thus notes that there is in fact no bright however, is simply that the common law has made important contributions, not that we need to accept its own (perhaps somewhat inflated) account of its influence. 97 Supra n. 22 at 24, discussing B Markensis (ed.), The Gradual Convergence: Foreign Ideas, Foreign Influences and English Law on the Eve of the 21st Century (Oxford: Clarendon Press, 1994) at 166; Geoffrey Marston, “The United Kingdom’s Part in the Preparation of the European Convention on Human Rights, 1950” (1993), 42 I.C.L.Q. 796. 98 Louis Henkin, “A New Birth of Constitutionalism: Genetic Influences and Genetic Defects” (1993), 14 Cardozo L. Rev. 533 and Louis Henkin et al. (eds.), Constitutionalism and Rights: The Influence of the United States Constitution Abroad (New York: Columbia University Press, 1989). 99 See, for instance, Lorraine Weinrib, “The Canadian Charter as a Model for Israel’s Basic Laws” (1993), 4 Const. Forum 85 and “Canada’s Charter: Rights Protection in the Cultural Mosaic” (1996), 4 Cardozo J. Int’l & Comp.L. 395; Richard J Goldstone “The South African Bill of Rights” (1997), 32 Tex. Int’l L.J. 451. 100 Koh, supra n. 11 at 2348. Similarly, he describes the transnational claim structure as one involving allegations of violation of domestic and international, public and private law, all alleged in a single action (at 2371).

682 M Moran line which separates domestic rules of decision from the rules of customary international law. Indeed, he points out that the principle of comity has roots in both common law and international law.101 And, in the absence of norms that are unambiguously domestic or foreign, the challenge that foreignness entails illegitimacy falls flat. A similar impulse is also apparent in the English context, where Murray Hunt and other commentators defend the use by domestic English courts of (non-executed or unincorporated) supranational treaty rights, in part by noting the complicated origin of the European Convention and, in particular, the input of English common law.102 In addition to pragmatic importance, this interactive conception of norm development also seems necessary if we are to generate the dialogical possibilities that so many commentators hold out as important in an increasingly interconnected world. Madam Justice L’Heureux-Dubé of the Supreme Court of Canada puts the point suggestively: “As courts look all over the world for sources of authority, the process of international influence has changed from reception to dialogue. Judges no longer simply receive the cases of other jurisdictions and then apply them or modify them for their own jurisdiction. Rather, cross-pollination and dialogue between jurisdictions is increasingly occurring.”103

This dialogical process of learning from each other is seen by commentators as crucial, a vital contribution to the quest for “the advancement not only of human rights but to the pursuit of justice itself, wherever we are”.104 Koh, too, points to at least one dimension of this in his suggestion that the integration of domestic and international law will be furthered through this kind of dialogue between American courts and other law-declaring institutions of the international system.105 In the adjudicative context, Brennan J.’s decision in Mabo illustrates the possibilities of a more interactive understanding of the process of norm building. Commenting on the refusal of Australia to recognise the rights and interests in land of the indigenous inhabitants of the settled colonies and the influence of the ICCPR, Brennan J. notes: “The common law does not necessarily conform with international law, but international law is a legitimate and important influence on the development of the 101

Harold Koh, “Is International Law Really State Law?” 111 Harv. L.R. 1824 at 1856. Supra n. 97. One can, of course, imagine a chauvinistic version of this understanding, but that is not germane to my point here. 103 Supra n. 12 at 17 (Emphasis in original). To a similar effect, see Madam Justice L’HeureuxDubé’s decision in Baker, supra n. 82 at 825. This approach is also noted by C. Scott, supra n. 82 at 100–1. 104 L’Heureux-Dubé, supra n. 12. at 40. 105 Koh, supra n. 11 at 2397. (note his discussion of the kinds of norms—federal common law, canons of domestic statutory construction, foreign or international court decisions CIL, HR treaties). Although Koh uses the terms “unitary” and “monist” to describe this goal, I am not convinced that this language is the most appropriate vehicle for conveying the kind of dialogical interaction which is his ambition. 102

Torture and Cosmopolitan Private Law 683 common law, especially when international law declares the existence of universal human rights. A common law doctrine founded on unjust discrimination in the enjoyment of civil and political rights demands reconsideration. It is contrary both to international standards and to the fundamental values of our common law to entrench a discriminatory rule which, because of the supposed position on the scale of social organisation of the indigenous inhabitants of a settled colony, denies them a right to occupy their traditional lands.”106

Brennan J.’s language here is significant: the norms of non-discrimination do not emanate exclusively from the international sphere. Instead, he points to congruity between international standards and the “fundamental values” of the common law. And his justification for the influence of international norms on the development of the common law is largely found in his insistence that international norms would hold the common law true to its own values. This reasoning suggests it is conceptually important to articulate a more adequate picture of the relationship between various kinds of norms so that the systems do not seem to exist in normative isolation, subject only to their own imperatives. But when legal regimes are understood as entirely or largely separate and dichotomous, this kind of normative interaction and development becomes difficult to account for and justify.

6 LOOKING FORWARD , LOOKING BACK

Ultimately, the torture cases give us a glimpse into the imperatives of a future in which law—including private law—is increasingly cosmopolitan. Here, as the torture cases and the surrounding debates reveal, the limitations of the traditional conception of binding law come to seem pressing to judges and others faced with a world where legal problems are profoundly multi-jurisdictional in nature. The task of the advocate, the judge, and the law-maker no longer seems adequately captured—if it ever was—by the notion of discrete mutually exclusive spheres of binding law, conjoined through a set of rules premised on conflict and choice. And indeed, what the torture cases show is a subtle, yet distinct, move away from this model and towards a more multi-faceted integrative understanding of sources and a broader persuasive approach to authority. The exact nature of the emerging cosmopolitan alternative is uncertain in its details but the adjectives and metaphors that recur in the struggle to develop a new vocabulary are telling: norms are “converging”, “interpenetrating” and “hybrid”; the interactions are not so much disputes about boundary as matters of “cross-pollinations”, “borrowings”, “transplants” and the ubiquitous “dialogue”; authority is persuasive and so the tasks of justification and judgment become critical. Judgment, on this view, is a more constructive task of articulating the commonalities and attending to the differences. And the essential act 106

Mabo and Others v. Queensland (No. 2) (1992) 175 CLR 1 at 42.

684 M Moran of justification has a particularistic, discursive quality inextricably bound up with the idea of persuasive authority. As Glenn notes, multiplying the sources of law “means multiplying the sources of legal dialogue. Law is less precise, but more communal and there are more possibilities of persuasion and adherence to law . . .”.107 Ultimately the move away from the traditional conception suggests that legitimacy will rest more on the ability to persuade than to command or coerce. The rule of law as a rearticulated normative ideal may thus have more to do with justification than with pedigree.108 And the task of justification requires constant reshaping of a common language to speak across difference and perhaps even indifference. It is not an easy task, but neither is it an unfamiliar one. In fact, the methodology of the common law and some of the lessons of its history may prove suggestive here. Characterised by a proliferation of sources and rules, focused on a particular problem and uniquely constituted through the act of judgment, the common law can be seen as a persistent instance of persuasive authority. Indeed, by disposition the common law is at home with a generous doctrine of sources and thus English common law has been described as a “shameless snapper-up of well-considered trifles of foreign law, from Rome, France, Germany, the United States” and most recently, of course, Europe.109 The authority of the case derives largely from its persuasiveness—the ability to distil and integrate complex and various sources into a coherent and convincing solution to a particular problem. Sir Edward Coke perhaps said it best when he noted the effects of a “substantial” judgment: “First, it openth the Understanding of the Reader and the Hearer; secondly, it breaketh through Difficulties; and thirdly, it bringeth home, to the Hand of the Studious, Variety of Pleasure and Profit.”110

And this particularistic and rhetorical method enables the common law to be responsive to its locale without losing its own personality—in the future as in the past, the development of the common law “will not be uniform but variegated.”111 Indeed, it is precisely this variegated character that makes the common law method particularly apt for the “process of learning from each other” that has always typified common law reasoning and that increasingly characterizes the incredibly dense web of jurisdictions which it inhabits.112 For this 107

“Persuasive Authority”, supra n. 29 at 297. David Dyzenhaus, “Law as Justification: Etienne Mureinik’s Conception of Legal Culture” (1998), 14 SAJHR 11 and “Recrafting the Rule of Law” in D Dyzenhaus (ed.) Recrafting the Rule of Law: The Limits of the Legal Order (Oxford: Hart Publishing, 1999). In that volume, see also Neil MacCormick, “Rhetoric and the Rule of Law” at 163–77. 109 Ibid. at 212. 110 Sir Edward Coke, Preface to the Ninth Part of the Reports (1613) quoted in Ernest J. Weinrib, Tort Law (Toronto: Emond Montgomery, 1997) at iii. 111 Lord Bingham, “The Future of the Common Law” supra, n. 13 at 217. 112 Ibid. referring to the impact of the law of Europe, the international human rights instruments, international trade and the “globalisation of almost all activity”. Indeed, he envisages a world of persuasive authority where jurisdictions borrow from each other, back and forth. 108

Torture and Cosmopolitan Private Law 685 reason Lord Goff suggests that the “fundamental ethos of the common law will remain intact” not only because of its commitment to the rule of law but also because its pragmatic discursive disposition is well-suited to represent that ambition in the modern world. As Gaius stated, in the Institutes, peoples are governed both by the law which is particular to them and by law which is common to humanity.113 And the challenge of justification that has long been the heart of the common law is to articulate the nature of these complex links between what is particular and what is common in a language that will persuade the listener of what Coke described as the “Right Reason of the Rule”.114 And it may be precisely these features of the common law, and the premium it inevitably places on the centrality of justification, that are illuminating. Whatever its many failings, the common law has always been uniquely constituted through the act of judgment—so much so that, here at least, it has always been implicit that authority is as much created through the act of justification as inherited from the state. In this sense, the particularistic, rhetorical method of the common law may provide a fruitful analogy in the quest to rethink the nature of law and authority. Oddly enough, our “post-modern” future may share some deep affinities with our pre-modern past.

113 114

Inst.Gaius 1.1, cited in Glenn, supra n. 29 at 298. Supra n. 110.

26

Private Law, Constitutionalism and the Limits of the Judicial Role OLIVER GERSTENBERG 1

“Aux Etats-Unis, la Constitution domine les législateurs, comme les simples citoyens (. . .), il est donc juste que les tribuneaux obéissent à la Constitution préférablement à toutes les lois.” “En Amérique, les théories politiques sont plus simples et plus rationnelles.” —Alexis de Tocqueville, De la démocratie en Amérique, I, I, ch. VI

1 INTRODUCTION : THE AUTONOMY OF PRIVATE LAW AND THE DEMOCRATIC PRINCIPLE

of a recent example from case law, I would like to illustrate the two questions which form the core of the following analysis—(1) the issue of whether and in which manner private actors are bound by constitutional principles; and closely related to this (2) the question of the role of the courts and of judicial enforcement of constitutional principles within the democratic process. Among other things, the well-known European Court of Justice (ECJ) Bosman judgment concerned a system of transfer rules according to which a professional football player might only be employed by a new club, even following the conclusion of his contract, where it paid his old club a transfer fee that could amount to a sum equal to eight times the annual income of the

W

ITH THE AID

1 Wissenschaftlicher Angestellter, University of Bremen; Jean Monnet Fellow, 1999–2000, European University Institute, Florence. Particular thanks are due to Gert Brüggemeier, Charles F. Sabel and Gunther Teubner for their valuable comments. This chapter is a translation from the original German, published as O Gerstenberg, “Privatrecht, Verfassung und die Grenzen judizieller Sozialregulierung” in (2000) ARSP Beiheft 74, 141. Draft translation by Michelle Everson. The chapter enlarges upon the various thoughts of the author in: “Privatrecht, Demokratie und die lange Dauer der bürgerlichen Gesellschaft”, (1997) Rechtshistorisches Journal 152; “Law’s Polyarchy: A Comment on Cohen and Sabel”, (1997) 3 European Law Journal 343; “Procéduralisation comme démocratisation: commentaire au Professor F. Michelman”, in J Lenoble, (ed.), Actes des Journées Juridiques Jean Dabin (Montechrestien, LGDF, 1999); “Ordinamento Privato, Intervento Publico e Pluralismo Sociale”, in J Derrida and G Vattimo, (eds), Diritto, Giustizia e Interpretazione (Latereza, Annuario Filosofico Europeo, 1998) 68.

