Tort Law: Challenging Orthodoxy 9781472561527, 9781849464710

In this book leading scholars from the United Kingdom, the United States and Australia challenge established common law

227 21 2MB

English Pages [509] Year 2013

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Tort Law: Challenging Orthodoxy
 9781472561527, 9781849464710

Citation preview

Foreword THE HONOURABLE JUSTICE THOMAS A CROMWELL SUPREME COURT OF CANADA

Judges, it has been said, are just law students who mark their own examinations.1 While some trial judges would perhaps prefer that they, rather than appellate courts such as the Supreme Court of Canada, had a larger role in ‘marking’ their ‘examinations’, the basic point of this adage is right. Judges remain law students throughout their careers as they try to do justice according to law in the seemingly limitless variety of disputes that find their way to the courts. Law students as we are, we need the help of the academy no less—and perhaps more—than we did when pursuing our law degrees. We need to have the benefit of the broad view and deep understanding of the principles and values in play that the best scholars bring to their work. There is a risk that this broader and deeper view may be lost amidst the docketdriven, daily pressures of judging. However, scholarship of this nature, if judged by the quality of the contributions to the Sixth Biennial Conference on the Law of Obligations, is flourishing throughout the common law world. This conference, hosted in this country for the first time courtesy of the Faculty of Law at Western University, was an important one for Canada. Ours is a vast, sparsely populated country with a small, widely dispersed and increasingly specialised legal community. These features create the risk of professional isolation, from each other within Canada, from colleagues abroad and even from colleagues working in different areas of the same field. A key accomplishment of Obligations VI was its contribution to keeping this sort of isolation at bay. Leading scholars, practitioners and judges from around the common law world tackled diverse subjects and in doing so challenged us to look beyond a horizon too often limited by jurisdiction, vocation and specialisation. This important international meeting brought leading Canadian scholars together with their international peers so that, at least for three days, the common law world appeared to us to be not only navigable but also intimate. I hope as well that it provided the international delegates with a more complete view of the depth and breadth of Canadian private law scholarship.

1 Henry Louis Mencken: see A Cooke (ed), The Vintage Mencken (New York, Vintage Books, 1955) at 233.

vi

Foreword

Professors Pitel, Neyers and Chamberlain have in their Introduction provided an excellent overview of the contributions to this volume of tort law scholarship, which, as they point out, represent only a sample of the riches offered to those of us fortunate enough to attend Obligations VI. Allow me to single out one broad area of special personal interest, which I hope will continue to be the subject of writing and discussion at future conferences. The relationship between common law and statutory obligations and remedies seems to me to be an increasingly important yet relatively unexplored field. To be sure, there were thoughtful and important contributions in this area as noted in the Introduction. However, there is room for a great deal more thinking and writing about this broad subject. Think of situations in which a claimant seeks to attach a private law duty to the exercise of a statutory power. The familiar question of when a regulator may also have a private law duty of care is of course one obvious example.2 But the field is much broader than that. Consider the claim that a regulator should be subject to fiduciary obligations in exercising its powers, or that judicial review of administrative action is a precondition to asserting a claim in negligence or abuse of statutory authority.3 These sorts of cases at the borderline—or perhaps, more precisely, in the areas of overlap—of private and public law throw up a host of substantive, remedial and even jurisdictional questions. One may ask whether the framework that we presently have, relying on notions such as collateral attack and the defence of statutory authority, is adequate to the task of responding to these questions. There is the other side of the coin, which is concerned with the impact of legislation and human rights law on the evolution of the common law. Does the common law evolve differently in countries with entrenched, judicially enforced bills of rights than in countries without such documents? Is it the case that the evolution of the common law may only be understood in the context of the evolution of the regulatory state?4 One thing is certain: there is considerable scope for discussion and debate at Obligations VII, the theme of which is divergence and convergence in the law of obligations in various common law jurisdictions. I hope that the interplay of private and public law will continue to be on the agenda. A final word: congratulations to the organisers at Western University and to all of the contributors. May private law scholarship continue to flourish to the benefit of all whose lives the common law so significantly affects. Ottawa, Canada 19 May 2013 2

See, eg, Fullowka v Pinkerton’s of Canada Ltd, 2010 SCC 5, [2010] 1 SCR 132. See, eg, Alberta v Elder Advocates of Alberta Society, 2011 SCC 24, [2011] 2 SCR 261; Canada (Attorney General) v TeleZone Inc, 2010 SCC 62, [2010] 3 SCR 585. 4 As suggested by Robert L Rabin, ‘The John G Fleming Lecture: A Brief History of Accident Law—Tort and the Administrative State’ (2012) 20 Tort Law Review 11. 3

Contributors Kit Barker, MA, Professor, TC Beirne School of Law, The University of Queensland Christine Beuermann, BCom, LLB, BCL, PhD, Lecturer, University of Tasmania Hazel Carty, MA, Reader in Law, University of Manchester Erika Chamberlain, LLB, PhD, Associate Dean (Academic) and Associate Professor of Law, Western University Paul S Davies, MA, Fellow and Tutor, St Catherine’s College, Oxford and University Lecturer, University of Oxford Matthew Dyson, MA, PhD, Fellow of Trinity College, Cambridge Neil Foster, BA, LLB, BTh, DipATh, LLM, Associate Professor, University of Newcastle (NSW) James Goudkamp, BSc, LLB, MA, BCL, MPhil, DPhil, Fellow and Tutor, Balliol College, Oxford and University Lecturer, University of Oxford Gregory C Keating, JD, PhD, William T Dalessi Professor of Law and Philosophy, University of Southern California Gould School of Law James Lee, BA, BCL, Senior Lecturer in Law, University of Birmingham Nicholas J McBride, BA, BCL, Fellow, Pembroke College, Cambridge Claire McIvor, LLB, MJur, PhD, Senior Lecturer in Law, University of Birmingham Jason W Neyers, BA, LLB, MST, Professor of Law, Western University Ken Oliphant, BA, BCL, Director of the Institute for European Tort Law, Vienna and Professor of Tort Law, University of Bristol Stephen GA Pitel, BA, LLB, LLM, PhD, Professor of Law and Goodmans LLP Faculty Fellow in Legal Ethics, Western University Elspeth Reid, MA, LLB, DipLP, FRSE, Professor of Scottish Private Law, University of Edinburgh Sandy Steel, BA, PhD, Lecturer in Law, King’s College London

x

Contributors

Graham Virgo, MA, BCL, Fellow, Downing College, Cambridge and Professor of English Private Law, University of Cambridge Mark L Wilde, LLB, LLM, PhD, Senior Lecturer in Law, University of Reading

Introduction STEPHEN GA PITEL, JASON W NEYERS AND ERIKA CHAMBERLAIN

T

HE CHAPTERS IN this book were initially papers presented at the Sixth Biennial Conference on the Law of Obligations, hosted by the Western University Faculty of Law from 17–20 July 2012. With the theme of ‘Challenging Orthodoxy’, the conference brought together academics, practitioners and judges from across the common law world. Presenters were encouraged to suggest new and innovative approaches to existing and emerging legal problems and to provide new interpretations and applications of traditional common law doctrines. This admittedly broad theme was chosen in part for its inclusiveness, aiming to expand rather than narrow the range of possible topics for analysis and discussion. For three days, attendees debated the continuing relevance of the common law and its ability to address the problems of the twenty-first century. The conference examined orthodoxies across the spectrum of private law, featuring some 50 papers in total. Limitations of space and time meant that we could not publish even a bare majority of these papers, and as a result this book includes only papers on the topic of tort law. Many of the other papers have been or will be published elsewhere.1 The chapters have been grouped together in accordance with three main themes.2 The initial chapters deal with larger, more encompassing questions including the purposes of tort law, the relationship of tort law to criminal law and the interaction of tort law with legislation. The second group of chapters challenges orthodoxies in the law of negligence, looking at causation, loss of chance, emotional distress and contributory negligence. The remaining chapters examine topics in the economic torts, nuisance and strict liability, including remedies for nuisance, the utility of the tort of conspiracy and ancillary doctrines and concepts such as accessory liability, malice and vicarious liability. In the opening chapter, ‘Against Certainty in Tort Law’, Ken Oliphant challenges the pursuit of certainty in tort law through the use of ‘bright line’ rules. He argues that such rules invariably create uncertainty, as judges

1 See, eg, (2013) 26 Canadian Journal of Law and Jurisprudence, which contains three of the papers. 2 We considered a random order to be too irreverent a challenge to orthodoxy.

xii

Stephen GA Pitel, Jason W Neyers and Erika Chamberlain

attempt to mitigate their seemingly arbitrary or unjust effects by recognising exceptions or qualifications. Further, since common law rules cannot be mechanically applied, the outcomes of cases can at times contradict the very principles that the rules are meant to espouse. Oliphant focuses his discussion on the duty of care in negligence, where the English courts have posited certainty as a rationalisation for the categorisation of cases and their incremental expansion.3 He is also critical of courts that have resigned themselves to admittedly unsatisfactory outcomes4 or have delegated the repair of troublesome issues, like recovery for psychiatric harm, to Parliament.5 Oliphant instead recommends an approach that embraces the exercise of judicial judgment: the relevant factors should be identified and assessed flexibly on the facts of individual cases. The role of appellate courts, he argues, is to define the parameters in which these factors should be considered. Although the outcomes of cases might then be less certain, they would be more closely aligned with justice. The next two chapters evaluate the purposes of tort law and the interests that it protects. In ‘Tort Law and Human Flourishing’, Nicholas McBride makes three central arguments. The first is that the ‘balanced view’ of how tort law determines what rights claimants have is a more plausible interpretative theory of English tort law than that offered by those writing in the Kantian tradition.6 The balanced view, which he has developed in detail elsewhere,7 is that tort law will impose a duty on defendants where a claimant’s interests in compelling defendants to do something outweigh, or are more important than, defendants’ interests in being left free to pursue their purposes. As examples of the balanced view, McBride examines cases relating to private nuisance, the frustration of education and business opportunities, pure economic loss, suicide, the detention of people and chattels, employment and inducing breaches of contract. His second argument is that the vision of human flourishing that currently underlies tort law closely matches the vision of human flourishing elaborated by Finnis and currently dominant in modern Western societies.8 The third argument is that this modern understanding of human flourishing is deficient in many respects. In its place, McBride sets out an alternative understanding of human flourishing and discusses what tort law would look like if it were 3 See, eg, Caparo Industries plc v Dickman [1990] UKHL 2, [1990] 2 AC 605, 628 (Lord Roskill). 4 Jain v Trent Strategic Health Authority [2009] UKHL 4, [2009] 1 AC 853. 5 White v Chief Constable of South Yorkshire Police [1998] UKHL 45, [1999] 2 AC 455. 6 See, eg, E Weinrib, The Idea of Private Law (Cambridge, Mass., Harvard University Press, 1995); E Weinrib, Corrective Justice (Oxford, Oxford University Press, 2012); A Ripstein, Force and Freedom (Cambridge, Mass., Harvard University Press, 2009). 7 NJ McBride, ‘Rights and the Basis of Tort Law’ in D Nolan and A Robertson (eds), Rights and Private Law (Oxford, Hart Publishing, 2011). 8 See, eg, JM Finnis, Natural Law and Natural Rights, 2nd edn (Oxford, Oxford University Press, 2011).

Introduction

xiii

based on that alternative understanding—suggesting, for example, that this new tort law would be less protective of pure economic loss and sharp practice (by business or government) but more protective of a claimant’s mental state. In ‘Private and Public: The Mixed Concept of Vindication in Torts and Private Law’, Kit Barker addresses both the orthodoxy that tort law serves to vindicate the plaintiff’s rights and the heterodoxy that there should be a distinctive head of ‘vindicatory damages’ to achieve this purpose. Taking the recent United Kingdom Supreme Court case of Walumba Lumba v Secretary of State for the Home Department9 as an example, Barker examines the case for vindicatory damages in situations where the plaintiff has suffered no objective loss, the defendant has suffered no gain, and the conduct is not so outrageous as to justify an award of exemplary damages. Proponents of vindicatory damages argue that they are warranted in such cases in order to ‘make good’ the plaintiff’s right, ‘mark’ its infringement, and fulfil the maxim ubi ius, ibi remedium.10 Barker argues, however, that a separate category of vindicatory damages is ‘functionally otiose’. He explores the plurality of public and private aims that are encompassed in the notion of vindication: preventing the infringement of rights, specifically enforcing rights, reversing the effects of infringement, declaring rights, and denouncing outrageous infringement of rights. Barker explains that so-called vindicatory damages are unnecessary, since they will never be able to prevent the infringement or specifically enforce a right, and since the remaining functions can be fulfilled through compensatory, restitutionary and exemplary damages awards. He concludes with a caution about recognising ‘public symbolism’ as a distinct purpose of tort law, a purpose that seems to undermine recent civil justice reforms and the incentives for out-of-court resolution of disputes. This analysis of vindication leads well into two chapters that examine the boundary between tort law and criminal law. First, in ‘“We Do This in the Criminal Law and That in the Law of Tort”: A New Fusion Debate’, Graham Virgo challenges the orthodoxy that the different functions of these two areas of law mean that consistency of interpretation and approach is not an appropriate objective. The orthodoxy is seen in Lord Scott’s statement in Ashley v Chief Constable of Sussex Police that ‘this plea for consistency between the criminal law and the civil law lacks cogency for the ends to be served by the two systems are very different’.11 Virgo considers the extent to which the interpretation of concepts in the law of tort should be

9

[2011] UKSC 12, [2012] 1 AC 245. See, eg, N Witzleb and R Carroll, ‘The Role of Vindication in Tort Damages’ (2009) 17 Tort Law Review 16; D Pearce and R Halson, ‘Damages for Breach of Contract: Compensation, Restitution and Vindication’ (2008) 28 OJLS 73. 11 Ashley v Chief Constable of Sussex Police [2008] UKHL 25, [2008] AC 962, [17]. 10

xiv

Stephen GA Pitel, Jason W Neyers and Erika Chamberlain

applied in the criminal law and vice versa. He also explores whether structural and theoretical developments in one system should legitimately be considered in the other to assist in the development and rationalisation of the law. Should it matter, for example, that a defendant might be prosecuted for doing something which he or she has a civil law right or even a duty to do? Is the orthodoxy a defensible answer to such a situation? Drawing on earlier imagery relating to common law and equity, Virgo looks to spark a new ‘fusion’ debate concerning boundary concepts relevant to both tort law and criminal law. He highlights the explicit recognition of the validity of the principle of consistency by McLachlin J in Hall v Hebert: ‘the law must aspire to be a unified institution, the parts of which—contract, tort, the criminal law—must be in essential harmony’.12 He illustrates his analysis with examples drawn from several areas including illegality, proprietary rights, conspiracy and self-help remedies. Similarly, in ‘Challenging the Orthodoxy of Crime’s Precedence over Tort: Suspending a Tort Claim where a Crime May Exist’, Matthew Dyson questions three orthodoxies governing the relationships of criminal law and tort law: the orthodoxy of partition (that the two areas of law are conceptually separate),13 the orthodoxy of priority (that tort law is subservient to criminal law in situations where the two conflict) and the orthodoxy of permanence (that the current operation of criminal law and tort law has always existed). He first provides a history of the ‘timing rules’, which mandate that a civil claim is prohibited or stayed when a criminal prosecution arises out of the same incident. Over time, the law has become more flexible and has allowed the courts greater discretion about whether to suspend a parallel civil claim.14 This has implications for the three orthodoxies outlined above. First, there are procedural and structural overlaps between criminal and civil proceedings, such as the availability of compensation in criminal courts, the admissibility of convictions as evidence in civil cases, and the consideration of criminal penalties in the assessment of punitive damages. Second, defendants now have the burden of showing that a parallel civil action will be detrimental to the defence of their criminal cases, which means that criminal law no longer has automatic precedence over tort.15 Finally, the evolution of the timing rules shows that their application has been anything but permanent. The next two chapters explore the interaction of tort law and legislation more generally. As James Lee observes in ‘Legislative Challenges to Orthodoxy’, whether seeking to reverse an individual decision, reaffirm an existing tenet, introduce limitation statutes or enact wholesale reform, 12 13 14 15

Hall v Hebert [1993] 2 SCR 159, [17]. See, eg, R v Hinks [2000] UKHL 53, [2001] 2 AC 241, 252–53 (Lord Steyn). Criminal Justice Act 2003, s 329(2). See generally JR Spencer, ‘Legislate in Haste, Repent at Leisure’ [2010] CLJ 19.

Introduction

xv

legislative interventions can have a significant impact on fundamental principles. He draws on statutes from the United Kingdom and Australia to illustrate the ways in which legislation can confirm, tolerate or challenge the orthodox law.16 Many of the examples he considers illustrate the ‘law of unintended consequences’: the courts interpret the statutes in ways the legislators did not consider or intend. His chapter also questions the manner in which Parliament does intervene, exploring the constitutional jurisprudence on legislative interference with existing causes of action. Unwilling to accept that the law ‘is a chaotic jumble of (surviving) common law rules and piecemeal, inconsistent and unsystematic statutory incursions’,17 he urges deeper understanding of ‘the continuing inter-relationship between the common law and statute and the policy choices that courts and legislatures … must make in their respective, but not mutually exclusive, roles of fixing and distributing the burden of liability’.18 Lee’s thesis is that even where the legislature seeks to challenge orthodoxy, its legislation should nevertheless be implemented with an understanding of the common law, in part because respecting coherence and understanding orthodoxy can lead to more effective legislation. Neil Foster explores a different relationship between private law and the legislature in his chapter, ‘Statutes and Civil Liability in the Commonwealth and the United States: A Comparative Critique’. Foster reviews the history of the use of statutes in those jurisdictions, noting that, in its formative years, the American law borrowed very heavily from the Commonwealth tort of breach of statutory duty. Despite this common inheritance, however, modern American jurisprudence tends to use breach of statute as merely evidence of ‘negligence per se’, thus establishing that a defendant has breached the standard of care in a negligence claim.19 Foster explores some of the pitfalls of this approach, including the tendency to conflate the issues of the duty and standard of care,20 the marginalisation of legislative intention in favour of judicial discretion,21 and the creation of a multitude of factually driven ‘excuses’ that tend to temper the otherwise expansive tendencies of the American law.22 Foster concludes that these weaknesses could be resolved by a return to the principles of the Commonwealth tort of breach of statutory duty.

16 Examples include the Compensation Act 2006, the Limitation Act 1980, the Damages (Asbestos-Related Conditions) (Scotland) Act 2009 and various Australian Defamation Acts. 17 J Dietrich, ‘What is “Lawyering”? The Challenge of Taxonomy’ [2006] CLJ 549, 573. 18 B McDonald, ‘Teaching Torts: Where to Start in an Age of Statutes?’ (2010) 18 Torts Law Journal 173, 186. 19 See DB Dobbs, The Law of Torts (St Paul, MN: West Publishing, 2000) § 133. 20 AM Linden, ‘Tort Liability for Criminal Nonfeasance’ (1966) 44 Canadian Bar Review 25. 21 Martin v Herzog 126 NE 814 (NY 1920). 22 DP Leonard, ‘The Application of Criminal Legislation to Negligence Cases: A Reexamination’ (1983) 23 Santa Clara Law Review 427.

xvi

Stephen GA Pitel, Jason W Neyers and Erika Chamberlain

Foster’s chapter concludes the part of the book dealing with broad thematic issues, and the next group of chapters focuses more particularly on orthodoxies in the law of negligence. In ‘The “Doubles the Risk” Test for Causation and Other Related Judicial Misconceptions about Epidemiology’, Claire McIvor examines the ‘doubles the risk’ test as it has been used in cases of probabilistic causation. In Sienkiewicz v Greif (UK) Ltd23 Smith LJ suggested in the Court of Appeal that this test should operate as the default test for causation in cases in which it is not scientifically possible to determine exactly how a particular harm has been brought about.24 On appeal, however, the Supreme Court rejected this approach. As McIvor demonstrates, in casting doubt on the general applicability of the doubling-of-the-risk approach the Supreme Court unfortunately thought that it was necessary to take issue with epidemiology.25 This is due to a judicial misconception that the ‘doubles the risk’ test is a specialist epidemiological principle. McIvor explains how the cases illustrate (i) the mistaken connection made by the courts between the doubling of the risk and the epidemiological concept of relative risk and (ii) the common misconception that epidemiologists deal only in naked statistics. She also demonstrates that there are serious legal problems underlying the orthodox judicial approach to assessing issues of probabilistic causation in tort. These relate primarily to a misapplication of the balance of probabilities standard of proof and stem from a judicial tendency, when faced with statistical evidence, to lose sight of the nature and purpose of the factual causation stage of the negligence enquiry. McIvor’s conclusion is that while the ‘doubles the risk’ test should be abolished, greater judicial engagement with the science of epidemiology would be highly beneficial. Continuing the theme of causation, Sandy Steel’s chapter, ‘Rationalising Loss of a Chance in Tort’, explores the challenging question of recovery for the loss of a chance. His aim is to provide a clear account of the existing rules and to then assess whether the current state of the law follows from tenable principles. In his view, much of the present law is open to question on grounds of inconsistency. Lost chances are recoverable as actionable loss where it is necessary to determine how a third party would have behaved, but not where it is necessary to determine how the claimant would have behaved.26 Recovery is allowed for lost chances of avoiding a financial loss but not lost chances of avoiding a physical loss.27 Increased risks of physical loss can

23 Sienkiewicz v Greif (UK) Ltd [2011] UKSC 10, [2011] 2 AC 229, aff’g [2009] EWCA Civ 1159, [2010] 1 QB 370. 24 Ibid, at [23] (CA). 25 Ibid, at [80]–[83] (SC). 26 See Allied Maples v Simmons & Simmons [1995] EWCA Civ 17, [1995] 1 WLR 1602. 27 See Gregg v Scott [2005] UKHL 2, [2005] 2 AC 176; Rothwell v Chemical & Insulating Co Ltd [2007] UKHL 39, [2008] 1 AC 281.

Introduction

xvii

be recovered when it comes to determining the extent of the defendant’s liability in damages but not in establishing the defendant’s liability. Having explained the state of the law, Steel offers two kinds of principle that might explain the law on recovery for losses of chances: the ‘inherent uncertainty principle’ and the ‘loss principle’. The former states that the court should apply a chance-based approach to recovery wherever the question of whether the defendant’s wrongful conduct caused the claimant’s traditional loss is insusceptible of proof. The latter is based on the notion that losing a chance of avoiding a traditional loss or of obtaining a traditional benefit is itself to suffer a loss. Steel argues the latter principle is normatively more appealing than the former and that losing a chance can indeed be to suffer a loss. The loss principle suggests, however, that the situations in which a person should be held liable in tort law simply for the destruction of a chance are limited and that even in such cases the current method of quantifying damages for such losses may be too generous. Moving away from causation, Gregory Keating next addresses the concept of ‘harm’ as it applies in the American law of negligent infliction of emotional distress (NIED). In ‘When is Emotional Distress Harm?’ he examines the problem of classifying such liability: is it a question of duty or proximate cause or a combination of the two? Contrary to the orthodox belief, Keating suggests that American tort law does not treat NIED as a matter of duty (in that it does not ask whether the plaintiff’s interest gives rise to an obligation on others to avoid carelessly jeopardising that interest).28 As a result, the law does not provide a sufficient account of the harm that the tort seeks to redress, nor does it provide a means of distinguishing those psychological harms that are recoverable in negligence from those that are not. Keating suggests that the relevant harm should be defined in terms of impairment.29 Just as physical harms impair the normal use of the plaintiff’s body, psychological harms impair the ability to function normally and damage the plaintiff’s agency. For example, childhood sexual abuse often causes lasting psychological damage by impairing the capacity to trust. Emotional distress is a legally recognised ‘harm’ when it is acute enough to impair the will. Thus, Keating criticises the common tendency to insist that recovery for emotional distress is inevitably arbitrary.30 In the last chapter in this group, ‘Rethinking Contributory Negligence’, James Goudkamp aims to change much of the orthodox thinking about the doctrine of contributory negligence. His core claim is that judges and scholars are mistaken in thinking that the doctrine is controlled by a two-stage

28

Boyles v Kerr 855 SW 2d 593, 597 (1992). See also S Shiffrin, ‘Wrongful Life, Procreative Responsibility, and the Significance of Harm’ (1995) 5 Legal Theory 117; JJ Thomson, The Realm of Rights (Cambridge MA, Harvard University Press, 1990) 227–48. 30 Thing v La Chusa 48 Cal 3d 644 (1989). 29

xviii

Stephen GA Pitel, Jason W Neyers and Erika Chamberlain

test that asks whether the claimant was careless with respect to his or her own safety and whether that carelessness was a cause of his or her injuries.31 He argues that the test really has three stages, with the third stage being an inquiry into whether any considerations of public policy exclude the doctrine.32 He also expresses concern about rules that determine the standard of care that defendants have to achieve for the purposes of the tort of negligence being automatically transferred into the contributory negligence setting. Because there are significant differences between the concepts of negligence and contributory negligence, there is no reason to think that rules that may be suitable in the domain of negligence are necessarily sound rules for the purposes of contributory negligence. Goudkamp also challenges one of the most basic orthodoxies about contributory negligence, namely that it is a doctrine that goes to the question of liability and so should be classified as a defence. In his view, the introduction by statute of the power of apportionment moved the doctrine out of the law of liability and into the law of remedies. Accordingly, it is properly discussed in chapters concerned with the assessment of damages rather than alongside rules such as the defences of voluntary assumption of risk and illegality. He also considers whether it is ever appropriate to apportion, for contributory negligence, 100%, 0% or a percentage closely approaching those two extremes (such as 98% or 2%). His arguments have the potential to change several basic elements of how we understand contributory negligence. Goudkamp’s reconceptualisation of contributory negligence as a form of remedy transitions into the final set of chapters, which deal with other nominate torts and their ancillary doctrines, beginning with Mark Wilde’s examination of remedies in nuisance. In ‘Nuisance Law and Damages in Lieu of an Injunction: Challenging the Orthodoxy of the Shelfer Criteria’, Wilde argues that public utility should be a factor taken into account by courts when deciding whether to exercise their discretion to award damages in lieu of an injunction.33 In order to justify this claim, Wilde examines the state of the common law immediately prior to the adoption of Lord Cairns’ Act,34 the reasoning in Shelfer v City of London Electric Lighting Company35 (which is taken to decisively outline the appropriate factors), and the subsequent judicial treatment of Shelfer in the English courts. His conclusion is twofold. First, the court in Shelfer never intended its outline 31 This two-stage test is endorsed in, eg, Lewis v Denye [1939] 1 KB 540 (CA) 554; Nance v British Columbia Electric Railway Co Ltd [1951] AC 601 (PC) 611. 32 This has parallels to the American Law Institute, Restatement (Third) of Torts: Liability for Physical and Emotional Harm (2010) § 7(a) (cmt h). See also E Bublick, ‘Comparative Fault to the Limits’ (2003) 56 Vanderbilt Law Review 977. 33 Enshrined in s 50 of the Senior Courts Act 1981. 34 Chancery Amendment Act 1858. 35 Shelfer v City of London Electric Lighting Company [1895] 1 Ch 287 (CA) [Shelfer].

Introduction

xix

of the appropriate factors to operate in a narrow manner and therefore that ‘there is no historic or doctrinal reason for a bright-line rule which excludes public utility considerations from any analysis of whether the court should exercise its discretion to award damages in lieu of an injunction’. Second, it is dangerous to arbitrarily exclude examinations of public utility at the remedy stage of the analysis since courts will be tempted to utilise the defence of statutory authority or the concept of the character of the neighbourhood36 to deny claimants’ nuisance claims altogether—thus ensuring that claimants ‘get nothing but the best by getting nothing at all’.37 Next, Hazel Carty tackles recent developments in the tort of conspiracy. In ‘The Tort of Conspiracy as a Can of Worms’, she argues that Commonwealth courts, through cases such as Total Network SL v Revenue and Customs Commissioners,38 Fatimi Pty Ltd v Bryant39 and Canada Cement LaFarge Ltd v British Columbia Lightweight Aggregate Ltd,40 have coalesced around a modern view of conspiracy. This view is that conspiracy: (i) is a vital tort which is available to remedy both direct and indirect harms which (ii) can operate outside of the economic torts’ traditional role of regulating trade competition disputes. Carty argues that this modern view has the potential to: (a) undo the rationalising of the economic torts undertaken by the House of Lords in OBG Ltd v Allan;41 (b) circumvent well-established rules relating to joint tortfeasance and the actionability of statutory breaches; and (c) lead to liability for merely unacceptable, rather than illegal, behaviour. She argues that, most disturbingly, these same Commonwealth courts have revitalised the tort of conspiracy and expanded the ambit of tortious liability without offering any compelling justification (in either principle or policy) for such a move. Moving from a specific instance of joint liability to a more general conceptualisation, Paul Davies examines accessory liability in private law in ‘Aid, Abet, Counsel or Procure?’. After analysing the various rationales for the doctrine (such as responsibility, culpability, rights-protection, property, deterrence and loss shifting) he concludes that the best understanding of accessory liability is that it is an amalgam of these various rationales. In essence, accessory liability ensures that the rights of innocent victims are protected against those who, at least partially, caused the rights violations and who are also blameworthy (and therefore culpable) for the state of affairs thus caused. After justifying the doctrine and examining the current 36

See, eg, Miller v Jackson [1977] QB 966 (CA) 980–82 (Lord Denning). S Tromans, ‘Nuisance—Prevention or Payment?’ (1982) 41 CLJ 87, 108. 38 Total Network SL v Revenue and Customs Commissioners [2008] UKHL 19, [2008] 1 AC 1174. 39 Fatimi Pty Ltd v Bryant [2004] NSWCA 140, 59 NSWLR 678. 40 Canada Cement LaFarge Ltd v British Columbia Lightweight Aggregate Ltd [1983] 1 SCR 452. 41 OBG Ltd v Allan [2007] UKHL 21, [2008] 1 AC 1. 37

xx

Stephen GA Pitel, Jason W Neyers and Erika Chamberlain

state of the law in equity,42 contract43 and tort,44 Davies then examines what a unified and coherent law of accessory liability across the private law should look like. His ultimate conclusion is that accessory liability should require a claimant to prove a conduct element (such as inducement or assistance) and a mental element (such as knowledge of the wrong) in order to create a prima facie culpability in the defendant. This culpability could then be displaced if the defendant could prove the defence of justification (modelled on that which exists for the tort of inducing breach of contract45)—namely that it was fair, just and reasonable for the defendant to act as he or she did. Elspeth Reid picks up on the theme of mental elements in her chapter, ‘“That Unhappy Expression”: Malice at the Margins’, where she examines the role that malice plays in fashioning the scope of tortious liability in both England and Scotland. After examining the various definitions of malice used by the courts and distinguishing between malice in law and malice in fact, Reid examines how malice figures into areas as distinct as defamation, private nuisance, the economic torts and misfeasance in public office.46 Her conclusion is that in the modern law, malice is found ‘only at the margins of established doctrines, or at the centre of marginal doctrines’. Thus she argues that malice as a concept might profitably be discarded in favour of a more focused examination of (1) the underlying legal and policy reasons why liability is imposed in each tort and (2) the modern concepts appropriate to matching the scope of liability for each tort with these underlying principles and policies. In the final chapter, ‘Dissociating the Two Forms of So-Called “Vicarious Liability”’, Christine Beuermann challenges the orthodox view that the vicarious liability imposed on employers for the wrongs of their employees is theoretically identical to the vicarious liability imposed on principals for the wrongs of their agents.47 Her argument is that each of these forms of strict liability is distinct because they respond to different potential risks that the relationships impose on third parties/strangers. Employer strict 42 See, eg, Royal Brunei Airlines Sdn Bhd v Tan Kok Ming [1995] 2 AC 378 (PC); Twinsectra Ltd v Yardley [2002] UKHL 12, [2002] 2 AC 164. 43 See, eg, Lumley v Gye (1853) 2 E & B 216, 118 ER 749 (QB); OBG Ltd v Allan, above n 41. 44 See, eg, CBS Songs Ltd v Amstrad Consumer Electronics plc [1988] AC 1013 (HL); Credit Lyonnais Bank Nederland NV (now Generale Bank Nederland NV) v Export Credit Guarantee Department [2000] 1 AC 486 (HL). 45 See, eg, Brimelow v Casson [1924] 1 Ch 302. 46 See, eg, Horrocks v Lowe [1975] AC 135 (HL) 149 (qualified privilege); Christie v Davey [1893] 1 Ch 316; Bradford Corporation v Pickles [1895] AC 587 (HL) (nuisance); Allen v Flood [1898] AC 1 (HL); Crofter Hand Woven Harris Tweed Co v Veitch [1942] 1 AC 435 (HL) (economic torts); and Three Rivers District Council v Governor and Company of the Bank of England [2001] UKHL 16, [2003] 2 AC 1 (misfeasance in public office). 47 See, eg, Bazley v Curry [1999] 2 SCR 534; Lister v Hesley Hall Ltd [2001] UKHL 22, [2002] 1 AC 215.

Introduction

xxi

liability responds to the risk of harm created by the nature of employment itself, which puts employees in a situation where there is a potential conflict between their duties in contract and the duties owed to strangers under the general law. This creates the risk that employees will violate the rights of strangers in order to complete employers’ tasks in a timely fashion. In contrast, principal strict liability responds to the risk of harm to strangers wishing to enter into contracts with principals. This risk of harm exists because strangers are not expected to confirm with principals the details of the contracts negotiated with agents and hence are vulnerable to being deceived by these agents. Beuermann’s ultimate argument is that this dual view of strict liability explains the contours of the doctrine and the results of the cases better than any traditional theory, and will therefore provide judges with greater guidance when dealing with these issues in the future. Taken together, the chapters in this book challenge a wide range of orthodoxies in tort law, from its overarching goals and its relationship to other areas of the law to specific principles in negligence, nuisance, vicarious liability, the economic torts and the law of remedies. They provide new perspectives on long-standing issues that have concerned the bar, bench and academy in the Commonwealth and the United States. By challenging traditional thinking in these areas, the authors hope to spur additional development of the common law of torts and ensure its continued relevance in the twenty-first century. *** Both the Obligations VI Conference and this book benefited from generous support and assistance. We thank the firms who sponsored the conference: Borden Ladner Gervais LLP, McCarthy Tétrault LLP, Hart Publishing, Carswell, Cohen Highley LLP, Lerners LLP and McKenzie Lake LLP. We are also grateful for the financial support of the Social Sciences and Humanities Research Council of Canada which, among other things, allowed us to host several distinguished keynote speakers: Chief Justice Beverley McLachlin and Justice Thomas Cromwell from the Supreme Court of Canada, Melvin Eisenberg (Berkeley), John Goldberg (Harvard), Andrew Robertson (Melbourne), Ernest Weinrib (Toronto), Richard Wright (Chicago-Kent) and Benjamin Zipursky (Fordham). We are also indebted to the faculty and staff at Western Law, and particularly to Dean W Iain Scott, to his predecessor Dean Ian Holloway, to our conference co-ordinator Michelle Bothwell, and to our student assistant Julia Bassett (who was funded by a Borden Ladner Gervais LLP Fellowship). We are grateful to Justice Cromwell for writing the foreword and for the editorial assistance of three talented Western Law students: Jordan Diacur, Adam Jacobs and Jordan McKie. We also thank Andrew Robertson for founding the Obligations series of conferences, which have become premier

xxii Stephen GA Pitel, Jason W Neyers and Erika Chamberlain events in the private law world. We were extremely honoured to be the first North American hosts for this distinguished gathering. Finally, we are grateful to Richard Hart and his excellent team at Hart Publishing. Richard’s support of the Obligations series of conferences has been a significant reason for its success. And what we had found in the past was as true in the present: Hart Publishing is a simply terrific collaborator in the creation of quality private law scholarship.