688 O Gerstenberg player.2 The ECJ concluded that this transfer system constituted an abuse of article 48 of the European Community (EC) Treaty on the free movement of workers: “Article 48 of the EEC Treaty precludes the application of rules laid down by sporting associations, under which a professional footballer who is a national of one Member State may not, on the expiry of his contract with a club, be employed by a club of another Member State unless the latter club has paid to the former club a transfer, training or development fee.”3

In so doing, the ECJ effectively developed an argument imbuing article 48 with the character of a “principle” generally applying to all citizens of the Community, and so restricting their freedom as employers in relation to all their cross-border economic activities, including even those conducted within a private association.4 On the other hand, however, the Court also recognized that such football clubs and associations possessed legitimate interests in relation to, for example, financial cross-subsidisation between clubs designed to protect sports competition against market mechanisms, as well as in relation to the development of young players. Nonetheless, the existing transfer system was not regarded as a proportional—i.e., protective of fundamental freedoms— mechanism to secure such prima-facie legitimate interests. The central feature of the decision, however, was the fact that the Court placed the duty to find a system of rules which would pay adequate attention to both of these colliding interests firmly back upon the private actors. Although the Court outlined various alternative and acceptable modes of organisation, it refused to specify a final regime. The clubs and associations themselves must “discover” a form of self-organisation that both realises the aims of the football industry and, at the same time, respects the principles contained within article 48. The Court regarded its role as one of promoting self-regulation experiments in private legal relations that will foster fundamental rights. The solution the ECJ identified in Bosman contrasts starkly with the current discussion in Germany on the effect of fundamental rights on private law. While the point of the ECJ’s solution was to address private actors as law-makers— responsible not only under the law but also for the integrity of the legal regime within which they themselves would act, so ensuring a concomitant constitu2 C–415/93, Jean Marc Bosman [1995] ECR I 4921. For extensive German academic commentary: (1996) JZ 248; (1996) NJW 505; (1996) EuZW 82; (1996) EuGRZ 17. 3 Point 2 of final judgment, Bosman, supra n. 2 at I–5081. 4 Cf., for comprehensive review: Ch Joerges, “The Impact of European Integration on Private Law: Reductionist Perspectives, True Conflicts and a New Constitutional Perspective”, (1997) ELJ 423; Ch Palme, “Kurzbeitrag/Das Bosman-Urteil des EuGH: Ein Schlag gegen die Sportautonomie” (1996) JZ 238; R Schulz and J Aulehner, “Die ‘3+2’ Regel und die Transfer-Bestimmungen des Fußballsports im Lichte des europäischen Gemeinschaftsrechts”, (1996) SpuRt 44; H-P Westermann, “Die Entwicklung im bezahlten Fußballsport nach dem ‘Bosman’-Urteil”, (1997) DZWir 485; H-P Westermann, “Erste praktische Folgen des ‘Bosman’-Urteils für die Organisation des Berufsfußballsports”, (1996) DZWir 82; R Blainpain and R Inston, The Bosman Case: The End of the Transfer System (Leuven, 1996).

Private Law and Constitutionalism 689 tionalisation and democratisation of the private regime—the current debate over the “third-party effect” (Drittwirkung5) of fundamental rights suggests that the relationship between private law, the Constitution and democracy is dilemma-ridden.6 On the one hand, there is agreement that, by virtue of its hierarchical precedence over “simple” or “lower-level” private law, the Constitution does have an effect, namely that the integrative function of constitutional rights must be efficiently secured within the realm of private law, and that there is a normative continuity between private law and the Constitution. In its Lüth judgment, which remains the determinative point of reference for the debate, the German Constitutional Court ruled that fundamental rights were “objective principles”, forming a value system that must be given effect in all areas of the legal system—although, at the same time, a conflict between private actors on the rights and duties arising out of a body of private law influenced by fundamental rights would remain, both substantively and procedurally, a private law conflict.7 As the German Constitutional Court was also to remind us in its Bürgschaft judgment concerning guarantees given without consideration, “private legal orders” are to be regarded as “a differentiated system of coordinated rules and constitutive mechanisms that must be inserted into the constitutional order”.8 On the other hand, however, current debate seems to suggest that a consequential realisation of the “narrow binding of private law to the Constitution”9 might itself lead to unconstitutional results. The concern is that if constitutional principles apply directly to private actors—without mediation through the formal legislative process—then private law would more and 5 Please note that the more common English translation of Drittwirkung as “horizontal effect” is not used here for reasons of analytical clarity. 6 Cf., for a comprehensive sense of the literature on the German debate, see: G Dürig, “Grundrechte und Zivilsprechung” in Th Maunz, (ed.), Von Bonner Grundgesetz zur gesamt deutschen Verfassung: FS. Nawiasky (München, 1956); W Leisner, Grundrechte und Privatrecht (1996); R Alexy, Theorie der Grundrechte (1985) 475 et seq.; K Hesse, Verfassungsrecht und Privatrecht (1988); CP-W Canaris, “Grundrechte und Privatrecht”, (1984) 184 AcP 201;CP-W Canaris, “Verstöße gegen das verfassungsrechtliche Übermaßverbot im Recht der Geschäftsfähigkeit und im Schadensersatzrecht”, (1987) JZ 993 and (1998) JZ 494; Canaris, “Grundrechtswirkungen und Verhähltnismäßigkeitsprinzip in der rechtlichen Anwendung und Fortbildung des privatrechts”, (1989) JuS 993; CP-W Canaris, “Die Verfassungwidrigkeit von §828 II BGB als rechts”, (1989) JuS 161; Canaris, “Verfassungs und europarechtliche Aspekte der Vertragsfreiheit in der Privatrechtsgesellschaft”, in F S Lerche (1993) at 873; CP-W Canaris, Die Bedeutung der iustitis distributiva im deutschen Vertragsrecht (München, 1997); D Medicus, “Der Grundsatz der Verhältnismäßigkeit im Privatrecht”, (1992) AcP 43; E Schmidt, “Verfassungsgerichtliche Einwirkungen auf zivilistische Grundprinzipien und Institutionen”, (1995) KritV 424; J Hager, “Grundrechte im Privatrecht”, (1994) JZ 373; P Lerche, “Grundrechtswirkungen im Privatrecht”, (1994) JZ 373; P Lerche “Grundrechtswirkungen im Privatrecht: Einheit der Rechtsordnung und materielle Verfassung”, in FS Odersky (1996) at 215; C-D Claassen, “Die Drittwirkung der Grundrechte in der Rechtsprechung des Bundesverfassungsgerichtes”, (1997) 122 AöR 65; M Fuchs, “Schuldrecht und Verfassungsrecht”, in FS Hans Zacher (1998) at 169; U Diederichsen, “Das Verfassungsgericht als oberste Zivilgericht”, (1998) 198 AcP 171; P A Windel “Ueber Privatrecht mit Verfassungsrang und Grundrechtswirkungen auf der Ebene einfachen Privatrechts”, (1998) 3 Der Staat 385. 7 BverfGE 7, 198. 8 JZ 1994, 408. 9 Medicus, supra n. 6.

690 O Gerstenberg more be drawn into complex processes of constitutional balancing. This, in its turn, would lead to a shift in the relationship between the judiciary and the legislature with very wide-ranging consequences. A judiciary committed to the direct—without recourse to simple legislation—binding of private law to the Constitution would inexorably be drawn into the centre of the law-making process, so that the role of the private law legislator would consequently decrease in importance and the principle of democracy would itself be undermined. The problem of the legitimacy of judicial balancing would be further augmented by the clear conflict between the indeterminate and indefinite nature of the judicial balancing process and the interest of private actors in clear and calculable legal rules (the formal rationality of private law). At its most extreme, the direct binding of private actors by fundamental rights would, in effect, make private law superfluous. The relevant literature is rich in drastic formulations of the problem: the “adjudicational hypertrophy” caused by “the constitutional colonisation of private law”10 threatens us with no less than a “farewell to private autonomy”.11 Indeed, there is even mention of a “methodological coup d’état”12 and of “an alteration in the fundamental principles of the current political system and the existing Constitution”.13 Impressed by this apocalyptic perspective, one might wonder whether, instead, the law of the Constitution itself demands that private law be only “loosely” bound to the Constitution. Does the prescribed respect for the principle of democracy require the recognition of the independence of private law from—or the privileged pre-political position with regard to—the Constitution? In this chapter, I will argue that any such questions must clearly be answered in the negative. From a constructivist perspective, the chapter aims to show that the solution identified in Bosman presents us with a generalisable model for the concomitant constitutionalisation and democratisation of private legal relationships. Additionally, I would like to defend the thesis of the directlydeliberative binding of private actors to fundamental rights and central constitutional principles. This thesis maintains that, when organizing their private legal relationships, private actors must orient themselves in accordance with a series of constitutionally validatable “justifications” (Gründen) produced in a process of dialogue. This process is direct since it is not dependent upon an intermediary, state-led legislative process, which necessarily takes the project of the creation of a functionally differentiated private regime that at the same time takes constitutional principles out of the hands of the public actors. Rather, it is the private actors themselves upon whom this original function is conferred in 10 St Oeter, “Drittwirkung der Grundrechte und die Autonomie des Privatrechts: Ein Beitrag zur funktionell-rechtlichen Dimension der Drittwirkungsdebatte”, (1994) 119 Aör 529. 11 W Zöllner, “Regelungsspielräume im Schuldvertragsrecht: Bemerkungen zur Grundrechtsanwendung im Privatrecht und zu den sogenannten Ungleichgewichtslagen”, (1996) 196 AcP 1; D Medicus, Abschied von der Privatautonomie im Schuldrecht? Ercheinungsformen, Gefahren, Abhilfen (Köln, 1995). 12 Diederichsen, supra n. 6. 13 Ibid.

Private Law and Constitutionalism 691 the form of a radically democratic process of law creation “from below”. This direct constitutional binding is deliberative to the degree that the private actors involved must orient themselves in line with ‘justifications’ that both further the project of interpretative constitutional development and secure the long-term future of this project across social subsystems. Such an altered analysis breaks with the traditional conception of the thirdparty effect of fundamental rights within private law. As Ludwig Raiser noted early on, this particular conception remains bound to the “liberal scheme of thought” which “sharply distinguishes state from society and which conceives of the Constitution as ordering the state and of private law as ordering society”.14 Even the concept of third-party effect, so Raiser argues, implies that the impact of the Constitution on private law embodies an “invasion of foreign territory”,15 and so is trapped, from the very outset, in a form of split thinking which relieves private actors of their social responsibility for the consequences of their action and for the realization of constitutional standards The following two theses stand in the forefront of this chapter’s subsequent considerations: (i) The constitutional impact upon private law (or, the constitutionalisation of private law) does not (only) concern the transposition of liberal fundamental rights and the principles of the welfare state (Sozialstaat) into private law;16 rather, it also always concerns the binding of private actors with antagonistic interests within deliberative justificatory processes—and, with this, the transposition into private regimes of the principle of democracy. Thus, “the constitutionalisation of private law” entails the confrontation of private law with a requirement to institutionalise rights for and possibilities of participation within social processes of norm creation. (ii) The legitimation of judicial intervention does not derive from the practice of furnishing private actors with—judicially constructed and disciplineinternal—“finite” solutions; instead, it is created by the mixture of activism and self-constraint that was so characteristic of the Bosman judgment, a mixture which fosters fora for justificatory processes steered by principles, and which thus also encourages regulatory experiments among private actors that are respectful of fundamental rights. I shall begin by summarising recent discussion on third-party effect (Section 2). In a second step, I shall develop the thesis of a directly deliberative creation of fundamental rights (Section 3), which is then tested in the light of an objection to it (Section 4). I shall conclude, in a forward-looking mode, with a reference to the self-contained discussion on a structural move beyond representative democracy. 14 L Raiser, “Grundgesetz und Privatrechtsordnung”, (1996), in Kronberg, Die Aufgabe des Privatrechts: Aufsätze zum privat- und Wirtschaftsrecht aus drei Jahrzehnten (1977) at 162. 15 Ibid. 16 Cf., in particular, M Fuchs, supra n. 6.

692 O Gerstenberg

2 THE STATE OF RECENT DEBATE ON THIRD - PARTY EFFECT

The debate over the relationship between private law and constitutional law is to be found within both constitutional law and private law circles. The doctrinal point of contention can be identified as article 1.III of the German Constitution (Grundgesetz, GG) stating that the legislature, executive and judiciary are bound by the fundamental rights catalogue (Grundrechte) of the GG. Article 1.III GG finds its equivalent in other legal orders.17 As should become apparent, the discussion highlighted here is characterised by a tendency towards the “reverticalisation” of the jurisdiction of fundamental rights. Hesse K. Hesse distinguished between the two functions that the Constitution performs for private law: on the one hand, protective, and on the other evolutionary.18 Within its protective function, the Constitution shields private law from any attempt to set it aside or erode it through law. In this respect, so Hesse argues, the Constitution furnishes a degree of protection for the fundamental bases of private law that private law itself could not provide—in particular, in the form of protection for marriage, for property and for personal integrity. By contrast, in its evolutionary function, the Constitution provides private law with development impulses and gives rise to new forms of law. Hesse maintains that, by virtue of its breadth and the openness of its norms, the Constitution is better placed than private law to respond and give effect to altered conditions and requirements within society. Constitutional law, so it is maintained, can react more promptly to real-world changes, and thus becomes the true catalyst of legal evolution. Hesse’s central thesis argues that a new balance needs to be constantly established between both functions. He also identifies this balancing requirement as both a result and an expression of the historical and normative emancipation of constitutional law from private law. Constitutional law, in its turn, can only have an impact on private law by virtue of this process of differentiation. Private law is not merely an instrument of self-determination; instead, it must also serve “social justice” and has thus evolved within a second dimension which conflicts with the first. 17 Cf., the instructive review in M Hunt, “The ‘Horizontal Effect’ of the Human Rights Act”, in (1998) Public Law 422; on the US debate, compare A Scalia, A Matter of Interpretation: Federal Courts and the Law (Princeton, Princeton UP, 1998) with comments by A Gutmann, G S Wood, L Tribe, M A Glendon and R Dworkin; see also C Sunstein, “Justice Scalia’s Democratic Formalism”, (1998) 107 Yale Law Journal 529. On the French debate, cf., N Molfessis, Le Conseil Constitutionel et le droit privé (1997 Paris, LGDJ); M Frangi, Constitution et droit privé: Les droits individuels et les droits économiques (Presses universitaires d’Aix-marseille, 1992). 18 K Hesse, supra n. 6; Hesse, Grundzüge des Verfassungsrechts der Bundesrepublik Deutschland (Heidelberg, 1991), 18th edition at 148 et seq.