1 Against Certainty in Tort Law KEN OLIPHANT*

I. INTRODUCTION

F

OCUSING ON ENGLISH law, this chapter argues that that pursuit of ‘bright line’ rules of tortious liability in the interests of certainty is counter-productive and results in incoherence and injustice. It is counter-productive because in practice it invariably creates more uncertainty than existed before, often as to the very rules to be applied. Rules whose consequences are sometimes arbitrary are adopted in the name of certainty, but courts attempt to mitigate their effects by recognising exceptions and qualifications which, because they lack any convincing basis in principle, are themselves of uncertain scope. The pursuit of certainty also results in incoherence because a weighing of the full set of relevant considerations cannot be reduced to the mechanical application of a rule; consequently, outcomes are attained that are at odds with underlying values and fundamental principles. That the approach is productive of injustice is evident from the number of occasions in recent years when British Supreme Court justices have felt the need to express their regret at the outcome of appeals over which they have presided and their sympathy for a claimant denied a remedy in damages. To be preferred is an approach based on the identification of relevant factors and their flexible assessment on the facts of individual cases. The exercise of judgment by the court—and the uncertainty that this necessarily entails—should be recognised as inherent in the judicial role, and as desirable rather than something to be distrusted and constrained.1 The role of the appellate courts should be conceived as setting the parameters within which this balancing exercise is conducted, and establishing the weight that

* Some parts of this paper develop ideas previously sketched in K Oliphant, ‘European Tort Law: A Primer for the Common Lawyer’ (2009) 62 CLP 440. 1 See also A Beever, Rediscovering the Law of Negligence (Oxford, Hart Publishing, 2007) 49–50, emphasising that judgment is required because judicial decision making cannot simply be based on the mechanical application of rules.

2

Ken Oliphant

is to be attached to the factors considered, rather than trying to lay down rules that lower courts are constrained to apply mechanically. Though the focus of this chapter is the English law of tort, it is hoped that the criticisms expressed of the current English approach will be of interest— and perhaps a cautionary tale—for those from other jurisdictions.

II. ON CERTAINTY IN LAW IN GENERAL

Certainty is almost universally seen as a quality to be desired in all law. It is ‘of the utmost importance’,2 and to some even essential: there is a ‘measure of certainty which is necessary to all law’;3 it is ‘demanded’ in the interests of practical convenience.4 Typically, two different aspects are emphasised, one ex ante, the other ex post. First, before any injury has occurred, it is said that certainty in the law allows individuals to plan their affairs with the knowledge of the potential liability costs attaching to the activities in which they engage or decide not to engage, thus contributing to a healthy entrepreneurial economy built on freedom of action.5 It is integral to this aspect of certainty that lawyers are able to give their clients accurate advice about their rights,6 and that insurers can estimate their potential liabilities with reasonable accuracy.7 Second, after an injury has occurred, certainty in the law is considered to contribute to the swift and economical settlement of any claim that results.8 It has long been recognised, however, that certainty may sometimes conflict with the overriding requirements of justice. Exactly 100 years ago, a study on that very topic began with the observation that ‘[t]here is in all modern states to-day a general conflict between certainty in the law and concrete justice in its application to particular cases’.9 The pursuit of legal certainty tends to produce rigidity and resistance to change, and the law

2 Leigh & Sullivan Ltd v Aliakmon Shipping Co Ltd (The Aliakmon) [1985] UKHL 10, [1986] AC 785, 817 (Lord Brandon) [The Aliakmon]. 3 Douglas v Hello! Ltd [2000] EWCA Civ 353, [2001] QB 967, 1002 (Sedley LJ). 4 White v Jones [1995] UKHL 5, [1995] 2 AC 207, 290 (Lord Mustill) (‘the certainty which practical convenience demands’). 5 Cf M Weber, Economy and Society: An Outline of Interpretive Sociology, vol 1 (G Roth and C Wittich (eds), Berkeley, University of California Press, 1978) 883 (‘capitalistic enterprise … cannot do without legal security’). 6 Candlewood Navigation Corporation Ltd v Mitsui OSK Lines Ltd (The Mineral Transporter) [1985] UKPC 21, [1986] AC 1, 25 (PC (Australia)) (Lord Fraser) [Candlewood]. 7 Cf Thing v La Chusa 48 Cal 3d 644, 647 (Eagleson J) (1989) [Thing] (‘uncertainty that has troubled lower courts, litigants, and, of course, insurers’). 8 Wells v Wells [1998] UKHL 27, [1999] 1 AC 345, 397 (Lord Clyde). 9 FR Coudert, Certainty and Justice: Studies of the Conflict Between Precedent and Progress in the Development of the Law (New York, D Appleton & Co, 1913).

Against Certainty in Tort Law 3 consequently loses touch with prevailing social standards.10 At the same time, the purportedly clearer and more predictable rules that are adopted in the name of certainty curtail the scope for the exercise of judgment on the facts of the individual case. It has been plausibly suggested that this reflects a ‘judicial abhorrence of responsibility’: the judge is able to ‘cover up behind a doctrine proclaiming to the world that in fact he has little or no personal discretion, and that he is compelled by ineluctable logic to the conclusions which he reaches’.11 The arbitrariness that results is accepted as ‘unavoidable’12 and the positive law thus comes to drift apart from the fundamental values and principles that are embedded in the legal system. Paradoxically, certainty is itself a rather nebulous concept. What sort of certainty is to be pursued: certainty as to the principles to be applied or certainty as to outcomes? And whose perspective is to be taken: the individual litigant embroiled in a concrete dispute, the forward-planner who wants to know his or her rights and liabilities in advance of engaging in certain conduct, or the insurer who aggregates liabilities incurred by a pool of insureds? Because of the many varied ways in which the questions can be both posed and answered, it seems likely that there are some forms of certainty that are desirable, but others whose pursuit may have unfortunate repercussions. Writing extrajudicially, Benjamin Cardozo astutely noted that ‘there is a certainty that is genuine and a certainty that is illusory’; the former is ‘worth attaining’, the latter ‘should be shunned’.13 As he explained further: judges strive at times after the certainty that is sham instead of the certainty that is genuine. They strive after a certainty that will keep the law consistent within their own parish, their little territorial jurisdictions, instead of the certainty that will keep it consistent with verities and principles as broad as the common law itself, and as deep and fundamental as the postulates of justice. The tendency is insidious and to some extent inevitable. Particular precedents are carried to conclusions which are thought to be their logical development. The end is not foreseen. Every new decision brings the judge a little farther. Before long he finds himself in a dilemma. He does not like the spot where he is placed, yet he is unwilling and perhaps unable to retreat from it. The certainty that is arrived at by adherence to precedent is attained, but there is a sacrifice of another certainty that is larger and more vital. This latter certainty is lost if we view the law in shreds and patches, not steadily and whole with a sweep that reaches the horizon. Often a spurious consistency is preserved by artificial and unreal distinctions. The idol is discredited, but he is honored with lip service, the rubrics of the ancient ritual.

10

Ibid, at 2. HWR Wade, ‘The Concept of Legal Certainty: A Preliminary Skirmish’ (1941) 4 MLR 183, 195. 12 Thing, above n 7, at 666. Cf McLoughlin v O’Brian [1982] UKHL 3, [1983] 1 AC 410, 442 (Lord Bridge) [McLoughlin] (‘a largely arbitrary limit of liability’). 13 BN Cardozo, The Growth of the Law (New Haven, Yale University Press, 1924) 17. 11

4

Ken Oliphant We must have the courage to unmask pretense if we are to reach a peace that will abide beyond the fleeting hour. The law’s uncertainties are to be corrected, but so also are its deformities.14

III. CERTAINTY IN TORT LAW

The focus of this chapter is the English law of tort, more specifically the role played by the duty of care concept in establishing the boundaries of liability for negligence. The argument presented is that the duty of care concept has become the principal mechanism for the pursuit of certainty in the modern law of tort, but that the certainty that is sought is illusory in the sense conceived by Cardozo in the passage quoted above. It goes beyond the desirable certainty found when it is clear what legal principles are to be applied in the individual case, and hence what factors are to be taken into account in the court’s exercise of judgment. Dissatisfied with this modest goal, some judges and commentators seek additional certainty—relating to the outcomes of individual cases. This certainty is spurious because it cannot be attained, and its pursuit is harmful because it places the positive law at odds with underlying values and principles of interpersonal justice. It entails the acceptance of arbitrariness, and the inflexibility of law to adopt to changes in society. We would therefore be wise to recognise the force of Lord Bridge’s cautionary words in McLoughlin v O’Brian: we should resist the temptation to try yet once more to freeze the law in a rigid posture which would deny justice to some who … ought to succeed, in the interests of certainty, where the very subject matter is uncertain and continuously developing.15

A. The Duty of Care as Control Mechanism The reason why the duty of care concept has been crucial to the pursuit of certainty in tort law is because it is seen as providing a mechanism for controlling the scope of liability for negligence both pre-emptively and ex post facto.16 It is a mechanism of pre-emptive control inasmuch as it enables courts to strike out ‘unarguable’ claims in preliminary proceedings on the basis that no duty of care arises on the facts the claimant alleges to have

14

Ibid, at 17–18. McLoughlin, above n 12, at 443. See also Canadian National Railway Co v Norsk Pacific Steamship Co [1992] 1 SCR 1021, 91 DLR (4th) 289 (McLachlin J). 16 For judicial recognition of this role of the duty of care concept, see eg Candlewood, above n 6, at 25; Page v Smith [1995] UKHL 7, [1996] AC 155, 189 (Lord Lloyd). 15

Against Certainty in Tort Law 5 occurred.17 Despite Lord Browne-Wilkinson’s prudent warning in Barrett v Enfield LBC that ‘it is not normally appropriate to strike out where the law is uncertain and developing’,18 judicial willingness to accede to striking-out applications appears to be undiminished—even in claims that are ‘unarguable’ only because of the weight of the policy considerations deemed to oppose the recognition of a duty of care.19 The duty concept also allows appellate courts to control decisions of trial judges who may be swayed by excessive sympathy for the injured claimants appearing before them. In strict law, an appellate court is entitled to substitute its own assessment of the proper inference to be drawn from primary facts (eg whether the damage was foreseeable or whether the defendant exercised reasonable care) for that of the trial judge.20 But appellate courts frequently choose to rely instead on the absence of a duty of care because a ruling on a point of law establishes a precedent that must be followed in future cases too. These two aspects have transformed the way in which the duty of care is conceived. Where it was once viewed as a question of mixed fact and law—in which questions of (factual) foreseeability and (legal) proximity were considered as part of one global inquiry21—the modern tendency is to separate out the factual and legal aspects of the duty and to treat them as independent hurdles to be surmounted.22 The argument presented here is that these attempts to ‘police’ the scope of the duty of care have had unfortunate consequences for the coherence of the law and its ability to do justice. Four particular aspects may be highlighted: a focus on categories of case rather than on principle; the instrumental recourse to policy arguments; an explicit policy of incrementalism; and a succession of regrettable outcomes that testify to a judicial abdication of responsibility in this area. These will now be addressed in turn.

17 CPR 1998, r 3.4. A court may also dispose of a claim summarily if the claimant has no real prospect of success (CPR, pt 24), and this applies where the claim lacks an adequate factual foundation even if it is well conceived as a matter of law. 18 Barrett v Enfield LBC [1999] UKHL 25, [2001] 2 AC 550, 557 [Barrett]. 19 See, eg, as regards the liability of childcare professionals, JD v East Berkshire Community Health NHS Trust [2005] UKHL 23, [2005] 2 AC 373 [JD] and, as regards the liability of the police, Brooks v Commissioner of the Police for the Metropolis [2005] UKHL 24, [2005] 1 WLR 1495 [Brooks] and Smith v Chief Constable of Sussex Police [2008] UKHL 50, [2009] 1 AC 225 [Smith]. 20 Benmax v Austin Motor Co Ltd [1955] AC 370, [1955] 2 WLR 418 (HL). 21 Cf Elguzouli-Daf v Commissioner of Metropolitan Police [1994] EWCA Civ 4, [1995] QB 335, 349 (Steyn LJ) (‘It does not seem to me that these considerations can sensibly be considered separately … inevitably they shade into each other’). 22 Clerk and Lindsell on Torts, 20th edn (London, Sweet & Maxwell, 2010) §§ 8-06 to 8-07, speaking of ‘factual duty’ and ‘notional duty’. Because factual duty involves inference from primary facts, it would seem to be inappropriate to strike out a claim by reference to this factor alone, because the primary facts necessarily remain to be ascertained.

6

Ken Oliphant

B. A Focus on Categories in Place of Principle A defining feature of the modern law of negligence has been its disintegration into a number of barely connected ‘pigeonholes’ or ‘pockets’ of liability.23 The duty of care is conceived as specific to particular situations,24 and legal argument focuses on clusters of decided cases on similar facts as the parties attempt to draw convincing analogies or, as the case may be, to identify persuasive grounds of distinction. The task is explicitly conceived of as pragmatic25 rather than principled. The positive law consequently loses contact with deep-rooted legal values and fundamental principles of interpersonal justice. Though the way was paved by other decisions of the House of Lords,26 Caparo v Dickman plc27 stands as the decisive authority. There Lord Bridge famously suggested that the courts should abandon ‘the modern approach’ of looking for a single general principle underlying the tort of negligence and revert to ‘the more traditional categorisation of different specific situations as guides to the existence, the scope and limits of the varied duties of care which the law imposes’.28 Concurring, Lord Roskill explicitly identified certainty as a key factor justifying the change in focus: ‘the traditional categorisation of cases as pointing to the existence and scope of any duty of care … is infinitely preferable to recourse to somewhat wide generalisations which leave their practical application matters of difficulty and uncertainty’.29 Though Lord Bridge pointed to three ‘necessary ingredients’ in any duty situation—namely, the foreseeability of damage, a relationship of proximity between the parties, and that it is fair, just and reasonable to recognise a duty of care30—he emphasised that every potential duty situation should be assessed independently and pragmatically. Each is ‘sufficiently distinct to require separate definition of the essential ingredients by which the

23 See, eg, J Stapleton, ‘Duty of Care and Economic Loss: a Wider Agenda’ (1991) 107 LQR 249 (referring to the ‘pockets of case law’ approach); KM Stanton, ‘Professional Negligence: Duty of Care Methodology in the Twenty-First Century’ (2006) 22 Professional Negligence 134; KM Stanton, ‘The Neighbour Principle in the 21st century: Yesterday’s Revolution’ (2012) 20 Tort Law Review 61. 24 For an analysis of the duty of care concept in terms of recognised ‘duty situations’, see Clerk & Lindsell on Torts, above n 22, at § 8-05. 25 Caparo Industries plc v Dickman [1990] UKHL 2, [1990] 2 AC 605, 618 (Lord Bridge), 628 (Lord Roskill) [Caparo]. 26 See especially Governors of Peabody Donation Fund v Sir Lindsay Parkinson & Co Ltd [1983] UKHL 5, [1985] AC 210; The Aliakmon, above n 2; Hill v Chief Constable of West Yorkshire [1987] UKHL 12, [1989] AC 53 [Hill]. 27 Above n 25. 28 Ibid, at 616, 618. 29 Ibid, at 628. 30 Ibid.

Against Certainty in Tort Law 7 existence of the duty is to be recognised’.31 This was not necessarily a complete turn away from principled analysis. Lord Bridge himself gave at least formal recognition to the importance of underlying general principles. In any case, calling off the pursuit of a single general principle that can determine the existence of a duty of care in individual cases leaves scope for the identification or development of more specific principles that can be called upon in particular types of case. As Lord Goff observed in Smith v Littlewoods Organisation Ltd, ‘having rejected the generalised principle, we have to search for special cases in which, upon narrower but still identifiable principles, liability can properly be imposed’.32 However, the way this has been accomplished in practice has tended towards the mere description of common elements of past cases in which a duty of care has been found to arise33 rather than a serious attempt to identify principles which express the underlying values of the law. There has been undue reliance on factual analogy without identifying the normative significance of those features of different cases that are perceived to be on all fours or, as the case may be, to provide grounds of distinction. And, in cases falling outside the existing precedents, the courts have too readily fallen back on policy argumentation rather than seeking to develop the new mid-level principles that Lord Goff had in mind.

C. Instrumental use of Policy Arguments While it is now generally accepted that policy issues are ‘justiciable’ in the modern law of negligence,34 the extent of the reliance on policy in recent decades has been excessive. In fact, one might reasonably say that, in the period in question, UK Supreme Court decisions on the duty concept in negligence (and House of Lords’ decisions before) have been typified by the ad hoc invocation of policy arguments to lend spurious support to whatever outcome is intuitively believed to be correct—without regard to the consistency with which they are deployed. Consider the following: —

31

First, the floodgates argument is perhaps the policy consideration that is most frequently relied on in the modern law of negligence, but it remains profoundly controversial. In McLoughlin v O’Brian,35 Lord

Ibid, at 616. Smith v Littlewoods Organisation Ltd [1987] UKHL 3, [1987] AC 241, 280. 33 Cf Hedley Byrne & Co v Heller & Partners [1963] UKHL 4, [1964] AC 465, 526 (Lord Devlin) [Hedley Byrne] (‘see how far the authorities have gone’). 34 McLoughlin, above n 12, at 428 (Lord Edmund Davies) (cf Lord Scarman at 430, maintaining the contrary proposition); A Robertson, ‘Justice, Community Welfare and the Duty of Care’ (2011) 127 LQR 370. 35 McLoughlin, above n 12. 32

8

Ken Oliphant

Wilberforce accepted that the fears aroused by the floodgates argument were often overstated, but nevertheless considered the argument of sufficient weight to make it necessary to impose proximity restrictions on liability for nervous shock ‘just because “shock” in its nature is capable of affecting so wide a range of people’.36 Conversely, Lord Bridge observed in the same case: ‘I believe that the “floodgates” argument ... is, as it always has been, greatly exaggerated’.37 — Second, the risk of overkill or detrimentally defensive conduct has led the courts to deny a duty of care in several high-profile cases,38 but several judges have nevertheless expressed scepticism about the weight properly to be attached to it.39 Most memorably, in Home Office v Dorset Yacht Co Ltd40 Lord Reid roundly dismissed the suggestion that British public servants might be so apprehensive, easily dissuaded from doing their duty, and intent on preserving public funds from costly claims, that they could be influenced negatively in their conduct: ‘my experience’, he commented, ‘leads me to believe that Her Majesty’s servants are made of sterner stuff’. — Lastly, in declining to recognise a duty of care on social services to the parent of a child taken into protective care, the House of Lords relied primarily on the potential conflict of duty between the interests of child and parent,41 yet equally obvious conflicts of interest in other contexts—for example, the conflict between the duty owed by an advocate or expert witness to the client and that owed to the court—have not been deemed sufficient to deny the existence of a duty of care.42 Each of these examples shows that, in the absence of real evidence about the effects of the imposition of a duty of care on potential defendants, reliance upon such policy arguments can only be speculative and intuitive, and contributes to the degeneration of law into a collection of fact-specific rules, without concern for principle or coherence. Further, as the courts seem content to reassess the policy question afresh in each new factual scenario, it becomes well nigh impossible to predict which way cases on novel facts will go, or indeed to identify any rational basis for reconciling 36

Ibid, at 421–22. Ibid, at 442. 38 See especially Lord Keith in Rowling v Takaro Properties Ltd [1987] UKPC 2, [1988] AC 473, 502 (PC (New Zealand)); Hill, above n 26; Marc Rich & Co v Bishop Rock Marine Co (The Nicholas H) [1995] UKHL 4, [1996] AC 211; Smith, above n 19. 39 See, eg, Barrett, above n 18; JD, above n 19, at [33] (Lord Bingham). 40 Home Office v Dorset Yacht Co Ltd [1970] UKHL 2, [1970] AC 1004, 1033. 41 JD, above n 19. By contrast, Lord Bingham, dissenting, at [44], thought that the interests of child and parent were generally consonant or at least not sufficiently dissonant as to justify a general denial of any duty of care towards the latter, while accepting that the child’s interests had to prevail if a conflict should on occasion arise. 42 Arthur JS Hall & Co (A Firm) v Simons [2008] UKHL 38, [2002] 1 AC 615; Jones v Kaney [2011] UKSC 13, [2011] 2 AC 398. 37

Against Certainty in Tort Law 9 divergent outcomes in past cases, except on the basis that—in the present legal climate—the conservative option normally prevails.

D. Incrementalism A further aspect of the currently strong attachment to certainty is the explicit policy of ‘incrementalism’ that has been adopted by the courts in applying the duty of care to new situations. This approach may be traced back to an oft-cited dictum of an Australian judge, who urged that the range of duty situations should only be extended ‘incrementally and by analogy with established categories’.43 However, ‘incrementalism’ does not lend itself to the coherent development of the law.44 First, to say that extensions of the scope of the duty of care should be allowed only incrementally is not to indicate the principles which determine whether such extensions should be allowed: it merely prescribes that those principles should be narrowly defined. The need to identify the relevant principles remains. As Lord Bingham has observed: ‘the incremental test is of little value as a test in itself, and is only helpful when used in combination with a test or principle which identifies the legally significant features of a situation’.45 Second, there is a danger that the principles by which step-by-step expansions of the duty are permitted will reflect arbitrary features of previously decided cases rather than the essential merits of the claimant’s action against the defendant; liability may turn upon history rather than justice. In short, as an Irish judge has astutely noted, ‘the verbally attractive proposition of incremental growth ... suffers from a temporal defect—that rights should be determined by the accident of birth’.46 The result has been a lack of sensitivity to the underlying merits of individual claims, and the introduction of undue rigidity into the law.

E. Judicial Passivity in the Face of Unsatisfactory Outcomes A final aspect to highlight is the passivity shown by the judiciary in the face of blatantly unsatisfactory outcomes, and its undue readiness to say that the responsibility for addressing problems lies with Parliament, not the courts.

43

Sutherland Shire Council v Heyman (1985) 157 CLR 424, 481 (HCA) (Brennan J). A Mullis and K Oliphant, Torts, 4th edn (Basingstoke, Palgrave Macmillan, 2011) 22. For further criticism, see L Dolding and R Mullender, ‘Tort Law, Incrementalism, and the House of Lords’ (1996) 47 Northern Ireland Law Quarterly 12. 45 Customs and Excise Comrs v Barclays Bank plc [2006] UKHL 28, [2007] 1 AC 181, [7]. 46 Ward v McMaster [1988] IR 337, 347, [1989] ILRM 400 (McCarthy J). 44

10

Ken Oliphant

Evidence that outcomes have indeed been unsatisfactory is provided by the all too frequent expressions of regret from the judges who have to apply the law. Let me give just one example, out of many possible candidates.47 In Jain v Trent Strategic Health Authority,48 the claimants lost their nursing home business after the local health authority unreasonably made an ex parte, without notice application to a magistrate for cancellation of their statutorily required registration. By the time the claimants were able to present their side of the story, in the appeal hearing six months later, their business had suffered irreversible damage. They then suffered the further indignity of losing their action for damages. The Human Rights Act 1998 was not effective at the relevant time, so the claimants had relied on common law negligence. But the Law Lords rejected their claim on the basis that the authority owed them no duty of care, the loss being purely economic. Every one of their Lordships expressed regret at the outcome.49 The judiciary’s response to such unreasonable outcomes has been sadly deficient. The decision of the House of Lords in White v Chief Constable of South Yorkshire Police,50 dealing with claims for post-traumatic stress disorder by police officers present at the Hillsborough football stadium disaster, represents a particular low point. The Law Lords considered the law of nervous shock to be so far beyond judicial repair that the only sensible strategy for the courts was—in Lord Steyn’s words—‘to say thus far and no further ... [and] by and large to leave any expansion or development in this corner of the law to Parliament’.51 As Lord Hoffmann observed, ‘the search for principle’ in this area of the law had been ‘called off’.52 Faced with such intransigence, it might be well to remind ourselves who created the mess in which the House of Lords found itself. Even a child knows the maxim, ‘You broke it—you fix it’. The story does not end there. Lord Steyn’s cautionary words were invoked subsequently by Lord Hope in the Pleural Plaques decision, which declined to extend the category of ‘primary victim’ to cover those suffering psychiatric illness as the result of their fear of developing cancer in the future in consequence of their past exposure to asbestos.53 This delegation of responsibility to the legislature looks decidedly odd when one considers that, just five months before, the Department of Constitutional Affairs 47 See also JD, above n 19, at [137] (Lord Brown) (‘legitimate grievances’, ‘paying the price of the law’s denial of a duty of care’); Rothwell v Chemical & Insulating Co Ltd [2007] UKHL 39, [2008] 1 AC 281, [59] (Lord Hope) [Rothwell]. 48 Jain v Trent Strategic Health Authority [2009] UKHL 4, [2009] 1 AC 853 [Jain]. 49 Ibid, at [40] (Lord Scott), [41] (Lord Rodger) (implicitly), [42] and [48] (Baroness Hale), [52] (Lord Carswell); [53] and [57] (Lord Neuberger). 50 White v Chief Constable South Yorkshire Police [1998] UKHL 45, [1999] 2 AC 455 [White]. 51 Ibid, at 500. 52 Ibid, at 511. 53 Rothwell, above n 47.

Against Certainty in Tort Law 11 had published a Consultation Paper on The Law on Damages, in which it recommended that the courts continue to take responsibility for developing the law of liability for psychiatric injury.54 The government’s expectation was that the courts would do so flexibly and incrementally; legislation, it was thought, would run the risk of imposing rigid requirements which would be unable to accommodate developments in medical knowledge and diagnoses. We seem therefore to be stuck in a game of legal pass-the-parcel in which the music never stops, and no one gets closer to unwrapping the fundamental principles upon which rational legal development depends.

IV. THE COUNTER-PRODUCTIVENESS OF THE SEARCH FOR CERTAINTY

The analysis above demonstrates that the pursuit of certainty through the predominant modern judicial approach to the duty of care—involving rigorous ‘policing’ of duty of care issues—in fact results in regrettable outcomes and a lack of fidelity to underlying principles and values. It is also counter-productive, as it has in fact produced a high degree of uncertainty, even obscurity. Issues are left unresolved; new doubts are introduced. This is evident from the repeated visits first to the House of Lords, and then to the UK Supreme Court, that have been necessary in many particular contexts in recent years, often without any final resolution of the disputed issues or clarity as to the relevant principles, sometimes with the introduction of new uncertainties through the suggestion of possible exceptions or qualifications. A few examples should make this clear.

A. The Liability of Childcare Professionals First, in a period of just over ten years around the turn of the millennium, the House of Lords was called on no fewer than four times to determine the liabilities of childcare professionals. In X v Bedfordshire; M v Newham,55 the Law Lords ruled that social workers owed no duty of care in considering whether to take into protective care children who were or might have been at risk of being abused in their homes. Conversely, in Barrett v Enfield London Borough Council,56 the decision was that the social workers did owe an (arguable) duty of care once a child had been taken into protective care, and might in principle therefore be liable for psychiatric harm resulting

54 Department of Constitutional Affairs, The Law on Damages (2007) (Consultation Paper CP 9/07), [94]. 55 X v Bedfordshire; M v Newham [1995] UKHL 9, [1995] 2 AC 633. 56 Above n 18.

12

Ken Oliphant

from his frequent changes of foster placement (nine in total, over 17 years). W v Essex County Council57 also found an arguable duty of care in respect of a child in a foster placement—but this time to the foster parents whose own children were abused by the boy they had taken in and who suffered psychiatric injury in consequence. Finally, in JD v East Berkshire,58 after the Court of Appeal had declined to follow X v Bedfordshire; M v Newham on the grounds that the introduction of the Human Rights Act 199859 had decisively changed the balance of the relevant policy considerations, and so ruled that a duty of care was owed in deciding to take a child into care, the Lords ruled that the scope of that duty did not extend to the child’s parents. Precluding such a duty, in particular, was the potential conflict between the duty owed to the child and that argued to be owed to the parents. Despite this flurry of cases, significant uncertainties remain. A first uncertainty is whether a duty of care is owed by a care worker to a child whom the care worker decides to place in protective care because of a perceived risk of abuse. As just mentioned, the House of Lords said not in X v Bedfordshire; M. v Newham, but the Court of Appeal concluded in JD v East Berkshire that the policy considerations had shifted following the enactment of the Human Rights Act 1998, and found that there was now a duty to a child taken into care. This seems to have been accepted when the case reached the House of Lords—where only the (alleged) duty to the parent was in issue. Yet the reasoning runs counter to the analysis of the House of Lords in other cases when considering the Act’s impact on the common law, in which the existence of a possible claim under the Act has been seen to reduce or even obviate the need to fashion a claim in negligence,60 and the existence of a duty of care to the child in such circumstances has been regarded as at least open to question.61 A second uncertainty is whether, even if it is conceded that a duty is owed to a child taken into care, a duty is owed to a child whom social services mistakenly conclude is not at risk of harm, and who is consequently left in the family home, where he or she experiences abuse. These were the actual facts of X v Bedfordshire, which was distinguishable for that reason from JD v East Berkshire. It cannot be taken for granted that the Court of Appeal decision in the latter case—in favour of a duty to the child—will be treated as applicable if the Court of Appeal, or a lower court, is faced with

57

W v Essex County Council [2000] UKHL 17, [2001] 2 AC 592 [W v Essex]. Above n 19. 59 A person whose rights under the Act are infringed by a public authority may also bring a claim for compensation under the mechanism provided by the statute, but the focus of the present chapter is on liability for negligence at common law. 60 See especially JD, above n 19, at [94] (Lord Nicholls); Smith, above n 19, at [82] (Lord Hope), [136] (Lord Brown) (‘it is quite simply unnecessary now to develop the common law to provide a parallel cause of action’). 61 Jain, above n 48, at [25] (Lord Scott). 58

Against Certainty in Tort Law 13 a claim based on a failure to remove a vulnerable child from an abusive home environment. On the contrary, it seems quite likely that even a first instance court might consider the point open to decision in the post-Human Rights Act legal world. As the law is generally reluctant to impose a duty of care that entails an affirmative obligation to intervene, a future ‘no-duty’ decision in this context is not to be discounted.

B. The Police Next, there is the liability of the police. In just over 20 years, three police negligence actions reached the House of Lords: Hill v Chief Constable of West Yorkshire,62 Brooks v Commissioner of Police for the Metropolis63 and Smith v Sussex Police.64 In all these cases, the House of Lords ruled that no duty of care arises in respect of the police’s investigation of crime, whether to a potential victim or (in Brooks) a witness, because the threat of litigation might induce officers to adopt a detrimentally defensive approach to criminal investigations, which would not be fair, just and reasonable.65 Again, significant uncertainties remain. The last case of the three leaves open the question whether the police owe a duty of care in respect of their performance of operational tasks, for example, concerning public safety on the roads.66 Established Court of Appeal authority says they do not,67 but that seems now to be open to challenge. And there is still a question mark over whether the police may assume a responsibility towards a particular person, for example a witness who agrees to testify in a criminal trial. The Court of Appeal has accepted that a duty of care may arguably arise on this basis,68 and this seems not yet to have been ruled out as a possibility by either the House of Lords or the Supreme Court.69 Nevertheless, it cannot be affirmed with any measure of certainty that the Supreme Court will adopt this reasoning if called upon to decide the matter.

62

Above n 26. Above n 19. 64 Above n 19. 65 In Hill, above n 26, the claim also failed for lack of the necessary relationship of proximity between the police and the victim. 66 Smith, above n 19, at [79] (Lord Hope), [109] (Lord Carswell). 67 Ancell v McDermott [1993] EWCA Civ 20, [1993] 4 All ER 355. 68 Swinney v Chief Constable of Northumbria Police [1996] EWCA Civ 1322, [1997] QB 464. A similar analysis was also proposed at first instance in Van Colle v Chief Constable of Hertfordshire [2006] EWHC 360, [2006] 3 All ER 963, [89] (Cox J) (QB). The claim was actually under the Human Rights Act, and was ultimately rejected by the House of Lords: [2008] UKHL 50, [2009] 1 AC 225 [Van Colle] (decided with Smith, above n 19). 69 In fact, in Van Colle, above n 68, at [120], [135] Lord Brown considered that the argument was correct in principle. 63

14

Ken Oliphant

C. Wrongful Conception Third, we come to wrongful conception. In just a few years at the turn of the millennium, we had three major decisions, two of them (McFarlane v Tayside Health Board70 and Rees v Darlington Memorial Hospital NHS Trust71) in the House of Lords. The first of these (McFarlane) decided that the liability of a health authority for the negligent performance of a sterilisation operation, which was consequently ineffective, is limited to losses directly attributable to the resulting pregnancy and birth and does not extend to the costs of the child’s upbringing; the latter costs fall outside the ambit of the health authority’s duty of care. In the second case (Rees), the Law Lords confirmed their earlier decision, and ruled that the same principle also applied where the costs of raising the child were increased by the pre-existing disability of the mother; no exception was to be admitted in respect of the additional costs she would incur as a result of her disability. Intervening between the two House of Lords decisions is the third case in the series, Parkinson v St James and Seacroft University Hospital NHS Trust.72 This did not reach the Lords, so the issue it raised—whether damages for wrongful conception extend to additional costs attributable to the disability of the ‘unwanted’ child—remains unresolved. The three Law Lords in the minority in Rees (Lord Steyn, Lord Hope and Lord Hutton) approved the decision that damages could be awarded. Of the majority, two (Lord Bingham and Lord Nicholls) were against it, and indeed opposed to any exception on grounds of the child’s disability, one (Lord Scott) thought that Parkinson was wrongly decided but that the award of damages might be appropriate where the very purpose of the sterilisation was to protect against the birth of a child with an inherited disability. The final Law Lord (Lord Millett) expressly declined to voice an opinion.73 So here too we have uncertainty, and a further trip to the Supreme Court will be necessary to resolve once and for all the issue of wrongful conception claims by disabled children.

D. Psychiatric Harm Finally, we may turn to the law of nervous shock. In just over 20 years, we have had six major House of Lords decisions.74 These have in no way 70

McFarlane v Tayside Health Board [1999] UKHL 50, [2000] 2 AC 59. Rees v Darlington Memorial Hospital NHS Trust [2003] UKHL 52, [2004] 1 AC 309. 72 Parkinson v St James and Seacroft University Hospital NHS Trust [2001] EWCA Civ 530, [2002] QB 266. 73 Rees, above n 71, at [9] (Lord Bingham), [18] (Lord Nicholls), [35] (Lord Steyn), [57] (Lord Hope), [91] (Lord Hutton), [112] (Lord Millett), [145] (Lord Scott). 74 Hicks v Chief Constable of the South Yorkshire Police [1991] UKHL 9, [1992] 2 All ER 65; Alcock v Chief Constable of South Yorkshire [1991] UKHL 5, [1992] 1 AC 310 [Alcock]; 71

Against Certainty in Tort Law 15 improved the law as previously laid down in McLoughlin v O’Brian,75 a decision in which the majority of the House of Lords in fact adopted a commendably flexible approach to establishing a duty of care,76 only for a later panel of Law Lords to reinterpret the decision to reflect Lord Wilberforce’s more rigid, minority approach in that case.77 According to that approach, a duty of care is generally owed to a person suffering psychiatric harm from an accident involving another person only if the claimant had a close tie of love and affection with the ‘primary victim’, and either witnessed the accident directly or came upon its immediate aftermath. Again, significant uncertainties remain. One is whether an exception to the usual requirements should be admitted if the events are especially horrific. In the leading case, one Law Lord thought that such an exception might be justified, and therefore expressly declined to exclude the possibility of liability where, for example, a passer-by witnesses a petrol tanker careering out of control into a school in session and bursting into flames.78 It is not self-evident that the imagined scenario would be any more horrific than the tragic events in the case actually before his Lordship,79 and it seems somewhat distasteful to engage in the measurement of different degrees of horror, which was one factor that induced the Court of Appeal subsequently to reject the contemplated exception.80 But no final court of appeal has yet ruled on the matter so—in theory at least—it remains unresolved. A further uncertainty relates to a second possible qualification to the general approach adopted in this area. Generally, no duty of care is owed to persons who suffer psychiatric harm as the result of watching television coverage of an accident in which they know a close relative is involved, as the interpolation of television coverage—even if live—takes away the

Page v Smith, above n 16; White, above n 50; W v Essex, above n 57; and Rothwell, above n 47. Some other decisions in the same period (eg Barber v Somerset County Council [2004] UKHL 13, [2004] 1 WLR 10) may be regarded as somewhat less important in terms of legal principle. Under Scots law, see also Simmons v British Steel plc, 2004 SC (HL) 94, [2004] ICR 585. 75

Above n 12. Every Law Lord except Lord Wilberforce considered that the reasonable foreseeability of psychiatric injury raised a prima facie duty of care, all but Lord Scarman accepting it might be rebutted in an appropriate case by countervailing policy considerations. See Mullis and Oliphant, above n 44, at 37–38. This interpretation is in line with that advanced by H Teff, Causing Psychiatric and Emotional Harm: Reshaping the Boundaries of Legal Liability (Oxford, Hart Publishing, 2009) 63–64. For a rival interpretation, see D Nolan, ‘McLoughlin v O’Brian (1983)’ in C Mitchell and P Mitchell (eds), Landmark Cases in the Law of Tort (Oxford, Hart Publishing, 2010). 77 Alcock, above n 74. 78 Ibid, at 403 (Lord Ackner). 79 The case arose out of the Hillsborough football stadium tragedy of 1989. 80 McFarlane v EE Caledonia Ltd [1993] EWCA Civ 27, [1994] 2 All ER 1. 76

16

Ken Oliphant

scenes’ sudden impact upon the senses.81 It has been suggested, however, that there might be occasions when television pictures would have at least as great an impact on observers as actual presence at the scene, an illustration being where parents are watching live television pictures of their children travelling in a hot-air balloon when the balloon suddenly bursts into flames and plummets to the ground.82 The dictum seems not to have received subsequent judicial attention, so the existence of the proposed exception remains undecided. Turning from secondary victims to primary victims, we find yet more uncertainty. In the first place, it is unclear even how the term ‘primary victim’ should be defined, and in particular whether it extends beyond the core category of persons who are themselves imperilled by the accident or other occurrence that the defendant has negligently caused.83 Lord Oliver once posited two further categories of primary victim: rescuers and those who are the unwitting mechanism through whom another person is killed or injured.84 The first of these suggested categories (rescuers) was rejected by the House of Lords when it was confronted by the issue shortly afterwards.85 A few years later, however, the Law Lords declined to strike out as unarguable a claim falling under the second of Lord Oliver’s proposed categories (unwitting agents of misfortune)—in a case where parents alleged they had suffered psychiatric harm after their young children were sexually abused by a foster child whom they had accepted into their household, after receiving (false) assurances from the council that the child they were fostering had no record of such conduct.86 Because the decision was only that a duty of care was ‘arguable’, it again cannot be treated as definitively resolving the matter. In any case, a more fundamental uncertainty must also be confronted, relating to the core category of primary victims—those actually imperilled by the accident. Under the approach adopted by the House of Lords in Page v Smith,87 the claimant establishes a duty of care by showing the foreseeability of any personal injury, as opposed to the foreseeability of psychiatric harm in particular. The decision thus departs from the test that has long been applied in respect of secondary victims, and for that reason its correctness was subsequently doubted in more than one decision of the House 81

Alcock, above n 74. Ibid, at 405 (Lord Ackner). 83 Cf W v Essex, above n 57, at 601, where Lord Slynn remarked that ‘the categories of those claiming to be included as primary or secondary victims are not as I read the cases finally closed. It is a concept to be developed in different factual situations.’ 84 Alcock, above n 74, at 407–08, referring to Chadwick v British Railways Board [1967] 1 WLR 912, [1967] 2 All ER 945 and Dooley v Cammell Laird & Co Ltd [1951] 1 Lloyd’s Rep 271, respectively. 85 White, above n 50. 86 W v Essex, above n 57. 87 Above n 16. 82

Against Certainty in Tort Law 17 of Lords.88 It seems likely that a full challenge to the decision—and hence to the making of any distinction at all between primary and secondary victims—will be brought to the Supreme Court in the coming years.

V. TOWARDS A MORE FLEXIBLE APPROACH

In truth, the search for certainty in an area of law where so much depends on judicial assessments of what is ‘reasonable’ is as likely to succeed as an attempt to catch a moonbeam. Uncertainty is the inevitable result of the open-textured nature, and hence the inherent instability, of the concepts employed. Apparently firm boundary walls erected by the courts are prone to subsidence because of the insecurity of their foundations. A prime example is provided by the many duty-restricting rules that, on thorough scrutiny, prove to be based on foreseeability or some variant upon it. For example, knowledge—which lies at the heart of the ‘narrow’ ratio decidendi of Caparo v Dickman89—is simply foreseeability of greater degree than that normally required, at least if one accepts that knowledge can be imputed on the basis of what was foreseeable with a sufficient degree of probability, which is a question of fact for the tribunal of fact in every case. A rule based on (actual or imputed) knowledge can never bring the hopedfor certainty. But Caparo v Dickman demonstrates that the power of a supreme court decision is not limited to its narrow ratio decidendi. The ‘message’ that it sends to lower courts and to litigants is equally, perhaps more, important. Thus, after Caparo, first instance judges and the Court of Appeal took a notably more restrictive approach to the scope of the duty of care in respect of the preparation and certification of company accounts than had previously prevailed.90 A more recent decision that has had a comparable impact is Tomlinson v Congleton Borough Council,91 whose narrow ratio decidendi relating to the statutory duty of care owed under the Occupiers’

88 White, above n 50, at 473–80 (Lord Goff); Rothwell, above n 47, at [52] (Lord Hope), [104] (Lord Mance). Cf Corr v IBC [2008] UKHL 13, [2008] 1 AC 884, [40] (Lord Walker) (‘a much simpler test’). 89 Above n 25. See especially Lord Bridge at 621: ‘the defendant knew that his statement would be communicated to the plaintiff, either as an individual or as a member of an identifiable class, specifically in connection with a particular transaction or transactions of a particular kind … and that the plaintiff would be very likely to rely on it for the purpose of deciding whether or not to enter upon that transaction or upon a transaction of that kind’. As I have remarked elsewhere, this language is so imprecise as to make us ask why the test was not satisfied on the facts of Caparo itself: Mullis and Oliphant, above n 44, at 67. 90 See, eg, James McNaughton Paper Group Ltd v Hicks Anderson & Co [1990] EWCA Civ 11, [1991] 2 QB 113; Berg Sons & Co Ltd v Adams [1992] BCC 661. 91 Tomlinson v Congleton Borough Council [2003] UKHL 47, [2004] 1 AC 46 [Tomlinson].

18

Ken Oliphant

Liability Acts92 seems in retrospect—and was probably envisaged by its authors at the time—to be less important than their general remarks about ‘compensation culture’93 and the need to be vigilant against a decline in notions of personal responsibility in addressing the issue of breach of duty and the required standard of care. Although the House of Lords’ analysis of the duty issue was rather technical, and may perhaps be considered unpersuasive,94 the main message the decision sent to lower courts and potential litigants was powerful and unambiguous—and has been influential in a wide range of contexts.95 This suggests that the role of a final court of appeal may most usefully be conceived of in terms of ‘setting the tone’ for lower court decisions, and steering the lower courts back into line when their decisions begin to tend in the wrong direction. It seems not to be productive for them to seek to constrain the exercise of judgment in those courts by laying down detailed legal rules that attempt to specify in detail the proper scope of the duty of care. Experience shows that gaps will inevitably remain, and it is more important to state the principles according to which the existence of a duty of care should be determined, and to engage in a flexible weighing up of all normatively relevant factors, than to strive for a certainty that can never be achieved, when the effort to attain it is liable to produce incoherence and injustice.

92 Following Lord Hoffmann, ibid, at [27], this may be said to be that the duty of care under the Occupiers’ Liability Acts is contingent on a risk of injury arising out of the state of the premises and not out of what the claimant chose to do on the premises. This was followed in Keown v Coventry Healthcare NHS Trust [2006] EWCA Civ 39, [2006] 1 WLR 953; Siddorn v Patel [2007] EWHC 1248 (QB) (the risk of falling through a skylight if one chooses to dance on a roof is not one arising from the state of the premises as such). A subtly but significantly different approach was taken by the majority of the High Court of Australia in Vairy v Wyong Shire Council, [2005] HCA 62, 223 CLR 422 and Mulligan v Coffs Harbour City Council [2005] HCA 63, (2005) 223 CLR 486, where it was held that the obviousness of the risk that attached to the claimant’s activity (diving into water of unknown depth) on the land was only one factor to consider in determining whether the occupier was required to take steps to prevent the risk from materialising, although it might be determinative in particular cases. 93 See, eg, Tomlinson, above n 91, at [81] (Lord Hobhouse): ‘The pursuit of an unrestrained culture of blame and compensation has many evil consequences and one is certainly the interference with the liberty of the citizen.’ 94 For criticism, see K Oliphant, ‘England and Wales’ in H Koziol and BC Steininger (eds), European Tort Law 2003 (Vienna, Springer, 2004) 132–35. 95 Though there are a few cases that apply the narrow ratio of the Tomlinson decision (see above n 91), these are outnumbered by the substantial number of cases in a variety of different contexts that apply the Law Lords’ more general remarks regarding breach of duty and the standard of reasonable care. See, eg, Clare v Perry [2005] EWCA Civ 39, (2005) 149 SJLB 114; Evans v Kosmar Villa Holidays Plc [2007] EWCA Civ 1003, [2008] 1 WLR 297; Poppleton v Trustees of the Portsmouth Youth Activities Committee [2008] EWCA Civ 646, [2009] PIQR P1.

2 Tort Law and Human Flourishing NICHOLAS J McBRIDE*

I. INTRODUCTION

F

OR A GOOD number of years now, academics on both sides of the Atlantic have been challenging the orthodoxy that tort law is simply concerned to determine when a defendant should be made to pay compensation to a claimant for harm that the defendant has caused the claimant to suffer. These rebel academics have tended to argue that tort law performs the far more fundamental role of determining what basic rights we have against each other and what remedies will be available when those rights are violated. This chapter is not about that battle, which for my money has been largely won by the rebels. Instead, I want to move on to another dispute which has broken out between the rebels over what basic rights tort law gives us against other people. The rebels all agree over the basic form of those rights. They take the form of Hohfeldian claim rights, where Claimant can be said to have a right against Defendant that Defendant do x if the law imposes a duty on Defendant to do x and it does so for Claimant’s benefit.1 Where the rebels disagree is over when tort law will impose such a duty on Defendant for Claimant’s benefit. What we can call the ‘Kantian school’ of thought argues that tort law will only impose a duty on Defendant for Claimant’s benefit where it is necessary to do so in order to protect Claimant’s independence as a

* My thanks to all the participants at the Obligations VI Conference, the members of the Cambridge Centre for Private Law, and members of the University of Hong Kong for their comments on this chapter, especially Roderick Bagshaw, John Goldberg, Jason Neyers, Sandy Steel, Rob Stevens and Jason Varuhas. Rajiv Shah also made a number of valuable comments on an earlier draft. All errors in this chapter are my responsibility. 1 For a more detailed elaboration of this point, see R Stevens, ‘Rights and Other Things’ in D Nolan and A Robertson (eds), Rights and Private Law (Oxford, Hart Publishing, 2011) 116–18.

20

Nicholas J McBride

person.2 In a previous article,3 I advanced what we can call a ‘balanced view’ as to when tort law will impose a duty on Defendant to do x for Claimant’s benefit: that tort law will normally do so when Claimant’s interests in Defendant’s being required to do x outweigh, or are more important, than Defendant’s interests in being left free not to do x. This view is appropriately called ‘balanced’ as it involves balancing Claimant’s interests against Defendant’s interests. In balancing these interests, tort law has to take a view as to what is in Claimant’s and Defendant’s respective interests. And that, in turn, requires tort law to take a position on what is involved in human flourishing. We cannot tell whether Defendant’s being made to do x will make Claimant better off (and if so, to what extent), or make no practical difference to Claimant, unless we can construct some sort of line of value for Claimant, one end of which will involve Claimant doing ‘really badly’ and the other end of which will involve Claimant doing ‘really well’, and place Claimant on that line of value and see what effect Defendant’s being made to do x would have (or is likely to have) on Claimant’s position on that line. But in order to construct such a line of value, we have to take a view as to what doing ‘really well’ and what doing ‘really badly’ means for someone like Claimant. The same applies when we have to determine whether Defendant’s being made to do x will make Defendant worse off (and if so, to what extent), or make no effective difference to Defendant. So if the balanced view is correct, tort law must rest on some (however roughly worked out) theory as to what human flourishing involves. In this chapter, I want to do three things. First, make the balanced view of how tort law determines what rights claimants have against defendants more plausible to its Kantian critics by pointing to a number of cases where it seems fairly obvious that some kind of balancing process between the interests of claimants and the interests of defendants is going on in determining what rights claimants have against defendants. Second, drawing on these cases, I want to spell out the vision of human flourishing that underlies the way tort law currently balances the interests of claimants 2 See, in particular, E Weinrib, The Idea of Private Law (Cambridge, Mass, Harvard University Press, 1995) chs 3–5; E Weinrib, Corrective Justice (Oxford, Oxford University Press, 2012) chs 1, 6 and pp 337–41; A Ripstein, ‘Beyond the Harm Principle’ (2006) 34 Philosophy & Public Affairs 215; A Ripstein, ‘Tort Law in a Liberal State’ (2007) 1 Journal of Tort Law 3; A Beever, ‘The Form of Liability in the Torts of Trespass’ (2011) 40 Common Law World Review 378. I call this view ‘Kantian’ not because Kant necessarily took this view (Kant scholars seem to find it impossible to reach much agreement over what Kant actually did think), but because Kant’s writings—in particular, his Doctrine of Right in the Metaphysics of Morals—are regularly invoked in support of this view. For a masterly reading of Kant’s Doctrine of Right that lays the groundwork for this view of tort law, see A Ripstein, Force and Freedom (Cambridge, Mass, Harvard University Press, 2009). 3 NJ McBride, ‘Rights and the Basis of Tort Law’ in Nolan and Robertson, above n 1.

Tort Law and Human Flourishing

21

against defendants. Unsurprisingly, we will find that that vision is a pretty conventional one. It closely matches the understanding of what human flourishing involves that is currently dominant in Western societies and that is most effectively elaborated in the writings of John Finnis. Third, I want to argue that this orthodox, Western understanding of what human flourishing involves is incorrect and that tort law in its current state rests on a false premise as to what is in our interests and what is not. In making this argument, I will set out an alternative understanding of what human flourishing involves, and discuss what tort law would look like if it were based on that alternative understanding.

II. BALANCING IN TORT LAW

Most tort academics on the English side of the Atlantic would agree with the following passage: There are several good things in life, such as liberty, bodily integrity, land, possessions, reputation, wealth, privacy, dignity, perhaps even life itself. Lawyers call these goods ‘interests’. These interests are all good, but they are not equally good. This is evident when they come into conflict (one may jettison cargo to save passengers, but not vice versa, and one may detain a thing, but not a person, as security for a debt). Because these interests are not equally good, the protection afforded to them by the law is not equal ...4

This passage implicitly endorses the balanced view of the basis of the rights tort law gives us against other people. The more important A’s interest in B’s not doing x, the more likely it is that the law will impose a duty on B not to do x for A’s benefit. The less important, the harder it will be to justify imposing such a duty on B. A number of cases seem to bear out this elementary point, at least to those on the English side of the Atlantic.5

A. Private Nuisance Cases Private nuisance cases are the most obvious ones which involve the courts balancing two people’s interests against each other and deciding which is more important and should be allowed to win out.

4 AW Weir, A Casebook on Tort, 10th edn (London, Sweet & Maxwell, 2004) 6 (emphasis in original). 5 For further explorations of this balancing process, see R Stevens, ‘The Conflict of Rights’ in A Robertson and HW Tang (eds), The Goals of Private Law (Oxford, Hart Publishing, 2009) (though I depart from Stevens in that he sees the balancing process as being between prima facie rights, and I do not).

22

Nicholas J McBride

The available case law indicates that the courts think that there is never any value in maliciously disturbing another’s use and enjoyment of land, and in the case of such a disturbance they will always side with the person whose use and enjoyment of land has been disturbed.6 Where the interference is non-malicious and a necessary incident of some legitimate pursuit of the interferer, the courts have to make a value judgment as to whose interests are more valuable: the interferer’s in achieving his or her goals, or the interests of the person whose quiet enjoyment of land is being interfered with. For example, in Christie v Davey a family gave music lessons and occasionally put on concerts in their home.7 All this music making disturbed their neighbour. His attempts to retaliate for the disturbance by practising on musical instruments himself (very badly), and (it was alleged) shrieking, hammering, knocking on tea trays and blowing whistles, were straightforwardly enjoined by North J. But the judge was also asked to decide whether to allow the defendant’s counterclaim asking for his musical neighbours’ lessons and concerts to be limited or shut down. North J rejected the counterclaim. It is hard to avoid the impression that the respectability of the musical family’s profession as music teachers played a large part in the judge’s decision to allow them to carry on with that profession in their home. At various points in his judgment, North J refers to the daughter of the family as ‘a young lady who has received a good musical education and has taken a very good musical degree’;8 notes that one of the musicians who would visit the family to practise her singing ‘is a very skilful musician and is a medallist of the Academy of Music, and she has come to London for the purpose of continuing her musical studies’;9 and emphasises that ‘The profession which the ... family practise is a perfectly legitimate and proper one’.10 There can be absolutely no doubt that had the neighbour in Christie been disturbed by an exactly equal quantity of noise produced by the mother and daughter next door having sex with men for money, an injunction would have been readily granted. Music making (the court would have thought) is important for human flourishing and therefore worthy of protection; prostitution is not.

6 See, eg, Christie v Davey [1893] 1 Ch 316 [Christie]; Hollywood Silver Fox Farm v Emmett [1936] 2 KB 468. Bradford Corporation v Pickles [1895] AC 587 (HL) is not to the point here as the form of interference in that case—blocking water coming onto the claimant’s land—was incapable of amounting to a private nuisance given that the claimant in that case had no right to have the water come onto his land. See, further, NJ McBride and R Bagshaw, Tort Law, 4th edn (Harlow, Pearson Education, 2012) 433–34. 7 Christie, above n 6. 8 Ibid, at 324. 9 Ibid, at 325. 10 Ibid, at 327.

Tort Law and Human Flourishing

23

That this is so is confirmed by the decision of the Court of Appeal in Thompson-Schwab v Costaki where a family’s mere knowledge that the house next door was being used for the purposes of prostitution was held to be sufficient to give rise to a claim in private nuisance.11 The interests of the defendants in being allowed to carry on their trade counted for nothing with the court: the only issue for the court was whether what the defendants were doing could be said to have had enough of an adverse effect on the claimants’ interests as to amount to an actionable nuisance. In his judgment, Romer LJ established such an adverse effect by focusing on the fact that the claimants: have young sons living there ... and one of them has three young servants. One can well imagine the effect of what [is] going on in the house ... next door ... is likely to have on the minds of those young people. One can imagine also the feelings of visitors who come to the houses ...12

It is clear that the judge thought that growing up in a healthy atmosphere, and not having one’s friends put off from visiting you, are important interests that should be protected by the law. A far more difficult case was Miller v Jackson where the interests of the claimants and the defendant seemed much more equally matched on the scale of value used by the court to determine whose interests should prevail.13 The defendant was a cricket club. Lord Denning MR’s speech in the case gives some flavour of the importance attached by him to the defendant’s activities: In summertime village cricket is the delight of everyone. Nearly every village has its own cricket field where the young men play and the old men watch. In the village of Lintz in County Durham they have their own ground, where they have played these last 70 years. They tend it well. The wicket area is well rolled and mown. The outfield is kept short. It has a good club house for the players and seats for the onlookers. The village team play there on Saturdays and Sundays. They belong to a league, competing with the neighbouring villages.14

This little Eden was—like the original Eden—threatened by an interloper: a newcomer who is no lover of cricket [and who] ... complains that when a batsman hits a six the ball has been known to land in his garden or on or near his house. His wife has got so upset about it that they always go out at week-ends. They do not go into the garden when cricket is being played. They say that this is intolerable. So they asked the judge to stop the cricket being played. And the judge, much against his will, has felt that he must order the cricket to be stopped: with

11 12 13 14

Thompson-Schwab v Costaki [1956] 1 WLR 335 (CA). Ibid, at 341–42. Miller v Jackson [1977] 1 QB 966 (CA) [Miller]. Ibid, at 976.

24

Nicholas J McBride

the consequence, I suppose, that the Lintz Cricket Club will disappear. The cricket ground will be turned to some other use. I expect for more houses or a factory. The young men will turn to other things instead of cricket. The whole village will be much the poorer.15

Given the above, it is no surprise that Lord Denning MR found for the defendants: In this case it is our task to balance the right of the cricket club to continue playing cricket on their cricket ground—as against the right of the householder not to be interfered with. On taking the balance, I would give priority to the right of the cricket club to continue playing cricket on the ground, as they have done for the last 70 years. It takes precedence over the right of the newcomer to sit in his garden undisturbed.16

Geoffrey Lane LJ took a different view. For him, avoiding the threat posed by the cricket club to the claimants’ bodies and property was far more important than preserving the more intangible benefits associated with allowing the cricket club to carry on operating.

B. Education and Business Sterilisation Cases In Phelps v Hillingdon LBC, the claimant was struggling at school.17 Her difficulties were due to the fact that she was dyslexic, but no one was aware that she was suffering from this disability. An educational psychologist who had been given the job of determining why the claimant was having educational difficulties carelessly failed to pick up on her dyslexia and suggested that the claimant merely needed her confidence boosted. The claimant continued to make no progress with her studies, and she left school with no qualifications. When she was later diagnosed as dyslexic, she sued for compensation for the damage that had been done to her education by the psychologist’s misdiagnosis. The House of Lords had no doubt that the psychologist had owed the claimant a duty of care in assessing her. Lord Nicholls remarked: Take a case where an educational psychologist is employed by an education authority. In the course of his work he assesses a pupil whose lack of progress at school has been causing concern ... He makes a diagnosis and advises the education authority. The diagnosis is hopelessly wrong ... In consequence, the pupil fails to receive the appropriate educational treatment and, as a result, his educational progress is retarded, perhaps irreparably. When carrying out the assessment ... did the psychologist owe a duty of care to the child?

15 16 17

Ibid, at 976. Ibid, at 981. Phelps v Hillingdon LBC [2001] 2 AC 619 (HL) [Phelps].

Tort Law and Human Flourishing

25

I confess I entertain no doubt on how that question should be answered. The educational psychologist was professionally qualified ... The purpose of his assessment was to enable him to give expert advice to the education authority about the child. The authority was to act on that advice ... Throughout, the child was very dependent on the expert’s assessment. The child was in a singularly vulnerable position. The child’s parents will seldom be in a position to know whether the psychologist’s advice was sound or not. This seems to me to be, on its face, an example par excellence of a situation where the law will regard the professional as owing a duty of care to a third party as well as his own employer.18

Phelps can be usefully contrasted with a string of English cases—Harris v Evans,19 Neil Martin Ltd v Commissioners for Her Majesty’s Revenue and Customs,20 and Jain v Trent Strategic HA21—where the courts have held that government bodies cannot be sued in negligence for getting in the way of someone’s carrying on a trade or business.22 So frustrating someone’s educational progress is actionable in negligence, but frustrating someone’s ability to make money will not be. In both cases, the claimant’s ability to pursue his or her purposes is dependent on the defendant public authority getting out of his or her way, but it is only in the educational case that a duty of care will be owed. Why is this? One explanation might be that in cases like Harris, Neil Martin and Jain the courts would have liked to find that a duty of care was owed, but stayed their hand out of a concern not to get in the way of public officials doing their job effectively by exposing them to a threat that doing their job might result in their being sued. Similar concerns could have led the House of Lords to refuse to find a duty of care in Phelps but in that case, the House of Lords did not let such concerns affect them.23 So this explanation just begs the question—Why did the claimant win in Phelps while the claimants in the trio of ‘business sterilisation’ cases mentioned above lost? The most straightforward and obvious answer to this question seems to be that the courts think that being educated properly is a much more important interest than being allowed to make money. The courts will therefore give greater protection to a claimant whose educational prospects have been frustrated by a defendant than they will a claimant whose business plans have been frustrated by a defendant.

18

Ibid, at 665–66. Harris v Evans [1998] 1 WLR 1285 (CA). 20 Neil Martin Ltd v Commissioners for Her Majesty’s Revenue and Customs [2007] EWCA Civ 1041, [2008] Bus LR 663. 21 Jain v Trent Strategic HA [2009] UKHL 4, [2009] 1 AC 853. 22 For discussion of these cases, see R Bagshaw, ‘Negligently Making Business Activities Illegal’ (2009) 17 Torts Law Journal 295 and McBride & Bagshaw, above n 6, at 192–95, 200. 23 Phelps, above n 17, at 667 (Lord Nicholls); at 672 (Lord Clyde). 19

26

Nicholas J McBride

C. Pure Economic Loss Cases The fact that the courts do not regard people’s interests in making money as being that important helps to explain why—in England at any rate—a duty of care not to cause someone pure economic loss will only normally be owed in a case where the defendant has ‘assumed a responsibility’ to the claimant to safeguard the claimant’s economic welfare. However, there are cases where—even in England—a duty of care not to cause someone pure economic loss will be owed even though the defendant has not ‘assumed a responsibility’ to the claimant. The two principal cases where such a duty will be owed are Spring v Guardian Assurance24 and Smith v Eric S Bush.25 In a Spring-type case Hopeful is applying for a job with Company, and Company asks Referee for a reference for Hopeful. Referee stymies Hopeful’s prospects by carelessly supplying a bad reference. In a Smith-type case Hopeful is buying a house at the bottom end of the market, and asks Bank for a loan to help Hopeful buy the house. Bank asks Surveyor to value the house to ensure that it will be sufficient security for the loan. Surveyor reports back to Bank that the house is sound, and Bank informs Hopeful that the house purchase loan has been approved. Hopeful deduces from this that Surveyor has said the house is problem-free and—as Surveyor will have anticipated would happen—Hopeful does not get another survey of the house done because Hopeful has little money to spare. Instead, Hopeful relies on Surveyor to have done a good job of surveying the house. It turns out that Surveyor did not, and the house suffers from fundamental problems which will cost a lot of money to sort out. In the meantime, the house is worth much less than Hopeful paid for it, and much less than Hopeful borrowed from Bank. What makes these cases so special that the courts will find that a duty of care was owed to Hopeful in these cases when the normal rule is that strangers will not owe each other duties of care to protect the other against suffering pure economic loss? It seems obvious that what accounts for the difference is that the losses that have been suffered by Hopeful in these cases are much more serious than those suffered in a run-of-the-mill pure economic loss case. Looking at Spring first, a job is not just a source of income. It is, for the vast majority of people, their primary source of income. It is also a source of important secondary benefits such as a sense of self-esteem, and contact with other people who could become important friends and associates. Robbing someone of their ability to get a job is far more serious than, say, cutting off the power supply to a factory for 14 hours.26 24 25 26

Spring v Guardian Assurance plc [1995] 2 AC 296 (HL) [Spring]. Smith v Eric S Bush [1990] 1 AC 831 (HL) [Smith]. Spartan Steel & Alloys Ltd v Martin [1973] 1 QB 27 (CA).

Tort Law and Human Flourishing

27

Turning to Smith, what makes the loss suffered in that kind of case so serious is that Hopeful’s reliance on Surveyor’s judgment has effectively resulted in Hopeful’s being made destitute. As Surveyor knew when carrying out the survey of the house (which was being sold at a rock bottom price), Hopeful did not have much money to begin with, and is now saddled with a house that needs substantial repairs, as well as a very large debt to Bank. Hopeful will almost certainly go bankrupt and may well be rendered homeless—an outcome that could have easily been avoided had Surveyor taken more care with his or her survey. The seriousness of the impact of Surveyor’s carelessness on Hopeful easily explains why the ‘no duty to protect strangers from pure economic loss’ rule is displaced in this case.

D. Defamation Cases It is commonplace to say that the law of defamation represents a compromise between two competing interests—the interest claimants have in preserving their reputation, and the interest defendants have in being allowed to say what they like free from fear of being sued. So, for example, on the occasions when a defendant’s statement is protected by qualified privilege, the law prioritises the defendant’s freedom of speech over the claimant’s interest in protecting his or her reputation. But if the defendant abuses this privilege and maliciously defames the claimant, then the balance of interests alters: the defendant’s malice will mean that his or her speech is less worthy of protection than the claimant’s reputation. All this may seem so obvious as to be hardly worth saying.27 However, one of the key cases on qualified privilege—Watt v Longsdon—is worth exploring in greater detail.28 In that case, the defendant Longsdon was a director of the Scottish Petroleum Company, and was based in England. He received a letter from an employee in the company’s Morocco office, accusing his boss, Watt, of various things including being drunk, dishonest and of having an affair with his housemaid (described in the letter as ‘an old woman, stone deaf, almost blind, and with dyed hair’!). Longsdon showed the letter to the chairman of the company. He also showed the letter to Watt’s wife, who was living in England, and was a friend of Longsdon’s. The accusations in the letter were untrue, and Watt sued Longsdon for defamation. The Court of Appeal held that Longsdon did no wrong—and could not be held liable for defamation—in showing the letter to the chairman of the 27 See, however, A Ripstein, ‘Your Own Good Name: Understanding Defamation’ to be published in A Ripstein, Private Wrongs (Cambridge, Mass, Harvard University Press, 2015) (forthcoming) which challenges this orthodox view of the law of defamation. 28 Watt v Longsdon [1930] 1 KB 130 (CA).

28

Nicholas J McBride

Scottish Petroleum Company. That publication was protected by qualified privilege. But the Court went on to hold that the publication to the wife was not protected by qualified privilege. Greer LJ argued that: no right minded man in the position of the defendant, a friend of the claimant and of his wife, would have thought it right to communicate the horrible accusations contained in Mr. Browne’s letter to the claimant’s wife. The information came to Mr. Browne from a very doubtful source, and in my judgment no reasonably right-minded person could think it his duty, without obtaining some corroboration of the story ... to pass on these outrageous charges ... to the claimant’s wife.29

The lack of corroboration did not stop the publication to the chairman of the company being privileged, but it did in relation to the publication to the wife. Why? It seems pretty obvious that the Court thought that marriages are more important than jobs, and that interfering with the claimant’s marriage required much more justification than interfering with his job. As Scrutton LJ observed, ‘as a general rule it is not desirable for any one ... to interfere in the affairs of man and wife’.30

E. Medical Negligence Cases The case of In re A concerned two conjoined twins, Jodie and Mary.31 Mary was the weaker twin, and her entire blood supply came from Jodie’s heart. Jodie’s heart was not strong enough to support the two of them, and it was inevitable that if the twins were not separated, Jodie’s heart would fail and both Jodie and Mary would die. It was also inevitable that if the twins were separated, Mary would die. The Court of Appeal was asked to decide whether it would be lawful for doctors to operate to separate the twins. Ward LJ held that at first sight, the doctors were on the horns of a dilemma, as they owed each of Mary and Jodie a duty to treat them with reasonable skill and care. If they operated, they would not act in Mary’s best interests as she would die as a result of the operation.32 But on the other hand, if they did not operate, they would fail to act in Jodie’s best interests.33 In deciding what the doctors should do in this case, the court had no alternative but to balance Mary’s interests in the operation not being performed against Jodie’s interests in the operation

29 30 31 32 33

Ibid, at 154. Ibid, at 150. In re A [2001] Fam 147 (CA). Ibid, at 190. Ibid, at 200.