Private Law and Constitutionalism 693 At the same time, Hesse details the various pathologies which might culminate in the “superseding” of private law by the Constitution: (i) the opening up of private law to complex processes of adjudication creates tension with the expectation of clear and detailed private law rules (legal certainty); (ii) the independence of private law is threatened by the direct recourse to fundamental rights, to the disadvantage of substantively appropriate regulation and the evolution of the various areas of private law; (iii) this results in a threat to “private autonomy”; and finally, (iv) the division of labour between constitutional and technical jurisdictions is blurred, so that the German Constitutional Court could become the highest legal instance in civil proceedings. The solution proposed by Hesse is directed to the legislature. This body, it is maintained, bears the primary responsibility for the enactment of differentiated and concretised fundamental rights that are directly applicable to the parties to a private legal relationship. The recourse to the legislature, Hesse argues, could lessen the imminent threat of the supplanting of private law by constitutional law: the independence of private law would be better guaranteed, while the civil jurisdiction would, in essence, be spared the need to engage in a wide-ranging and direct application of fundamental rights that generally requires complex processes of constitutional adjudication. Thus, according to Hesse, the interpolation of statutes is the appropriate means to guarantee the protection of fundamental rights against encroachments and threats from a non-statal sphere. With the passage of time, the question of third-party effect should—or so it is hoped—lose its urgency to the degree that the legislature fulfils its function of delineating spheres of freedom through the enactment of new private law statutes. A primary characteristic of this solution is the protective competence of the state: a duty is placed upon the (formal) legislative process to furnish private actors with adequate problem-solving mechanisms. In this respect, the premises underlying the public/private distinction are not prejudiced. Little regard is paid to the dynamic of the decentralised self-organisation of private actors, or to the notion of the disciplining of such decentralised processes through the integration within them of a structural imperative to orient themselves in accordance with constitutionally legitimate justifications. Reciprocal norm production by private actors is not envisaged.

Canaris The article by Canaris, hinges upon the stark distinction between the “acts of the state” on the one hand, and the “acts of private actors” on the other.19 This distinction is prescribed by article 1, paragraph 3 GG, whereby the catalogue of fundamental freedoms only take effect, as directly applicable law, in relation to the acts of the legislature, the executive and the judiciary. Thus private actors 19

Canaris, “Grundrechte und Privatrecht”, supra n. 6.

694 O Gerstenberg might only be directly bound to the fundamental freedoms at the cost of “the evolution of law beyond the linguistic terms [of the constitution]”, an operation rejected by Canaris.20 With reference to the prevailing opinions within constitutional law, Canaris considers the fundamental freedoms to be, on the one hand, defensive rights of the citizen and, on the other, regards them as duties placed upon the state to provide protection. In their character as defensive rights, fundamental freedoms “indirectly” bind private law statutes and jurisprudence. In this manner, an effort is made to prevent the exemption of private law from all requirements of conformity with fundamental freedoms. Both statutory private law creation and the legal principles that guide the judges in matters of legal interpretation or development must be directly measured against the fundamental freedoms. By contrast, in their second character as duties placed upon the state to provide protection, fundamental freedoms should come into effect where the issue is one of the “protection of a citizen . . . from other citizens”21—although, in this case, as Canaris hurries to point out, “the addressee of the norm is clearly not the individual legal subject, but the legislator . . . , who is required, by virtue of the Constitution, to create the requisite protective norm.”22 In this function, the legislator is “supported and complemented by the judges.” The matter is, thus, one of an indirect effect—an effect mediated through the legislature and the judiciary—of fundamental freedoms within the private sector. Canaris refers to a “minimal standard of protection” which is to be concretised in the light not only of the importance and form of the legal good under threat, but also with regard to the intensity of the threat and the ability of the individual who is under threat to offer resistance. Canaris considers the notion that fundamental freedoms can only take effect through the general clauses of private law to be “an unjustified simplification of the analysis, since norms that are precise in their material content may also be imbued with a character protective of fundamental rights and accordingly provide us with an analogous basis for the application and the extension of protective fundamental rights”.23 Here Canaris refers to the threat that private law might limit the protection afforded by fundamental rights: such protection might thus not be made dependent upon the fortuitous existence or otherwise of a requisite mediating norm in private law—or so he chides the supporters of third-party effect.24 A further central thought within Canaris’s reasoning is the fact that private law is often concerned with conflicting interests which can, generally speaking, be overridden, but that must nonetheless be brought into equilibrium within the process of practical concordance amongst these interests. Neither may the degree of protection afforded drop below the constitutionally required minimum, nor may it “disproportionately” restrict the fundamental rights of the 20 21 22 23 24

Canaris, “Grundrechtswirkungen . . .”, supra n. 6. Ibid., at 163. Ibid. Ibid. Canaris, “Grundrechte . . .”, supra n. 6.

Private Law and Constitutionalism 695 second legal subject. Here, Canaris highlights the importance of the openness of the process of the creation of practical concordance, since there must never be only one constitution-conforming solution regarded as the prescribed result of the combined effect of the protective norm and the duty to protect. In this manner, simple law retains room for manoeuvre between the imperative minimum standard of protection for the first fundamental right and the disproportionate and unconstitutional limitation of the second. Canaris’s analysis is structured by the notion that private legal subjects are not directly bound by fundamental rights. In this manner, the scope of the application of fundamental rights is again understood in terms of the vertical citizen–state relationship. However, the conclusion that this structural reference point is prescribed by article 1, paragraph 1 GG is based on an indirect line of reasoning and may be criticised. Seen from the perspective of constitutionconforming processes of private law creation, this formulation can be censured to the degree that fundamental rights can only have an effect in the form in which they are originally given application within a horizontal citizen–citizen relationship.25 Legislation and jurisprudence can only (re)construct those duties which exist between the affected parties as an expression of reciprocity in the sphere of fundamental rights, reciprocity which can be discursively expounded by the parties on their own initiative. As a consequence of Canaris’s reassumption of fundamental rights within a vertical scope of application, private legal subjects only appear in the guise of strategically oriented participants within the market, or in the character of the objects of state policies of protection and are thus never afforded the role of responsible (co)authors of law or solvers of problems. As Mestmäcker noted critically in an early article, the attempt to compensate for the imagined pollution of private law through the doctrine of third-party effect inexorably re-establishes the primacy of the vision of an unpolitical and value-neutral private legal order.26

Hager In his critical engagement with the thesis of an indirect application of fundamental rights to private law, Hager places the notion of the precedence of constitutional law over private law, as a “lower legal order” (passim), in the centre of his considerations and thus arrives at a theory of the direct effect of fundamental rights within private law.27 He argues: “In contrast to prevailing opinion, fundamental rights exert direct effect upon private law within the social sphere (bürgerlich-rechtlichen Gesetze), as well as upon legally-binding 25

For similar analysis, see P Lerche, supra n. 6, with reference to Alexy, supra n. 6. E-J Mestmäcker, “Über das Verhältnis des Rechts der Wettbewerbsbeschränkungen zum Privatrecht”, (1968) AcP 235 at 239. 27 Hager, supra n. 6. 26

696 O Gerstenberg agreements.”28 The point of reference for the systematic analysis is shared with Canaris and derives from the assertion that article 1 paragraph 3 GG also binds the private law legislator and, thus, elevates fundamental rights to standards against which private laws must be assessed. Further, fundamental rights do not merely act as standards against which precise legal rules must be measured, but must also be noted in the case of general clauses, such as paragraphs 133, 138 and 157 of the German Civil Code (Bürgerliches Gesetzbuch, BGB). With this, however, fundamental rights would become the direct standard for the assessment of all legal transactions which, for their part, must conform with the aforementioned BGB provisions. Since these sections—as provisions of simple law—cannot overrule or modify constitutional prescriptions, they are consequently of no value in themselves as regards the Constitution and thus simply the formal legal mechanism upon which the solution, or “cloak”, may be hung, and under which the necessary adjudication may take place. Private law may thus only claim for itself a role independent from the constitutionally determined minimum standard to the extent that the Constitution opens up alternative regulatory options. With this, however, the accent would shift away from the mediation of fundamental rights to the demands of the Constitution, and so to the process of the adjudication of constitutionally protected interests through judges who would thus gain a central importance. In a further step, Hager is concerned to demonstrate that the common distinction between the defensive and protective functions of fundamental rights is, in the final analysis, redundant, since each function simply stands in a mirror relationship to the other. In both the traditional constitutional and private law perspective, the defensive function of fundamental rights is “uncontested” (and thus justiciable with relative ease), while, in sharp contrast, the protective function is indeterminate and open.29 This perspective, or so Hager argues, affords an unacceptable advantage to those who transgress the legal spheres of others: the affected party can only invoke the protective function, while the transgressor has recourse to the defensive function. Yet Hager’s argument fails because the same facts may be regarded in both their transgressionary and their protective aspects, so that this distinction does not entail consequences for the structure of adjudicative processes. When reviewing the substance of contracts, we can reckon with the fact that the less favoured party consented to the contractual rules. Hager’s thesis may, on the one hand, be understood as a radicalization of Canaris’s position. He shares with Canaris the assumptions that private law may be allowed to limit the scope of protection for the fundamental rights of citizens, while the precedence of constitutional law above “simple” private law 28

Hager, supra n. 6 at 383. On this reading, see E-W Böckenförde, “Grundrechte als Grundsatznormen: Zur gegenwärtigen Lage der Grundrechtsdogmatik” in E-W Böckenförde, Staat, Verfassung, Demokratie (Frankfurt am Main, 1991). 29

Private Law and Constitutionalism 697 must be elevated to an analytical point of departure. However, Hager nonetheless radicalises Canaris’s position to the extent that he argues that any analysis based upon the precedence of the Constitution must inevitably result in a theory of the direct binding of private actors by fundamental rights. On the other hand, however, Hager remains loyal to Canaris’s position in that, in placing the entire project of adjudication in the hands of the judiciary, he retains the distinction between “acts of state” and “acts of private legal subjects”: in the final analysis, the judiciary presents private actors with externally determined solutions to problems while, for their part, private actors are not integrated with projects of norm creation.

Diederichsen The resulting problem of the democratic legitimation of judicial social regulation constitutes the point of reference for Diederichsen’s attempt to reinstate the doctrine of direct third-party effect.30 His self-styled “statutorily-positivistic” (gesetzespositivistisch) thesis is based upon variants of well-rehearsed arguments: the danger of an alteration in the balance of powers to the advantage of the courts, including the Constitutional Court; the exploding of the principle of the separation of powers; and the blending of the distinction between constitutional and simple law. Further, the opening up of private law to adjudication amongst constitutionally protected goods is deemed to endanger the formal rationality of private law—or, to be an absorption of private law into constitutional law: “[T]he specific values of codification that derive from the combination of a clear logical structure, precise conceptual constructions and elasticity would be destroyed by indeterminate legal concepts.”31 Diederichsen’s argument validates “the fundamental independence and independent jurisdiction of private law as regards the Constitution”,32 or, alternatively expressed, “the special position of private law within the fundamental value system”.33 Diederichsen attempts to relativize the critique of the precedence of the Constitution over private law with the aid of a metaphor of simple constitutional impact through “radiation”: “[W]here the values of the fundamental rights catalogue ‘radiate’ into private law, they have a lesser impact than they would have had had they been made directly applicable in a second order area of law”.34 The transposition of fundamental rights seemingly occurs “with the aid of the legal institutions of private law and the formulations contained therein”. Private law stands alongside the Constitution and is deemed to be equivalent to it. According to Diederichsen, the applicability of fundamental 30 31 32 33 34

Diederichsen, supra n. 6. Ibid. at 229. Diederichsen, supra n. 6 at 231. Ibid. at 232. Ibid. at 234 , n. 290.

698 O Gerstenberg rights in civil law can only be explained as the outcome of both constitutional and civil law jurisprudence; seen from the perspective of the theory of the sources of law, the extent of the jurisdiction of fundamental rights can thus only reach as far as the preconditions for the creation of customary law allow.

3 SYSTEMATISATION : PRIVATE LAW , THE CONSTITUTION AND THE PRINCIPLE OF DEMOCRACY

At the heart of the normative issue of the relationship between private law, the Constitution and democracy lies the problem of how spheres of social autonomy might be made compatible with one another and how competing legal positions might be brought into an equilibrium. Mutual agreement between spheres of social autonomy is not a given fact, but must constantly be re-established by society in a continuous and constitution-conforming process. While traditional understandings of private law are founded upon a vision of a pre-political private law with static normative content identifiable through syllogistic reasoning and, further, attempt to protect the split-thinking between the Constitution and private law against the presumed “politicisation” arising out of demands for adjudication, modern analyses concentrate more upon the rationality of the procedures through which the equilibrium between spheres of autonomy must be established in the first place.35 Such procedures assume a legitimising power of their own when all affected parties have equal rights of participation, thematicisation and argumentation; in their ideal form, they must make possible forms of problem-solving that will be accepted by all affected parties in line with their common interests. Traditional concepts based upon the third-party effect of fundamental rights are distinguished from this latter approach in that they continuously rephrase the issue of legal effect within the vertical state–citizen relationship and thus relieve private actors of social responsibility. From the traditional third-party effect perspective, the discussion on the relationship between private law and the Constitution necessarily ends in the two—dilemma-ridden—options presented. On the one hand, the Constitution’s claim for supremacy over private law is effected through the theory of a simple “indirect” third-party effect of fundamental rights. In technical legal terms, this determines that the application of fundamental rights must also take place beyond and above the indeterminate legal concepts and general clauses of private law. As a result, the primary concerns here become the state’s protective duties and, as their flip-side, the (presumed) threat of an explicit and potentially limitless juridification of private relationships. The primary characteristic of this state-centred perspective is, thus, the issue of the legitimation of “statutorily-representative judge-made 35 Similarly, the contractual legal analysis of E Schmidt, supra n. 6, who emphasises the dimension of common autonomy usage.