Tort Law and Human Flourishing

29

being performed. Ward LJ observed that in conducting this balancing exercise: it is legitimate … to bear in mind the actual quality of life each child enjoys and may be able to enjoy. In summary, the operation will give Jodie the prospects of a normal expectation of relatively normal life. The operation will shorten Mary’s life but she remains doomed for death.34

Given this, the ‘balance [was] heavily in Jodie’s favour’.35 While the facts of In re A may seem relatively unique, they merely present in dramatic form the same kind of dilemma that doctors are constantly faced with in deciding which of their patients should be treated first, or which of them should benefit from a life-saving treatment that can be given to only one patient, such as an organ transplant. In determining what their duty to each of their patients requires them to do—who goes first, and who goes second; who gets the life-saving organ, and who does not36—doctors have to balance the respective interests of their patients and decide whose interests are more pressing.37 A doctor who acted unprofessionally in determining which patient to favour—deciding, for example, to treat the pretty blonde who has a sore throat before attending to the elderly gentleman presenting with severe chest pains—will breach the duty of care he or she owed the disfavoured patient and could be sued by that patient for medical negligence. In determining whether a doctor could be sued in negligence for wrongly favouring one patient over another, the courts would inevitably have to engage in the same kind of balancing exercise between the patients’ interests that the Court of Appeal engaged in in In re A.

F. Cases Involving Suicide and Self-Harm An individual’s interest in not being allowed to needlessly throw away the benefits of being alive accounts for why prison and police authorities have a duty to take reasonable steps to prevent those in their custody from 34

Ibid, at 197. Ibid. 36 In the UK, the rules on who gets an organ transplant are administered by the NHS Blood and Transplant Authority. Under the rules governing kidney transplants (where the kidney has been obtained from a stranger) children have priority over adults. While this can be accounted for as in part based on the difficulties involved in treating children with dialysis, it may also reflect In re A-type reasoning: a child can live longer with a new kidney than an adult would. See www.organdonation.nhs.uk/ukt/about_transplants/organ_allocation/organ_allocation.asp (last accessed January 2013). 37 The possible bases for discriminating between patients when medical time or resources are limited are well discussed in JK Mason and GT Laurie, Mason and McCall Smith’s Law and Medical Ethics, 8th edn (Oxford, Oxford University Press, 2010) 389–400. See also C Newdick, Who Should We Treat? Rights, Rationing, and Resources in the NHS, 2nd edn (Oxford, Oxford University Press, 2005). 35

30

Nicholas J McBride

committing suicide,38 and why doctors will have a duty to take reasonable steps to save the life of someone who has attempted to commit suicide. Where, however, an individual (C) has made it clear that it is his or her settled wish that he or she be allowed to kill themselves, the balance may shift and it may be the case that anyone laying hands on C in an attempt to stop C committing suicide will commit a legal wrong.39 C’s interest in being allowed to kill him- or herself, deliberately and not on a whim or depressive impulse, will trump his or her interest in being allowed to enjoy the benefits of being alive. And it will also trump the interests of C’s loved ones who will foreseeably suffer psychiatric illness in the aftermath of C’s suicide; they will not be allowed to claim that C owed them a duty of care not to kill him- or herself.40 Where C carelessly harms him- or herself, C’s interest in being allowed to determine his or her own fate will not count in the balance against finding that he or she owed a duty of care to members of his or her family who were foreseeably made mentally ill by C’s harming him- or herself. However, it has been held that in such a case, the courts will still refuse to find that C owed members of his or her family a duty of care not to harm him- or herself. The courts’ desire to encourage people not to put those close to them through an emotional wringer will be trumped by their desire to avoid encouraging destructive litigation between members of the same family.41 G. Abatement of Nuisance Cases In answering the question of when a defendant was entitled to destroy, or go onto, the claimant’s property in order to abate a nuisance that the claimant had created or allowed to continue, the courts have found themselves having to balance a number of different interests. There is, first of all, a line of cases where a defendant has torn down a house in which the claimant was living, and which was erected either on the defendant’s land or on land over which the defendant had some right or interest. In Perry v Fitzhowe,42 the house was on common land on which 38

Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360 (HL). This seems to be the implication of B v An NHS Hospital Trust [2002] EWHC 429, [2002] 2 All ER 449 (Fam), where it was held that the doctors in that case would commit the tort of trespass to the person if they continued to treat a patient whose medical condition was such that she no longer wanted to live and wanted treatment discontinued. In the case of Kerrie Wooltorton—who was taken into hospital in October 2008 suffering from an overdose but was allowed to die (at the age of 26) by her doctors when it was discovered that she had signed a document before taking the overdose indicating that she wanted no measures taken to save her life—the coroner ruled that Wooltorton’s doctors had not acted unlawfully in respecting her wishes. For further discussion, see S Michalowski, ‘Trial and Error at the End of Life—No Harm Done?’ (2007) 27 OJLS 257. 40 Greatorex v Greatorex [2000] EWHC 223, [2000] 1 WLR 1970 (QB) 1984–85. 41 Ibid, at 1985. 42 Perry v Fitzhowe (1846) 8 QB 757, 115 ER 1057. 39

Tort Law and Human Flourishing

31

the defendant had a right to graze his sheep. It was held that the defendant had acted unlawfully in pulling down the house as the claimant and his family were inside the house at the time it was pulled down. As between the claimant’s interest in his and his family’s health and safety, and the defendant’s interest in being allowed unimpeded access to the common land on which the claimant’s house had been built, the claimant’s interest prevailed and meant that the defendant acted unlawfully in pulling down the house. The balance of interests was different in Burling v Read, where the property that was pulled down was a workshop which the claimant had occupied and which was on the defendant’s land.43 Lord Campbell held that the fact that the property that had been pulled down was a workshop ‘and not the dwelling house of the’ claimant and the fact that the workshop was on the defendant’s land was sufficient to distinguish the case from Perry v Fitzhowe.44 Perry v Fitzhowe was further limited in Davies v Williams, where it was held that a house unlawfully erected on common land could be pulled down even if the claimant occupier of the house was still inside the house so long as the claimant had been given adequate notice that the house was going to be pulled down.45 If the claimant was still in the house at the time the wrecking crew arrived, he only had himself to blame. That was the balance that was struck in the nineteenth century. But today, although there is no authority on the point, it seems beyond doubt that the courts would not be so hard headed in allowing people to protect their property rights, and would not hold that it was lawful to abate a nuisance by pulling down a property while there were still people inside the property. The priority given in Perry v Fitzhowe to protecting lives over protecting property rights would apply across the board. So far as nuisances coming from the claimant’s land are concerned, a defendant is not allowed to go onto the claimant’s land to abate the nuisance unless it is an emergency where the ‘security of lives or property’ are at stake.46 If there is no emergency, then ‘persons should not take the law into their own hands, but … appeal to a Court of Justice’.47 Where a nuisance coming from the claimant’s land can be abated without the defendant leaving their own land—as is the case where branches from a tree on the claimant’s land overhang the defendant’s land—then the courts have ruled that the defendant can take steps to abate the nuisance without giving the claimant any notice that the defendant is going to do this. But mindful of the aggravation between neighbours that can be caused by such unilateral

43 44 45 46 47

Burling v Read (1850) 11 QB 904, 116 ER 711. Ibid, at 713. Davies v Williams (1851) 16 QB 546, 117 ER 988. Earl of Lonsdale v Nelson (1823) 2 B & C 302, 107 ER 396 (KB) 400 (Best J). Ibid, at 312.

32

Nicholas J McBride

action, the concession was made grudgingly. In the leading case of Lemmon v Webb,48 Lord Herschell LC warned that: It ... probably is ... a very unneighbourly act to cut down the branches of overhanging trees unless they are really doing some substantial harm … and it certainly might call for the intervention of the Legislature if it became at all a common practice for neighbours to exercise what may be their legal rights in thus cutting off what would frequently be a considerable portion of the trees which grow on the other side of their boundary.49

In other words, Lord Herschell thought that, for the time being, the interests of landowners in saving themselves money by taking direct action to protect their property rights rather than going to court should be allowed to prevail over their neighbours’ interests in avoiding the aggravation and distress that such direct action might cause. But were too many landowners to take advantage of this liberty, the balance of interests might shift in favour of placing limits on landowners’ abilities to resort to self-help to protect their property rights.

H. Cases Involving the Detention of a Person or Chattel Even if a restaurant posts a prominent notice at its entrance, warning customers that they will not be allowed to leave the restaurant until they pay the bill, it will still be held liable for false imprisonment if it implements its threat.50 As Williams observes: ‘A creditor cannot imprison his debtor in the absence of statutory authority. It makes no difference that the debtor is on the creditor’s premises … and the creditor is merely barring the way out’.51 But until recently, it was also the case that if the restaurant posted a notice in its car park saying that non-customers parking their cars in the car park are liable to have their cars clamped and will have to pay £100 to have the clamp released, then the restaurant would do nothing wrong in implementing this threat—at least in relation to a non-customer who has seen the notice and chosen to take the risk of having his or her car clamped.52 What is the difference between 48

Lemmon v Webb [1895] 1 AC 1 (HL). Ibid, at 4. 50 Sunbolf v Alford (1838) 3 M & W 248, 150 ER 1135 (Exch). See, in particular, the judgment of Lord Abinger CB, excoriating the argument that an innkeeper is allowed to detain a non-paying customer as ‘monstrous’ and ‘utterly bad’ (at 1138). 51 G Williams, ‘Two Cases on False Imprisonment’ in RHC Holland and JFM Schwarzenberger (eds), Law, Justice and Equity: Essays in Tribute to GW Keeton (London, Pitman, 1967) 47. For a very detailed discussion of the history behind the one case that might be thought to go the other way on this point—Robinson v Balmain New Ferry Co Ltd [1910] AC 295 (PC)—see M Lunney, ‘False Imprisonment, Fare Dodging and Federation—Mr Robertson’s Evening Out’ (2009) 31 Sydney Law Review 537. 52 Arthur v Anker [1997] QB 564 (CA); Vine v Waltham Forest LBC [2000] EWCA Civ 106, [2000] 1 WLR 2383. See now s 54 of the Protection of Freedoms Act 2012 which makes it an offence to immobilise a vehicle without legal authority, and specifies that ‘The express or 49

Tort Law and Human Flourishing

33

these two situations that explains why there is liability in the first, but (until recently) there was no liability in the second? The answer seems obvious: people are more important than things, and you need more justification for locking people up than you do for locking up people’s property. I. Cases Involving Work that an Employee is Unable to Do Safely Suppose Employee who works for Employer developed a medical condition that meant that Employee was unable to do his or her work safely. In such a case, the courts have had to consider whether the duty of care that Employer owed Employee to see that Employee was reasonably safe in working for Employer meant that Employer was duty bound to dismiss Employee. Resolving that question has involved the courts in balancing three interests: (1) Employee’s interest in not being physically injured; (2) Employee’s interest in keeping his or her job; and (3) Employee’s interest in being allowed to determine for him- or herself what risks he or she will run in his or her life. At least where the risk of Employee being injured is very small, it seems that there will be no duty on Employer to dismiss Employee: it will be up to Employee to decide for him- or herself whether he or she wants to run the risk of being injured. In Withers v Perry Co Ltd, the claimant developed dermatitis as a result of being exposed to grease while working in the defendants’ factory.53 The defendants had the claimant work in another part of the factory where the risk of her coming into contact with any oil that might aggravate her dermatitis was minimised. However, the work she did resulted in her from time to time coming into contact with oily fluids, and each time her dermatitis flared up as a result. She sued the defendants for compensation for these further attacks of dermatitis, arguing that the defendants had owed her a duty to dismiss her when it became clear that she could not work for them without coming into contact with grease. The Court of Appeal dismissed the claim. Devlin LJ held that: It cannot be said that an employer is bound to dismiss an employee rather than allow her to run a small risk. The employee is free to decide for herself what risks she will run … if the common law were to be otherwise it would be oppressive to the employee, by limiting his ability to find work, rather than beneficial to him.54

Where the risk of injury is greater, then the balance of interests shifts and the need to protect Employee from that risk of injury may start to outweigh

implied consent … of a person otherwise entitled to remove the vehicle to the immobilisation … is not lawful authority for the purposes of [the Act]’. 53 54

Withers v Perry Chain Co Ltd [1961] 1 WLR 1314 (CA). Ibid, at 1320.

34

Nicholas J McBride

Employee’s interest in continuing in employment, and in being allowed to decide for him- or herself what risks he or she will run. This was the case in Coxall v Goodyear GB Ltd, where the claimant’s job involved him in spray painting tyres in the defendant’s factory.55 It was discovered that the claimant had a predisposition to asthma that made it very likely that his kind of work would result in his becoming ill. Inevitably, he did become ill—complaining of severe headaches and giddiness—but nothing was done to stop him spray painting cars. Eventually, the claimant collapsed, and sued the defendant in negligence. It was held that the risk of injury to the claimant had been so serious that the defendant had owed the claimant a duty to stop him working for them as a spray painter.

J. Inducing a Breach of Contract Cases Inducing a breach of contract is not always a tort. If a defendant can raise a defence of justification then the defendant’s inducing a breach of the claimant’s contract with someone else will not amount to a tort. One of the leading cases on justification is Brimelow v Casson.56 In that case, an association representing the interests of chorus girls induced various theatre proprietors to break their contracts with a promoter who was employing chorus girls on such low wages that they were forced to turn to prostitution to survive. Russell J found that the association’s conduct was justified. The promoter’s interest in having his contracts respected counted for very little when weighed against the association’s interest in bringing to an end the practice of paying ‘less than a living wage to chorus girls [which] frequently drives them to supplement their insufficient earnings by indulging in misconduct for the purpose of gain, thus ruining themselves in morals and bringing discredit on the theatrical calling’.57 III. TORT LAW’S ACCOUNT OF HUMAN FLOURISHING

A. Tort Law’s Table of Values Looking at cases such as the ones above allows us to construct a rough table of values that seems to underlie the balance tort law—at least in England and Wales58—strikes between the interests of claimants and defendants in 55

Coxall v Goodyear GB Ltd [2002] EWCA Civ 1010, [2003] 1 WLR 536. Brimelow v Casson [1924] 1 Ch 302. 57 Ibid, at 312. 58 It is entirely conceivable that other common law jurisdictions might employ a different table of values (for reasons explored below, text accompanying n 85), which is why my examples of balancing in tort are drawn exclusively from UK case law. 56

Tort Law and Human Flourishing

35

determining what basic rights claimants have against defendants. Such a table is set out below. Very important

Important

Not very important

Not at all important

Life

Liberty

Making money

Health

Property

Not being upset

Malicious activities

Good education

Having a job

Intimate relationships (friendship, marriage) with other people

Not being made destitute

Degrading activities (eg prostitution)

Playful activities (eg musicmaking, cricket)

Where Claimant’s interest in Defendant’s not doing x counts as ‘very important’ or ‘important’ and Defendant’s interest in being allowed to do x counts as ‘not very important’ or ‘not at all important’ the courts are likely to impose a duty not to do x on Defendant for Claimant’s benefit so long as everything else is equal.59 In the same way, where Claimant’s interest in Defendant’s not doing x counts as ‘not very important’ or ‘not at all important’ and Defendant’s interest in being allowed to do x counts as ‘important’ or ‘very important’ the courts are likely to refuse to find that Claimant has a right against Defendant that Defendant not do x. So, for example, in Miller v Jackson,60 Lord Denning MR cast the dispute in that case as being one between the defendant’s interest in being allowed to continue with its playful activity (important) and the claimants’ interest in not being upset (not very important), and therefore found for the defendant. At no point did Lord Denning note that the defendant’s activities exposed the claimants to a risk of physical injury.

59 The qualification ‘so long as everything else is equal’ is very important. Institutional and policy concerns may lead the courts to deny that Defendant owed Claimant a duty to do x even though Claimant’s interest in Defendant’s being made to do x outweighed Defendant’s interest in being left free to do x. So, for example, we would expect—on the basis of the above table of values—that English law would routinely impose duties of easy rescue on defendants for the benefit of claimants in mortal danger: claimants’ interests in being kept alive will count for more than defendants’ interests in being left free to go about their business. However, institutional concerns about not making the law uncertain, and policy concerns about the law’s not deterring people from attempting to rescue others, or putting themselves in a position to rescue others, by threatening them with being held liable if they are not successful in rescuing people in danger kick in to discourage the law from imposing duties of easy rescue on defendants for the benefit of claimants. For further discussion, see McBride & Bagshaw, above n 6, at 213–17. 60 Above n 13, and accompanying text.

36

Nicholas J McBride

Instead, he said that the claimants found their situation ‘intolerable’ and that the wife was ‘so upset about [the prospect of cricket balls landing in the garden during weekend matches] that they always go out at weekends’61 and that she was ‘a very sensitive lady who has worked herself up into ... a state’.62 In contrast, Geoffrey Lane LJ saw this as a dispute between the defendant’s interest in being allowed to continue with its playful activity (important) and the claimants’ interest in avoiding damage to their property (also important) and damage to their bodies (very important), emphasising that there ‘is no way in which damage to the claimants’ property can reasonably be prevented except by ceasing to play cricket on this ground’63 and that there ‘is here a real risk of serious injury’.64 Given this way of looking at the case, his decision in favour of the claimants was unsurprising.

B. Liberty as a Value There is one item in the above table of values—liberty—that has not really been mentioned so far. People’s interests in liberty—being able to choose for themselves how to live their lives—can figure on either the claimant’s side, or the defendant’s side, or both, in determining what rights a claimant has against a defendant. On the defendant’s side, liberty is always an interest that is in play. Finding that the defendant owes the claimant a legal duty to do x will deprive the defendant of the freedom to choose not to do x. Depriving the defendant of that freedom will always require some justification precisely because the defendant has an interest in being free to choose how to live his or her life.65 On the claimant’s side, the importance of the claimant’s interest in being able to choose for him- or herself how to live his or her life can give rise to the claimant having a right against the defendant that the defendant not deprive him or her of that freedom.66 A whole host of rights recognised under the law of tort can be seen as operating to protect this interest, most notably rights arising under the law of trespass. Even the law of negligence—which seems mainly concerned to protect material interests such as life, health, property

61

Ibid, at 976. Ibid, at 982. 63 Ibid, at 985. 64 Ibid, at 986. 65 See HLA Hart, ‘Are There Any Natural Rights?’ (1955) 64 Philosophical Review 175. 66 This paragraph of my chapter owes a large debt to discussions I have had with Jason Varuhas. 62

Tort Law and Human Flourishing

37

and not being made destitute—works on occasion to impose liberty-protecting duties of care on defendants. One thinks of the duty of care in Chester v Afshar67 requiring a doctor to inform a patient of the significant risks associated with the patient’s upcoming operation so that the patient can decide for herself whether or not to have the operation; or the duty that a doctor owes a patient to carry out a sterilisation operation on him with reasonable skill and care so that he and his sexual partner can be free to have sex without their having to worry about the risk of their conceiving a child.68 In the table of values set out above, liberty is classified as ‘important’. However, that is too simplistic. Some liberties are more important than others. As we have seen, an individual’s freedom to decide whether or not to kill himself is regarded as so important that it must be given priority over an individual’s interest in not being deprived of the benefits of being alive. An individual’s interest in being allowed to go about as she pleases is certainly not as important as that—though, as we have seen, it is important enough that it easily trumps a restaurant’s interest in being allowed to secure payment for its meals by imprisoning non-paying customers.69

C. Finnis’s Account of Human Flourishing The table of values that underlies tort law’s judgments about what rights claimants should have against defendants is not hard to discern or construct. What is much more difficult to understand is what vision of human flourishing underlies that table of values. Why does tort law regard such things as playful activities, or the maintenance of marriages and friendships, or the ability to choose for yourself what life to lead as being important human interests that are worthy of protection, while regarding someone’s desire to have the satisfaction of seeing his or her enemy brought low as being of no importance at all and not something that the law should be interested in protecting? I will argue that in making these kinds of judgments, tort law is giving effect to the vision of human flourishing that we find set out in the works of Finnis—most notably, his masterpiece Natural Law and Natural Rights.70 In this part, I will briefly set out Finnis’s account of human flourishing. In the following part, I will show how this account helps to make sense of

67

Chester v Afshar [2004] UKHL 41, [2005] 1 AC 134. See, eg, Rees v Darlington Memorial Hospital NHS Trust [2003] UKHL 52, [2004] 1 AC 309. 69 See text accompanying n 50 above. 70 JM Finnis, Natural Law and Natural Rights, 2nd edn (Oxford, Oxford University Press, 2011). 68

38

Nicholas J McBride

the table of values that tort law gives effect to in determining what rights claimants have against defendants. For Finnis, human flourishing involves participating in a range of basic goods, which give us reasons for choosing and acting that require no further support or justification.71 As such, the reasons for choosing and acting that the basic goods give us ‘lay down for us the outlines of everything one could reasonably want to do, to have, and to be’.72 The 1980 first edition of Natural Law and Natural Rights set out a comprehensive list of what these basic goods are: (i) life (defined as ‘every aspect of the vitality ... which puts a human being in good shape for selfdetermination’);73 (ii) knowledge; (iii) play; (iv) aesthetic experience; (v) sociability (friendship); (vi) practical reasonableness (defined as ‘the basic good of being able to bring one’s own intelligence to bear effectively ... on the problem of choosing one’s actions and lifestyle and shaping one’s character’);74 (vii) ‘religion’ (defined as thinking ‘reasonably and (where possible) correctly’ about the issue of whether there ‘is a transcendent origin of the universal order-of-things and of human freedom and reason’).75 Over time, this list evolved. The 2011 second edition of Natural Law and Natural Rights endorsed as ‘best’76 the following list of basic goods from an article published in 1997: (1) knowledge (including aesthetic appreciation) of reality; (2) skilful performance, in work and play for its own sake; (3) bodily life and the components of its fullness, viz. health, vigour and safety; (4) friendship or harmony and association between persons in its various forms and strengths; (5) the sexual association of a man and a woman which ... should be acknowledged to be a distinct basic human good, call it marriage; (6) the good of harmony between one’s feelings and one’s judgments (inner integrity), and between one’s judgments and one’s behaviour (authenticity), which we can call practical reasonableness; (7) harmony with the widest reaches and most ultimate source of all reality, including meaning and value.77

A flourishing life will not necessarily participate in all of these goods. But someone’s flourishing will be impaired if they do not acknowledge that all

71 JM Finnis, J Boyle and G Grisez, Nuclear Deterrence, Morality and Realism (Oxford, Clarendon Press, 1987) 278. 72 Finnis, above n 70, at 97. 73 JM Finnis, Natural Law and Natural Rights (Oxford, Clarendon Press, 1980) 86. 74 Ibid, at 88. 75 Ibid, at 89–90. 76 Finnis, above n 70, at 448. 77 JM Finnis, ‘Commensuration and Public Reason’ in JM Finnis, Reason in Action— Collected Essays Volume I (Oxford, Oxford University Press, 2011) 244, n 25 (emphasis in original).

Tort Law and Human Flourishing

39

of the basic goods are goods, and are not open to acting on the reasons for choosing and acting that they supply: Some scholars may have little taste or capacity for friendship, and may feel that life for them would have no savour if they were prevented from pursuing their commitment to knowledge. None the less, it would be unreasonable for them to deny that, objectively, human life ... and friendship are good in themselves. It is one thing to have little capacity and even no ‘taste’ for scholarship, or friendship, or physical heroism, or sanctity; it is quite another thing, and stupid or arbitrary, to think or speak or act as if these were not real forms of good.78

So someone cannot be said to be flourishing as an individual if he chooses to perform ‘any act which of itself does nothing but damage or impede a realization or participation of any one or more of the basic forms of human good’.79 Such a choice involves a denial that the basic good being damaged is a good, and is therefore not something to be damaged but something to be promoted. A crucial claim in Finnis’s account of human flourishing is that all of the basic goods are equally basic in the sense that ‘each can reasonably be focused upon, and each, when focused upon, claims a priority of value. Hence there is no objective priority of value amongst them.’80 As such the basic goods are incommensurable. So, for example, we have no reason to think that enjoying a certain amount of play is better than enjoying a certain amount of knowledge. Given this, Finnis argues, there is no one right way of flourishing as a human being. We cannot say a flourishing life will involve x amount of play, y amount of study, and z amount of friendship. A flourishing life will simply participate in those goods—or, at least, be open to the possibility of participating in those goods—in proportions and intensities to be determined by the individual leading that life. And because each of the basic goods ‘can be participated in, and promoted, in an inexhaustible variety of ways and with an inexhaustible variety of combinations of emphasis, concentration and specialization’,81 there are an inexhaustible variety of ways in which someone can lead a life that can truly be said to be flourishing.

D. Applying Finnis’s Account to Tort Law The table of tort law’s values set out above seems to correlate quite closely with Finnis’s account of human flourishing.

78 79 80 81

Finnis, above n 70, at 105. Ibid, at 118 (emphasis in original). Ibid, at 93. Ibid, at 100.

40

Nicholas J McBride

States and activities that either amount to basic goods (life, health, intimate relationships with other people, playful activities), or are very closely related to someone’s ability to participate in those basic goods (good education, property, having a job, not being made destitute) are regarded by tort law as ‘very important’ or ‘important’. The fact that there is no one right way of flourishing as a human being, which gives rise to the need on the part of each individual to decide for themselves how they will participate in the various basic goods, accounts for the importance that tort law attaches to personal liberty. In contrast, states and activities that do not amount to basic goods and are only remotely connected with one’s ability to participate in those goods (making money, not being upset) are rated as ‘not very important’ by tort law and are not given very much protection. Activities that involve a negation of Finnis’s vision of human flourishing are not rated as important at all by tort law, with the result that they count for nothing when tort law balances the interests of claimants against those of defendants. The two activities that are picked out in the table of values above as being ‘not at all important’ to tort law are: (i) maliciously seeking to harm someone else, and (ii) engaging in a degrading activity such as prostitution. On Finnis’s theory, someone who maliciously sets out to harm someone else impairs his or her own flourishing by choosing to become a practically unreasonable person; that is, someone who is willing to do something which of and in itself does nothing but damage someone else’s ability to participate in the basic human goods.82 According to Finnis, prostitution damages people’s abilities to participate in the basic good of marriage in that choosing to have sex for money, and choosing to pay for sex, ‘sets the wills of the choosers, willy-nilly, against the good of marriage’.83 This is because sex with a marital partner can only be expressive of one’s commitment (fides) to one’s partner if one is only willing to have sex with one’s marital partner. Someone who has sex for money, or pays for sex, and does not subsequently repent of that choice indicates that they would be willing under the right circumstances to have sex with a non-marital partner and cannot as a result invest sex with their marital partner with expressive significance. In so doing, Finnis argues, they seriously damage their ability to participate in the good of marriage: a fundamentally important part of marriage (sexual intercourse) becomes detached from that good.84

82 See, further, JM Finnis, ‘Intention in Tort Law’ in JM Finnis, Intention and Identity— Collected Essays Volume II (Oxford, Oxford University Press, 2011). 83 JM Finnis, ‘Marriage: A Basic and Exigent Good’ in JM Finnis, Human Rights and Common Good—Collected Essays Volume III (Oxford, Oxford University Press, 2011) 323. 84 Ibid, at 320–23; and JM Finnis, ‘Sex and Marriage: Some Myths and Reasons’ in the same volume.

Tort Law and Human Flourishing

41

It might be thought that tort law’s rating some basic goods (or goods that are strongly related to one’s ability to participate in those goods) as ‘very important’ and others as merely ‘important’ is incompatible with Finnis’s vision of human flourishing, according to which the basic human goods are incommensurable and give us no reason to prefer one good over another. But this is not so. Finnis does not deny that a legal system can decide to rate one basic good as being more important than another. But that decision will be based on arational factors, rather than the fact that one basic good is more important than another: in the life of a community, the preliminary commensuration of rationally incommensurable factors is accomplished not by rationally determined judgments but by decisions (themselves presumably based ultimately on commensuration of alternative options by feelings) ... Is it fair to impose on others the risks inherent in driving at more than 10 mph or in planting trees near the roadside? Yes, in our community, since our community has by custom and law decided to treat those risks and harms as not too great. Have we a rational critique of a community which decided to limit road traffic to 10 mph ...? Or to have no trees along the road? ... No, we have no rational critique of such a community. In short, the decision to permit road traffic to proceed faster than 10 mph, or to plant trees along the verge ... was rationally underdetermined.85

So in a case like Miller v Jackson,86 the fact that Geoffrey Lane LJ regarded the claimants’ physical safety as being more important than allowing the defendant cricket club to carry on playing cricket merely reflects the fact that we live in a community that has chosen to treat life and physical safety as being more important than playful activities. It could have been otherwise. We could live in a community that chose to regard sports and games playing as more important than physical safety. Such a community would regard the extra thrill involved in playing a game like ‘Objects in the Dark’87 (as opposed to the game ‘Grab’88) as ‘worth’ the extra physical risk involved. But, as it happens, we do not live in such a community—as is made clear by decisions such as The

85 Finnis, above n 77, at 248 (emphasis in original). See also JM Finnis, Fundamentals of Ethics (Oxford, Clarendon Press, 1983) 90: ‘But don’t we, as individuals and societies, have some stable and intelligent relative evaluations of solutions to practical (moral) problems as “better” and “worse” solutions? Indeed we do. Just as we adopt systems of weights and measures in terms of which we can then carry out commensurations, comparisons and computations of quantities, so we as individuals and societies adopt sets of commitments that bring the basic human values into a relation with each other sufficient to enable us to choose projects and, in some cases, to undertake a cost-benefit analysis to identify better and worse (and even, sometimes, best) solutions’ (emphasis in original). 86 Discussed above, text accompanying n 13. 87 n people run around a room in the dark trying to pick up n-1 blocks of wood. Whoever does not pick up a block of wood is eliminated from the game, and the game starts again with one less player and one less block of wood, and so on and so on until we get to a winner. 88 The same game as ‘Objects in the Dark’ except played in the light, not the dark.

42

Nicholas J McBride

Scout Association v Barnes, which has effectively outlawed the playing of ‘Objects in the Dark’ in England.89

IV. AN ALTERNATIVE ACCOUNT OF HUMAN FLOURISHING

I hope I have said enough by now to make plausible my claims that: (1) tort cases involve the courts making judgments about what human flourishing involves; and (2) the vision of human flourishing that seems to underlie tort law at the moment is very similar to Finnis’s account of what human flourishing involves.90 If (1) is true, (2) should not surprise us. The reason is that Finnis’s account of human flourishing represents the current orthodoxy in Western societies on what human flourishing entails.91 Finnis would go further and argue that it represents the orthodoxy in all societies: All human societies show a concern for the value of human life ... All human societies show a concern for truth ... all societies show a favour for the values of co-operation, of common over individual good, of obligation between individuals, and of justice within groups. All know friendship ... All value play, serious and formalized, or relaxed and recreational ... All display a concern for powers or principles which are to be respected as suprahuman; in one form or another, religion is universal.92

Certainly, when I ask my students if they can improve on Finnis’s account— either by adding to his list of basic goods, or subtracting from it, or by coming up with a completely different account of human flourishing—they find themselves unable to do so. And why should they? People living in Western societies tend to think that the good life consists precisely in living a form of life that allows us to enjoy and participate in Finnis’s list of basic goods.93 89 The Scout Association v Barnes [2010] EWCA Civ 1476. The claimant was part of a troupe of scouts playing ‘Objects in the Dark’ when he was injured, running towards a wall. It was held that it had been negligent to have the scouts play the game in the dark, given the risks of injury involved. 90 It should be noted that I make no further claim (3) that tort law pursues the perfectionist goal of encouraging people to live lives that are flourishing. That is not tort law’s concern. But where claimants’ and defendants’ interests come into conflict, my claim is that tort law resolves that conflict by reference to the relative importance of those interests, and draws on a particular vision of human flourishing to determine how important those interests are. 91 Similar visions of human flourishing can be found in MC Murphy, Natural Law and Practical Rationality (Cambridge, Cambridge University Press, 2001); P Foot, Natural Goodness (Oxford, Clarendon Press, 2001); and R Skidelsky and E Skidelsky, How Much Is Enough? The Love of Money and the Case for the Good Life (London, Allen Lane, 2012). The annual ‘Human Development Reports’ issued by the United Nations Development Programme (especially the first, 1990, report and the 2010 report) identify human flourishing with health, knowledge, and enjoying a decent standard of living. For a survey of the reports’ views on human flourishing, see S Alkire, ‘Human Development: Definitions, Critiques, and Related Concepts’ Oxford Poverty & Human Development Initiative Working Paper No 36 (2010). 92 Finnis, above n 70, at 83–84. 93 Compare the list of the ‘good things of life’ set out by Geoffrey Faber—Fellow of All Souls College, Oxford, and co-founder of the publishing house Faber & Faber—in a letter to

Tort Law and Human Flourishing

43

But I dissent.94 I think Finnis’ account of human flourishing—the account of human flourishing that is now dominant in the West—is radically defective. In this part of my chapter, I will try to set out an alternative vision of human flourishing that transcends the defects that I think afflict Finnis’s account. I will then look at what implications my account of human flourishing has for the balance tort law currently strikes between the interests of claimants and the interests of defendants in determining what rights the first has against the second.