Private Law and Constitutionalism 699 law”.36 Alternatively, in the interest of the limitation and disciplining of judicial power, emphasis is laid upon the independence and special position, as regards the Constitution, of a private law which controls its own level of openness to constitutional values, to the degree that it only recognises the legitimate application of those fundamental rights that, doctrinally speaking, have been mediated through general clauses. The democratic principle nonetheless remains a primary element, although only to the degree that its function is defined negatively through the effort to restrict the potential for judicially-led legal development. In normative terms, the principle of democracy is not concretised. In one model, which elevates the independence of private law to the position of primary analytical building-block (the theory of the private law society), democracy loses its value as a fundamental—expressive of the republican ideal of self-determination—legitimising principle and is relegated to the position of one constitutional value among others. Private law is viewed as a collection of pre-political values and freedoms which do not merely act as a barrier to the interventionism of the modern welfare state, but also form the core substance of modern liberal society, as per G. Böhmer’s formative construction, “Öffentliches Recht vergeht, Privatrecht besteht” (“Public law changes, private law remains constant.”).37 Modern analyses argue that value-pluralism is compatible, not with a “normatively thick” democratic principle, but only with the competitive principle, a principle which, on the one hand, makes possible the free pursuit of self-defined goals through self-determined action and, on the other, might dilute social power.38 Competition numbers among those mechanisms which place the individual in a position to make use of more knowledge than he/she might ever possess: a fact which, at the same time, constitutes a limitation upon the individual’s degree of responsibility for the consequences of his/her actions. By the same token, competition rules cannot be grounded upon the explicit knowledge that an individual has of the effects of his/her actions upon a competitor. Such rules transcend the individual horizons of participants, expressing society’s implicit knowledge and so taking on a pre-political status.39 In contrast to a purely institutional principle of democracy that, on the one hand, attracts constitutional distrust but, on the other, is given instrumental value as a means of limiting judicial decision-making power, the modern analysis is characterised by its normative—argumentation-theoretical (argumentationstheoretisch)—“thickening” of the democratic concept and by its radical 36

Ch von Bar, Gemeineuropäisches Deliktsrecht (München, 1996) 594. G Böhmer, Grundlagen der Bürgerlichen Rechtsordnung, 1. Buch 1 (Tu– bingen:1950), at IX ,with reference to Gustav Radbruch. 38 Richard Epstein, most recently in Mortal Peril: Our Inalienable Right to Health Care? (Reading, MA.: 1997); Mestmäcker, supra n. 26; Mestmäcker, “Die Wiederkehr der bürgerlichen Gesellschaft und ihres Rechts”, (1991) RJ 177. 39 Mestmäcker, supra n. 26. Cf. also the highly ambivalent formulation: “The private legal order is far more a system of norms anterior to the Constitution, which, in its own manner, plays a part in the realisation of the constitutional order and safeguards it.” 37

700 O Gerstenberg socialisation of private law through the proceduralisation of fundamental rights.40 In the manner exemplified in Bosman, the modern analysis is concerned with the binding of private actors through self-directed projects of justification, which may very well be motivated by the pragmatic-institutional goal-setting of the actors but which, at the same time, remain responsive to constitutional requirements. Fundamental rights are released from their state straitjacket and are reconstructed as “trumps” in a continuous argumentation game that reintegrates, at the legal and political level, private actors who are situated within conflicting interest constellations and societal subrationalities. The performative goal of the validation of subjective rights is to confront the functionally differentiated self-regulation experiments within society with requirements of justification, thus converting the internal decision-making process into broader public deliberation. Viewed from the democratic and constitutional theory perspective, this operation concerns justificatory processes which should lead to solutions—i.e., the concrete and mutual congruence of positions of freedom—that are consistent with the mutual recognition of equal and free citizens.41 The role of the judges is one of ensuring the openness of discursive justificatory projects within society in the face of imminent tendencies of self-induced inertia, selfclosure to minority interests and simple backsliding into cheap talk.42 The judicial role is, on the one hand, an activist one, to the degree that it ensures the continuity of the project of constitutional renewal within society through the systematic safeguarding of opportunities for deliberative participation. Judges confront private actors with visions of abstract normative goals which are, on the judges’ part, the outcome of successful—though not finite—learning processes. On the other hand, however, the judicial role also encompasses selfrestraint, in that the private actors themselves are institutionally called upon to pursue such normative visions through experimental forms of innovative selforganisation. The argumentation-theoretical concept of democracy allows us to understand the practice of the judicial application of law as a permanent dialogue between judges and private actors. The judges extend an experimental autonomy to private actors, bounded by abstract normative goals; in return, private actors furnish the judges with the record that they require in order cumulatively to reform and adjust the goals they, the judges, have set. The argumentation-theoretical and experimental model breaks down the sharp distinction between “acts of state” and “acts of the individual” upon which the classical doctrine of third-party effect is founded. To the extent that 40 Cf., for such a thesis in US constitutional law, M C Dorf and Ch Sabel, “A Constitution of Democratic Experimentalism” (1998) 98 Columbia Law Review 267. 41 J Cohen, “Procedure and Substance in Deliberative Democracy”, in S Benhabib (ed.), Democracy and Difference (Princeton, Princeton University Press, 1996) at 95. 42 The embedding of “pluralistic democratic legal development” in societal discussion and learning processes in the context of tort law is emphasised in G Brüggemeier, “Tort Law: Judicial Politics, Democratic Legitimation, Transnationality: An (un)Civil Law Perspective”, in Ch Joerges and O Gerstenberg (eds), Private Governance, Democratic Constitutionalism and Supranationalism (EC Commission, 1998) at 117.

Private Law and Constitutionalism 701 it “thickens” and socialises the democratic concept in an argumentationtheoretical manner, it overcomes that fixation with the state-institutional law-giving process which is characteristic of the concept of “statutorily representative judge-made law” (gesetzesvertretenden Richterrecht). In this way, the model also avoids the common trade-off between legislative and judicial law-making, a trade-off that is made at the inevitable cost of the substantive content of the Constitution and whose outcome cannot but be constitutional revision. Consequently, a republican conception of private law places great emphasis upon the truly “horizontal” validity of fundamental rights, understood as “trumps” in a discursive project encompassing private actors. The (independent) role of private law entails the structuring of these decentralised discourses in order to enable private actors to develop constitutional values with reference to the standard of directly deliberative constitutional conformity.

4 A CRITIQUE

One possible criticism of the understanding of private law outlined above derives from the threat that the envisaged interconnection between autonomy and democratic experimentalism is, in the final analysis, unstable: opportunism and weak motivation on the part of private actors ultimately leads the argumentation-theory analysis back to judicial interventionism.43 This shortcoming has recently been formulated by Friedrich Kübler in a critical evaluation of the discursive theory of law.44 Kübler takes as his starting point Josef Essers’ realization that argumentative rationality exists outside the dogmatic system and its methodologies—to the degree that the establishment, within the legally prescribed alternatives, of a consensus on the rationality of a solution constitutes its own persuasive procedures above the law. Kübler concentrates all his considerations upon the problem of the legitimacy of “an everexpanding process of judicial law-making”, as well as upon the question of how “obedience of thought” might be secured among the judges and their ancillary lawyers. He confronts well-known discursive-theoretical positions with the assertion that the effort to anchor the decisionally relevant social understandings of legal experts within society as a whole is a mere “fiction”; he calls for support from experience with a European jurisdiction that, to date, has found no partner legal community—“partner” in the sense of a community integrated through “a comprehensive ethical-political debate” As a consequence, Kübler 43 The complex relationship of this model to the instructive analysis of Gunther Teubner may not be explored here. Cf. B Graber and G Teubner, “Art and Money: Constitutional Rights in the Private Sphere?”, (1998) 18 OJLS 61; G Teubner, “Vertragswelten: Das Recht in der Fragmentierung von Private Governance Regimes”, (1998) 17 Rechtshistorisches Journal 234; cf., also, O Gerstenberg, “Justification (and Justifiability) of Private Law in a Polycontextural World”, (2000) Social & Legal Studies. 44 F Kübler, “Ideologieverdacht und universale Diskursverpflichtung” in E Schmidt and H-L Weyers, (eds), Liber Amicorum Josef Esser (Heidelberg, 1995) at 91.

702 O Gerstenberg situates the dialogue on competing social models and legal paradigms that are demanded by the “communicative theoretical approach” within the disciplineinternal discourse of the legal system itself. He makes reference to the responsibility of lawyers to engage in interdisciplinary self-interpretation and so lays great value upon the autonomy of the legal system. Seen from the background of “the permanent competition between conflicting value-visions and social models” which characterises a liberal political culture, the reference to the autonomy of law cannot be dismissed out of hand. At the same time, however, a model that places almost all of its faith in the ability of the judiciary to engage in rational self-programming must nonetheless be criticized to the degree that its focus on the political system once again relieves private actors of social responsibility and re-establishes the distinction between public and private spheres or, between “acts of the state” and “acts of the individual”. In that it re-assigns to the political system—mediated through the judiciary (presumably the “least dangerous branch”)—the necessary task of apportioning, in a self-directed and discipline-internal process, spheres of responsibility, as well as the competence to hand down to society praetorially conceived solutions to problems, it effectively relieves private actors of their desirable—especially with regard to privatisation tendencies—duty to consider the consequences of their actions. Kübler’s critique—as a precept of judicial action—threatens to become a self-fulfilling prophecy: efficiency-oriented private actors will no longer be required to take on the role of problem-solvers before an extended public of citizens and to act, through innovative exercises of experimental autonomy, as practical law-makers and as co-interpreters of the Constitution. Equally, were the fundamental distrust of societal problemsolving processes justified, it would remain unclear why we, the citizens, should in turn be called upon to trust in the problem-solving capacities of a judicial elite, which, in firm accordance with its own professional beliefs, produces “right answers”.45

5 CONCLUSION : INTERNAL AND EXTERNAL PERSPECTIVES

The radical horizontalisation of the applicability of fundamental rights (which must be matched by a horizontalisation of the duty to comply 46) through an argumentation-theoretical understanding of fundamental rights as a “trump” in an argumentation game encompassing antagonistic interests, is itself connected to the political science thesis of a general and fundamental alteration in forms of political organisation. The separation of powers model of representative democracy was founded upon the assumption that the legislature would be able comprehensively to anticipate and grasp difficulties in the application of its 45

O Gerstenberg, “Procéduralisation . . .”, supra n. 1. On this point, see the valuable consideration by M Zürn and D Wolf, “European Law and International Regimes: the Role of Law Beyond the Nation State”, (1999) 5:3 ELJ. 46

Private Law and Constitutionalism 703 laws. The envisaged goal was the definitive and finite solution of social problems. From a functional perspective, this goal is linked to the premise of transparency and the limited complexity and stable nature of the subject matter to be regulated; and, from a normative perspective, it draws upon the vision of a distinction between a private-efficiency orientation and public responsibility. The realisation that the legislator has forfeited timely and problem-suited interventionist capacities in none other than the areas of innovative regulatory policymaking, and that these areas have witnessed the creation of new forms of autonomous and decentralised problem-solving, lends support to the suspicion that we are dealing with a structurally conditioned, polyarchical decentralisation of representative democracy.47 A new argumentation-theoretical model has, in its turn, reacted to such suspicions, no longer subsuming emerging societal problem-solving mechanism under the concept of a delegation of the powers of a state subject, and instead denoting this trend to be one of autonomous norm production. The concept of deliberative polyarchy attempts to respond to the idea of functional differentiation through the idea of decentralised directly deliberative participation. In this manner, the constitutionalisation of private law as a means of enabling directly deliberative forms of participation becomes one module within a novel and far-ranging concept of a (potentially) transnational democracy.

47 Cf. A Fung, B Karkkainen and Ch Sabel, “After Backyard Environmentalism: Towards a New Model of Information-Based Environmental Regulation” , November 1998.