A. Pascal’s Renunciation I will begin my challenge to Finnis’s account of human flourishing— and, by extension, our own, Western, views of what human flourishing entails—with a story about the genius mathematician and philosopher Blaise Pascal. In 1654, when Pascal was only 31, he and Pierre de Fermat (of ‘Fermat’s Last Theorem’ fame) made a huge breakthrough in the theory of probability, by solving ‘the problem of points’.95 Peter Bernstein continues the story: Less than a month later, Pascal underwent some kind of mystical experience. He sewed a description of the event into his coat so that he could wear it

TS Eliot on 15 September 1927 (see V Eliot and J Haffenden (eds), The Letters of TS Eliot, Volume 3 1926–1927 (London, Faber & Faber, 2012) 707): ‘A comfortable house, a car, good food, some sport, domestic interests, pleasant companionship, practical curiosities’. I do not, of course, claim that Finnis’s particular positions on sexual and personal morality (criticised in N Bamforth and DAJ Richards, Patriarchal Religion, Sexuality and Gender: A Critique of New Natural Law (Cambridge, Cambridge University Press, 2008)) are currently widely accepted in Western society: that is clearly not the case. But Finnis’s deeper claims as to what the good life for a human being might look like do seem to be widely accepted. Even my students who express themselves sceptical as to the status of marriage as a basic human good spend most of their spare time searching for ‘the one’ with whom they can hope to share the rest of their life. 94 As did TS Eliot, responding to Geoffrey Faber (see the preceding footnote) on 18 September 1927 that ‘My own ideal Good Life (state) would be to be such a person that I should, by my nature and without special effort, inspire other persons towards heroism and saintliness’ (Eliot and Haffenden, ibid, at 711). While not endorsing Eliot’s vision of human flourishing, this chapter agrees with him that human flourishing is to be identified more with a state of being than a set of life activities. 95 A and B each put £100 into a pot. They then play a number of rounds of a particular game of chance (where the chances of anyone’s winning are 50-50), agreeing that the first one to score 10 wins gets the £200 pot. When the score stands 6-3 in A’s favour, the game has to be called off. How should the pot be divided between A and B? Pascal and Fermat realised that the question turned on what the probability was that A would score 10 wins before B from a position of being 6-3 up. So if the probability that A would score 10 wins before B having been 6-3 up in the game was n%, then A should get n% of the pot and B the rest. They then went on to find ways of calculating what this probability was. (In A and B’s game, the probability that A would have gone on to score 10 wins before B from a position of being 6-3 up was 83%, and so A should get 83% of the pot.)

44

Nicholas J McBride

next to his heart, claiming ‘Renunciation, total and sweet.’ He abandoned mathematics and physics, swore off high living, dropped his old friends, sold all his possessions except for his religious books, and, a short while later, took up residence in the monastery of Port-Royal in Paris ... In July 1660, Pascal took a trip to Clermont-Ferrand, not far from Fermat’s residence in Toulouse. Fermat proposed a meeting ‘to embrace you and talk to you for a few days’, suggesting a location halfway between the two cities ... Pascal wrote back in August: I can scarcely remember that there is such a thing as Geometry [ie mathematics]. I recognise Geometry to be so useless that I can find little difference between a man who is a geometrician and a clever craftsman. Although I call it the best craft in the world it is, after all, nothing else but a craft ... It is quite possible I shall never think of it again.96

Pascal never wrote to Fermat again. If we adopt Finnis’s account of human flourishing, Pascal’s life took a wrong turn from 1654 onwards. (He died eight years later, in 1662.) He deliberately turned his back on Finnis’s basic goods of knowledge, friendship and play, and adopted an ascetic lifestyle that is—on Finnis’s account—incompatible with human flourishing. But obviously Pascal did not feel the same way. For Pascal, human flourishing involved living a life of the type that he adopted from 1654 onwards, and required him to reject the life that he had lived up until then. Pascal took the view that human flourishing does not involve the pursuit and attainment of a variety of goods such as friendship and knowledge and play. In his Pensées—a collection of notes for a book in defence of the Christian religion that his death prevented him writing— Pascal insisted that human flourishing involved attaining a certain mental state: all our dignity consists in thought. It is on thought that we must depend for our recovery, not on space and time, which we could never fill. Let us then strive to think well: that is the basic principle of morality.97 It is not in space that I must seek my human dignity, but in the ordering of my thought.98

Who is right about human flourishing—Pascal or Finnis? I will argue that Pascal was much closer to the truth about what human flourishing involves than Finnis is. In order to reach this conclusion, I want to look at certain mental states which even we in the West will readily acknowledge

96 PL Bernstein, Against the Gods: The Remarkable Story of Risk (New York, John Wiley & Sons, 1996) 68. 97 B Pascal, Pensées (London, Penguin, 1995) (trans AJ Krailsheimer) 66. This pensée is listed as number 200 in the Krailsheimer edition, and as number 347 in the French 1897 edition of the Pensées prepared by Leo Brunschvicg. 98 Ibid, at 29. (Pensée number 113 in Krailsheimer, number 348 in Brunschvicg).

Tort Law and Human Flourishing

45

are incompatible with human flourishing. By thinking about these mental states—and, by extension, what human flourishing does not involve—we can arrive at some interesting conclusions about what human flourishing does involve.

B. Non-Flourishing Mental States There are four mental states that I think we can all agree are incompatible with human flourishing in that anyone who suffers from any one of these mental states for long enough cannot be said to be flourishing as a human being. In order to simplify the discussion, in this part I will be discussing what flourishing entails for a concrete individual named Florence. (i) Misery The first mental state that is incompatible with human flourishing is misery. If Florence is miserable all the time, she cannot be said to be flourishing as a human being. Doubtless, Finnis would accept this, but would argue that Florence will not be miserable if she is, on his terms, flourishing as a human being. Florence will recognise certain activities and relationships as being intrinsically worthwhile and worth pursuing, and as such will enjoy participating in those activities and relationships. The good life, for Finnis, will not only be good but feel good.99 However, there is a problem. Finnis’s account of human flourishing is purely objective in nature: Florence will flourish as an individual the more she participates in the various basic goods that he identifies. But this does not take into account the effect that Florence’s caring about something can have on her flourishing as a person. To understand this point, we need to distinguish between ‘cares’ and ‘mere desires’. If you merely desire that x happens, then you will be happy enough if x happens, but if x does not happen, it is no big deal. It would have been nice for x to happen, but its not happening is nothing to get worked up about. It is different if you care that x happens. If you care that x happens then you identify your well-being with x happening, so that if x does not happen, you experience that as a serious blow to your own well-being.100 99 See, further, D Parfit, Reasons and Persons (Oxford, Oxford University Press, 1984) 499–502; S Darwall, Welfare and Rational Care (Princeton, Princeton University Press, 2002) ch 4; and S Kagan, ‘Well-Being as Enjoying the Good’ (2009) 23 Philosophical Perspectives 253. My thanks to Sandy Steel for the last two references. 100 Just to be clear about the relation between caring and well-being, it is not the case that you care about x happening because x’s happening contributes in some way to your wellbeing. It is the case that if (for whatever reason) you care about x happening, then if x does not happen, you experience a diminution in your own well-being.

46

Nicholas J McBride

As Harry Frankfurt—the world’s leading philosopher on the psychology of caring about something—observes: A person who cares about something is, as it were, invested in it. He identifies himself with what he cares about in the sense that he makes himself vulnerable to losses and susceptible to benefits depending upon whether what he cares about is diminished or enhanced.101

So if Florence cares that x happens, she will not flourish as an individual if x does not happen. Florence will be miserable because, from her point of view, her well-being is dependent on x’s happening. And if Florence is miserable then it is hard to say that she is flourishing as an individual. There are two lessons about the nature of human flourishing that we can draw from all of this. The first lesson is that an important component of human flourishing is that what we care happens, does in fact happen. But the second lesson is that caring about things is dangerous for human flourishing. If we care too much, then we can get in the way of our own flourishing. For example, suppose that Florence cares about the flourishing of every single person on the planet, so that if there is one person alive out there who is not flourishing, Florence takes that personally as she experiences any harms to another’s well-being as a harm to her own well-being. Florence seems doomed to be perpetually miserable, as it is hard to imagine a state of affairs where everyone in the world will be flourishing. So by caring about the flourishing of everyone on the planet, Florence dooms herself to a non-flourishing existence where she is perpetually miserable. In order to flourish, she needs to scale back on what she cares about: she needs to stop caring that everyone on the planet flourish and try to reach a state of mind where she merely desires that everyone on the planet flourish. Let us draw on all this to start building up from scratch a picture of what a flourishing individual will look like. I think we can all accept that a flourishing individual will have certain survival needs (for things like health, food, drink, shelter and warmth) that are being met. But flourishing is obviously about more than just survival. We have just seen that caring about things that do not happen can impair human flourishing, as can caring about too many things. So on that basis we can say that: A flourishing individual (1) will have certain survival needs (for health, food, drink, shelter, warmth) that are being met (2) and what he or she cares happens, does happen (3) but there are only a limited number of things that he or she will care happens.

101 H Frankfurt, The Importance of What We Care About: Philosophical Essays (Cambridge, Cambridge University Press, 1988) 83 (emphasis in original).

Tort Law and Human Flourishing

47

(ii) Anxiety The second mental state which—if experienced for a prolonged time—is (I hope we can all agree) incompatible with human flourishing is anxiety. We cannot say that Florence is flourishing as a human being if she is perpetually worrying. To understand the root causes of anxiety, we again have to draw on the distinction between ‘cares’ and ‘mere desires’. Someone experiences anxiety when: (i) they care that x happens; and (ii) whether or not x happens is not entirely under their control; and (iii) there is a significant chance that x is not going to happen. Mere desires do not give rise to feelings of anxiety: as I have already observed, if you merely desire that x happen, then you will not be particularly bothered if x does not happen. But caring that x happens does give rise to feelings of anxiety when you are not in control over whether x happens or not and there is a significant chance that x is not going to happen.102 So far as I know, Finnis does not address the issue of what the state of anxiety will be of someone who is, by Finnis’s standards, living a flourishing life—but we have good reason to believe that it will be relatively high. This is because Finnis’s vision of what is involved in human flourishing is both complex and contingent. To explain: on Finnis’s view, human flourishing can be compared with the performance of a symphony.103 A symphony cannot be performed without assembling into a coherent whole a variety of different elements that all contribute to the successful performance of the symphony. These elements simply cannot be made to cohere without someone caring about putting them together into a coherent whole. Someone who merely desired that the symphony be performed would not be willing to invest the time and effort required to bring the project of having the symphony performed to a successful conclusion. The complexity of the project would defeat anyone who was not personally invested in making it work. But a person who did care that the symphony be performed could be expected to experience a great deal of anxiety about whether the performance will come off, as there are so many contingencies that are beyond his or her control that might stop the symphony being performed—the concert hall burning down; a number of members of the orchestra, or the conductor, falling ill; a last-minute attack of nerves among the orchestra.

102 To be clear about the relation between anxiety and caring: I do not think that if you get anxious about whether x happens, you are necessarily anxious because your well-being is contingent on whether x happens. When you get anxious over something you care about, you are not necessarily anxious for yourself. The better analysis is that the thing that has led you to attach such importance to x happening that you identify your well-being with x happening, is the same thing that also makes you anxious that x happen. 103 The construction of a house would be another valid analogy.

48

Nicholas J McBride

The same sort of reasoning indicates that if Finnis’s account of what is involved in human flourishing is correct, then it will be impossible to lead a flourishing life without experiencing a great deal of anxiety. Florence could not construct a life that participates in Finnis’s basic goods without caring that she lived such a life—that is, without caring about her own flourishing.104 But given that there are so many things that are outside Florence’s control that could easily get in the way of her living such a life,105 if she did set her cap at living a life that was flourishing by Finnis’ standards, she could be expected to experience a great deal of anxiety as to whether she would achieve her goal. If this is right, then we may end up drawing a pessimistic conclusion: that human flourishing is beyond Florence. She cannot construct for herself a life that is flourishing, according to Finnis’ standards, without experiencing the sort of high levels of anxiety that will prevent her ever enjoying an existence that can be said to be flourishing. But this pessimistic conclusion only follows if Finnis’s vision of what human flourishing involves is correct. If we dispense with that assumption we can construct an anxiety-free vision of what human flourishing involves. We observed above that there are only a limited number of things that a flourishing individual will care about. Let us limit that further by saying that if Florence is to flourish as an individual, the only things that she will care about are things that are under her control. Whatever is beyond her control, she will merely have desires about. This stipulation avoids our pessimistic conclusion: Florence is not someone who is doomed to suffer levels of anxiety that would stop her flourishing. However, our stipulation has two implications. The first is that a condition of Florence’s flourishing as an individual is that she will not care about her survival needs being met. (Though she will still merely desire that her survival needs be met.) Whether or not her survival needs are met is not something that is ultimately under her control, and so caring about whether those needs are met is liable to expose her to

104 Finnis, above n 70, argues that ‘In order to be sufficiently open to all the basic forms of good in all the changing circumstances of a lifetime … one must have a certain detachment from all the specific and limited projects which one undertakes. There is no good reason to take up an attitude to any of one’s particular objectives, such that if one’s project failed and one’s objective eluded one, one would consider one’s life drained of meaning’ (at 110). That may be so, but it seems likely that one would have to care deeply about achieving the overarching objective of constructing a ‘rational plan of life’ which provides one with a ‘harmonious set of purposes and orientations’ (at 103, emphasis in original) that participate in Finnis’s basic goods if that objective is ever to be attained. 105 For further exploration of this point, and philosophical attempts to redefine human flourishing in a way that makes its achievement non-contingent on outside factors, see M Nussbaum, The Fragility of Goodness (New York, Cambridge University Press, 1986) especially chs 11–12.

Tort Law and Human Flourishing

49

flourishing-impeding levels of anxiety. The second is that as the only thing that is really under any individual’s control is their mental state, the only thing Florence can care about—if she is to flourish as an individual and not experience undue anxiety—is cultivating and maintaining a certain kind of mental state. Both of these implications are hard to accept. It is hard to accept that if we want to flourish as individuals, we must stop caring about an important aspect of our flourishing—whether our survival needs are met. And it is hard to accept that what we often think of as one of our best aspects—the intense care we feel for our families and friends—may be incompatible with our flourishing. But a failure to accept either of these implications opens the door to the pessimistic conclusion that flourishing is beyond us and is not something to which we can aspire. It may be that I will lose some readers to pessimism at this point. But for those who are willing to press on and explore the notion of human flourishing further, here is what we have learned so far about what human flourishing involves from what we know it does not involve: A flourishing individual (1) will have certain survival needs (for health, food, drink, shelter, warmth) that are being met (2) but he or she will not be overly concerned whether those survival needs are met. (3) Instead, the only thing he or she will care about is cultivating and maintaining a certain mental state, and he or she will be successful in cultivating and maintaining that mental state. (iii) Hate We can make short work of this mental state. I think all reasonable people would agree that hating even just one other individual is incompatible with human flourishing. Certainly, Finnis would agree: Since each of the basic goods is as good in the lives of others as in the life of the person deliberating, [practical reasonableness directs us towards] integral human fulfilment, that is, the flourishing of all human purposes and groups, considered not as an end ... but rather as a kind of ideal of reason against which plans of action can be measured. Thus the requirements of practical reason ... are to be understood as specifications of the most general ... moral principle: all one’s choices and other kinds of willing should be open to integral human fulfilment.106

106

Finnis, above n 70, at 419–20 (emphasis in original).

50

Nicholas J McBride

Goodwill to all men is an essential component of human flourishing. It needs to be built into the picture of human flourishing that we are gradually building up: A flourishing individual (1) will have certain survival needs (for health, food, drink, shelter, warmth) that are being met (2) but he or she will not be overly concerned whether those survival needs are met. (3) Instead, the only thing he or she will care about is cultivating and maintaining a certain mental state, and he or she will be successful in cultivating and maintaining that mental state. (4) Having that mental state will involve, or result in, him or her having a desire that all other people flourish. (iv) Boredom The final mental state that I think we can all agree is incompatible with human flourishing (if prolonged for long enough) is boredom. The view that boredom is incompatible with human flourishing is most closely associated with Marx. As Leopold notes, Marx condemned ‘the repetitiveness of labour, the “monotony” of the modern factory’ and ‘the machine-like (“mechanical”) character of labour’ in industrial societies which results in an ‘absence of judgment and control as the worker is “depressed both intellectually and physically to the level of a machine”’.107 However, the root cause of boredom is not—as Marx thought—the lack of opportunities to express one’s personality or creativity. Even being in The Beatles proved boring after a while. Boredom results from one’s thinking that one’s life lacks any purpose or meaning.108 This is, I think, a big problem for Finnis’s vision of human flourishing. It is not hard to imagine that even if Florence succeeded in constructing a life for herself that participated in Finnis’s basic human goods, she would eventually become bored with her life. Is this it? she might wonder. A few years of enjoying these goods, and that is it? What is the point of that? As Finnis himself observes, Florence’s participation in his basic goods can only ever be temporary: Our health fails, our stock of knowledge fades from recall, our making and appreciation of play and art falters and finishes, our friendships are ended by distance,

107

D Leopold, The Young Karl Marx (Cambridge, Cambridge University Press, 2007) 232. There are, of course, more short-term and milder forms of boredom that are experienced when someone has no idea what to do with themselves for the next hour, or the rest of the day, or when someone loses interest in what they are doing but for one reason or another is required to carry on what they are doing. But it is still a sense of purposelessness or meaninglessness that triggers the feeling of boredom in all cases. My thanks to Roderick Bagshaw for pointing this out. 108

Tort Law and Human Flourishing

51

time, death; and death appears to end our opportunities for authenticity, integrity, practical reasonableness, if despair or decay have not already done so.109

Given this, Florence is bound to start to wonder what is the point of pursuing all these goods. What difference does it make, she might think, if I enjoy a flourishing life or a non-flourishing life if it makes no difference in the end? And with that wondering comes in dissatisfaction, restlessness and boredom. Finnis attempts to escape this problem by suggesting that Florence’s participation in his basic goods will not prove to be temporary at all, all appearances to the contrary notwithstanding, and that her participation in those goods here and now is an essential preliminary to her enjoying those goods in a more intense form after she has shuffled off this mortal coil.110 But this is unconvincing. As CS Lewis observed, in talking about those who believe that we will enjoy in an afterlife the same sort of existence that we had here on Earth (but better): ‘it rings false. We know it couldn’t be like that. Reality never repeats. The exact same thing is never taken away and given back.’111 Our account of human flourishing must take account of the fact that flourishing is not boring. It can do so if we stipulate that: A flourishing individual (1) will have certain survival needs (for health, food, drink, shelter, warmth) that are being met (2) but he or she will not be overly concerned whether those survival needs are met. (3) Instead, the only thing he or she will care about is cultivating and maintaining a certain mental state, and he or she will be successful in cultivating and maintaining that mental state. (4) Having that mental state will involve, or result in, him or her having a desire that all other people flourish (5) and he or she will find some purpose or meaning for his or her life in cultivating and maintaining that mental state. C. Elaborations of the Basic Picture So far we have been building up a picture of what human flourishing must involve given what we know it cannot involve. Enough has been said by now to give us a good idea of what sort of things human flourishing will involve, and to allow us to think that whatever human flourishing involves,

109

Finnis, above n 70, at 372. See JM Finnis, ‘Action’s Most Ultimate End’ in JM Finnis, Reason in Action—Collected Essays Volume I (Oxford, Oxford University Press, 2011) 159, 171. 111 CS Lewis, A Grief Observed (London, Faber & Faber, 1961) 23. Cf 1 Corinthians 2:9, quoting Isaiah 64:4. 110

52

Nicholas J McBride

it is something very different from the account of human flourishing offered to us by thinkers like Finnis.112 Interestingly, the general picture of human flourishing that we have built up—simply on the basis of what we know human flourishing cannot involve—is consistent with the principal competitors to the current Western accounts of what human flourishing involves. For example, according to the Christian account of human flourishing Blaise Pascal adopted from 1654 onwards, the mental state cultivated and maintained by a flourishing individual involves their choosing to seek to be united with God in their dispositions and actions. This choice will give rise to a large number of desires, one of the most important of which will be a desire that all human beings flourish. And a flourishing individual will think that the purpose or meaning of their life on Earth is to give them the chance to choose to seek unity with God, with the consequence that they will be united with God after they die. The Buddhist account of what human flourishing entails is completely different from the Christian account but still fits the general picture of human flourishing that has been developed here. On the Buddhist account, an individual can only be said to be flourishing when they achieved an enlightened mental state—normally referred to as nirvana—which frees them from the suffering that normal existence here on Earth entails. Achieving this enlightened state requires one to develop, among other things, an attitude of loving kindness (mettã) towards all living beings, which attitude will lead one to desire that all human beings flourish. The goal of achieving nirvana gives purpose or meaning to people’s lives as it helps them break out of samsãra—the vicious cycle whereby suffering in one’s life here on Earth condemns one to be reborn after death into another life on Earth that involves more suffering. For the purposes of this chapter, we do not have to adjudicate between these (and any other) rival elaborations of our basic account of what human flourishing entails. In the next part of my chapter, I will merely spell out the implications that our unelaborated account of human flourishing will have for tort law. A tort law that was based on our alternative account of human

112 Finnis attacks the sort of picture of human flourishing being developed here on the basis that it suggests that there is only one way for a human being to flourish. Finnis, above n 70, condemns all such positions, arguing that ‘Only an inhumane fanatic thinks that man is made to flourish in only one way or for only one purpose. If a religious person says that man is made simply for the glory of God ... we must reply by asking whether the glory of God may not be manifested in any of the many aspects of human flourishing ... and be expressed in ... any of the inexhaustibly many life-plans which conform to the requirements [of practical reasonableness]’ (at 113, emphasis in original). It is not clear how Finnis thinks he is proving his point here. If he is right, and the ‘glory of God’ may be manifested in any of the many aspects of human flourishing, then one can think that the only way to flourish is to manifest the ‘glory of God’ in one’s life without falling into the inhumane fanatic’s trap of thinking that there is only one concrete form of life that can count as flourishing.

Tort Law and Human Flourishing

53

flourishing (henceforth, the ‘Alternative Account’) would look somewhat different from the tort law we have today.

V. IMPLICATIONS FOR TORT LAW

According to the Alternative Account, what really matters for an individual at a physical level is that their survival needs are catered for; and what really matters for an individual at a mental level is that they enjoy a particular (and as yet undefined) mental state that (whatever else it involves) will include freedom from anxiety and hatred for others. Adopting the Alternative Account of human flourishing would have two major implications for the way tort law balances the interests of claimants and defendants.

A. First Implication: Pure Economic Loss A tort law that was based on the Alternative Account could still be expected to place a great deal of importance on life, health, education, personal liberty, property, having a job and not being made destitute. Protecting all those values is intimately linked to the job of helping ensure that people’s survival needs are met. However, one would expect a tort law that was based on the Alternative Account to downgrade ‘making money’ from being a ‘not very important’ interest (as it is at the moment) to one that is ‘not at all important’. So, for example, it is hard to imagine a tort law that is based on the Alternative Account going out of its way to protect the claimants in White v Jones from the risk that they would fail to receive an inheritance as a result of the defendant solicitors’ negligence.113 Not receiving the inheritance would not leave the claimants worse off in any way that the Alternative Account would recognise, and so a tort law based on the Alternative Account would not regard it as at all important that the claimants received that inheritance. Were the defendants’ negligence to result in the claimants not receiving their intended inheritance, they would be encouraged to ‘get over it’, get on with their lives, and focus on what really mattered.

B. Second Implication: Pure Distress In contrast with its treatment of pure economic loss cases, a tort law that was based on the Alternative Account could be expected to upgrade not

113

White v Jones [1995] 2 AC 207 (HL).

54

Nicholas J McBride

being upset from being a ‘not very important’ interest (as tort law regards it at the moment) to being an ‘important’ interest—at least in cases where someone’s being upset involves their falling, or creates the danger of their falling, into non-flourishing states of anxiety or hatred for others. (i) Anxiety Taking anxiety first, a tort law that was based on the Alternative Account might well have allowed a remedy to the claimants in Rothwell v Chemical & Insulating Co Ltd.114 In that case, the claimants had been exposed to excessive quantities of asbestos by the defendants. At the time Rothwell was heard, the claimants had not developed any asbestos-related diseases. However, the pleural membrane around their lungs showed signs of thickening here and there; the areas of thickening are known as ‘pleural plaques’. This thickening indicated that asbestos fibres had penetrated the membrane, with the result that there was a chance that the claimants would develop asbestos-related diseases in future, including mesothelioma, a cancer which is always fatal to whoever develops it. The claimants’ actions against the defendants were dismissed. The only duty of care the defendants had owed the claimants was geared towards protecting them from physical harm, and as the claimants had not yet suffered any physical harm as a result of the defendants’ breach of that duty of care, they had not suffered any actionable harm. The House of Lords held that the development of pleural plaques did not count as physical harm because they had no effect on the claimants’ health. The anxiety felt by the claimants as to whether or not they would develop a disease in future was not something the claimants could sue for. As the Court of Appeal remarked: ‘The law does not recognise a duty to take reasonable care to avoid anxiety’.115 A tort law that was based on the Alternative Account would not be so categorical in dismissing the possibility that a defendant could owe a claimant a duty of care geared towards protecting the claimant from suffering anxiety. It would recognise that people have an important interest in avoiding anxiety. While our interest in avoiding anxiety requires us to work on ourselves and to try not to be so self-preoccupied and to try to take a cooler, more detached view of the various setbacks and defeats that we might be subject to in the future, a tort law that was based on the Alternative Account would recognise that we are naturally timorous creatures and are constantly exposed to temptations to worry about the future. Given this, it might well impose on defendants duties to take care not to expose claimants 114 Rothwell v Chemical & Insulating Co Ltd [2007] UKHL 39, [2008] 1 AC 281, aff’g [2006] EWCA Civ 27, 90 BMLR 88. 115 Ibid, at [63] (CA).

Tort Law and Human Flourishing

55

to such temptations when there is absolutely no need to do so. Rothwell might well have been one such case.116 Imposing a duty on the defendants in Rothwell to take care not to cause the claimants anxiety by exposing them to excessive quantities of asbestos dust would not have been any more burdensome to observe than the physical injury-avoiding duty of care that the defendants already owed the claimants. So the defendants would have had little interest in not being subjected to an anxiety-avoiding duty of care, and the claimants would have had a substantial interest in such a duty of care being imposed on the defendants. The balance of interests as between the claimants and the defendants might therefore have favoured imposing such a duty of care on the defendants—at least in the eyes of a tort law based on the Alternative Account. (ii) Hate Turning to hate, a tort law that was based on the Alternative Account would recognise that we are not just naturally timorous; we are also choleric creatures—quick to anger and slow to forgive. Given this, we have an important interest in others not treating us in ways that would give rise to our experiencing reasonable feelings of anger towards them, as that anger can easily turn into hatred and destroy our capacity to flourish as human beings. This is not an interest that is regarded as particularly important by tort law in its current state, though there are areas of the law of tort which, as a matter of fact, do proscribe and sanction some forms of behaviour that are objectively anger making. Such behaviour includes some forms of private nuisance, defamation of character, fraud, malicious falsehood, harassment and unwarranted discrimination. All of these forms of wrongdoing could be expected to survive under a tort law based on the Alternative Account. However, such a tort law could be expected to go further and seriously consider giving people: (1) rights that others not treat them in a humiliating or degrading manner;117 (2) rights that others not attempt to kill them (as opposed to actually killing them);118 (3) rights that others not maliciously interfere with their earning their livelihood;119 and (4) rights that others not needlessly destroy their marriages.120

116 With the emphasis very heavily on ‘might’. Even if the law thought that in principle imposing an anxiety-avoiding duty of care on someone for another’s benefit would be justified, there would still be formidable practical objections to imposing such a duty of care (Will we make the law too uncertain? Will the law become too oppressive?) that would have to be overcome before the existence of such a duty of care could finally be recognised. 117 Rejected in Wainwright v Home Office [2003] UKHL 53, [2004] 2 AC 406. 118 Rejected in Mbasogo v Logo Ltd [2006] EWCA Civ 1370, [2007] QB 846. 119 Rejected in Allen v Flood [1898] AC 1 (HL). 120 Annulled by Law Reform (Miscellaneous Provisions) Act 1970, s 4.

56

Nicholas J McBride

A tort law that was based on the Alternative Account would also be sensitive to the possibility that its operations can create a sense of grievance that would not have existed in the absence of tort law. After all, there is little more anger making than being denied one’s legal rights; a fact that tort law already acknowledges in awarding aggravated damages to those whose rights have been intentionally and outrageously violated. So tort law can create anger where none would have existed before, or at least not to the same degree. Given this, a tort law based on the Alternative Account would be cautious about developing a law on invasion of privacy, given that one of the effects of developing such a law would be to legitimate and accentuate someone’s feelings of anger when his or her privacy is violated, and possibly create such feelings of anger when had no law on invasion of privacy existed, someone’s reaction to his or her privacy being violated would have been to accept it as something one just has to put up with. In particular, it is doubtful that a tort law based on the Alternative Account would take the momentous step that was taken by the Court of Appeal in Imerman v Tchenguiz, in ruling that merely finding out private information can amount to an actionable wrong.121 It is hard to see how A’s merely finding out private information about B can make B worse off in any way that the Alternative Account would acknowledge. So giving B a right that A not find out such information will not make B better off in any way and has the potential to make B worse off by giving B an incentive to feel angry towards A if A ever violates that right. For similar reasons, a tort law that was based on the Alternative Account would applaud the efforts of the House of Lords (and now the Supreme Court) since the late 1980s to scale back people’s expectations as to what they can expect as of right from public authorities. The exaggerated expectations created by decisions such as Anns v Merton LBC could only have ended up embittering the British people against their government and other public bodies.122 Having said that, a tort law that was based on the Alternative Account would regard as unduly restrictive the current position in English law on public authority liability for omissions, according to which a public body will only be held to owe another a duty to save them from harm in the sort of cases where the law would impose a duty to act on a private person.123 There are some examples of failures to act on the part of public authorities that display such a coolly calculating indifference to human welfare that any reasonable person affected by those omissions

121 Imerman v Tchenguiz [2010] EWCA Civ 908, [2011] 2 WLR 592. For further discussion, see McBride & Bagshaw, above n 6, at 614–16. 122 Anns v Merton LBC [1978] AC 728 (HL). 123 Stovin v Wise [1996] AC 923 (HL); Gorringe v Calderdale MBC [2004] UKHL 15, [2004] 1 WLR 1057.

Tort Law and Human Flourishing

57

could be expected to be driven to rage as a result.124 Such omissions would be readily proscribed by a tort law based on the Alternative Account in order to ensure that people do not end up nursing damaging feelings of resentment about the way they have been treated by those charged with looking after their interests.

VI. CONCLUSION

In the 1960s and 1970s, tort law must have seemed like a very boring area of law. It was viewed as an offshoot of the state’s social security programmes, doling out compensation to those who had been injured as a result of another’s fault. Obviously, there has been a significant rebellion against this view from the 1980s onwards, with the rebels arguing that tort law plays a fundamentally important role in our legal system in that it defines what basic rights we have against each other. However, some of the rebels would like tort law to retain its boring character, arguing that tort law does, and should, give effect to a very limited number of basic rights centred on protecting and preserving just one basic value, such as independence. I disagree. Tort law is not so limited in its interests and ambitions. In balancing the interests of claimants against defendants, it cannot but take a position on fundamental questions about what is a good life for us and what social arrangements will best enable each of us to lead a good life. I would suggest that thinking about these questions ourselves will lead us to conclude that tort law in its current state is not so much in need of reform, as of a reformation.

124 An example: I was recently told of a hospital that has a target for processing patients that come into its Accident & Emergency (A&E) department. The target requires that the patient be admitted to the hospital, transferred, or discharged within four hours of admission. (This is a standard target for all NHS hospitals, though there has been talk of abolishing it.) The hospital has a policy that if a patient has been waiting more than four hours for processing, and there are other patients awaiting processing who have been waiting for less than four hours, priority should be given to the patients who have been waiting for less than four hours. The reason is that there is a chance of meeting the target of processing those patients under four hours, but no chance anymore of meeting the target in relation to the patient who has been waiting more than four hours. The consequence is that at very busy times, nurses and doctors employed in the hospital’s A&E department are positively required not to attend to a patient who has been waiting to be treated for more than four hours, which includes a requirement not to use up time giving the patient food and water.

3 Private and Public: The Mixed Concept of Vindication in Torts and Private Law KIT BARKER*

I. INTRODUCTION

T

HIS CHAPTER EXAMINES the meaning of vindication in private law in general and the law of torts in particular. In doing so, it addresses both what is often characterised as a well-accepted orthodoxy and an associated idea that currently remains heterodox. The orthodoxy is that one aim and effect of the law is to ‘vindicate’ rights, though there is too little understanding, I shall suggest, of what this actually means. The heterodoxy is that there is, or should be, a distinct head of damages in private law called ‘vindicatory damages’, existing alongside other heads such as nominal, contemptuous, compensatory, restitutionary and punitive awards. Modern supporters of vindicatory damages suggest that plaintiffs whose rights have been infringed are (or should be) entitled to a conventional, substantial (ie non-nominal) monetary sum even when they have sustained no objective loss as a result of the infringement, even where the infringer has made no objective gain from it and even where the infringer’s conduct is not so outrageous as to warrant exemplary or punitive damages.1 * I am grateful for critical insights provided by Robert Stevens, Jenny Steele, Robyn Carroll, Rob Burrell, and colleagues. The normal caveat applies. 1 D Pearce and R Halson, ‘Damages for Breach of Contract: Compensation, Restitution and Vindication’ (2008) 28 OJLS 73; N Witzleb and R Carroll, ‘The Role of Vindication in Tort Damages’ (2009) 17 Tort Law Review 16, 28–31; SM Waddams, The Law of Damages (Toronto, The Law Book Company, 2011) [10.50]; R Stevens Torts and Rights (Oxford, Oxford University Press, 2007) 59–91; R Stevens ‘Damages and the Right to Performance: A Golden Victory or Not?’ in R Bronaugh, J Neyers and S Pitel (eds), Exploring Contract Law (Oxford, Hart Publishing, 2009) 171; and R Stevens, ‘Rights and Other Things’ in D Nolan and A Robertson (eds), Rights and Private Law (Oxford, Hart Publishing, 2012) 121–22. Note that, for Stevens, it is not so much that the law should recognise an additional head of ‘vindicatory damages’ as recognise that a variety of existing damages awards (including

60

Kit Barker

The purpose of such an award, it is said, is that it ‘makes good’ the right, that it ‘marks’ the right’s infringement,2 and/or that it fills a remedial gap left by other judicial remedies, thereby ensuring that where there is a wrong, there is always an adequate remedy: ubi ius, ibi remedium.3 In the pages that follow, I suggest that the orthodox view that private law vindicates rights is entirely correct, but that most4 existing accounts of what vindication means are deficient in failing to account for the plurality of public (social) and private (individualistic) aims that together constitute the concept of ‘vindication’. In short, I suggest that vindication in private law is a mixed concept—there is nothing singularly conceptually distinctive about the idea and it is in fact invoked in the implementation of a variety of important private and public aims. This contrasts with an apparently popular view that vindication has but a single meaning and steers us towards just one end. A secondary suggestion, which follows on from the first, is that once the concept of vindication is properly unpacked and understood in this way, it becomes clear that there is no need for any separate head of ‘vindicatory’ damages in private law. This is because the distinct aims which such awards might be thought necessary to implement can all be met through the use or adaptation of existing remedial principles and devices. ‘Vindicatory damages’ are functionally otiose. In making these two points, I have chosen to focus for illustrative purposes upon the law of torts, where the language of vindication is particularly prevalent and where the debate about vindicatory damages has recently troubled the United Kingdom Supreme Court. Nonetheless, it is my contention that these points hold true right across the

nominal, punitive and ‘substitutive’ awards) serve a vindicatory purpose. This suggests a reduction, not an increase, in remedial categories. See, similarly, Lord Scott, ‘Damages’ [2007] Lloyd’s Maritime and Commercial Law Quarterly 465 (proposing but two types of damages awards based on two, distinct purposes—compensation and vindication). 2 Witzleb and Carroll, above n 1, at 18 (‘make good’ the right—this is just one definition of vindication the authors identify, but represents their understanding of the term’s most common judicial use); Walumba Lumba v Secretary of State for the Home Department [2011] UKSC 12, [2012] 1 AC 245, [213], [216]–[217] (Lady Hale), [178], [180] (Lord Hope) [Lumba] (recognise the importance of the right and the wrong done); Pearce and Halson above n 1, at 74 (‘provide a measure of recognition’ for the right). 3 Plenty v Dillon [1991] HCA 5, 98 ALR 353, 356–57 (Gaudron and McHugh JJ) [Plenty]; Mosley v News Group Newspapers [2008] EWHC 1777 (QB) [216] (Eady J) [Mosley]; Pearce and Halson, above n 1, at 73; Witzleb and Carroll, above n 1, at 18. This idea has been used to justify a wide variety of outcomes and awards, not just ‘vindicatory damages’: Ashby v White (1703) 2 Ld Raym 938, (1703) 92 ER 126 (KB) [White] (determination of jurisdiction); Chester v Afshar [2004] UKHL 41, [2005] 1 AC 134, [87] (Lord Hope) [Chester] (substantial compensatory damages for negligence); Ashley v Chief Constable Sussex Police [2008] UKHL 25, [2008] 1 AC 962, [22] (Lord Scott) [Ashley] (refusal to strike out action for battery). It has even been proffered as the very rationale of the Human Rights Act 1998: D Feldman (ed), English Public Law, 2nd edn (Oxford, Oxford University Press, 2009) 825. 4 An important exception is to be found in the work of Dan Priel, who draws attention to the potential public dimensions of ‘vindicating’ rights in ‘A Public Role for the Intentional Torts’ (2011) 22 King’s Law Journal 183.