Appendix 1 Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment Adopted 10 Dec. 1984, GA Res. 39/46, UN GAOR, 39th Sess., Supp. No. 51, UN Doc. A/39/51 (1985), 1465 UNTS 85 (entered into force 26 June 1987) The States Parties to this Convention, Considering that, in accordance with the Principles proclaimed in the Charter of the United Nations, recognition of the equal and inalienable rights of all members of the human family is the foundation of freedom, justice and peace in the world, Recognizing that those rights derive from the inherent dignity of the human person, Considering the obligation of States under the Charter, in particular Article 55, to promote universal respect for, and observance of, human rights and fundamental freedoms, Having regard to article 5 of the Universal Declaration of Human Rights and article 7 of the International Covenant on Civil and Political Rights, both of which provide that no one may be subjected to torture or to cruel, inhuman or degrading treatment or punishment, Having regard also to the Declaration on the Protection of All Persons from Being Subjected to Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, adopted by the General Assembly on 9 December 1975 (resolution 3452 (XXX) ), Desiring to make more effective the struggle against torture and other cruel, inhuman or degrading treatment or punishment throughout the world, Have agreed as follows: Part I Article 1 1. For the purposes of this Convention, torture means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person in an

706 Appendix 1 official capacity. It does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions. 2. This article is without prejudice to any international instrument or national legislation which does or may contain provisions of wider application. Article 2 1. Each State party shall take effective legislative, administrative, judicial or other measures to prevent acts of torture in any territory under its jurisdiction. 2. No exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political instability or any other public emergency, may be invoked as a justification of torture. 3. An order from a superior officer or a public authority may not be invoked as a justification of torture. Article 3 1. No State Party shall expel, return (“refouler”) or extradite a person to another State where there are substantial grounds for believing that he would be in danger of being subjected to torture. 2. For the purpose of determining whether there are such grounds, the competent authorities shall take into account all relevant considerations including, where applicable, the existence in the State concerned of a consistent pattern of gross, flagrant or mass violations of human rights. Article 4 1. Each State Party shall ensure that all acts of torture are offences under its criminal law. The same shall apply to an attempt to commit torture and to an act by any person which constitutes complicity or participation in torture. 2. Each State Party shall make these offences punishable by appropriate penalties which take into account their grave nature. Article 5 1. Each State Party shall take such measures as may be necessary to establish its jurisdiction over the offences referred to in article 4 in the following cases: 1. When the offences are committed in any territory under its jurisdiction or on board a ship or aircraft registered in that State; 2. When the alleged offender is a national of that State; 3. When the victim was a national of that State if that State considers it appropriate. 2. Each State Party shall likewise take such measures as may be necessary to establish its jurisdiction over such offences in cases where the alleged offender is present in any territory under its jurisdiction and it does not extradite him pursuant to article 8 to any of the States mentioned in Paragraph 1 of this article. 3. The Convention does not exclude any criminal jurisdiction exercised in accordance with internal law. Article 6 1. Upon being satisfied, after an examination of information available to it, that the circumstances so warrant, any State Party in whose territory a person alleged to have committed any offence referred to in article 4 is present, shall take him into custody or take other legal measures to ensure his presence. The custody and other legal

Convention against Torture 707 measures shall be as provided in the law of that State but may be continued only for such time as it necessary to enable any criminal or extradition proceedings to be instituted. 2. Such State shall immediately make a preliminary inquiry into the facts. 3. Any person in custody pursuant to paragraph 1 of this article shall be assisted in communicating immediately with the nearest appropriate representative of the State of which he is a national, or, if he is a stateless person, to the representative of the State where he usually resides. 4. When a State, pursuant to this article, has taken a person into custody, it shall immediately notify the States referred to in article 5, paragraph 1, of the fact that such person is in custody and of the circumstances which warrant his detention. The State which makes the preliminary inquiry contemplated in paragraph 2 of this article shall promptly report its findings to the said State and shall indicate whether it intends to exercise jurisdiction. Article 7 1. The State Party in territory under whose jurisdiction a person alleged to have committed any offence referred to in article 4 is found, shall in the cases contemplated in article 5, if it does not extradite him, submit the case to its competent authorities for the purpose of prosecution. 2. These authorities shall take their decision in the same manner as in the case of any ordinary offence of a serious nature under the law of that State. In the cases referred to in article 5, paragraph 2, the standards of evidence required for prosecution and conviction shall in no way be less stringent than those which apply in the cases referred to in article 5, paragraph 1. 3. Any person regarding whom proceedings are brought in connection with any of the offences referred to in article 4 shall be guaranteed fair treatment at all stages of the proceedings. Article 8 1. The offences referred to in article 4 shall be deemed to be included as extraditable offences in any extradition treaty existing between States Parties. States Parties undertake to include such offences as extraditable offences in every extradition treaty to be concluded between them. 2. If a State Party which makes extradition conditional on the existence of a treaty receives a request for extradition from another State Party with which it has no extradition treaty, it may consider this Convention as the legal basis for extradition in respect of such offences. Extradition shall be subject to the other conditions provided by the law of the requested State. 3. States Parties which do not make extradition conditional on the existence of a treaty shall recognize such offences as extraditable offences between themselves subject to the conditions provided by the law of the requested state. 4. Such offences shall be treated, for the purpose of extradition between States Parties, as if they had been committed not only in the place in which they occurred but also in the territories of the States required to establish their jurisdiction in accordance with article 5, paragraph 1. Article 9 1. States Parties shall afford one another the greatest measure of assistance in connection with civil proceedings brought in respect of any of the offences referred to in article 4, including the supply of all evidence at their disposal necessary for the proceedings.

708 Appendix 1 2. States parties shall carry out their obligations under paragraph 1 of this article in conformity with any treaties on mutual judicial assistance that may exist between them. Article 10 1. Each State Party shall ensure that education and information regarding the prohibition against torture are fully included in the training of law enforcement personnel, civil or military, medical personnel, public officials and other persons who may be involved in the custody, interrogation or treatment or any individual subjected to any form of arrest, detention or imprisonment. 2. Each State Party shall include this prohibition in the rules or instructions issued in regard to the duties and functions of any such persons. Article 11 Each State Party shall keep under systematic review interrogation rules, instructions, methods and practices as well as arrangements for the custody and treatment of persons subjected to any form of arrest, detention or imprisonment in any territory under its jurisdiction, with a view to preventing any cases of torture. Article 12 Each State Party shall ensure that its competent authorities proceed to a prompt and impartial investigation, wherever there is reasonable ground to believe that an act of torture has been committed in any territory under its jurisdiction. Article 13 Each State Party shall ensure that any individual who alleges he has been subjected to torture in any territory under its jurisdiction has the right to complain to and to have his case promptly and impartially examined by its competent authorities. Steps shall be taken to ensure that the complainant and witnesses are protected against all ill-treatment or intimidation as a consequence of his complaint or any evidence given. Article 14 1. Each State Party shall ensure in its legal system that the victim of an act of torture obtains redress and has an enforceable right to fair and adequate compensation including the means for as full rehabilitation as possible. In the event of the death of the victim as a result of an act of torture, his dependants shall be entitled to compensation. 2. Nothing in this article shall affect any right of the victim or other person to compensation which may exist under national law. Article 15 Each State Party shall ensure that any statement which is established to have been made as a result of torture shall not be invoked as evidence in any proceedings, except against a person accused of torture as evidence that the statement was made. Article 16 1. Each State Party shall undertake to prevent in any territory under its jurisdiction other acts of cruel, inhuman or degrading treatment or punishment which do not amount to torture as defined in article 1, when such acts are committed by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. In particular, the obligations contained in articles 10, 11, 12 and 13 shall apply with the substitution for references to torture or references to other forms of cruel, inhuman or degrading treatment or punishment.

Convention against Torture 709 2. The provisions of this Convention are without prejudice to the provisions of any other international instrument or national law which prohibit cruel, inhuman or degrading treatment or punishment or which relate to extradition or expulsion. Part II Article 17 1. There shall be established a Committee against Torture (hereinafter referred to as the Committee) which shall carry out the functions hereinafter provided. The Committee shall consist of 10 experts of high moral standing and recognized competence in the field of human rights, who shall serve in their personal capacity. The experts shall be elected by the States parties, consideration being given to equitable geographical distribution and to the usefulness of the participation of some persons having legal experience. [article 17(2)-33 omitted]

Appendix 2 Alien Tort Claims Act & Torture Victim Protection Act 28 U.S.C. §1350 UNITED STATES CODE ANNOTATED TITLE 28. JUDICIARY AND JUDICIAL PROCEDURE PART IV—JURISDICTION AND VENUE CHAPTER 85—DISTRICT COURTS; JURISDICTION

Alien Tort Claims Alien’s action for tort The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States. Torture Victim Protection “Section 1. Short Title. “This Act may be cited as the ‘Torture Victim Protection Act of 1991’. “Sec. 2. Establishment of civil action. “(a) Liability.—An individual who, under actual or apparent authority, or color of law, of any foreign nation— “(1) subjects an individual to torture shall, in a civil action, be liable for damages to that individual; or “(2) subjects an individual to extrajudicial killing shall, in a civil action, be liable for damages to the individual’s legal representative, or to any person who may be a claimant in an action for wrongful death. “(b) Exhaustion of remedies.—A court shall decline to hear a claim under this section if the claimant has not exhausted adequate and available remedies in the place in which the conduct giving rise to the claim occurred. “(c) Statute of limitations.—No action shall be maintained under this section unless it is commenced within 10 years after the cause of action arose. “Sec. 3. Definitions. “(a) Extrajudicial killing.—For the purposes of this Act, the term ‘extrajudicial killing’ means a deliberated killing not authorized by a previous judgment pronounced by a

712 Appendix 2 regularly constituted court affording all the judicial guarantees which are recognized as indispensable by civilized peoples. Such term, however, does not include any such killing that, under international law, is lawfully carried out under the authority of a foreign nation. “(b) Torture.—For the purposes of this Act— “(1) the term ‘torture’ means any act, directed against an individual in the offender’s custody or physical control, by which severe pain or suffering (other than pain or suffering arising only from or inherent in, or incidental to, lawful sanctions), whether physical or mental, is intentionally inflicted on that individual for such purposes as obtaining from that individual or a third person information or a confession, punishing that individual for an act that individual or a third person has committed or is suspected of having committed, intimidating or coercing that individual or a third person, or for any reason based on discrimination of any kind; and “(2) mental pain or suffering refers to prolonged mental harm caused by or resulting from— “(A) the intentional infliction or threatened infliction of severe physical pain or suffering; “(B) the administration or application, or threatened administration or application, of mind altering substances or other procedures calculated to disrupt profoundly the senses or the personality; “(C) the threat of imminent death; or “(D) the threat that another individual will imminently be subjected to death, severe physical pain or suffering, or the administration or application of mind altering substances or other procedures calculated to disrupt profoundly the senses or personality.”

Appendix 3 Private International Law (Miscellaneous Provisions) Act (U.K.) 1995 (c 42) An Act to make provision about interest on judgment debts and arbitral awards expressed in a currency other than sterling; to make further provision as to marriages entered into by unmarried persons under a law which permits polygamy; to make provision for choice of law rules in tort and delict; and for connected purposes [Tort choice of law sections only] Part III Section 9 (1) The rules in this Part apply for choosing the law (in this Part referred to as “the applicable law”) to be used for determining issues relating to tort or (for the purposes of the law of Scotland) delict. (2) The characterisation for the purposes of private international law of issues arising in a claim as issues relating to tort or delict is a matter for the courts of the forum. (3) The rules in this Part do not apply in relation to issues arising in any claim excluded from the operation of this Part by section 13 below. (4) The applicable law shall be used for determining the issues arising in a claim, including in particular the question whether an actionable tort or delict has occurred. (5) The applicable law to be used for determining the issues arising in a claim shall exclude any choice of law rules forming part of the law of the country or countries concerned. (6) For the avoidance of doubt (and without prejudice to the operation of section 14 below) this Part applies in relation to events occurring in the forum as it applies in relation to events occurring in any other country. (7) In this Part as it extends to any country within the United Kingdom, “the forum” means England and Wales, Scotland or Northern Ireland, as the case may be. (8) In this Part “delict” includes quasi-delict. Section 10 The rules of the common law, in so far as they— (a) require actionability under both the law of the forum and the law of another country for the purpose of determining whether a tort or delict is actionable; or (b) allow (as an exception from the rules falling within paragraph (a) above) for the

714 Appendix 3 law of a single country to be applied for the purpose of determining the issues, or any of the issues, arising in the case in question, are hereby abolished so far as they apply to any claim in tort or delict which is not excluded from the operation of this Part by section 13 below. Section 11 (1) The general rule is that the applicable law is the law of the country in which the events constituting the tort or delict in question occur. (2) Where elements of those events occur in different countries, the applicable law under the general rule is to be taken as being— (a) for a cause of action in respect of personal injury caused to an individual or death resulting from personal injury, the law of the country where the individual was when he sustained the injury; (b) for a cause of action in respect of damage to property, the law of the country where the property was when it was damaged; and (c) in any other case, the law of the country in which the most significant element or elements of those events occurred. (3) In this section “personal injury” includes disease or any impairment of physical or mental condition. Section 12 (1) If it appears, in all the circumstances, from a comparison of— (a) the significance of the factors which connect a tort or delict with the country whose law would be the applicable law under the general rule; and (b) the significance of any factors connecting the tort or delict with another country, that it is substantially more appropriate for the applicable law for determining the issues arising in the case, or any of those issues, to be the law of the other country, the general rule is displaced and the applicable law for determining those issues or that issue (as the case may be) is the law of that other country. (2) The factors that may be taken into account as connecting a tort or delict with a country for the purposes of this section include, in particular, factors relating to the parties, to any of the events which constitute the tort or delict in question or to any of the circumstances or consequences of those events. Section 13 (1) Nothing in this Part applies to affect the determination of issues arising in any defamation claim. (2) For the purposes of this section “defamation claim” means— (a) any claim under the law of any part of the United Kingdom for libel or slander or for slander of title, slander of goods or other malicious falsehood and any claim under the law of Scotland for verbal injury; and (b) any claim under the law of any other country corresponding to or otherwise in the nature of a claim mentioned in paragraph (a) above. Section 14 (1) Nothing in this Part applies to acts or omissions giving rise to a claim which occur before the commencement of this Part. (2) Nothing in this Part affects any rules of law (including rules of private international law) except those abolished by section 10 above. (3) Without prejudice to the generality of subsection (2) above, nothing in this Part—

Private International Law Act 1995 (c 42) 715 (a) authorises the application of the law of a country outside the forum as the applicable law for determining issues arising in any claim in so far as to do so— (i) would conflict with principles of public policy; or (ii) would give effect to such a penal, revenue or other public law as would not otherwise be enforceable under the law of the forum; or (b) affects any rules of evidence, pleading or practice or authorises questions of procedure in any proceedings to be determined otherwise than in accordance with the law of the forum. (4) This Part has effect without prejudice to the operation of any rule of law which either has effect notwithstanding the rules of private international law applicable in the particular circumstances or modifies the rules of private international law that would otherwise be so applicable. Section 15 (1) This Part applies in relation to claims by or against the Crown as it applies in relation to claims to which the Crown is not a party. (2) In subsection (1) above a reference to the Crown does not include a reference to Her Majesty in Her private capacity or to Her Majesty in right of Her Duchy of Lancaster or to the Duke of Cornwall. (3) Without prejudice to the generality of section 14(2) above, nothing in this section affects any rule of law as to whether proceedings of any description may be brought against the Crown. ...