The Mixed Concept of Vindication 61 field of private law. Indeed, I suggest that they may well hold true in public law as well. The argument proceeds in three stages. The first part provides a brief illustration of both the orthodoxy and the heterodoxy in a recent false imprisonment case considered by the United Kingdom Supreme Court, Walumba Lumba v Secretary of State for the Home Department.5 The purpose here is to introduce the current puzzle over the meaning of vindication in private law and to identify what is at stake in the associated debate about ‘vindicatory damages’. The second part then deconstructs the idea that private law vindicates rights in more detail, exploring the meaning(s) of vindication and identifying some of the possible justifications that might be offered for it. Finally, we return to the remedial question about vindicatory damages: knowing what vindication means and what its purposes might be, we consider whether private law needs either a general or exceptional head of monetary relief bearing that name. Throughout the text, I make constant reference to the terms ‘public’ and ‘private’. The public–private distinction is notoriously difficult, primarily because those who have made it have tended to be referring not to any single distinction, but rather to a variety of different distinctions, and for a variety of different purposes.6 I take the view that such distinctions remain analytically useful, provided one is clear precisely which distinction one is making and why. For the sake of clarity, where I refer below to ‘private’ law in contrast to ‘public’ law, I am referring to the law (both common law and statutory) which formally governs relationships between private individuals,7 as opposed to the relationship between individuals and the state acting in its capacity as guardian of the public good. A claim in false imprisonment is a private law claim on this view, even if it is brought (as such actions customarily are) against an agent of the state. Where I refer to ‘private’ aims or values as opposed to ‘public’ ones, I am alluding instead to purposes or values which are individualistic as opposed to social in orientation. These are not the only two ways in which it is possible to distinguish public from private, or even the ‘best’ ways, simply the ways I have chosen and the ways which serve best to illustrate the point I wish to make about the mixed character of the concept of vindication as it currently exists in private law.

5

Above n 2. For some of these difficulties and distinctions, see W Lucy, ‘Public and Private: Neither Deep, nor Meaningful?’ in K Barker and D Jensen (eds), Private Law: Key Encounters with Public Law (Cambridge, Cambridge University Press, 2013) (forthcoming); P Cane, ‘Public Law and Private Law: A Study of the Analysis and Use of a Legal Concept’ in J Eekelaar and J Bell (eds), Oxford Essays in Jurisprudence (Third Series) (Oxford, Clarendon, 1987); D Oliver, Common Values and the Public/Private Divide (London, Butterworths, 1999). 7 Or corporations. 6

62

Kit Barker II. ORTHODOXY AND HETERODOXY ILLUSTRATED: LUMBA

A topical, recent case illustrating both the orthodox view that tort law vindicates rights and the more contentious heterodoxy about vindicatory damages is the decision in Lumba.8 Mr Lumba and Mr Mighty had been held in detention for two and five years, respectively, pending deportation from the United Kingdom. Their incarceration was supposedly made in accordance with a published policy of the Home Secretary in which there existed a presumption in favour of their release into the community, but they were in fact held under a different, deliberately unpublicised policy according to which there was a ‘near blanket’ presumption against release. Crucially for our purposes, both accepted that if the correct policy had been applied to their cases, they would have been lawfully held in custody in any event. This meant that, although their rights in both public and private law had been infringed, they could prove no loss flowing from the infringement. From the public law point of view, they had been denied procedural justice. From the private law point of view, a majority of the Supreme Court held that their detention had been unlawful and they had been falsely imprisoned.9 All six members of the majority rightly took the view that the absence of provable loss was irrelevant to the plaintiffs’ causes of action in tort. In doing so, they affirmed the orthodox view that one of the key functions of the tort of false imprisonment is to vindicate rights to physical liberty, not simply to compensate a victim’s loss. When it came to the appropriate remedy, however, the majority split. Three judges favoured granting nominal damages of £1,10 while the other three preferred a substantial (non-nominal), conventional award of either £500 or £100011 on the basis that only a sum of this greater order of magnitude would fully ‘vindicate’ the plaintiffs’ rights. Overall, the court therefore narrowly elected to award only £1, but it is fair to say that the arguments danced somewhat on a pinhead. Much debate focused on a body of precedent in both public and private law in which sometimes very sizeable damages awards have apparently been used to ‘vindicate rights’ even in the absence of provable personal harm, or in circumstances in which any such harm has already been made good through a compensatory award.12 The Supreme Court ultimately took the view that 8

Lumba, above n 2. Lords Phillips, Brown and Rodger dissented. 10 Lords Dyson, Collins and Kerr. 11 Lady Hale would have awarded £500; Lords Hope and Walker would have awarded £1000. 12 Public law cases include: Attorney General of St Christopher, Nevis and Anguilla v Reynolds [1980] AC 637 (PC (West Indies)) ($50,000EC = approx £11,000); Attorney General of Trinidad and Tobago v Ramanoop [2005] UKPC 15, [2005] 2 WLR 1324 (PC (Trinidad and Tobago)) [Ramanoop] (assessment remitted); Merson v Cartwright and Attorney General [2005] UKPC 38 (PC (Bahamas)) ($100,000 BAH = approx £64,000); Taunoa v Attorney General [2007] NZSC 70, [2008] 1 NZLR 429, [109], [255], [299] [Taunoa]; Takitota v 9

The Mixed Concept of Vindication 63 these authorities provided an insufficient basis for departing from existing conventions regarding the calculation of damages in a tort case.13 For current purposes, the case is intriguing because it raises a puzzle about the meaning and rationale of the idea of vindication. Given that the majority all agreed, in effect, that the tort of false imprisonment exists to vindicate rights, not simply to compensate losses, why did they differ in their translation of this important normative commitment into remedial relief? Why did one body of opinion regard nominal damages as sufficient to ‘vindicate’ the plaintiffs’ rights to liberty, but the other take the view that a distinct, substantial head of damages needed to be recognised in English law to achieve this aim? The most likely answer, I shall suggest, is that the two different bodies of opinion entertained different conceptions of what vindication means, of the purposes it serves, and of the relative importance of these purposes.

III. UNPACKING THE ORTHODOXY: THE VINDICATION OF RIGHTS IN PRIVATE LAW

A. Vindication as a Private Law Concern When discussing the meaning of vindication, it is customary to begin with an English dictionary.14 I avoid doing so here only because I am immediately

Attorney General [2009] UKPC 11, 26 BHRC 578 (PC (Bahamas)) ($100,000 BAH = approx £64,000); Vancouver (City of) v Ward, 2010 SCC 27, [2010] 2 SCR 28, [2010] 9 WWR 195 [Ward] ($5,000 CAN = approx £3,000—a combined compensatory, vindicatory and deterrent award). For a recent public law case in which a vindicatory award was denied for a mere procedural error without evidence of deliberate wrongdoing, see Dennis Graham v Police Service Commission and the Attorney General of Trinidad & Tobago [2011] UKPC 46 (PC (Trinidad and Tobago)). Private law cases include: Cole v Tucker 6 Tex 266 (1951) [Cole] (retrial ordered); The Blair Iron and Coal Co v Lloyd, 1 Walk 158 (Pa 1874) (venire facias de novo); Kuchenmeister v Home Office [1958] 1 QB 496 (£150); Plenty, above n 3 (damages to be assessed); TCN Channel Nine v Anning [2002] NSWCA 82 [TCN Channel Nine] ($25,000 AUS general damages including vindicatory element); Rees v Darlington Memorial Hospital NHS Trust [2003] UKHL 52, [2004] 1 AC 309 [Rees] (£15,000 ‘conventional’ award); Chester, above n 3 (substantial award for paralysis upheld despite causation difficulties); Mosley, above n 3 (£60,000 vindicatory and compensatory award for defamation); Grosse v Purvis [2003] QDC 151 ($15,000 vindicatory damages for intentional PTSD); Bunnings Group Ltd v Chep Australia Ltd [2011] NSWCA 342 [Bunnings Group] (compensatory damages vindicating personal property rights—leave to parties granted to make submissions on proper sum). Another recent example not available to the court at the time of its decision is Jones v Tsige [2012] ONCA 32, 108 OR (3d) 241 ($10,000 CAN—vindicatory conventional award for invasion of privacy—‘moral’ or ‘symbolic’ damages). 13 The conclusion was affirmed on similar facts in Masimba Kambadzi v Secretary of State for the Home Department [2011] UKSC 23, [2011] 1 WLR 1299, Lady Hale and Lord Hope both deferring to the majority position in Lumba. 14 See eg Witzleb and Carroll, above n 1, at 17: ‘to clear from censure, criticism, suspicion, or doubt … to assert, maintain, make good by means of action ... to defend against encroachment

64

Kit Barker

concerned with the concept’s legal meaning in private law, not with its broader etymology, though it would certainly be surprising if there were no overlap at all between the term’s institutional and broader public use. The legal language of vindication also appears prominently in public law, of course15 but the idea that the law vindicates rights is certainly not exclusively a public law phenomenon. Indeed, it is somewhat ironic that public law actions (such as those under the Human Rights Act 1998 (HRA) and actions for the violation of constitutional bills of rights and judicial review) are quite so regularly paraded as the primary means for vindicating our rights in the modern age, when these devices are relative newcomers on the block by comparison with the ancient writ of trespass. They are also often considerably less robust in the remedies they offer to victims than private law. Damages under the HRA are hence more limited than their common law relative and, as Professor Steele has astutely pointed out, less likely in fact to provide relief to the victims of rights violations in the absence of proof of some material damage.16 Furthermore, as the Supreme Court of South Australia has reminded us only recently, the judicial review of administrative acts that have resulted in a person’s wrongful imprisonment does not vindicate that person’s ‘interest in his or her freedom’ as such, or ‘compensate’ that person for ‘harm caused’.17 Far, then, from being a junior partner in the enterprise of protecting fundamental rights, tort law is often more effective in its vindication of such rights than existing public law

or interference’. According to the Oxford Etymological Dictionary of the English Language (Oxford, Oxford University Press, 2006), the Latin roots are vindicare (to lay claim, maintain by force) and, correspondingly, vindex (a plaintiff), not to be confused with (but actually often historically confused with!) vindicta (revenge). 15 Eg Tunbridge Wells Corp v Baird [1896] AC 434, 437 (Lord Halsbury LC) (vesting of street in urban authority under Public Health Act 1875 ensured ‘complete vindication of the rights of the public’); Aldred (Guy) v Miller [1924] JC 117, [1924] SLT 613 (Clyde LJG) (conviction of offender for street obstruction vindicated public right); R (on the application of K) v Secretary of State for the Home Department [2009] EWCA Civ 219, [2009] UKHRR 973, [91] (Elias LJ) (declaration that independent inquiry ought to have been ordered by Secretary of State vindicated public law rights of complainants); Zhang v Commissioner, Australian Federal Police [2009] FCA 1170, [87] [Zhang] (vindication of rights of judicial review). See also the public law cases on actions for the breach of constitutional provisions, above n 12. 16 J Steele, ‘Damages in Tort and Under the Human Rights Act: Remedial or Functional Separation?’ [2008] CLJ 606, 630–31. Compensatory awards under the Act are not available as of right and are typically less generous, on account of the way courts have interpreted s 8(4). For a powerful critique of this interpretation, see J Varuhas, ‘A Torts-Based Approach to Damages under the Human Rights Act 1998’ (2009) 72 MLR 750. Exemplary awards are also unavailable (formally at least—see D Fairgrieve, ‘The Human Rights Act 1998, Damages and Tort Law’ (2001) PL 695, 704–5) and the limitation period is shorter. The contrast between tort damages and those for violation of ‘constitutional’ provisions in other countries such as New Zealand and Canada is less pronounced. 17 State of South Australia v Lampard-Trevorrow [2010] SASC 56, [287]. It could equally be said, of course, that private law does little to effectively vindicate citizens’ public law rights, such as rights to procedural justice.

The Mixed Concept of Vindication 65 mechanisms, at least in some very important sense of the word ‘vindication’ that public law currently ignores or marginalises. The result is that recent judicial assertions that public law actions are to be contrasted with private law because their aim is to ‘vindicate’ rights,18 must simply be incorrect, unless the word vindication is being used in some specialised, idiosyncratic way that is entirely peculiar to public law and entirely absent from private law. This is one logical possibility, of course, and one of the questions that this chapter is intended to help unpack—but it would be more than a little surprising. Foreshadowing my conclusions somewhat, I shall suggest below that the term vindication in fact bears several different legal meanings, all of which are shared to a greater or lesser degree between public and private law, rather than being starkly divided between them. Public and private law give rise to very different types of right, but nonetheless share some basic understandings of what it means to vindicate them.19 For now, it is enough simply to observe that unrefined statements to the effect that tort is distinct from public law because (only) public law ‘vindicates’ rights are demonstrably false. The observation is important

18 R (on the application of Greenfield) v Secretary of State for the Home Department [2005] UKHL 14, [2005] 1 WLR 673, [19] (Lord Bingham) (actions under the HRA are not akin to tort actions, as the mere finding of a violation of the Convention can be an important vindication of the plaintiff’s rights); Watkins v Secretary of State for the Home Department [2006] UKHL 17, [2006] 2 AC 395, [8]–[9] (Lord Bingham) [Watkins] (contrasting private law with ‘other and more appropriate ways of bringing public officers to account’); Van Colle v Chief Constable of Herts Police [2008] UKHL 50, [2009] 1 AC 225, [138] (Lord Brown) (actions under the HRA); Dunlea v Attorney-General [2000] NZCA 84, [2000] 3 NZLR 136, [66]–[67]; Taunoa, above n 12, at [363]–[368] (Elias CJ, Blanchard, Tipping, McGrath and Henry JJ) (actions under the New Zealand Bill of Rights Act 1990). 19 Some commentators (Stevens is one) take the view that it is only ever private law that vindicates rights. This, on the basis that public law gives rise to enforceable duties without individual, correlative Hohfeldian claim-rights. The point seems strongest in respect of actions for judicial review or habeas corpus, where individual applicants are sometimes said to have ‘standing’ and ‘interest’ in proceedings, not ‘rights’ as such (see, explicitly, AXA General Insurance Ltd v The Lord Advocate (Scotland) [2011] UKSC 46, [2012] 1 AC 868, [62] (Lord Hope), [170] (Lord Reed)). But there are other cases in which public law provides real, individual entitlements amounting to claim-rights—most obviously in actions under the HRA and under constitutional provisions: see eg A v B [2009] EWCA Civ 24, [2009] 3 WLR 717, [23] (Laws LJ); A v Essex County Council [2010] UKSC 33, [2011] AC 280, [210]; Rabone v Pennine Care NHS Trust [2012] UKSC 2, [2012] 2 AC 72, [108] (Lady Hale). There is also some support for the view that actions for judicial review vindicate individual rights (Watkins, above n 18, at [68] (Lord Walker); S&G v Local Authority X [2003] EWHC 551 (Fam), [2003] 2 FLR 42, [39]; Zhang, above n 15, at [87]; Delta Electricity v Blue Mountains Conservation Society Inc [2010] NSWCA 263, [90]). Note that even if public law duties do not always correlate to individual claim rights, some commentators apply Hohfeld to collective rights: M Kramer, N Simmonds and H Steiner, A Debate Over Rights (Oxford, Oxford University Press, 1998) 58–60; J Morss, ‘The Legal Relations of Collectives: Belated Insights from Hohfeld’ (2009) 22 Leiden Journal of International Law 289. There are also other models of collective public rights that do not rely upon Hohfeld: G Rainbolt, The Concept of Rights (New York: Springer, 2006) ch 7; D Newman, Community and Collective Rights: A Theoretical Framework for Rights Held by Groups (Oxford, Hart Publishing, 2011); M Jovanovic, Collective Rights: A Legal Theory (Cambridge, Cambridge University Press, 2012).

66

Kit Barker

because if such statements were truly taken to heart they could have serious implications for the way in which society chooses strategically to deal with rights violations. They could, for example, lead to the over-prioritisation of the role of ‘public’ mechanisms of redress whose vindicatory effects for victims are simply too weak to be morally acceptable. Members of Australia’s Stolen Generations could be forgiven for thinking that this is precisely what has happened in that jurisdiction, where long-overdue political apologies and ex gratia administrative compensation schemes have to date formed society’s front-line, but limited response to grave, historic wrongs.20 Those unfortunate victims who were wrongly separated from their families by the state as children might accept that such mechanisms are an appropriate starting point for the vindication of their rights, but rather fewer, I suspect, would regard themselves as fully vindicated by them. Neither a public apology from the state, nor a sum of ex gratia monetary reparation purports to address the full harm suffered by such individuals, even if it vindicates them in some less tangible way by acknowledging a wrong done to them, or by meeting some of their immediate needs. Indeed, this is precisely the type of case in which victims might be tempted to think that the types of ‘vindication’ currently offered by public law and public administration fail to cut the mustard and that a good old-fashioned private law remedy is needed if their rights are really to be made good. The vindication of rights is therefore a key part of the function of private law, not something peculiar to public law. Nor is it the preserve of only those parts of private law, such as the tort of false imprisonment or trespass to land, that make rights infringements actionable ‘per se’. Indeed, at one level of analysis, the whole of private law is ‘rights based’ for the simple reason that whenever a court imposes legal duties on one party to a bilateral dispute, it necessarily creates rights for another. This is an undeniable Hohfeldian truth which means—as Professor Stevens has recently highlighted in his excellent book Torts and Rights—that one may legitimately view even actions in negligence as being ‘rights based’—rather than ‘loss based’ in an important sense.21 Taking this point seriously, the law of torts is therefore divided between those ‘rights-based’ torts that require proof of damage and those ‘rights-based’ torts that do not; not as between ‘rightsbased’ and ‘loss-based’ torts.

20 See A Durbach, ‘The Cost of a Wounded Society: Reparations and the Illusion of Reconciliation’ (2008) 12 Australian Indigenous Law Review 22. In Tasmania, awards were dubiously limited to a paltry $5,000, which verges on insult. 21 Stevens, above n 1. The types of right that negligence law protects are all qualified rights to the protection of particular interests—physical, mental, proprietary and (on my own view) economic. See S Perry, ‘The Role of Duty of Care in a Rights-Based Theory of Negligence Law’ in A Robertson and Tang Hang Wu (eds), The Goals of Private Law (Oxford, Hart Publishing, 2009). Note that none of this says anything about the foundations of these types of qualified right, which are a blend of both moral and ‘policy-based’ reasoning.

The Mixed Concept of Vindication 67 B. Vindication—a Taxonomy of Meanings The meaning of vindication in private law can only meaningfully be understood by ascertaining what it is that judges think is entailed by the idea in determining private disputes.22 When approached in this way, vindication is, I suggest, no single thing, but a set of events occurring within judicial practice and in positive response to legal rights, though not necessarily in response to their past violation. This conception of vindication is drawn from a wide variety of cases in which the term ‘vindicate’ has been used by courts and effectively characterises as ‘vindicatory’ all judicial remedies in private law, as well as some acts (such as the giving of reasons for judgment, awards of interest on a judgment sum, or awards of litigation costs) that are not normally thought of as ‘remedies’ at all. While this definition is broad and stated at a high level of abstraction, it over-arches four, more distinct and concrete meanings that the idea of vindication bears in particular contexts. All of these meanings are shared to a greater or lesser degree with public law, although there is one particular form of vindication, to which I shall allude further below, that is predominantly a private law phenomenon. Before examining these more specific meanings, three swift observations should be made about what the above, general definition of vindication excludes. First, it clearly discounts from the concept of vindication judicial responses to phenomena other than rights. This gives a slightly false impression of the way in which judges use the term because courts do sometimes claim a role in vindicating not just rights, but a plaintiff’s factual assertions—such as a claim that he or she has told the truth about events.23 To the extent that courts make factual findings of this type and state definitive official histories, they are sometimes understood as acting rather like ad hoc public inquiries or ombudsmen, exposing past events to the light of day and exercising an important forensic role. I exclude such examples from the idea of vindication for now, only because our immediate concern is to capture what it means in private law to vindicate rights. That was the key issue in Lumba. Second, the definition excludes from the idea of vindication judicial responses to moral as opposed to legal rights. This is not to say that it is impossible to think of judges as vindicating moral rights when they judge, but simply that judges themselves tend not to see things in quite this way. They generally prefer a more conservative conception of the judicial role in

22 A range of purely normative theories may, of course, provide a priori definitions of vindication which cast existing judicial conceptions into illuminating critical light, but such definitions, being external to the law, are less useful, in my respectful view, as a starting point for deliberation. 23 See eg W v W; J v Bell [1999] UKPC 2 (New Zealand), [9] (Lord Hoffmann) (actions enable plaintiffs to vindicate their ‘version of the facts before a judge’).

68

Kit Barker

which they act to interpret and affirm existing legal rights, rather than to create new ones in response to the vagaries of moral argument. In providing an explanatory account of vindication’s legal conception, it is important, I believe, to take account of this internal point of view—a good explanation of the idea must reflect judges’ own understandings of what they do. Finally, the definition excludes as examples of vindication positive responses to legal rights that occur outside judicial practice. This is also questionable, since it could quite legitimately be said that defendants’ voluntary apologies, admissions of legal wrongdoing, and pretrial, out-of-court settlements all positively affirm a plaintiff’s legal rights to a greater or lesser degree. The importance of such events should not be underestimated and indeed they are expressly taken into account by judges themselves in deciding how and to what extent to vindicate a plaintiff’s right legally in subsequent proceedings.24 Some writers have gone so far as to suggest that non-institutional forms of vindication (private apologies in particular) ought to be further institutionalised by extending to judges the power to order them.25 The only reason for excluding such forms of vindication from the following account is the weak one that our purpose is to provide an account of the idea’s institutional conception, not its broader social practice. While a defendant’s voluntary apology may indeed vindicate a plaintiff’s legal right in one or more senses of the word we examine below, such vindication is not in itself institutional in design or operation. Until some account of it is taken by judicial practice, it remains a purely personal response. At the highest level of generality, then, a court vindicates private rights when it acts positively to affirm them. It does so, I suggest below, when it (i) prevents their infringement, when it (ii) declares them publicly, when it (iii) enforces them specifically and when it (iv) reverses the effects of their infringement. It is less clear that it does so in any way that is distinct from those just mentioned when it punishes those who violate them, although there is some debate about this issue to which we shall come in due course. Collectively, these four events comprise private law’s conception of vindication. 24 See eg Ronaldo v Telegraph Media Group Ltd [2010] EWHC 2710 (QB); Hays v Hartley [2010] EWHC 1068 (QB). Both cases concerned applications for the strike-out of defamation actions (the latter successful, the former not). Defamation damages awards also routinely take account of prior occurrences ‘vindicating’ a plaintiff’s right, such as apologies made by a defendant. 25 B White, ‘Say You’re Sorry: Court-Ordered Apologies as a Civil Rights Remedy’ (2006) 91 Cornell Law Review 1261; R Carroll, ‘Beyond Compensation: Apology as a Private Law Remedy’ in J Berryman and R Bigwood (eds), The Law of Remedies: New Directions in the Common Law (Toronto, Irwin Law, 2010). Currently, the instances in which courts have the power to order apologies are limited, being confined in the main to cases involving the breach of anti-discrimination statutes. In Australia, there are proposals to introduce this power in relation to actions for serious invasions of privacy: Australian Law Reform Commission, For Your Information: Australian Privacy Law and Practice (Report No 108) (August 2008) [74]–[75]. It also exists in respect of infringements of moral rights under the Copyright Act 1968 (Cth), s 195AZA(1)(d).

The Mixed Concept of Vindication 69 (i) Preventing Infringement The prevention of an infringement is an affirmative judicial response to the existence of legal rights and clearly conceptualised by courts as a species of vindication.26 Examples of such responses include the issue of interlocutory relief to prevent a right’s initial violation (quia temet), as well as judicial measures designed to prevent further violations, such as final, prohibitory injunctions, or orders for the delivery up or destruction of material offensive to the right in question.27 Such preventative responses are not particular to private law. The issue of a restraining order upon a perpetrator of domestic violence is hence a judicial measure made available in public law to prevent serious rights violations ex ante, as are prohibitory injunctions flowing from acts of judicial review. Vindication by prevention remains the exception, not the rule, in both public and private law. The reasons for this are partly practical (the time for prevention will often be past) but are also partly grounded in traditional liberal objections to the idea of prior restraint. (ii) Declaring (‘Marking’) the Right Courts vindicate rights in a second sense when they provide an affirmative, institutional acknowledgement of the right. This can again occur either before or after it has been infringed. A plaintiff can thus obtain a judicial declaration of rights in advance of a person’s interference with them, as, for example, where she seeks a declaration of her interest in property held by a bankrupt defendant in order to protect that property against the competing claims of third-party creditors. Marking the right in such instances is a useful way of simultaneously preventing its infringement in the sense just examined—it stakes a claim. Alternatively and more commonly, the judicial acknowledgement of rights post-dates their infringement, announcing the status of plaintiff and defendant as right-holder and wrongdoer ex post facto and thereby marking both the fact of the right and the fact of the wrong. Of the various senses of the word ‘vindication’ current in the law, it is this type of ‘declaratory’ vindication that is closest to that which the judicial supporters of vindicatory damages in Lumba appear to have had in mind28 and it is therefore central to our later discussion in Part V. It turns out to be more slippery than one might think. The most obvious instantiation of a post-infringement ‘declaration of right’ is a declaratory order, but money judgments can indirectly serve similar ends. The monetary remedy most commonly considered equivalent 26 See eg Jordeson v Sutton, Southcoates and Drypool Gas Co [1899] 2 Ch 217 (CA) 238 (Lindley MR) (nuisance); Imerman v Imerman [2010] EWCA Civ 908, [2011] 2 WLR 592, [143] (Lord Neuberger MR) [Imerman] (breach of confidence). 27 Imerman, above n 26. 28 See further Part V, below.

70

Kit Barker

to a judicial declaration is an award of nominal damages,29 but the institutional recognition of rights and wrongs can also be achieved through compensatory awards (as Lord Scott expressly recognised in Ashley v Chief Constable of Sussex Police),30 through restitutionary or exemplary awards31 and through a court’s provision of reasons for judgment,32 which reasons may advert directly to the nature and importance of the right and the basis for protecting it. In his recent work Torts and Rights, Professor Stevens suggests that there is a further category of damages—‘substitutive damages’—which are different to compensatory awards in so far as they are calculated to reflect the value of the right infringement itself, rather than the factual losses or gains flowing from it.33 The existence of this extra

29 See eg A Burrows, Remedies for Torts and Breach of Contract, 3rd edn (Oxford, Oxford University Press, 2004) 589; Lumba, above n 2, at [236] (Lord Collins) (both declarations and nominal damages can have a ‘vindicatory purpose and effect’). Cf Hyde Park Residence Ltd v Yelland [1999] RPC 665, 670, [2000] 3 WLR 215 (Jacob J) [Hyde Park Residence]. If both remedies indeed have the same purpose, nominal damages should be abolished because they provide a less explicit form of declaration of right than a declaratory order. 30 Above n 3, at [22]. See also Mosley, above n 3, at [216] (Eady J); Lumba, above n 2, at [100] (Lord Dyson), and at [236] (Lord Collins). 31 Cole, above n 12, at 268 (Hemphill Ch J) (punitive awards ‘set a public example’); Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29, [2001] 2 WLR 178, [91] (Lord Hutton) [Kuddus] (exemplary awards ‘mark disapproval’ of the wrong). See, however, section (v) below—the main focus of exemplary awards is on denouncing particularly heinous wrongdoers, not really on declaring the right as such. 32 Purnell v Business Magazine Ltd [2007] EWCA Civ 744, [2008] 1 WLR 1, [27] (Laws LJ) [Purnell]; Lumba, above n 2, at [256] (Lord Kerr). 33 Stevens, above n 1, at 59–91. Cf A Burrows, ‘Damages and Rights’ in Nolan and Robertson, above n 1. Stevens has sometimes been interpreted as saying that substitutive damages (which he alternatively refers to as ‘vindicatory’) represent the pre-infringement value of the right itself, but more recently he has clarified that they instead represent the ‘value of the infringement’ (‘Rights and Other Things’, above n 1, at 127. See also T Cutts, ‘Wrotham Park Damages: Compensatory, Restitutionary or a Substitute for the Value of the Infringement of the Right?’ (2010) Lloyd’s Maritime and Commercial Law Quarterly 215). With respect, the latter position is deeply problematic. An ‘infringement’ has no value in itself—it is simply conduct violative of a right and has no negative index other than by reference to the right violated. In circumstances in which a right is infringed, all that is logically capable of being valued is: (a) the value of the prior right (albeit that this may be hard to assess); (b) the value associated with the plaintiff’s power to insist on the right being respected where this exists in law (what might be regarded as the autonomy, negotiation or ‘insistence’ value of the right); (c) some actual diminution in the right’s value (the difference between the right’s pre- and post-infringement value); (d) some other loss (including losses or devaluations of other rights) suffered by a plaintiff in consequence of the violation; or (e) some gain (including acquisitions of other rights) made by a defendant as a consequence of the violation. Every right that the law is prepared to enforce specifically has an insistence value of type (b) inherent in it and this value can itself be diminished when a person invades the right, even if no other harm is caused. This preliminary analysis leads me tentatively to conclude that, contrary to Stevens’s stated view (n 35 below), damages ‘reflecting the infringement value’ must simply be morally symbolic. The analogous concept of ‘substitutionary’ damages is supported by S Smith, ‘Substitutionary Damages’ in C Rickett (ed), Justifying Private Law Remedies (Oxford, Hart Publishing, 2008). Unlike Stevens, however, Smith rationalises such awards as eliminating or preventing a plaintiff’s losses (at 95). In my view, this is an orthodox compensatory rationale, requiring no new label and not inviting the same degree of controversy.

The Mixed Concept of Vindication 71 species of damages is doubtful to my mind, if only because the category of compensatory damages seems to me to be perfectly analytically capable of capturing diminutions in the value of rights, as well as the consequential factual losses flowing from their violation.34 If, however, substitutive damages do exist as a distinct remedial head, they could also be construed as declaring rights symbolically by placing a monetary value upon them (or, as Stevens would say, upon interferences with them) and then announcing that value institutionally to the world.35 They might have an affirmative, expressive effect. Contemptuous damages are altogether different. They ‘mark’ the infringement of a plaintiff’s right in a technical sense, but simultaneously declare a jury’s disapproval of the plaintiff’s reasons for bringing an action in reliance upon it. This sends a very mixed message, which is far from being positive overall. It affirms the right, while at the same time strongly disaffirming the right-holder for abusing that right. The same mixed message is sent by nominal damages awards whenever a plaintiff is simultaneously denied the costs of his or her action, for in such cases reasonable observers are as likely to think of the plaintiff as having lost the action, as having won it.36 Strictly speaking, this is a mistake, but there is a fine line between being told that one should not have acted upon one’s rights, and being told that one simply does not have them. Where a plaintiff has been left seriously out of pocket by a court’s refusal to award costs, the public perception is likely to be that the denial of costs represents an institutional denial of the right itself. To the extent that contemptuous and nominal damages awards ‘mark’ the existence or infringement of rights at all, they therefore do so in a very muted way. Given that they serve no other, detectable ‘vindicatory’ function, there is a very good case for abolishing them altogether and relying more directly on declaratory orders and judicial reasons to say whatever it is that a court wants to say.37

34 See above n 33; below n 48 and accompanying text. One of Stevens’s objections to this, stated in correspondence, is that it obscures the fact that different quantification principles may apply to consequential and non-consequential types of loss. Personally, I doubt that courts will be as confused about this as Stevens suggests. 35 This is not Stevens’s view. In correspondence, he has indicated that there is nothing that is symbolic about such awards, they merely provide, by way of remedy, the ‘next best thing’ to the original right. 36 Hyde Park Residence, above n 29, at 670 (Jacob J). Costs are no longer guaranteed where awards are nominal: Anglo-Cyprian Trade Agencies v Paphos Wine Industries Ltd [1951] 1 All ER 873. 37 Burrows suggests that contemptuous awards should be maintained precisely because they alone are capable of sending this type of ‘mixed message’ (above n 29, at 590). But there is no reason why a declaration could not be combined with a denial of costs and adverse comment upon a plaintiff’s motives for action in a court’s judgment. Moreover, the distinction between contemptuous and nominal damages is so thin in many instances (the difference between 1p and £1?) that the case for distinguishing between them is hard to maintain.

72

Kit Barker

Vindication in the declaratory sense is not a phenomenon particular to private law and can equally be found in public law, particularly in cases involving the infringement of constitutional rights. Indeed, it is largely upon a series of such cases that the argument for recognising ‘vindicatory damages’ was built in Lumba.38 Its exact justifications are unclear. They could include the possible deterrent effects of institutional acknowledgements of right (though one suspects that the behavioural incentives created by nominal and contemptuous damages and post-infringement judicial declarations are weak at best); the sense of psychological or ‘therapeutic’ relief provided to a plaintiff by authoritative public assurances; the contribution a declaration makes to what has been referred to in a different context as ‘norm projection and reinforcement’ (public education about rights and the construction or reinforcement of social values);39 or, less empirically, public declarations of right might be thought simply to be conceptually required in any Kantian legal system, where freedom is the ultimate value. That is, judicial declarations could be construed as forming part of the essential condition of ‘public right’ that is necessary for private rights to be determinate and for individual freedom to exist in the Kantian sense.40

38 See cases at n 12 above. See also Manga v Attorney-General [2000] 2 NZLR 65, (1999) 5 HRNZ 177 (declaration granted—Bill of Rights Act 1990 (NZ)); James v Attorney General of Trinidad and Tobago [2010] UKPC 23 (Trinidad and Tobago) (declaration granted— violation of section 4 of the Constitution of Trinidad and Tobago). Such declarations are also achieved in the criminal law (for example via jury verdicts and judges’ reasons in sentencing) and administrative law (judicial declarations and other orders flowing from judicial review). Note that in some public law proceedings, it is the public right or authority of the court that is declared, as for example, where a party is punished for contempt of court. 39 The phrase is lifted from M Gallanter and D Luban’s seminal work on punitive damages: ‘Poetic Justice, Punitive Damages and Legal Pluralism’ (1993) 42 American University Law Review 1393, 1429–32. See, similarly, S Smith, ‘Rule-Based Rights and Court-Ordered Rights’ in Nolan and Robertson, above n 1, at 243–44, (court orders sometimes vindicate private rights for public reasons which include ‘informing citizens of their rights’). For a recent case in which public education was cited as a key justification for the ordering of a corrective statement, see Eatock v Bolt [2011] FCA 1103, [2011] 284 ALR 114, [15] (racial discrimination). 40 There are hints of this type of argument in E Weinrib, ‘Two Conceptions of Remedies’ in Rickett, above n 33, at 14–15, 27. On Weinrib’s conception of rights and remedies (itself grounded in Aristotle and Kant), all remedies have a dual function, the first of which is to eliminate the injustice between the parties, but the second of which is to provide ‘public and authoritative confirmation’ of the injustice. Such public confirmation is seen as an essential condition of the existence of a system of rights, without which the plaintiff’s individual right would remain ‘ineffective and insecure’. On the deontological relationship between private rights and the public systems necessary for their actualisation (‘public right’) in Kant’s philosophy, see further E Weinrib, ‘Private Law and Public Right’ (2011) 61 University of Toronto Law Journal 191; A Ripstein, ‘Private Order and Public Justice: Kant and Rawls’ (2006) 92 Virginia Law Review 1391; A Ripstein, Force and Freedom (Cambridge, Mass, Harvard University Press, 2009), especially 23–28, 145–81. As Ripstein explains in the latter work at 23, private rights in the Kantian scheme conceptually require assurance, determinacy and a mechanism of acquisition and co-ordination that overcomes problems of unilaterality. Court pronouncements of right and wrong clearly fit this normative niche.

The Mixed Concept of Vindication 73 These potential justifications clearly span the division between public and private aims. The Kantian justification for judicial declarations makes no empirical claim based on betterment of the public good. It simply asserts that for private rights to constitute a system of rights in which each is his own master and free from the oppressive purposes of others, open court adjudications are necessary from a deontological point of view.41 That appears to be a credible argument even from a more empirical perspective— how, after all, can a system of individual rights work in practice without institutional announcements of right (and wrong)? Whether a declaration of rights affects the behaviour of others or contributes to the construction of ‘right-respecting’ social attitudes is then, strictly speaking, irrelevant. By contrast, most of the other justifications for the institutional ‘marking’ of rights mooted above are public (social) in nature and depend for their validity upon empirical assumptions that are untested. The possible exception is the argument that declarations of right provide a degree of psychological relief for a victim, but while this claim seems intuitively feasible (don’t we all feel the better for being publicly affirmed?), it in fact posits a different type of vindication to the one we are currently considering. The argument is then not that the acknowledgement of a right is important in itself as an institutional public statement, but simply that it helps to reverse the effects of the wrong, where these effects take the form of harm done to a plaintiff’s mental welfare, dignity or autonomy. It is a reparative healing tool, not a public symbol constructed for any social ends.42 We come to this important sense of the term vindication in section (iv) below. (iii) Specifically Enforcing the Right Perhaps the strongest meaning of the term vindication in private law is the specific enforcement of a right. Specific enforcement entails the literal provision to a plaintiff of that in which the right consists. Obvious examples of judicial remedies which do this are orders for the specific performance of a contract, awards of an agreed price and mandatory or prohibitory injunctions. In Civilian systems, there is another, excellent example known as the vindicatio—an action via which a plaintiff asserts (‘vindicates’) his legal title to property as a basis for seeking its return.43 For reasons which

41 As Ripstein points out, legal rules and institutions are not, in the Kantian conception, valuable because they are empirically likely to bring freedom about—they are essential preconditions of that freedom in an avowedly non-instrumental account (Force and Freedom, above n 40, at 9–11). 42 See further R Carroll and N Witzleb, ‘It’s not just about the Money—Enhancing the Vindicatory Effect of Private Law Remedies’ (2011) 37 Monash University Law Review 216 (drawing upon therapeutic jurisprudence). 43 On the Roman origins, see P Birks, ‘The Roman Law Concept of Dominium and the Idea of Absolute Ownership’ (1985) Acta Juridica 1, 5 (‘This thing is mine’).

74

Kit Barker

are mainly historic and influenced by the values of nineteenth-century market liberalism, the specific vindication of rights remains the exception rather than the rule in English contract law.44 Similarly, rather than adopting a continental-style vindicatio, English law and the common law systems based upon it tend to ‘vindicate’ property rights only indirectly by making interference with them wrongful and then undoing the effects of the wrong ex post facto through monetary awards. On one view, much of the law of restitution is engaged in an analogous exercise, restoring in monetary guise the value of property rights which have been lost in circumstances in which they should not have been.45 Vindication by restoration, not by specific enforcement, is thus the predominant approach to rights protection in common law systems. There are several clear examples of specific enforcement in public law, such as the writ of mandamus and the ancient action of habeas corpus, the latter of which specifically enforces a person’s right to have the legality of his or her detention determined by a court of law. The substance of the rights in such cases is different—they are procedural rather than substantive in nature and import a normative relationship between individuals (or groups) and the state—but the idea of ‘specific vindication’ is exactly the same. The normal effect of a valid challenge to the procedure that has been used to make a public law decision is hence that the decision has to be made again, this time using the right procedure. There are, I suggest, few clearer examples of the literal enforcement of rights anywhere in the law. (iv) Reversing the Effects of Infringement The most common way in which private law ‘vindicates’ rights is by ‘setting right’ the effects of their violation as between the right-holder and the 44 It is often argued that damages for breach of contract (or an account of the profits of such a breach) are equivalent to specific performance, in the sense that they are calculated in such a way as to provide the defendant with strong incentives to perform, or give the plaintiff the same satisfaction or utility (‘consumer surplus’) as if the contract had been performed. See eg Pearce and Halson, above n 1; C Webb, ‘Performance and Compensation: An Analysis of Contract Damages and Contractual Obligation’ (2006) 26 OJLS 41. The authors’ observations about the effects and purposes of such awards are flawless, but the awards nonetheless involve a priced substitute for the right, or an incentive for its observance, not its literal enforcement. For this reason, in so far as such monetary awards ‘vindicate’ rights in any of the senses identified in this chapter (‘incentivising’ awards clearly go beyond this), they are conceptualised as mechanisms for reversing the effects of an infringement—section (iv), below. 45 In The Law of Quasi Contract, 2nd edn (Sydney, The Law Book Co, 1989), S Stoljar appears to have gone one step further, arguing in effect that the bulk of restitutionary rights were based on a plaintiff’s existing title to property, but that analysis is flawed for reasons set out in Burrows, above n 29, at 29–30. For the more limited claim that some restitutionary rights are based on continuing title to property, see G Virgo, The Principles of the Law of Restitution, 2nd edn (Oxford, Oxford University Press, 2006) 11–17; D Fox, ‘Legal Title as a Ground of Restitutionary Liability’ (2000) Restitution Law Review 465; R Grantham and C Rickett, Enrichment and Restitution in New Zealand (Oxford, Hart Publishing, 2000) 32–34.

The Mixed Concept of Vindication 75 infringer ex post facto, in so far as this can practically be done. This type of vindication is almost entirely alien to modern public law, although it is not unheard of.46 It is achieved primarily through monetary awards and it is necessarily inexact, particularly where personal injury, mental harm, reputational injury, or harm to pride, dignity or personal autonomy is involved. Reversing the effects of rights violations upon intangible financial interests is an easier and more exact process, since monetary remedies then effectively substitute like for like. Compensatory awards (including aggravated damages47) make good the loss an infringement has caused to a plaintiff, albeit that sometimes the ‘loss’ may consist in the loss of an opportunity to gain, the loss of a ‘chance’ to avoid economic harm, a loss of personal autonomy or dignity, a devaluation of the right itself, or the loss of a (factual) opportunity or (legal) power to insist upon, or negotiate the right.48 46 See, eg, criminal compensation orders. Compensatory damages under the HRA also ‘go some distance towards’ vindication in this sense: (Ramanoop, above n 12, at [18]–[19] (Lord Nicholls)). See also awards for the breach of constitutional provisions in other countries. In Ward, above n 12, at [28] McLachlin CJ indicated that Charter damages ‘vindicate’ by making good the harm done to society by a breach, rather than the harm done to any particular individual. Since it is not society that actually receives such damages, this posits a theory of vindication based on deterring institutional harm, not reversing the infringement’s private effects. 47 The modern view is that such awards compensate for intangible injury to feelings or (distinctly) dignity, suffered in consequence of the circumstances surrounding a rights infringement: Thompson v Commissioner of Police of the Metropolis [1997] EWCA Civ 3083, [1998] QB 498, 512 (Lord Woolf MR); Malcolm v Ministry of Justice [2010] EWHC 3389 (QB) [83], [131] (Sweeney J); Wayne Quirk v New South Wales [2011] NSWSC 341, [178] (Grove AJ); Law Commission, Aggravated, Exemplary and Restitutionary Damages, Law Comm Report No 247 (1997), [1.42]; A Beever, ‘The Structure of Aggravated and Exemplary Damages’ (2003) 23 OJLS 87, 89; J Berryman, ‘Reconceptualizing Aggravated Damages: Recognizing the Dignitary Interest and Referential Loss’ (2004) 41 San Diego Law Review 1521; J Murphy, ‘The Nature and Domain of Aggravated Damages’ [2010] CLJ 353. 48 There is much debate as to whether the category of ‘compensatory’ awards is capable of including awards for rights violations which are unaccompanied by material ‘factual’ loss. The classic case is where the plaintiff is granted a reasonable fee for the defendant’s use of the plaintiff’s chair, when the plaintiff would not otherwise have sat on the chair or put it to profitable use. Is such an award compensatory? It seems perfectly arguable that in such a case the property right to the chair includes the right to negotiate its use with others, and that the plaintiff’s power to insist on this right is substantially diminished as a consequence of the infringement, even if the market value of the chair itself remains unaffected and even if the plaintiff would never, as a matter of fact, have chosen to negotiate its use by the defendant. See above n 33. Part of the value of any right which the law will specifically enforce is hence the power to choose whether to insist upon it and this power is effectively ‘lost’ in such a case, even if the underlying right to the chair itself continues to subsist. While it is therefore historically true to say (as Thomas J observed in Daniels v Thompson [1998] 3 NZLR 22, 70) that compensatory damages do not recognise the value of the power to choose whether to insist on a right, there is no reason why they should not. For cases in which compensatory damages have arguably been awarded so as to capture losses of choice associated with rights, see Rees, above n 12 (loss of power to choose whether or not to have a child); Chester, above n 3 (loss of power to choose whether to undertake medical risk); TCN Channel Nine, above n 12 (loss of power to exclude defendant from land and associated distress); Bunnings Group, above n 12 (loss of power to negotiate a rate for the use of personal property); Force India Formula One Team Ltd v 1 Malaysia Racing Team SDN Bhd [2012] EWHC 616, [2012] All ER (D) 198 (Ch) (loss of power to negotiate use of confidential information).

76

Kit Barker

Restitutionary awards eradicate gains the defendant has made by requiring him to pay them to the right-holder, whether or not the latter has suffered a loss.49 Awards of interest also serve to reverse the effects of infringement, reflecting as they do either the loss a plaintiff has suffered by being kept out of compensation to which he is entitled, or the gain a defendant unjustly obtained50 by being allowed to retain benefits it should not have received. Likewise, costs awards aim primarily to reverse the adverse financial effects that have been forced upon a plaintiff by virtue of having to litigate to protect his or her rights.51 Monetary awards operate not just to reverse the pretrial effects of an infringement, but also, by way of temporal projection, to address its anticipated post-trial effects. Most often this is done by a court making an informed guess as to what those effects will be and then attaching an approximate money value to them in advance of their occurrence, but in at least one well-known case, pecuniary awards themselves help both to repair damage done and prevent further future losses. In the tort of defamation, it is thus routinely accepted that one of the functions of a general damages award is to ‘vindicate’ the plaintiff’s reputation. Vindication here means more than simply providing a public acknowledgement of the right, although that may be part of it, as suggested in Part III B (ii) above. It means cleaning the plaintiff’s name of its tarnish. Since reputational damage actually consists in a deleterious change in the public perception of a person, its remediation entails an equivalent change for the better in that perception and monetary sums are calculated symbolically with this objective in mind.52 Whether a given award is successful in restoring a plaintiff’s name for the future depends entirely, of course, on the extent to which it reaches the public consciousness and scepticism can legitimately be expressed about this, particularly when awards are routinely granted so long after the original defamatory slur was made that little to no publicity attaches to them. Speedy, well-publicised

49 For a recent example of a judicial statement to the effect that restitution vindicates rights, see VFS Financial Services (UK) Ltd v Euro Auctions (UK) Ltd and others [2007] EWHC 1492, [2007] All ER (D) 346 (QB) [79]. Note that only a minority of restitutionary (gainbased) awards respond to rights ‘infringements’ in the sense of civil wrongs—most respond to events such as mistakes that involve no ‘wrongdoing’ on the part of the defendant. The characterisation of the rights at stake in the latter types of case is much contested and beyond the scope of the current work. Note also that some gain-based awards for wrongdoing go beyond reversing the effects of the wrong as between plaintiff and defendant and engage in pure profit-stripping. Where this is the case, their rationale is deterrent. 50 Sempra Metals Ltd v IRC [2007] UKHL 34, [2008] 1 AC 561. 51 Australian Receivables Ltd v Tekitu Pty Ltd and Ors [2011] NSWSC 1425, [22] (Ward J): ‘An order for costs … is compensatory in nature in order to reflect the vindication of … [plaintiff’s] successful claim.’ 52 See Witzleb and Carroll, above n 1, at 34–35; Uren v John Fairfax & Sons Ltd (1966) 117 CLR 118, [513] (Barry J); Broome v Cassell & Co Ltd [1972] UKHL 3, [1972] AC 1027, 1071 (Lord Hailsham) [Broome]; Cornes v The Ten Group Pty Ltd [2011] SASC 104, [122–27] (Peek J).

The Mixed Concept of Vindication 77 judicial declarations of the slur’s falsity would undoubtedly clear a person’s name more effectively. Fully reasoned judgments can sometimes do the same job,53 as can other, non-judicial actions such as an apology on the part of a defamer, an offer of amends,54 or the publication of a plaintiff’s ‘reply’ to the defamatory imputation. In theory, nonetheless, the ‘vindicatory’ element of defamation damages is restorative in purpose, however inefficient such awards may actually be in achieving this end. Vindication in defamation law forms part of the process whereby the effects of defamatory matter upon a person’s name are undone. To the extent that this is achieved, the normative force of the idea is exhausted.55 This is very different to the idea underlying the views of judicial supporters of vindicatory damages in Lumba, for whom the idea of vindication continues to demand concerted judicial action even when a plaintiff has suffered—and stands to suffer—no harm at all. The vindication of rights in the restorative sense is most often—and most easily—justified by reference to Aristotelian conceptions of corrective justice.56 Such justifications are predominantly individualistic in nature. They suggest that in reversing the effects of rights infringements, the law restores a prior moral and legal equality between those who commit and those who suffer injustice. Some of the effects of infringement that monetary awards ‘reverse’ consist not in consequential factual losses (though these are most common), but in losses of power or choice associated with the legal right that was violated, such as the loss of a power to exclude a person from one’s land, goods or private affairs. And in some cases involving the infliction of harm to dignity or emotional distress, the reversal of an injustice’s effects can be achieved through symbolic institutional acts other than awards of

53 Rackham v Sandy & Ors [2005] EWHC 1354, [2006] 1 Costs LR 34, [13] (Gray J); Purnell, above n 32, at [27] (Laws LJ). So can a grant of judicial leave to a plaintiff to make a statement in open court publicising a defendant’s substantial payment into court: Charlton v EMAP Plc, The Times, 11 June 1993 (QB). 54 Whether the defendant is actually sorry is irrelevant to the restorative effect of the apology or offer. 55 The point is well put by Rothman J in Cantwell v Sinclair [2011] NSWSC 1244, [162]: vindication may be a purpose of defamation damages, but it ‘does not give rise to greater damages than would otherwise be the case for consolation and compensation’. 56 The literature is too expansive to set out here, but all accounts must now begin with the influential work of Professor Weinrib, with its characteristic Kantian spin: E Weinrib, The Idea of Private Law (Cambridge, Mass, Harvard University Press, 1995); E Weinrib, Corrective Justice (Oxford, Oxford University Press, 2012). See further, J Gardner, ‘What is Tort Law For? Part 1. The Place of Corrective Justice’ (2010) 30 Law and Philosophy 1. For more explicit consideration of how restitutionary liabilities feature in this framework, see E Weinrib, ‘Restitutionary Damages as Corrective Justice’ (2000) 1 Theoretical Inquiries in Law 1. For three counter-points amongst the many that are available, see R Posner, The Economics of Justice (Cambridge, Mass, Harvard University Press, 1983) (economic analysis); B Zipursky, ‘Civil Recourse, not Corrective Justice’ (2003) 91 Georgetown Law Journal 695 (civil recourse theory); H Dagan, The Law and Ethics of Restitution (Cambridge, Cambridge University Press, 2004) (distributive justice).

78

Kit Barker

damages, such as corrective statements, court-ordered apologies, or widely publicised reasons for judgment. While such symbolic acts may therefore be regarded as important for public reasons, as suggested in Part III (B) (ii), they can also, on occasion, have a private, restorative function in remediating the personal harm that an infringement has done. Symbols can heal individual victims. There may be still other devices that can usefully be developed in the implementation of such restorative aims that the law has yet to embrace and the law should certainly remain open to all such possibilities.57 (v) Punishing Infringement Punishment consists in the imposition upon a defendant of an unwelcome consequence which goes beyond the obligation to compensate a plaintiff’s loss and beyond stripping her of any net gain made from a rightsinfringement.58 In contrast to public law, private law only rarely punishes wrongdoers and it is certainly not the norm nowadays for judges to construe punishment as an event by which they seek to vindicate private law rights. The main focus of punitive awards is upon the outrageousness of the wrong, the public power of the wrongdoer, or the wrongdoer’s cynical and exploitative intentions and on the need in each instance to deter the relevant conduct, not on the plaintiff’s right as such.59 Such awards are more concerned to ‘denounce’ the wrongdoer, one might say, than to ‘announce’ the right. Moreover, they almost invariably place the plaintiff in a better position (both as a matter of fact and as a matter of right) than that which she occupied prior to the infringement, so that the award goes further than simply reversing the infringement’s effects in the sense discussed in the previous section. Perhaps for these reasons, Lord Scott in Ashley chose

57

See further Carroll and Witzleb, above n 42. Customarily, punitive consequences to the defendant are more serious than a liability for either the plaintiff’s loss or the defendant’s net gain. But is this a necessary feature of punishment conceptually? Perhaps one could still ‘punish’ the defendant by slapping her wrist, even where she has wrongly caused the plaintiff a £1 million loss, or wrongly made a £1 million gain. If, however, punishment is to achieve either of its commonly mooted ends (retribution and/or deterrence), it seems likely that the consequences imposed will indeed have to be more serious. Retribution requires at least a proportionate relationship between consequence and wrong (if not indeed between consequence and wrongful damage—an eye for an eye, if not necessarily five oxen for one ox killed—Exodus 22:1). And while deterrence could be achieved by leaving the defendant in a neutral position vis-à-vis her starting point (ie by simply removing net gains), it is likely to be more forcefully achieved by higher penalties. 59 See, eg, Gray v Motor Accident Commission [1998] HCA 70, 196 CLR 1, [15], [31] (Gleeson CJ, McHugh, Gummow and Hayne JJ) [Gray]; H McGregor, McGregor on Damages, 18th edn (London, Sweet & Maxwell, 2009) 11-012. But see Stevens, Torts and Rights, above n 1, at 85–88 (awards vindicate rights through retribution—eradicating a more serious moral wrong). 58

The Mixed Concept of Vindication 79 expressly to contrast the concepts of punishment and vindication, implying their mutual incompatibility.60 Despite this, some judges undoubtedly continue to insist that punitive awards ‘vindicate’ in one way or another. The most commonly expressed view is that they vindicate the ‘strength of the law’,61 the ‘rule of law’,62 or that they ‘mark the court’s disapproval’63 of a defendant’s conduct, but Lord Nicholls has also described them as ‘emphatically vindicating’ the plaintiff’s right to physical welfare in a case involving crass medical negligence.64 Similarly, in Australia, some judges have openly said that exemplary awards ‘vindicate the victim’.65 In what sense or senses of the word, then, could it be said that punitive awards vindicate? One way in which they might do so is in the first sense already identified above—namely, by providing a strong acknowledgement or ‘marker’ of the plaintiff’s right. They would then operate rather like ex post facto judicial declarations, only as declarations that shout loudly rather than simply announcing the right in more sober tones. This analysis is superficially congruent with the alternative label often given to punitive awards: ‘exemplary’ damages. Examples are, after all, public exhibitions and fundamentally expressive in both purpose and effect. However, while punitive awards probably do have an incidental effect in ‘marking’ rights on some occasions,

60

Ashley, above n 3, at [22]. Almost all such statements are citations of Lord Devlin in Rookes v Barnard [1964] UKHL 1, [1964] AC 1129, 1223 [Rookes] (intimidation): see, eg, Australian Consolidated Press v Uren [1969] 1 AC 590, [1967] 3 WLR 1338 (PC (Australia)) (defamation); Millington v The Commissioner of Police The Times, 28 May 1983 (QB) (Forbes J) (false imprisonment); Holden v Chief Constable [1987] QB 380, [1986] 3 WLR 1107 (CA (Civ Div)) (Purchas LJ) (wrongful arrest); Pettigrew v Northern Ireland Office [1990] NI 179 (Hutton CJ) (assault); AB v South West Water Services [1993] QB 507, [1993] 2 WLR 507 (CA (Civ Div)) (StuartSmith LJ) (public nuisance); Kuddus, above n 31, at [52] (Lord Nicholls), [75], [91] (Lord Hutton) (misfeasance in public office); Borders (UK) Ltd and others v Commissioner of Police of The Metropolis and another [2005] EWCA Civ 197, [2005] Po LR 1, [52] (Sedley LJ) (conversion); Muuse v SS Home Dept [2010] EWCA Civ 453, (2010) 107(19) LSG 24, [77] (Sir Anthony Morritt C, Thomas LJ, Sir Scott Baker) [Muuse] (false imprisonment). It is hard to find a single statement to this effect prior to Rookes. Note, also, that at least one of the judges who sat with Lord Devlin in that case (Lord Reid) later expressly disagreed with his view that punitive awards serve any useful function in vindicating the law’s strength: Broome, above n 52. 62 Kuddus, above n 31, at [79] (Lord Hutton). These comments were picked up in New South Wales v Ibbett [2006] HCA 57, 231 ALR 485, [40] (Gleeson CJ, Gummow, Kirby, Heydon and Crennan JJ) [Ibbett] (trespass to land) and Whitbread v Rail Corporation New South Wales [2011] NSWCA 130, [23] (McColl JA) [Whitbread] (wrongful arrest, false imprisonment). 63 Watkins, above n 18, at [26] (Lord Bingham); Whitbread, above n 62, at [215] (Whealey JA). 64 A v Bottrill [2002] UKPC 44, [2003] 1 AC 449, [30] (New Zealand). 65 James v Hill [2004] NSWCA 301, [69], (Tobias JA, Sheller and Hodgson JJA agreeing); Harris v Digital Pulse Pty Ltd [2003] NSWCA 10, 56 NSWLR 298, [55] (Spigelman CJ) (‘appeasement of the plaintiff’), [195] (Mason P) [Harris] (‘vindicating a person who is the victim of wrongdoing’). 61

80

Kit Barker

this does not really capture what judges mean when they talk about punishing ‘to set an example’. This is because their intention is primarily to stigmatise the defendant’s conduct, not simply to paint the plaintiff’s right in brighter colours.66 That is why they refer to punitive awards as marking their ‘disapproval’ of the defendant or society’s ‘outrage’, rather than as marking their ‘approval’ of the right violated. It also explains why awards are available as an exceptional head, not as a general head of damages available for the violation of particular types of right. If exemplary awards were designed to highlight the importance of rights of a particular genus—rights to physical liberty, for example—one would expect them to be available in all cases in which rights of that type are violated. In fact, the purpose of a court marking its disapproval through an exemplary award is pretty clearly to achieve forward-looking public aims of deterrence or prevention.67 Indeed, it is no coincidence that many of the cases in which awards are made involve the infringement of private rights by high-handed agents of the state and some cases point directly to judges’ belief that it is only such awards that will force government to implement the systemic changes necessary to ensure that disgraceful conduct of the type in question does not occur again.68 The deterrent rationale of awards explains why courts have dropped the old terminology of ‘vindictive’ damages and why awards are available against private employers69 and the state70 for the outrageous acts of their employees or agents. If the aim were vindictively to strip a pound of flesh from the wrongdoer, this would make little sense. If it is to provide those in positions of power and control with strong incentives to stamp out

66 Gray, above n 59, at [15], [31] (Gleeson CJ, McHugh, Gummow and Hayne JJ); Whitbread, above n 62, at [21] (Giles JA). These are two, distinct aims, not a single aim viewed from different points of view, albeit that the pursuit of one may have the other as an incidental side-effect. 67 According to Gallanter and Luban, above n 39, at 1429–32, ‘prevention’ is a better word to use than ‘deterrence’, because it conveys the way punitive awards change social behaviour not simply by instilling the fear of sanctions into people (‘deterring’), but also by projecting and reinforcing important social norms. See also A Sebok, ‘What did Punitive Damages Do? Why Misunderstanding the History of Punitive Damages Matters Today’ (2003) 78 ChicagoKent Law Review 163, 202–3, 206. In Harris, above n 65, at [55], Spiegelman J likewise distinguished deterrence from ‘denunciation’. This, it is submitted, is an oblique reference to the same distinction, denunciation comprising the critical public censure of a person’s departure from an expected social norm, with a view to preventing further such departures. For cases endorsing a deterrent conception of punitive awards, see eg Muuse, above n 61, at [77]; Watkins, above n 18, at [26] (Lord Bingham); Zorom Enterprises Pty Ltd v Zabow and Ors [2007] NSWCA 106, (2007) 71 NSWLR 354 [Zorom]. 68 Muuse, above n 61, at [77] (Thomas LJ, The Chancellor and Sir Scott Baker agreeing). 69 Zorom, above n 67. The liquidation of the employer in this case was considered irrelevant, precisely because the award might deter others. 70 Ibbett, above n 62; State of New South Wales v Landini [2010] NSWCA 157; Rowlands v Chief Constable of Merseyside Police [2006] EWCA Civ 1773, [2007] 1 WLR 1065; Muuse, above n 61.

The Mixed Concept of Vindication 81 undesirable practices, it is more readily explicable.71 It also explains why courts are reluctant to make more than a single award of punitive damages, no matter how many plaintiffs have been adversely affected by a defendant’s wrongdoing. A second possibility is that punitive awards could be construed as vindicating plaintiffs’ rights by reversing some of the effects of rights violations that are not captured by traditional compensatory and restitutionary awards. This is the fourth sense of the term vindication alluded to above. Historically, there is some truth to the idea because the types of harm that were traditionally conceived of as constituting actionable ‘loss’ for the purposes of compensatory damages awards were limited and omitted some types of intangible harm, in particular what we would nowadays conceive of as harm to personal dignity or autonomy. Punitive awards served in part to fill the compensatory gap72 and thereby engaged in a form of corrective justice. But if the law has historically been too slow to compensate injuries of this type, there are clear signs that it is changing course and persuasive arguments that it should.73 Nowadays, the job is partly done by aggravated compensatory damages awards, although there are admittedly still some gaps. Similarly, to the extent that punitive damages have ever played a role in stripping cynical wrongdoers of their gains (a function clearly implied by Lord Devlin’s second category in Rookes v Barnard74), that function could now easily and more precisely be accommodated by restitutionary awards75 and/or by the confiscation of criminal gains by the state in public law. When all of this is considered, it is very hard to conceive of punitive awards as making any sensible contribution to the restorative aims of corrective justice in the modern day, despite a surprising recent suggestion to this effect in Tasmania.76 It is possible, one must suppose, that a particular 71 Cf Stevens, above n 1, at 87 (arguing that there is no reason to incentivise an employer if it has itself behaved reasonably). The liability of employers and the state is also consistent with the aim of compensating plaintiffs. 72 See M Redish and A Mathews, ‘Why Punitive Damages are Unconstitutional’ (2004) 53 Emory Law Journal 1, 13–19 (exemplary damages compensated for ‘immeasurable but real’ harms in the 19th century, including dignitary harm); Sebok, above n 67, at 195–202 (19thcentury awards served a range of functions including the compensation of emotional suffering and insult as well as the ‘return of lost honour’); Cooper Indus Inc v Leatherman Tool Group Inc 532 US 424 (2001) 437, n 11, 121 S Ct 1678. The confusion may also have been promoted by the historical failure of philosophers such as Aristotle, Aquinas and Grotius to properly distinguish corrective from retributive justice: A Beever, ‘Justice and Punishment in Tort: A Comparative Theoretical Analysis’ in Rickett, above n 33, at 253–67. 73 See above n 47; Rees, above n 12; Chester, above n 3; TCN Channel Nine, above n 12; Bunnings Group, above n 12. 74 Rookes, above n 61, at 1226 (cases in which the ‘defendant’s conduct has been calculated to make a profit for himself which may well exceed the compensation payable to the plaintiff’). 75 Kuddus, above n 31, at [109] (Lord Scott); McGregor on Damages, above n 59, at 11-029. 76 Gunns Ltd v Alishah (No 4) [2010] TASSC 24, [58] (Porter J).

82

Kit Barker

legal system might be so doctrinally underdeveloped as to be unable at this point in time properly to remedy the personal side-effects of rights infringements through compensation or restitution, but if this is so, the message is that these systems need to work harder to expand the reach of compensatory and restitutionary remedies, not persist in the use of punitive awards to oblique, restorative ends. While there is therefore probably some historical truth to the idea that punitive damages achieved justice for victims by restoring their moral equality with their offenders and by remedying injuries to their dignity, this aim can nowadays be achieved simply by increasing compensatory awards in appropriate instances to account for harm to dignity and emotional harm and by simultaneously declaring victims’ rights.77 This does not exclude punitive awards altogether from the modern private law—it simply means that if they are to be retained, their function must be expressly aligned with public purposes such as deterrence,78 not with the private purpose of personal restoration. This, it seems to me, is the view ultimately taken by Spiegelman J in the Digital Pulse case. For although his Honour did make the suggestion that punitive damages might in the modern age continue to serve to ‘vindicate’ plaintiffs by ‘appeasing’ them, he went on explicitly to identify the purpose of such appeasement as being a public one (the preservation of a peaceful social order) not any private purpose of restoring a plaintiff’s dignity, or compensating his or her emotional distress.79 The conclusion to which we are inevitably drawn is that there is no sense in which private rights are ‘vindicated’ by punitive awards that is not already accounted for by the other legal meanings of the term to which

77 Note that for some writers, the dignity argument is not simply historical, but a basis for defending the modern punitive damages system and/or the idea of retribution itself: see A Harris, ‘Re-Reading Punitive Damages: Beyond the Public/Private Distinction’ (1989) 40 Alabama Law Review 1079, 1102 (punitive damages protect the ‘honour or dignity of the individual’ for both private and public reasons); J Hampton, ‘Correcting Harms versus Righting Wrongs: The Goal of Retribution’ (1992) 39 UCLA Law Review 1659, especially 1685–86, 1698 (punishment is one form of ‘retribution’ affirming the moral equality of the wrongdoer and the wronged in a Kantian sense—reacting to the offender’s denial of a victim’s (moral) ‘value’, and making good the moral injury between them); Gallanter and Luban, above n 39, at 1432–33 (largely adopting Hampton’s analysis); A Nezar, ‘Reconciling Punitive Damages with Tort Law’s Normative Framework’ (2011) 121 Yaw Law Journal 678 (adopting but tweaking Hampton’s analysis slightly—punitive damages ensure that the defendant ‘accounts’ for culpably denigrating the plaintiff’s moral status and ‘pays off a moral debt’ to the plaintiff); P Lee, ‘Contract Damages, Corrective Justice and Punishment’ (2007) 70 MLR 887 (punitive awards repair ‘moral injury’ and can hence be justified in corrective justice terms). Some have gone further still in their claims for the explanatory power of the concept of honour in private law: see N Oman, ‘The Honor of Private Law’ (2011) 80 Fordham Law Review 31. For a defence of punitive damages through the lens of ‘civil recourse’ theory, see B Zipursky, ‘A Theory of Punitive Damages’ (2005) 84 Texas Law Review 105 (such awards have a ‘civil aspect’ giving effect to a plaintiff’s ‘right to be punitive’, as well as a criminal aspect). 78 See eg J Edelman, ‘In Defence of Exemplary Damages’ in Rickett, above n 33. 79 Harris, above n 65, at [56].