Appendix 4 Revisiting Human Rights in the Private Sphere, Postscript: Developments Related to Pinochet as of January 2001 ANDREW CLAPHAM1

1 INTRODUCTION A number of developments related to the pursuit of justice against Senator Pinochet have occurred since the completion of Chapter 19.2 The publisher and editor of this volume have taken the view that these developments warrant a special mention in order to keep the coverage of the ‘Pinochet affair’ as up-todate as possible before going to press. To that end, this postscript has been prepared, and added as an appendix, in January 2001.

2 PINOCHET’S HEALTH AND JUDICIAL REVIEW APPLICATIONS BY BELGIUM AND NON-GOVERNMENTAL ORGANISATIONS The present narrative follows on from that found in Chapter 19. It will be recalled that the House of Lords judgment in March 1999 determined that Senator Pinochet had no immunity with regard to the charges of torture committed after 8 December 1988 (the date when the United Kingdom ratified the Convention against Torture).3 Having received representations from the Chilean Government concerning Pinochet’s health, the Home Secretary asked for Senator Pinochet to submit to medical tests. The Home Secretary concluded 1 Associate Professor of Public International Law, Graduate Institute of International Studies, Geneva. 2 A Clapham, “Revisiting Human Rights in the Private Sphere: Using the European Convention on Human Rights to Protect the Right of Access to the Civil Courts”, chapter 19 of this volume. 3 R v. Bow Street Metropolitan Stipendiary Magistrate and others, ex parte Pinochet Ugarte (No. 3), [1999] 2 All E.R. 97 (Pinochet No. 3).

718 Appendix 4 in January 2000 that, in the light of those tests, Pinochet was unfit to stand trial, and that no change to that position could be expected. The Home Office stated that the Home Secretary “is minded, subject to any representations he may receive, to take the view that no useful purpose would be served by continuing the present extradition proceedings and he should therefore decide not to extradite Senator Pinochet.”4 This announcement was met with legal challenges by a number of human rights groups and by Belgium.5 The claim by the human rights groups included a specific argument by Justicia, an organisation working for the victims of torture in Chile, that a decision not to extradite on medical grounds “would amount to at least an arguable denial of access to the courts in breach of art 6(1) of the European Convention on Human Rights and that this illustrates why the Secretary of State is bound to consider the interests of the victims as claimants for compensation in Spain when exercising discretion under s 12 of the Extradition Act.”6

Kay J was not convinced that the rights of the victims under the European Convention on Human Rights (Convention) were relevant in such a decision on extradition: “I do not accept that it is arguable that an as yet prospective decision not to extradite pursuant to s. 12 of the Extradition Act, or an actual such decision if it comes to be taken, must take account of the civil claims which it is proposed should attach to the criminal prosecution in Spain. Mr Sumption [counsel for the Home Secretary] put it somewhat laconically: extradition is not concerned with civil justice in a foreign (to which I might add, friendly) country. I have no doubt he is correct.”7

This application for permission to apply for judicial review was denied both for the human rights groups and for Belgium.8 However, on appeal, the application for judicial review succeeded, and the medical reports were shown to the states that had requested extradition. 4 Statement of 11 January 2000, reproduced in Guardian Unlimited, This postscript relies heavily on the reports usefully collected at http://www.guardianunlimited.co.uk/Pinochet_on_trial. 5 In a parallel development a Belgian judge had asked through letters rogatory for the appointment of a panel of independent medical doctors to examine Senator Pinochet’s actual state of health, and for the presence of a Belgian forensic doctor at such an examination. Further letters rogatory requested access to the existing medical reports. The failure of the Home Secretary to assure Belgium that no decision with regard to Senator Pinochet would be taken before these requests had been considered led to Belgium’s application for judicial review. It was also reported that Belgium might ask the International Court of Justice to order provisional measures against the United Kingdom in order to stop any move to allow Pinochet to return to Chile. ‘Belgium moves to halt Pinochet flight’, Guardian Unlimited, supra n.4, 26 January 2000. 6 R v. Secretary of State for the Home Department, ex parte Amnesty International and others, Unreported Judgment of 31 January 2000, Kay J, Queen’s Bench Division (Transcript on file with the author). 7 Ibid. 8 It should be noted that this was not a case in which the obligations under the Convention could be directly invoked before the English courts, as the Human Rights Act 1998 would only enter into force in England on 2 October 2000. See G Virgo, “Characterization, Choice of Law, and Human Rights”, chapter 12 in this volume, for a discussion of the impact of the Human Rights Act on transnational civil claims.

Revisiting Human Rights in the Private Sphere, Postscript 719 For present purposes we should note how the European Convention argument regarding the victims’ rights was dealt with by the judges on appeal. Note that, by the stage of the hearing on appeal, the human rights organisations were not asking to see the reports, only for the reports to be disclosed to the four requesting States. While the Court in this appeal did in fact order disclosure of the medical reports on a confidential basis, it reiterated that Convention rights were not really engaged. Simon Brown LJ did not see any Convention obligation as relevant to the Home Secretary’s proposed decision: “A number of torture victims represented by Justicia propose to bring civil claims in Spain which would be dependent upon the successful prosecution there of Senator Pinochet, and that this prospect ought not to be curtailed by the Secretary of State’s proposed decision without the process being attended by greater procedural safeguards than have thus been afforded. Of course one sympathises deeply with the victims of torture: their concerns that justice in the extradition process be seen to be done is readily understandable. Their civil claims, however, to my mind add nothing to their urgent plea that, if at all possible, Senator Pinochet should stand trial.”9

Similarly, Latham LJ saw no merit in the argument that the return of Pinochet to Chile trespassed on rights to access to a court under article 6 of the Convention: “Whilst the interests of victims are capable of being a consideration relevant to the exercise of the Secretary of State’s discretion, a refusal to order Senator Pinochet’s return [extradition] in no way interferes with the rights of those victims in the United Kingdom nor does it preclude them from bringing proceedings wherever they wish to do so.”10

5 APPLICATIONS TO THE EUROPEAN COURT OF HUMAN RIGHTS CHALLENGING THE UNITED KINGDOM’S DECISION TO RETURN PINOCHET On 2 March 2000, the Home Secretary, having considered the further representations he received as a result of the Court’s disclosure ruling, announced that he would not order the extradition of Senator Pinochet to Spain. Senator Pinochet left for Chile later that day. The earlier-mentioned human rights NGO, Justicia, had filed an application with the European Court of Human Rights alleging violations of articles 6(1) and 13 with regard to the process culminating in the Home Secretary’s decisions not to extradite or allow the prosecution of Senator Pinochet. This application 9 R v. Secretary of State for the Home Department, ex parte The Kingdom of Belgium; R v. Secretary of State for the Home Department, ex parte Amnesty International and others, Unreported Judgment of 15 February 2000, Queen’s Bench Division (Transcript on file with the author). 10 Ibid.

720 Appendix 4 (Application no. 54674/00) remains on the Court’s docket at the time of writing of this postscript in January 2001. The irony in the present context is that the Home Secretary’s reasons for his decision specifically state that any other decision could result in a violation of Senator Pinochet’s article 6(1) rights under the European Convention as he would not be able to enjoy a fair trial in Spain.11 In effect, any article 6 right to access to court that the victims might have was overridden by the priority which was given to Pinochet’s own article 6 rights with regard to a future trial. At stake from the victim’s perspective was a choice between the rights of two private parties with both sides relying on the European Convention. Whether or not any victims’ rights under the European Convention were violated by the decision of the Home Secretary, or whether Pinochet’s Convention rights would have been violated by Spain in the event of a trial, are questions which do not lend themselves to easy answers. But we should not confuse the complex English domestic law questions of standing and the reasonableness of a Ministerial decisions on extradition (including the reasonableness of the balancing of what may have been two competing article 6 rights) with the international law question that was the subject of Chapter 19: does article 6 of the European Convention assist victims in their quest to get access to the civil courts for complaints about torture? If we see the right to complain about torture as a civil right and a question of civil jurisdiction, two questions remain to be resolved. First, does the Convention demand that the national legal order recognise the application of human rights in the private sphere and provide a remedy? The question whether the UK judges will create such a common law remedy as a result of the Human Rights Act 1998 has generated a good deal of commentary,12 but in the end the issue will not only depend exclusively on the judicial creativity of the British judges. It will also depend on whether the European Court of Human Rights will judge that the absence of a civil remedy for any human rights complaint against a private individual represents a violation of article 6(1) (access to court) or article 13 (effective remedy) of the European Convention. In his article on the application of human rights standards in private disputes, Jonathan Cooper highlights a decision of the European Commission of Human Rights, just prior to the Commission’s abolition as a result of Protocol 11, with regard to a complaint brought by Earl and Countess Spencer regarding their right to private life in the context of press reports about the Countess’ health.13 11 ‘Straw tells why charges were not possible, Edited text of Jack Straw’s speech to MPs’, Guardian Unlimited, supra n. 4, 3 March 2000. 12 See M Hunt, “The Horizontal Effect of the Human Rights Act”, (1998) Public Law 423, and also J Cooper “Horizontality: The Application of Human Rights Standards in Private Disputes” in P Havers and R English, eds., An Introduction to Human Rights and the Common Law (Oxford, Hart, 2000) 53-69; D Oliver, “The Human Rights Act and Public Law/ Private Law Divides”, (2000) EHRLR 343-61; T Raphael, “The Problem of Horizontal Effect”, (2000) EHRLR 493-511. 13 Cooper, ibid. at 63.

Revisiting Human Rights in the Private Sphere, Postscript 721 Although the application was dismissed, the Commission’s decision includes the following important passage: “On the facts as presented by the parties, the Commission would not exclude that the absence of an actionable remedy in relation to the publications of which the applicants complain could show a lack of respect for their private lives. It has regard in this respect to the duties and responsibilities that are carried with the right of freedom of expression guaranteed by Article 10 (Art. 10) of the Convention and to Contracting States’ obligation to provide a measure of protection to the right of privacy of an individual affected by others’ exercise of their freedom of expression (see, mutatis mutandis, No. 10871/84, Dec. 10.7.86, D.R. 48, p. 158 and No. 31477/96, Dec. 15.1.97, unpublished).”14

Of course the Pinochet victims’ situation is not as straightforward as the Spencers. There is an added layer of complication to the right to a civil remedy as the original violence for the Pinochet victims is seen as having occurred outside the jurisdiction. This leads us to our second remaining question: does the Convention help where the victims of torture demand civil jurisdiction over events which originally took place abroad? This issue has been at the heart of a parallel series of cases concerning torture in Kuwait and the recognition of state immunity for Kuwait in the English courts.

4. AL-ADSANI AND KUWAITI IMMUNITY AT THE EUROPEAN COURT OF HUMAN RIGHTS Mention has been made in several chapters of the Al-Adsani case, brought in the UK, in which torture allegations against the state of Kuwait were dismissed on the basis of Kuwait’s state immunity from the jurisdiction of domestic courts for conduct occurring in Kuwait.15 Subsequent to the English Court of Appeal’s dismissal of the case, Mr Al-Adsani has taken his claim to the European Court of Human Rights. The application in Sulaiman Al-Adsani v. United Kingdom has been declared admissible by the Court. The ruling being sought is a finding that the reliance on state immunity law by the English courts denied the applicant access to a court to complain about torture by the Kuwaiti authorities.16 As the case was declared admissible with regards to articles 3, 6(1) and 13 of the European Convention, it is likely that there will be a crucial international ruling 14 European Commission of Human Rights, Applications 28851/95 and 28852/95, Decision of 16 January 1998, available at http://www.echr.coe.int/. 15 Al-Adsani v. Government of Kuwait and Others (1996), 107 ILR 536 (Eng., CA). 16 European Court of Human Rights Application 35763/97, deliberation of 1 March 2000. The assessment of the Court leaves the issues open: “The Court notes that the Government, in arguing that the torture alleged did not fall within Article 1 of the Convention and that the proceedings in question did not concern a ‘right’ recognised under domestic law, are pleading that the application is incompatible with the provisions of the Convention under Article 35§. However the Court is of the view that these arguments are closely linked with the substance of the applicant’s complaints under Articles 3, 6§1 and 13 of the Convention.” Available at http://www.echr.coe.int/.

722 Appendix 4 on the relationship between torture and access to courts regarding foreign torturers in civil litigation. The applicant’s arguments are summarised in the Court’s admissibility decision: “As regards Article 6 § 1 of the Convention, the applicant argues that this provision applies to the proceedings in question. Torture is a civil wrong in English law and the United Kingdom asserts jurisdiction over civil wrongs committed abroad if the plaintiff is resident in the United Kingdom and has a sufficient interest in the matter. The applicant has that status and the domestic courts accepted jurisdiction over his claims against the individual defendants. The applicant’s claim against the Kuwait Government was not defeated because of its nature but because of the identity of the defendant. By conferring immunity on foreign Governments the domestic courts removed a category of persons from the jurisdiction of the courts. This cannot escape the control of the Court. As regards the Government’s submissions concerning the non-justiciability of acts of State, the applicant argues that torture is not an act of State. The doctrine of act-of-State non-justiciability is based on a judicial policy of self-restraint, which, as opposed to a statute, cannot prevail over international law. In any event, the doctrine of act of State, which is unclear, does not apply to all action by a foreign State within its territory. The applicant argues that restricting the right of access to court in his case does not serve a legitimate aim. The House of Lords in ex parte Pinochet (No. 3) accepted that the prohibition of torture had acquired the status of a jus cogens norm in international law and that torture had become an international crime before the UN Convention against Torture. It follows that the applicant’s torture in Kuwait was an offence prohibited under jus cogens. In the applicant’s view, in these circumstances there can be no immunity from criminal jurisdiction. The applicant argues that there is no rational basis for allowing sovereign immunity in a civil action when immunity would not be a defence in a criminal action arising from the same set of facts. Sovereign immunity is trumped by jus cogens across the board. Otherwise, there would be no forum where torture can be redressed other than the forum of the locus delicti.”17

We shall have to wait and see how the Court deals with this application but its judgement will be of critical importance for future developments in this field.

5 ONGOING DEVELOPMENTS: THE HAGUE CONFERENCE AND PROCEEDINGS AGAINST PINOCHET IN CHILE Closely related to the European Convention issue of access to a court in cases with a transnational dimension are parallel developments within the Hague Conference on Private International Law on the question of national jurisdiction over civil cases where the injury took place abroad. The negotiations over the Draft Convention on Jurisdiction and Foreign Judgments in Civil and Commercial Matters have not yet concluded to what extent restrictive rules on civil jurisdiction over events abroad should contain exceptions with regard to human rights violations. The treaty is due to be finalised in 2002. However, it is 17

Ibid.

Convention against Torture 723 noteworthy that the provisional text already allows, in certain circumstances, for access to the civil courts to complain about acts of torture committed abroad.18 At the time of writing there is no point in further speculation on how these questions will be answered by the Hague Conference. But in the long run it seems that we will see a recognition that international law will at least permit, and to some extent encourage, the courts of one country to allow civil claims for torture – even where the torture was committed abroad. As for Senator Pinochet, the Supreme Court of Chile on 8 August 2000 upheld by 14 votes to 6 the Court of Appeal’s decision to lift the immunity which attached to Senator Pinochet as Senator for life.19 Senator Pinochet agreed to medical tests in January 2001,20 thus opening up the possibility that he could be questioned by the Chilean investigating magistrate, Judge Guzman, who has been pressing for Pinochet to be prosecuted under Chilean law now that he is back in Chile. At the time of writing Senator Pinochet was due to be questioned in relation to these charges which concern 19 people executed by the military with no due process as part of the Caravan of Death. There are also reported to be over 200 other criminal cases moving through the Chilean courts. Even if Pinochet were to be eventually found unfit for trial under Chilean law with regard to the criminal charges, this would not necessarily prevent other actions related to the freezing of his assets. Nor would it preclude an ongoing parliamentary investigation into drug smuggling by the Chilean armed forces in 1988.21

18 Preliminary draft Convention on Jurisdiction and Foreign Judgments in Civil and Commercial Matters, adopted by the Special Commission of the Hague Conference on Private International Law on 30 October 1999, amended version (new numbering of articles):

Article 18 (3) Nothing in this Article shall prevent a court in a Contracting State from exercising jurisdiction under national law in an action [seeking relief] [claiming damages] in respect of conduct which constitutes – [Variant One:[a) genocide, a crime against humanity or a war crime[, as defined in the Statute of the International Criminal Court]; or] [b) a serious crime against a natural person under international law; or] [c) a grave violation against a natural person of non-derogable fundamental rights established under international law, such as torture, slavery, forced labour and disappeared persons]. [Sub-paragraphs [b) and] c) above apply only if the party seeking relief is exposed to a risk of a denial of justice because proceedings in another State are not possible or cannot reasonably be required.] Variant Two: a serious crime under international law, provided that this State has established its criminal jurisdiction over that crime in accordance with an international treaty to which it is a party and that the claim is for civil compensatory damages for death or serious bodily injury arising from that crime.] Available at http://hccch.net/e/conventions/draft36e.html Le Monde, 10 August 2000, at 2 and 9. 20 ‘Pinochet agrees to health tests’, Guardian Unlimited, supra n. 4, 10 January 2001, . 21 Ibid. 19

724 Appendix 4 6 CONCLUSION As we have seen, using the European Convention to protect access to the civil courts for foreign torture has not yet proven conclusive or even particularly persuasive. But what is clear is that the Convention is likely to continue to generate expectations that new remedies and new forms of victim participation should be provided by States party to the European Convention. In short article 6 of the Convention looks likely to continue to be a powerful tool for demanding access to justice for human rights disputes in the private sphere.

Index Access to court, right of under European Convention on Human Rights (ECHR), 513–35 Acquiescence, and doctrine of tacit approval, 504 Act of state, 16–17 amnesties, 358–9, 589–95 Anglo-Canadian law, 125–9, 343–71 Australia, 625–8 bindingness of judgment, 59 choice of law doctrine, 16 interaction with state immunity, 343–4, 363, 368–71 Israel, 653–6 non-justiciability, 16 United States, 78–81, 360–3 Actionability, see causes of action Adjudicative jurisdiction, 157–96 amnesties, 589–95 Canada, 119 injunctions, 601–29 See also forum non conveniens, Local remedies and Service Agency: attribution, 21, 24, 36, 500–1 , 504 derivative liability, 450 Agent Orange litigation, 204, 210 Al-Adsani litigation, 337, 722–3 Alien Tort Claims Act (ATCA), 3, 14, 68–70, 73, 74–6, 85–102, 110, 113, 506 act of state, 16 causes of action, 407–10 command responsibility, 448–50 See also under United States Alternative fora, 10, 29, 39, 189, 398–9 See also forum non conveniens Amnesties: act of state, 358–9, 589–95 as shields in foreign litigation, 111, 26, 566–600 duty to compensate, 286–7 International Criminal Court, 26 treatment under the Convention Against Torture (CAT), 13 Applicable law, see Governing law and Choice of law Arone litigation, 6, 8, 24, 33–44, 130–4 Attribution: agency, 500–1, 504 state action requirement in US, 92–102 Australia, and human rights injunctions, 27, 28, 601–29

Authority, binding versus persuasive, 58–60, 676–685 Belgium, re Pinochet, 719 Bhopal litigation, 10, 198, 202, 203, 205–9, 492 Canada, torture in Somalia, 6, 8, 24, 33–44, 130–4 Capitalism: globalisation of conflicts values, 10, 11, 12, 21, 201, 203, 306–7 neoliberal ideology, 239–43 state responsibility, 494, 504–5, 507–8 CAT, see Convention Against Torture Categories, as overly restrictive, 45, 666–9 Causes of action: Alien Tort Claims Act (ATCA), 74–6, 407–10 characterisation of, 62–3 Convention Against Torture (CAT), 535–50 for torture in transnational law, 373–400 international human rights law, 389–99, 401–38 reasonableness, 33–44 relevance of interpretive principle of harmonisation, 258 sufficiency of existing torts, 109–18, 334–6 transnational law, 17, 62 Certainty, and public law, 687–703 Chad, case against former leader, 114 Characterisation, in conflict of laws, 14–16, 62–3, 308–9, 320, 325–42 Chile, amnesty in, 571–3, 724 Choice of law: amnesties, 595–600 Canada, 123–5, 292–4, 296–301, 311–15 international human rights law, 401–38 limitations periods, 38–44 tort in UK, 294–6, 319–21, 323 tort in US, 85–9, 124, 316–18 tort (general), 291–323 See also Governing Law Civil law: judicial role, 31, 57 reception of forum non conveniens, 194–5, 205 Civil liability: corporations in home state courts, 491–512 derivative liability of leaders, 442–63 implications of criminal proceedings for, 523–9

726 Index Civil redress: article 14 of the Convention Against Torture (CAT), see Convention Against Torture (CAT), article 14 torture prevention, 146–8 Civil remedies, under the Convention Against Torture (CAT), 535–50 Classification, see characterisation Colonialism, 497–8 Comity: act of state, 16, 353 adjudicative jurisdiction, 162–3, 166, 180 amnesties, 592 choice of law, 301–6 discourse of, 9, 204 See also under territoriality See also Extraterritoriality human rights injunctions, 621–3, 628–9 Command responsibility, doctrine of, 19, 94, 441–50, 459–61 Commercial activity exception, to state immunity, 11, 71, 213–45 Common law judicial reasoning, 30, 34, 41 Common law method, and transnational law, 30, 34, 41, 684–5 Common law, reception of customary international law, 417–29 Communitarianism, versus transnational tort premises, 554–9, 564–5 Compensation: justice of amnesties, 587–8 under international human rights law, 112 under the Convention Against Torture (CAT), see Convention Against Torture, article 14 Competence, discourse of, 204 Complicity, see State responsibility and Multinational corporations Conflict of laws, see Private international law Consent, and commercial activity exception, 228–32 Constitutional torts, and Canada, 423–9 Constitutionalism: European Convention on Human Rights (ECHR), 23, 530–1 private law in Germany, 687–703 Contract law, and state immunity, 11 See also Capitalism Contract, relationship to state immunity law, 218–20, 228–32 Control of private actors, duty of, 505–11, 516–21 Convenience, discourse of, 204 Convention Against Torture (CAT): article 14, 13, 18, 23, 24, 37, 119–20, 123, 131, 174, 253, 260–4, 286–7, 305–6, 310, 398, 406, 430–1, 535–50 civil remedies, 535–50

criminal law focus, 251 derivative liability, 458–9 extraterritorial application of, 541–9 interaction with private international law, 540–1, 549–50 nationality jurisdiction, 167 See also Convention Against Torture, article 14 Cooperation theory, and immunity doctrine, 232–9 Corporate accountability, 45–63, 54, 491–512 Corporate codes of conduct, 60, 494 Cosmopolitanism, 16, 661–85 Council of Europe, admission to, 149 Courts, role of, see Legitimacy and Justiciability Criminal law: analogy for civil jurisdiction, 13, 169 benefits compared to civil law, 8, 109, 115–18 Convention Against Torture (CAT), 276, 281–7 implications for civil liability, 13, 23, 41, 276, 281–7, 523–9, 645–50 universal jurisdiction, 3, 5, 171–4 Crown prerogative, and act of state, 359 Cultural differences, 25, 27, 28, 380, 601–29 See also under Legitimacy Customary international law: civil causes of action, 76, 91, 375–6, 417–29 relationship to common law, 50, 417–29 Death penalty, and foreign court injunctions, 601–29 Dédoublement fonctionnel, concept of, 179–81 Democracy, 258, 691, 702–3 Derivative liability, two forms of, 19, 442–63 Deterrence, 8, 22 Dialogue: cross-cultural, 27 in legal academy, 5 national/international, 114 private law adjudication, 32, 687–703 transnational judicial, 27, 29, 681–3 Diplomatic immunity, 277 Diplomatic protection: corporations, 50, 492, 497–501 relevance to civil jurisdiction, 174 Direct effect, of treaty law, 50 Diversity jurisdiction, 73 Domestic remedies, exhaustion of, see Local remedies Double aspect, in state immunity doctrine, 220–2, 236–9 Drittwirkung, horizonatal effect doctrine in Germany, 31, 32, 48, 687–703 Duty to ensure, and private law, 48–9

Index 727 Economic benefits, and immunity doctrine, 232–9 ECtHR, see European Court of Human Rights Effective remedies, 23, 42 and article 13, 23 right under ECHR, 532–3 Enforcement of foreign judgments: Israel, 654–6 United States, 209 Enforcement: as paradigm, 30 through injunctions, 601–9 Enterprise liability, doctrine of, US, 452–4 Enterprise risk, doctrine of, Canada, 454–6 Erga omnes obligations, 172–4, 378–9, 397–8, 400, 434–5, 502 Ethical discourse in law, 10, 25 European Convention on the Prevention of Torture (European CPT), approach of, 8, 136–53 European Court of Human Rights (ECtHR): article 6 and extradition, 719–22 civil remedy under, 721–2 interaction with UK Human Rights Act, 15, 340 powers of judgment, 59 relationship to European Convention on the Prevention of Torture, 143, 145, 146 right to a court, 22, 513–35 European CPT, see European Convention on the Prevention of Torture European Union, Brussels Convention, 31, 52, 120 Exhaustion of local remedies, see Local Remedies and see also under Forum non conveniens Extradition, impact on civil rights of victims of crime, 523–9 Extraterritorial torture by states, duty to redress under CAT, 535 Extraterritoriality: adjudicative jurisdiction, 162–3 injunctions as remedy, 601–29 jurisdiction and regulation, 18, 21, 27, 36, 53–5, 56 power/duty to exercise, 492–512, 535 See also territoriality and comity and Convention Against Torture, article 14 Federal question jurisdiction, 72–3 Field operations, UN, 465–89 Formal invalidity of foreign law, and act of state, 346–50, 352 Forum non conveniens, 9, 10, 29, 39, 51 as common law export, 194–5, 205 Australia, 623–5 Canada, 121–3 , 194–5 Ontario, 188–95

local remedies rule, 176–9 responsibility versus interest in, 207 United States, 83–4, 197–212 See also Adjudicative jurisdiction General principles of law, 47, 50 Geographies, underlying conflict of laws, 203 Germany, doctrine of drittwirkung in, 48, 687–703 Globalisation, see Capitalism Governing law, 14, 18, 38, 55, 63, 123–5 Alien Tort Claims Act (ATCA), 87 forum non conveniens analysis, 191 international law as, 87 jus cogens, 378–9 lex fori interaction with international law, 98 lex forism, 211 limitations periods, 38–44 relationship to international human rights law, 389–99, 401–38 See also choice of law Guzman, proceedings against Pinochet, 724 Hague Conference, Draft Convention on Jurisdiction, 723–4 Harmonisation, as principle of interpretation, 12, 13, 24, 40, 41, 50, 129, 256–60, 268–71, 376, 387–9, 421–2 Hazardous wastes, transfer of, 501 Head of state immunity, 11, 84, 277, 282–6, 343–4, 363 See also Pinochet Home state responsibility, for corporate conduct, 10–11, 21, 491–512 Horizontal effect, 19, 22, 48, 341 in EC law, 687–8 in German law, 687–703 Human Rights Act (UK), 15, 339–42, 719 Human rights, see under Jus cogens, Erga Omnes, Torture and Slavery Humanitarian law, 20, 33, 99–102, 131–2 Hybridity, and transnational tort adjudication, 661–85 ICTFY, see International Tribunal for the Former Yugoslavia Immunity: act of state, 17 amnesties, 566–600 civil cases under CAT, 549 diplomatic, see Diplomatic immunity European Convention on Human Rights (ECHR), 20, 22 heads of state, see Head of state immunity intergovernmental organisations, 20, 475–85 of police from suit, 22, 521–3 sovereign immunity, see State Immunity state immunity, see State Immunity

728 Index Impunity: CAT, 275–87 distinguished from amnesty, 577 Indirect responsibility, of individuals, 442–63 Individual responsibility, see Criminal law, international, and Civil liability and redress Inhuman and degrading treatment or punishment: as evolving concept, 146 US law, 98 See also Torture Injunctions, 27 as remedy for human rights violations abroad, 601–29 Australia, 107, 601–629 United States, 107 Interest versus responsibility, in forum non conveniens, 207 Intergovernmental organisations, liability and responsibility of, 465–89 International civil court, desirability of, 595 International Criminal Court, 26 amnesties, 26, 398–9 International criminal law, see Criminal law, international International criminal responsibility, for acts during UN operations, 487–9 International human rights law: as basis for cause of action, 14 as governing law, 389–99, 401–38 International Law Commission, 500, 507, 512 International Tribunal for the Former Yugoslavia (ICTFY), 19, 114, 138 Interpretation: harmonisation principle, see Harmonisation of international treaties, 24 role of acquiescence, 57–8 Interrogation techniques, as torture, 632–58 Interstate remedies, relevance to civil jurisdiction, 116, 174–81 Intervention: by foreign courts, see Judicial imperialism into affairs of foreign states, see Legitimacy and Act of State Islamic law, regarding death penalty, 601–29 Israel, and interrogation practices, 28, 632–58 Judges: legalism and resistance by, 10, 12, 30, 31, 34 role of, see Legitimacy and Judicial imperlism Judicial ideology, 46–47, 197–2 12 Judicial imperialism, 5, 12, 25, 396–9, 553–6 Judicial Jurisdiction, see Adjudicative jurisdiction and forum non conveniens Judicial role, compared to legislative role, see Justiciability

Jurisdiction: duty/power to exercise, 55–6, 174 forms of, 9, 54–5 territorial, see Territoriality See also under Adjudicative jurisdiction and Universal jurisdiction and Nationality Jus cogens, 405, 434 Alien Tort Claims Act (ATCA), 91–2 amnesties, 576 choice of law, 301–6, 378–9 common law, 422–3 implied exceptions in domestic statutes, 12 interpreting immunity statutes in light of, 253, 255, 265–74, 723 move from power to duty, 504 nationality jurisdiction over corporations, 492 norms and human rights, 6, 12, 14, 18, 21 relationship to discretionary service ex juris, 121, 187–8 relationship to universal adjudicative jurisdiction, 172–4 Justice, of amnesties, 566–600 Justiciability, 247–54, 268–74 amnesties before foreign courts, 593–5 concerns re transnational law, 375, 383–9 distinct from normativity, 56–7 incentives, 594–5 judicial legitimacy in developing private law, 701–2 legislative role in Canada, 118–23 political questions, 81 role of courts versus legislatures, 12, 25 timeliness of foreign court role, 29 See also Legitimacy and non-justiciability Kuwait, torture in, 337, 722–3 Laws of war, see Humanitarian law Leaders, liability of, 442–63 Legalism, as danger, 239–45, 553–66 Legislative role, compared to judicial role, see Justiciability Legitimacy, and transnational torts, 5, 225, 553–66 Lex causae, see Governing law Lex fori, see Governing law Lex loci delicti, see Governing law Liability: allocation between states and the United Nations, 484–7 of leaders, 442–63 Limitation periods, 34, 38–44 Limitations: and choice of law, 38–44 in Arone case, 132 in the United States, 103–6 Local remedies, exhaustion of, 9, 83–4, 174, 176–9

Index 729 Market, see Capitalism Multilateral Agreement on Investment, 496 Multinational corporations, and accountability for human rights harms, 7, 10, 19, 21, 32, 491–512 Multiplicity, of sources, 15, 661–86 National Defence Act (Canada), 39 Nationality jurisdiction, 10, 33, 34–44, 167–9, 492–512 Necessity, as legal defence to torture, 632 Negligence, and torture, 21, 34–7, 131 Neoimperialism, 243–5 See also Colonialism Neoliberalism, as ideology, 239–43 Nominate torts, see Torture, characterisation as tort and Tort, assault and battery Non-justicability: Al-Adsani, 723 as distinct from act of state, 16, 345, 356–9, 370 transnational doctrine of, 593–5 United Nations law, 20 Non-state violence, duty to protect against, 19, 516–21 Nuremberg and Tokyo, 19, 128, 279, 443–4 Objective territoriality jurisdiction, 170 Official conduct, 16, 17, 20, 39 On-site visits, and torture prevention, 141–8 Ordre public, of Europe, 530–1 Overriding norms, see Jus cogens Pardon, amnesty functioning like, 584–5 Passive personality jurisdiction, 169 Peremptory norms, see Jus cogens Pinochet affair, 5, 13, 16, 20, 22, 23, 39, 110, 114, 115, 282–7, 569–74 Pinochet: implications for act of state doctrine, 16, 344, 365–71 implications of non-extradition for civil rights, 523–33 relevance to Israeli context, 643–50 update, 718–25 Police liability and immunity, 22 Political economy, see Capitalism Political question doctrine, 81–2 Predictability, and concern re public law, 687–703 Prevention: as objective, 8, 136–53 models of, 136–9 under CAT, 173 Principled instrumentalism, in choice of law, 431–8

Private International Law Act (UK), 14, 15, 319–21, 323, 337–9 Private international law, pathologies of, 10, 197–212 Private international/public international law relationships, see Public international/ private international Private law theory, and public law values, 4, 661–703 Private law: autonomy from public law, 687–703 democracy, 687–703 Private sphere, and human rights accountability, 513–35 Private/public law relationships, see Public/ private Protective jurisdiction, 171 Public Authorities Protection Act (Ontario), 38 Public international law, relation to private international law, 9, 15, 18, 51–3, 163, 196, 197, 333–4, 354–6, 429–437, 540–1, 549–50 Public policy exception, 16, 17, 59, 63, 125, 198, 209, 350–6 Public/private law relationships, 4, 32–4, 45, 48–9, 89–3 99, 661–703 Punishment, as objective, 136–53 Reasonable cause of action, 33–44 Reception, of international law, 49–50 in Westminster systems, 248–60, 271 Recognition and enforcement of foreign judgments, 29, 209, 654–6 Recognition of new torts, 411–13 Recognition: foreign amnesty laws, 576–7 foreign laws, 346–50, 352 torture and Israeli law, 632–58 Remedies: duty to provide, 511–12 human rights injunctions, 601–29 justice of amnesties, 587–8 limits in United Nations law, 21 United States cases, 106–7 See also under Compensation and Injunctions and Reparation Reparation, evolving right to, 41–3, 418–9, 429–31 Repugnancy, see Public Policy Respondeat superior, doctrine of, 19, 35, 441–2, 450–7, 461–3 Responsibility versus interest, in forum non conveniens, 207 Restatement of Conflict of Laws, 103 Restatement of Foreign Relations Law, 93, 95, 97, 99, 100, 129 Restorative justice, as standard for evaluating amnesties, 26, 579–86

730 Index Right to a court, European Convention on Human Rights (ECHR), see European Convention on Human Rights and Civil liability and redress Right to compensation, under the Convention Against Torture (CAT), see Convention Against Torture, article 14 Right to reparation, in evolving UN law, 41–3, 418–19, 429–31 Risk allocation, global, 10, 19, 201, 203 Saudi Arabia, and death penalty, 27, 221, 601–29 Self-amnesties, 13 Self-determination: as relevant value in immunity doctrine, 225–6, 245 legitimacy, 12, 553–66 Senegal, action against Chad leader in, 114 Separation of powers, judges and legislatures, see Justiciability Service ex juris, 9, 77–8, 120–1 tort action in Ontario, 170, 183–9 Service, 157, 181–8 Sex tourism, 55–6 Shari’a law, 27, 601–30 Shin Bet, 28 Slavery, 102, 158–9, 171, 173 Somalia, torture by Canada in, 6, 8, 24, 33–44, 130–4 Sources of law, multiplicity of, 15, 30, 41, 85, 661–85 South Africa: amnesty in, 573–5 horizontal effect in, 48 Sovereignty, see Legitimacy and Neoimperialism and Intervention Standing to sue, 33, 34, 36, 37, 103, 130 State action requirement, and human rights actions in US, 92–102 State immunity: act of state, 16 Canada, 125–9 commercial activity exception, 213–45 Convention Against Torture (CAT), 251, 275, 279–80, 284, 286 in personae, 343–4, 368–71, 650–2 interpreting in light of international law, 129, 247–74, 723 jus cogens, 723 United States, 70–2 State liability, for acts abroad, 130–3 State responsibility: and complicity with conduct of nationals, 21 for corporate conduct, 21, 47–8, 492–512 See also Home state responsibility Statutory interpretation, see Interpretation, of domestic law in light of international law

Statutory right of action for torture, in Yukon, 417, 431 Statutory tort, in UK, 336–9 Strict liability, 19, 21 See also under Risk, global allocation of Substantive invalidity, and act of state, 346–50, 354–6 Sudan, 3, 20, 158–9, 194, 220, 457–63 Symbolic value, of tort awards, 8, 112 Territoriality: as spatial fixation, 201–3 basis for adjudicative jurisdiction, 9, 160–4, 169–71 choice of law, 301–6 jurisdiction, 53–5 See also under Extraterritoriality and Comity Third country legal action, 8, 109 Third party effect, German horizontal effect doctrine, 687–703 ‘Ticking bomb’ situations, 28, 631–60 Tokyo, see Nuremberg and Tokyo Tort: as disincentive to treaty participation, 150–3 as vehicle for effective protection of treaty law, 394–9 assault and battery, 109–18, 131, 334–6 cause of action in international law, 74–6, 87, 90–1 choice of law, see Choice of law, in tort duty of care, 34–7 intentional torts, 35, 62 recognition of new torts, 411–13 relationship to human rights law in Canada, 390–3, 423–9 relationship to prevention and punishment objectives, 136–53 relationship to public law values, 30–1, 423–9, 687–704 relationship to statutory human rights law, 17 remedial virtues, 111–4 rights of family members, 37 theory and study of transnational torture claims, 661–85 treatment in state immunity doctrine, 223–4 Torture: as nominate tort in Canada, 336–9, 374–438 as tort: notions of translation, 7 overview of issues, 3–32 problems of dichotomies, 7 by Canada in Somalia, see Torture, in Somalia by Canada characterisation as tort, 109–18, 329–39 in interrogation in Israel, 632–58 in Somalia by Canada, 6, 8, 24, 33–44, 130–4

Index 731 in United States law, 97 prevention versus punishment as objectives, 136–53 relationship to death penalty, 614–20 scope of prohibition under international law, 402–7 Torture Victim Protection Act (TVPA), 7, 41, 74, 86, 103, 110, 113 Translation, of torture and tort, 45, 60–3 Transnational common law, notion of, 437–8 Transnational corporations, see Multinational corporations Transnational law, 52–3, 374–5 as regulative ideal, 40 causes of action, 14, 17, 62, 374–400 common law method, 684–5 culture, 601–29 growing use of international law in domestic law, 17, 19 judicial discourse, and shaping of private law, 662–6, 681–3 non-justiciability, 26 relationship to lex loci delicti, 377–82 right of access to court, 529–35 right of access to justice, 23 Transparency, and torture prevention, 148–50 Treaty procedures, as alternative remedies, 175–6 Truth and reconciliation commission processes, 13, 26, 278, 571–5 See also Amnesties and restorative justice Turkey, 144–5, 149 TVPA, see Torture Victim Protection Act

UN, see United Nations United Nations: civil claims procedures, 20 civil liability of, 475–89 Committee against Torture, see Convention Against Torture (CAT) evolving right to reparation in, 41–3, 418–19, 429–31 field operations of, 465–89 immunities of, 20, 475–85 international responsibility of, 471–8 state liability for acts of, 20, 33, 484–7 United Kingdom, and horizontal effect, 339–41 United States, human rights litigation experience in, 4, 7, 65–107, 165 Universal jurisdiction: Alien Tort Claims Act (ATCA) as basis, 74–6 civil, 13, 14, 24, 260–4, 723–4 criminal, see Criminal law, international, and Pinochet case Validity, of foreign laws, 16, 346–50, 352–6 Waiver: and state immunity, 71–2, 265–8 commercial activity exception, 228–32 War crimes, and immunities, 279–80 Westminster systems, 4, 17, 49–50, 248–60, 271, 417–29 World Trade Organisation, 59, 496 Yukon, statutory right of action for torture, 417, 431