The Mixed Concept of Vindication 83 we have previously adverted. They either act as dramatic public ‘markers’ (albeit of particular instances of wrongful behaviour, rather than of the importance of private rights generally); or they have played a historic role in effecting restoration within a private law system whose conception of loss has simply been too slow to develop. Their ‘marking’ and deterrent roles are clearly shared with public law and will remain current for as long as public law and public agencies prove incapable of perfectly policing undesired behaviours. They are no longer required in any restorative role.

C. The Mixed Conception of Vindication It is evident from the preceding analysis that the institutional (judicial) conception of vindication in private law in general (and in tort law in particular) is mixed. It is mixed in the sense that there are at least four different judicial understandings of what it means to vindicate. Almost all of these are shared to a greater or lesser degree with public law, although the types of right that public and private law vindicate are different and it is only private law that makes a determined and systematic commitment to vindicating rights in the important sense of reversing the effects of their infringement as between right-holder and wrongdoer. While declaratory vindication is necessarily entailed to a greater or lesser degree by all of the other forms we have examined, the reverse is not the case. Publicly preventing a right’s infringement, reversing the infringement’s effects or specifically enforcing a right inevitably also entail declaring it, either expressly or impliedly. Conversely, the mere declaration of a right does not necessarily entail vindicating it in any of the other ways we have identified, although it might sometimes assist in preventing the right’s infringement (by warding at least some potential infringers away) or in reversing its adverse emotional or dignitarian effects. The concept is also mixed in the second, distinct sense that the purposes which the various forms of vindication serve, even within private law, potentially span the divide between public and private ends. Preventing rights infringements, declaring rights, specifically enforcing them and reversing the effects of their violation are all types of response that are consistent with the individualistic goals of corrective or interpersonal justice80—the sphere of justice that gives effect to bilateral obligations of right and duty 80 I take the latter term from A Robertson, ‘Rights, Pluralism and the Duty of Care’ in Nolan and Robertson above n 1, at 435. It helpfully captures in plain language the insight that not all individualistic conceptions of justice entail ‘correcting’ or rectifying wrongs ex post facto. For similar reasons, others have long preferred the term ‘commutative justice’ (J Finnis, Natural Law and Natural Rights (Oxford, Clarendon Press, 1980) 178–79, drawing on Aquinas) or ‘interactive’ justice (R Wright, ‘The Principles of Justice’ (2000) 75 Notre Dame Law Review 1859, drawing on Finnis).

84

Kit Barker

between private citizens. There are, however, clear signs that vindication in the specific sense of judicially ‘declaring rights’ or ‘marking wrongs’ is also understood by some judges in tort cases as having a significant public dimension. That is, the declaration of particular (often ‘fundamental’) rights and the expressive denunciation of particularly heinous instances of rights violation are thought of as being important symbolically for the purposes of public education, social construction and deterrence—what one might collectively refer to as ‘social messaging’ functions—as well as sometimes being important in relieving a plaintiff of the psychological effects of a wrong and restoring his or her dignity for the future. Although there may be Kantian justifications (justifications of ‘public right’) for maintaining a system of public adjudication in which disputes about private rights are publicly resolved,81 that deontological system logically precludes justifications for declarations of right which are based on broader ‘social messaging’ functions such as these. That does not seem, however, to prevent judges alluding to them. Unsurprisingly, such judicial allusions to public purposes have proven most common in the field of punitive damages, where the commitment to deterrence remains strong, but they are now also bubbling to the surface in a variety of other cases, such as Lumba. The symbolic value of judicial actions is being accorded increased importance not just in public, but also, it seems, in private law. This symbolism is thought particularly apposite in tort cases involving abuses of state power, but is not necessarily confined to such cases. It gives the law in such cases a key public function in constructing and reinforcing particular social norms. Since all such public messaging functions rely for their validity upon untested empirical assumptions, it is only right to question them. Most judicial declarations of right will reach only a narrow audience, unless it is made a condition of an order that they are more broadly publicised. The extent to which monetary judgments are capable of conveying precise public messages is also seriously questionable and indeed, in the majority of cases, the provision of clear reasons for judgment may be a more exact educational tool when it comes to explaining and validating the values upon which society is based. Money is a powerful incentive, but an inarticulate mode of expression. There are also, I think, real issues about whether it would be appropriate to make a defendant pay, through private litigation costs and damages awards, simply for the public to be better informed about rights generally. The point is illustrated by the facts of the Ashley case,82 where the United Kingdom Supreme Court surprisingly refused to strike out a plaintiff’s claim against the police for battery despite the fact that the police had already agreed to compensate him fully for the injuries he had

81 82

Above, n 40 and text. Above n 3.

The Mixed Concept of Vindication 85 suffered. Could one really justify making a defendant pay the costs of a show trial, if the purpose of this was simply to capture the public imagination or contribute to our social education about the importance of rights? Publicly funded information systems would seem to be more efficient and wide-reaching mechanisms for social messaging than the cumbrous and randomised processes of tort litigation. To claim that social messaging is a beneficial effect of judicial decision making is one thing (even if the evidence of this is currently sparse); to claim that it is the justification for judicial action, quite another. It is not simply Kantians that would be up in arms at the idea. Empiricists would also be banging on the table, demanding the evidence upon which assumptions about the efficacy of judicial messaging in promoting the public good are based. If there is a justification for decisions such as Ashley, it therefore seems more likely to lie in the idea that plaintiffs have a (public law) right against courts to obtain a resolution of their legal claims,83 than in the idea that the action contributes to the dissemination of some important public announcement about the value of the right infringed. That does not necessarily mean that Ashley is right—there must come a point at which the assertion of even a public right to adjudication becomes so disproportionately expensive to effect that a question mark should be raised against it. As one court has put it, the ‘game may not be worth the wick, let alone candle’.84 Indeed, the idea that it is appropriate to pursue such public rights whatever the associated cost seems to be in real tension with current trends in our procedural law, according to which private litigants are strongly incentivised to mediate, negotiate and resolve their disputes ‘off the public record’. It is nonetheless marginally easier to think of the action in Ashley as a mechanism for realising a public right to a decision (about private rights), I suggest, than as a mechanism for some broader goal such as educating the public, or conducting some form of privatised public inquiry into police behaviour.85 There is much more to be said about this and in some ways it is crucially important to our understanding of the role of private law in a modern age 83 See, analogously, S Smith, ‘The Rights of Private Law’ in Robertson and Tang Hang Wu, above n 21, at 118–19, 121 (‘action rights’). While Smith ultimately chooses to describe such rights as ‘private’ (because they reinforce the private rights a plaintiff has against another private person), he concedes that the classification is not obvious and that, on at least one view, the right is a public one. 84 Jameel v Dow Jones [2005] EWCA Civ 75, [2005] QB 946, [67]–[69] (Lord Phillips MR, Sedley LJ, Jonathan Parker LJ) (action in defamation struck out as abuse of process despite fact that it might vindicate minimally). See, analogously in public law: MD v Secretary of State for the Home Department [2011] EWCA Civ 453 [MD] (judicial review action seeking declaration struck out on basis that the costs of continuing proceedings disproportionate to vindicatory potential of the remedy). 85 Cf Priel, above n 4, at 199, who argues that the type of vindication in Ashley was of a public type, namely the provision of a ‘public forum for declaring rights’. Note that this function is described as ‘broad’ and ‘societal’ (at 184) and is related to increasing the ‘accountability of public authorities’ (at 207).

86

Kit Barker

fixated on the rhetoric of rights. It ties into a much broader debate, current in the public law literature, about the role of law as a mode of public expression.86 For our current purposes, however, we need not be further detained by the debate about the precise nature of the justifications for the concept of vindication in torts and private law, other than to observe that, on current judicial understandings, they are decidedly mixed. The reason for thus suspending the debate is simply that, however these aims are ultimately framed, they can be achieved without introducing a new genus of damages known as ‘vindicatory damages’. IV. THE HETERODOXY: VINDICATORY DAMAGES87

We now return to the controversial proposition canvassed by three members of the majority in the Lumba case, that tort law needs a distinct type of damages—neither nominal, compensatory, restitutionary, nor exemplary—in order to ‘vindicate’ rights, including the right to personal liberty, which is currently protected by the tort of false imprisonment. What is the case for recognising such an independent head of relief in private law, either as a general or exceptional head of damages? The logical consequence of all we have learned about the concept of vindication in the preceding section is that no such separate remedy is required. At one level of analysis, all damages awards—indeed all judicial remedies—are ‘vindicatory’ of rights in the sense of affirming their existence in one way or another. From this point of view, the very label ‘vindicatory damages’ is confusing as a designator for any specific ‘quantum’ or ‘type’ of monetary relief. More fundamentally than this, however, when we examine the various different types of vindication which the law endorses, there is no single subspecies of vindication—and no aim underlying that subspecies—that the law cannot already implement through the use or adaptation of existing remedial devices. A distinct remedial ‘measure’ of ‘vindicatory’ damages is not just conceptually difficult, but functionally superfluous. Such damages cannot logically ‘prevent’ an infringement taking place; nor can they ‘specifically enforce’ a right. To the extent that there is a need to vindicate a right by reversing the effects of its violation as between the parties to a dispute, this can be achieved through the use of compensatory awards which give due allowance for a broad range of losses, including affronts to a victim’s dignity, losses of opportunity (‘chance’) and diminutions in 86 For a critical overview of expressive theories of public law, see M Adler, ‘Expressive Theories of Law: A Skeptical Overview’ (2000) 148 University of Pennsylvania Law Review 1363. 87 For cases in which it is said that ‘vindicatory damages’ have been awarded, see above n 12.

The Mixed Concept of Vindication 87 personal autonomy (‘choice’); and through the use of restitutionary awards which strip wrongdoers of their wrongful gains. There again, if vindication is construed as a concern about ‘marking’ or ‘declaring’ rights generally, then this can surely be done either directly through judicial declarations and the provision of emphatic and well-publicised reasons for judgment, or (if language fails us, which I personally doubt) through the existing range of monetary awards. Finally, if courts’ concerns are not actually to declare the importance of rights generally, but more specifically to mark particular instances of outrageous infringement that need to be denounced and deterred, then exemplary damages are currently available to this end. All members of the Court in Lumba seem (rightly, I think) to have assumed that such awards are available in principle even where a plaintiff has suffered no provable loss. Whether an exemplary award should have been made on the facts of Lumba itself is a matter on which Lord Dyson thought that opinions could reasonably differ.88 The state’s wrongdoing had some very distasteful trappings, including a deliberate attempt to conceal practices which it knew might be unlawful; but at the same time the plaintiffs were not the individuals most likely to elicit sympathy or to be seen to deserve the windfall of an exemplary award. The key point is nonetheless that if the deterrence and denunciation of abuses of state power were thought by the court to be a sufficiently important priority on the facts of the case, exemplary damages could surely have been used to do the job, as indeed they were in another recent false imprisonment case, Muuse v Secretary of State for the Home Department.89 It also seems most unlikely that awards as small as £500 or £1,000 would effectively denounce or deter the types of deliberate misconduct that were at stake in Lumba, which leaves the underlying rationale of the sums suggested a mystery. The reasoning of those members of the majority in Lumba that favoured vindicatory damages in tort unfortunately does little to assist in solving the conundrum. No judge thought an award necessary to reverse the effects of the infringements of the plaintiffs’ rights to liberty, since all made the point that the plaintiffs were no worse off as a result of the infringements of those rights than they would otherwise have been. At one point in his very brief judgment, Lord Hope indicated that the function of the award was

88 Above n 2, at [166]. Lord Dyson found at [155] that there was ‘undoubtedly evidence’ to support the view that the Secretary of State actively discouraged disclosure of the hidden policy; that the decision not to publish it was deliberate and she or her officials knew or were reckless as to the unlawfulness of their actions, preferring for political reasons to leave it to the courts to remedy any illegality. He ultimately deferred to the view of lower courts, pointing to the risk of large numbers of plaintiffs and the fact that no oral evidence had been forthcoming from government ministers in proceedings. All other members of the court agreed, despite Lord Walker finding the conduct ‘deplorable’ (at [195]). 89 Above n 61.

88

Kit Barker

to ‘recognise the importance of the right to the individual’ and to ‘assert’ it as a ‘valuable one as to whose enforcement the plaintiff has an interest’.90 He also later suggested (rather tenuously, given what we have just said about the paltry nature of the sum) that it might operate so as to deter future breaches.91 Excluding the deterrence argument for the moment, this suggests that he understood the primary ‘vindicatory’ aim of the award to be one of either ‘declaring’ or ‘enforcing’ the right. But literal enforcement was clearly not possible on the facts ex post facto and a declaration of the right to liberty was equally made and equally broadly publicised, surely, by the terms of their Lordships’ judgments and by a nominal monetary award. How could an award of £1000 realistically add anything meaningful to the mix? Lady Hale’s suggestion that an award would ‘mark the law’s recognition that a wrong has been done’,92 or ‘mark the invasion of important rights’93 similarly focuses on the declaratory function of the award and is subject to precisely the same criticism. Lord Walker’s explanation was briefest of all, being confined to the observation that the common law has long recognised that more than nominal damages should be granted ‘to vindicate an assault upon an individual’s person or reputation’ and that on these facts such an award was justified by way of analogy.94 He relied less on public law cases to justify the award than on these limited private law examples in tort, but his judgment sadly still gives no clue at all as to the precise sense in which he thought the award might vindicate, or indeed why that might be important. Looking beyond the terms of these judgments, one possible rationale for awarding a substantial measure of damages might have been to compensate the insult to dignity or sense of humiliation caused to the plaintiffs by the denial of their public law rights to have their detention determined on a proper procedural footing. Unlike the loss of liberty they suffered, this loss would presumably not have been incurred, had they been lawfully detained under the correct policy. If this is correct, it suggests that public law failings ought in themselves to give rise to a measure of compensation in some instances, irrespective of whether a cause of action is also available in private law. That possibility has certainly been canvassed in other contexts95 and it is effectively what courts are able to offer plaintiffs under the Human Rights Act in some instances. Indeed, Merris Amos has expressly suggested that procedural failings could and should give rise to damages awards for non-pecuniary harm (including ‘feelings of injustice’) under the Human 90

Ibid, at [178], [180]. Ibid. 92 Ibid, at [213]. 93 Ibid, at [216]. 94 Ibid, at [195]. 95 See P Cane, ‘Damages in Public Law’ (1999) 9 Otago Law Review 489; M Amos, ‘Extending the Liability of the State in Damages’ (2001) 21 LS 1. 91

The Mixed Concept of Vindication 89 Rights Act where a state breaches Article 6 of the European Convention.96 Unfortunately, however, while this explanation for giving modest sums seems potentially viable, there is no hint anywhere in the judgments in Lumba that ‘vindicatory’ damages were regarded as a way of compensating violations of the plaintiffs’ public law rights. Nor does it seem to have been argued that the plaintiffs lost any ‘chance’ of avoiding imprisonment by having the wrong policy applied to them—any such chance presumably being regarded as altogether too slim to register on a compensable scale, or indeed to be evidentially precluded by their admission that they would have been imprisoned under the published policy in any event. How, then, is the current confusion about the existence of ‘vindicatory damages’ as a distinct remedy in private law to be explained? The pressure to recognise them has stemmed, I suggest, from two, distinct sources. The first, to which we have already adverted in our discussion of exemplary awards, is the historical failure of the law sufficiently to broaden its recognised heads of compensable loss. If the law of torts had always recognised as ‘compensable’ the losses of autonomy, power, dignity and opportunity that often attend rights infringements, there would be considerably less incentive in the modern age for judges to seek to create a separate damages entity to reverse the effects of such infringements on plaintiffs. Compensatory damages would cover all of the open ground and there would be no remedial ‘gap’. This is not to say that every rights infringement should give rise to a substantial award, but it is to acknowledge that some rights infringements entail significant intangible losses, even where no material consequential loss is present and that the law is perfectly adequately equipped to deal with such events.97 Ironically, the capacity of private law to meet such challenges was demonstrated some time ago in a case with facts remarkably similar to those in Lumba: Roberts v Chief Constable of the Cheshire Constabulary.98 In that instance, Mr Roberts was again unlawfully imprisoned in circumstances in which he could have been perfectly lawfully detained. He was awarded a sum of £500 (coincidentally the same sum as suggested by Lady Hale in Lumba), but the award was referred to straightforwardly as compensatory, without any distinct reference to ‘vindicatory damages’. The correctness of that award has now, of course, been questioned by the majority approach in Lumba, but the point made here survives that criticism: if sums of money are required to redress the personal effects of affronts to dignity or losses of autonomy caused by misuses of public or private power, the category of compensatory damages is already available to that end.

96

Amos, above n 95, at 8–9, citing McMichael v United Kingdom (1995) 20 EHRR 266. See further, D Nolan, ‘New Forms of Damage in Negligence’ (1997) 70 MLR 59; A Ripstein, ‘As If It Had Never Happened’ (2007) 48 William and Mary Law Review 1957. 98 Roberts v Chief Constable of the Cheshire Constabulary [1999] EWCA Civ 655, [1999] 1 WLR 662. 97

90

Kit Barker

The second source of pressure that may have led to the current focus on vindicatory damages is the traditional (and in my view unfortunate) association of punitive damages in private law with retributive purposes. At several points in Lumba and in many of the public law cases involving vindicatory damages, close analogies are drawn between ‘vindicatory’ and punitive awards, the key point of distinction being identified as the fact that punitive damages aim to punish (exact retribution), whereas vindicatory awards do not, serving instead simply to ‘mark’ the right or wrong in question and to deter its recurrence.99 This is indeed seen to be a virtue of vindicatory awards and a possible reason why they may be available in circumstances in which punitive awards are not. In the modern day, however, ‘punishment’ in the specific sense of vindictive retribution upon the wrongdoer cannot realistically be regarded as the purpose of ‘punitive’ awards either, for reasons already discussed. Such awards are not an exercise in vengeance100 and it would be far better, for the avoidance of confusion, to recast and rebrand them universally as ‘exemplary’ awards. Their aim is better understood in terms of deterrence via example setting—a purpose which they potentially share with ‘vindicatory’ awards. If the function of punitive awards in private law were properly stipulated as purely deterrent, the perceived pressure to recognise vindicatory damages for similar purposes would immediately be relieved—it would be clear that whatever deterring needs to be done in cases involving the abuse of state power can already be done through exemplary awards, if courts are so minded. The phenomenon of vindicatory damages is thus in truth a parasite upon declaratory, compensatory and deterrent purposes that already flow through the veins of existing private law remedies. They have no normative life of their own. Indeed, this seems to have been true from the very earliest days when the expression ‘vindicatory’ damages was used by courts. In one of the US cases cited by Lord Collins in Lumba as being first to refer to the concept of vindicatory damages, Cole v Tucker, Hemphill Ch J hence explained the idea in the following way: That in actions for malicious trespasses on personal property, vindicatory, or corrective damages may be awarded, is a rule as old as it is firmly established. Compensation, in the legal and technical signification of the term, is not deemed a sufficient recompense for injuries of this character. Compensatory damages are given where the injury is not tainted with fraud, malice, or wilful wrong;

99 Lumba, above n 2, at [214]–[215] (Lady Hale), at [178] (Lord Hope), at [100] (Lord Dyson), and at [233] (Lord Collins) (doubting the distinction); Ramanoop, above n 12, at [19] (Lord Nicholls); Taunoa, above n 12, at [255]. 100 Some courts still refer to retributive purposes in a mixed rationalisation, but if punitive awards are to be retained to this end in private law, they should be clearly distinguished from ‘exemplary’ awards, the purpose of which should be deterrent only. The point in the text then remains valid. Provided exemplary awards are available purely to deter, we do not need vindicatory awards to the same end.

The Mixed Concept of Vindication 91 but where either of these elements intervene, another ingredient is added to the ordinary constituents of injury, viz: the sense of wrong and insult; and damages are given as well for compensation to the sufferer, as for the punishment of the offender. In the language of Judge Baldwin, where trespass is committed in a wanton, rude or aggravated manner, indicating a desire to injure, a jury ought to be liberal in compensating the party injured, in all he has lost in property, in expenses for the assertion of his rights, in feeling or reputation; and even this may be exceeded, by setting a public example, to prevent the repetition of the act. In such cases there is no certain fixed standard; for a jury may properly take into view, not only what is due to the party complaining, but to the public, by inflicting what are called, in law, speculative, exemplary or vindictive damages.101

The word vindication in the first, highlighted, sentence of this passage is used in association with—indeed perhaps as a synonym for—correction (reversing the effects of a rights infringement). Compensation in a ‘technical sense’ is understood as one component of correction, but not to be sufficient to achieve it where there is a ‘sense of wrong or insult’ or an attempt by a defendant to degrade the plaintiff—what today might be regarded as loss of dignity or humiliation compensable by an aggravated compensatory damages award. The second association that appears in the second, highlighted section of the passage is between vindicatory damages and the deterrent purposes of punishment—purposes that are clearly met by exemplary awards. As it turns out, then, ‘vindicatory’ damages did not, even in their original conception, set out to ‘vindicate’ rights in any way that is not already familiar to us, or currently within the law’s relatively easy reach. The result is that the Supreme Court in Lumba was right to reject the heterodox of ‘vindicatory damages’ in tort law. Ironically, this is not because the idea poses some radical new vision of the purposes of remedies in private law that is provably unacceptable, but because it adds nothing to our existing vision of those purposes and nothing to our capacity to fulfil them.

V. CONCLUSIONS

Vindication in common parlance is an emotive term. It is powerful for precisely that reason. Like all powerful instruments, however, the idea must be used responsibly and with due reflection upon its meaning and purposes, not simply because rights-speak is currently de rigeur. In the above account, I have sought to provide a description of the concept which is true to its institutional (legal) use, without necessarily committing to a particular normative theory of how the idea should be understood. The first main conclusion to be drawn is that there is no singular conception in either private or public law of what it means to ‘vindicate’ rights, 101

Cole, above n 12, at 268 (emphasis added).

92

Kit Barker

nor indeed any singular understanding of what the purpose of vindicating rights might be. In this sense, although the orthodox view that private law vindicates rights is often characterised as clear and uncontroversial, it desperately needs to be regarded with rather closer attention to detail. The idea bleeds heavily when its skin is scratched. Our understanding of the orthodoxy is best advanced, I have suggested, by recognising and discriminating in a rigorously disciplined manner between four different types of ‘vindication’ effected by private law; and by acknowledging that, while most of these sit comfortably with mainstream objectives of interpersonal justice, some judges regard some types of vindication as important for reasons which are overtly public. It was thus a public, symbolic role of private law in ‘marking rights’ that was foremost in Lord Hope’s mind in Lumba. Such uses of tort law for the purposes of symbolic gesture are not without controversy and their implications and costs need to be very carefully considered. In one way, of course, everything the law does in respect of rights leaves a footprint in the sand, demonstrating to the world that justice has walked by. There is nothing wrong in that and it is to be regarded as both welcome and reassuring. The symbolic effects of judicial acts are inevitable and no more (or less) to be regretted than the acts themselves. Constructing public symbolism as a distinct purpose of tort law (or private law remedies in general) is, however, a much more controversial enterprise and requires some very serious thinking before it is embraced. In this piece, I have not so much fully explored this path of thinking, as raised a flag to some of its potential pitfalls and controversies. Symbols are always subject to misinterpretation (might a £500 award for false imprisonment not be thought blatantly insulting, not affirming?) and money is rarely the most articulate mode of expression, even if it talks powerfully to some people in a motivational sense. There is also an obvious way in which the idea of using private law for symbolic, public messaging purposes blows directly in the face of modern civil justice reforms in almost all jurisdictions. The general aim and increasingly dramatic effect102 of these is to keep cases out of court whenever possible and to achieve speedy, privatised, informal solutions to particular disputes. The de-formalisation of justice that is so often applauded in the name of efficiency and equality of access in the twentyfirst century is surely in acute tension with any symbolic use of private law as an institutional, judicial expression of right and wrong. The more space we provide for institutional declarations of right, the more we risk undermining incentives for parties to resolve their disputes out of court and the more we risk wasting public money. This ought, I think, to be borne carefully in mind by any who suggest that private law has a symbolic purpose.

102 On the extent of these effects in the United States, see J Langbein, ‘The Disappearance of Civil Trial in the United States’ (2012) 122 Yale Law Journal 522.

The Mixed Concept of Vindication 93 In public law, it may be that things are different and that public symbols are more clearly central to the enterprise in question.103 It may also be that there are costs associated with the de-formalisation of justice that have not yet been properly assessed and which need to be, if a proper balance is to be struck between doing ‘public’ justice on the one hand and achieving private dispute resolution on the other. I deliberately make no comment on these questions. My intent here is not so much to stake out a place in the debate as to make the obvious point that it needs to be resolved before we can possibly commit to a view that the formal declaration of right and wrong is itself a purpose of private law that actually extends beyond—or supersedes—the interests of individuals using the system. The second main conclusion I have drawn is that the majority of the Supreme Court in Lumba were correct to reject the heterodox of ‘vindicatory’ damages as a distinct remedy in private law because, whatever type of vindication needs to be done, our law already has the wit to do it. If there is any remedial ‘gap’ in tort law’s arsenal, it can be made good through the development of perfectly traditional means—in particular, by broadening our conceptions of actionable loss for the purpose of awarding compensatory damages in a way that is better attuned to the types of interest the law needs to protect in the modern day; and perhaps also by clarifying the purely deterrent role of exemplary damages awards in tort cases, such that their potential for non-retributive but meaningful use is made more apparent. There is no reason to doubt that either of these developments is possible. Indeed, there seems much to commend them. Any further multiplication of concepts or ‘types’ of damages seems ill advised, however benevolent the intention. The closing irony is then this: while the orthodox idea that private law vindicates rights probably ought to be more controversial to most readers than it currently is, the heterodoxy of ‘vindicatory damages’ raised in recent tort cases founders on what is for all revolutionaries the most embarrassing of reasons—the idea’s own functional redundancy. Surely, nothing takes the wind from one’s sails faster than the realisation that one is really not needed in the achievement of one’s own ends.

103 For consideration of analogous issues arising in public law in the context of judicial review and actions under the HRA, see J Miles, ‘Standing under the Human Rights Act 1998: Theories of Rights Enforcement and the Nature of Public Law Adjudication’ [2000] CLJ 133. Writing before the Act came into force, the author argues that widening the standing requirements under the Act in accordance with a communitarian model of rights enforcement could (and perhaps should) give greater scope to the ‘expository’ function of courts in public law proceedings. Note, however, that some considerable scepticism has been expressed about ‘expressive’ theories of public law—see Adler, above n 86. See also the concerns expressed about the disproportionate expense of judicial review proceedings on some facts in MD, above n 84.

4 ‘We Do This in the Criminal Law and That in the Law of Tort’: A New Fusion Debate GRAHAM VIRGO

I. INTRODUCTION

T

HE OBJECTIVES OF the criminal law and the law of tort are fundamentally different. Criminal law is public, punitive law, which exists to maintain security through the control of certain forms of behaviour.1 The law of tort, on the other hand, is concerned with the identification of personal rights and the award of remedies to vindicate those rights. As Coffee recognised, ‘tort law prices, while criminal law prohibits’.2 But despite these obvious differences of objective, the concepts deployed by criminal and tort law have much in common. There are, for example, several wrongs which have a common origin and which can trigger both criminal and tortious liability, such as assault, battery and theft/conversion. Crimes involving interference with property rights clearly depend on civil notions of property and proprietary rights and a significant number of other offences are defined with reference to concepts used in tort law.3 Equally, there are situations where the definition of a cause of action in tort has been influenced by concepts as they are defined in the criminal law.4 It might, therefore, be thought obvious that a rational legal system should seek to ensure, at the very least, consistency between the two legal regimes. But orthodoxy states that the different function of criminal law and the law of tort means that the search for consistency of interpretation and

1

G Williams, ‘The Definition of Crime’ (1955) 8 Current Legal Problems 107. JC Coffee, ‘Does “Unlawful” Mean “Criminal”?: Reflections on the Disappearing Tort/ Crime Distinction in American Law’ (1991) 71 Boston University Law Review 193, 194. 3 See, eg, the English Fraud Act 2006, s 3 of which creates an offence of dishonestly failing to disclose information where there is a legally enforceable civil obligation to do so. 4 Most notably dishonesty. See Part VI below. 2

96

Graham Virgo

approach is not an appropriate objective of the law. This was recognised explicitly by Lord Scott in Ashley v Chief Constable of Sussex Police,5 who said: this plea for consistency between the criminal law and the civil law lacks cogency for the ends to be served by the two systems are very different. One of the main functions of the criminal law is to identify, and provide punitive sanctions for, behaviour that is categorised as criminal because it is damaging to the good order of society ... The function of the civil law of tort is different. Its main function is to identify and protect the right that every person is entitled to assert against, and require to be respected by, others. The rights of one person, however, often run counter to the rights of others and the civil law, in particular the law of tort, must then strike a balance between the conflicting rights.

Similarly, Coleman recognised that the ‘differences between torts and the criminal law are so fundamental that the net result of applying one’s understanding of the criminal law to torts is bad philosophy and total confusion’.6 A different approach to the relationship between criminal and civil law had been advocated by JC Smith in a seminal article published 40 years ago.7 Smith considered the impact of civil law concepts on the interpretation of certain crimes, notably the property offences, and concluded that, for those offences at least, ‘the civil law of property must be accepted as the essential background and raison d’être of the criminal statute which should not be construed so as to alter civil law rights’, and that ‘the criminal law must not impose duties which are inconsistent with rights, stricto sensu, or duties in the civil law’.8 Most significantly he recognised that: A complete separation between the civil law and the criminal law is neither possible nor desirable. If the law is not to be brought into disrepute, there must at least be such consistency between the two branches of the law that the Civil Division of the Court of Appeal does not find itself holding a transaction to be enforceable by a plaintiff whilst the Criminal Division is sending him to prison for entering into it. It would be intolerable if the criminal law, on the one hand, and the law of tort and contract, on the other, were to impose conflicting duties on the subject.9

Forty years on, and in the light of Lord Scott’s contradictory position in Ashley, it is important to consider whether orthodoxy can be defended, 5 Ashley v Chief Constable of Sussex Police [2008] UKHL 25, [2008] AC 962, [17]–[18] [Ashley]. See also Lord Carswell, [76]. 6 J Coleman, Risks and Wrongs (Oxford, Oxford University Press, 1992) 222. See also GH Treitel, ‘Contract and Crime’ in C Tapper (ed), Crime, Proof and Punishment: Essays for Sir Rupert Cross (London, Butterworths, 1981) 82 for similar views about the relationship between criminal and contract law. 7 JC Smith, ‘Civil Law Concepts in the Criminal Law’ (1972) 31 CLJ 197. 8 Ibid, at 224. 9 Ibid, at 197.

A New Fusion Debate

97

namely whether criminal and tort law are so different that seeking consistency of approach is not a legitimate aim. Crucially, to what extent should the interpretation of concepts in the law of tort be applied in the criminal law and, just as important but much less analysed, to what extent should the interpretation of concepts in the criminal law be applied in tort law? But this chapter is not simply concerned with the interpretation of common concepts. It seeks to examine the more difficult question as to whether structural and theoretical developments in one system should legitimately be considered in the other to assist in the development and rationalisation of the law. In particular, should it matter that the defendant might be prosecuted for doing something which he or she has a civil law right or even duty to do? This chapter aims to start a new debate about fusion of criminal and tort law. Any suggestion that there should be complete fusion of these systems would be absurd, due to their fundamentally different objectives and distinct procedures. But the division between criminal and tort law is too easily stated without careful justification. It is no longer sufficient to say that criminal and tort law are just different, without explaining why those differences exist. When such matters are considered by the courts, such reasons are rarely identified and, when they are, tend to be unconvincing. If no such reason can be identified then, despite Lord Scott’s dictum in Ashley, consistency of interpretation and reasoning should be sought between criminal and tort law. This is what actually happens in another context where there has been a fusion debate, namely that which concerns the substantive relationship between common law and equity. To the extent that common concepts are used at common law and in equity, it has sometimes been assumed that any rational body of law would interpret those concepts in the same way, so that fusion of interpretation is essential. But the actual relationship between the two regimes is more subtle.10 It requires differences to be justified and, if they cannot, then assimilation is appropriate, with equity prevailing where there is conflict or variance between the rules.11 So, for example, differences in the interpretation of tracing rules at common law or in equity cannot be justified, and neither can differences relating to the operation of the remedy of rescission at common law and equity,12 so in both situations assimilation is appropriate. But it does not follow that fundamental differences relating to the nature of proprietary interests at common law and in equity must also be removed, so that, for

10

See A Burrows, ‘We Do This in the Common Law but That in Equity’ (2002) 22 OJLS 1. Judicature Act 1873, s 25(11); see now Senior Courts Act 1981, s 49. 12 See Halpern v Halpern (No 2) [2007] EWCA Civ 291, [2007] 3 All ER 478, [70] (Carnwath LJ). 11

98

Graham Virgo

example, the trust should disappear. The vital and distinct functions of the trust and equitable proprietary interests can be justified.13 Of course, the fusion debate relating to common law and equity is fundamentally different from any debate involving ‘fusion’ of criminal and tort law, because common law and equity have been procedurally fused for many years and the different objectives of criminal and tort law mean that any complete procedural fusion would be inappropriate. Also, equity developed as a qualification of the common law, whereby the strict common law rules were tempered by equity to avoid injustice. The distinct functions of criminal and tort law militate against one body of law tempering the operation of the other. But, like a Venn diagram, there is undoubted scope for their operation to overlap and for one to influence the development of the other. The real issue in this chapter is to examine the extent of that overlap and to determine, within this area of overlap, whether the interpretation and operation of the law is consistent. In reality the issue is not one of fusion or assimilation but one of cross-fertilisation and influence.

II. THREE MODELS

There are three possible models for the relationship between criminal and tort law. They are not necessarily exclusive in their operation but exemplify different modes of thinking about the relationship between them.

A. Separate Bodies of Law The most extreme model is that, by virtue of their different objectives, neither criminal nor tort law should be of any relevance to the other as regards the interpretation and development of the law. It is this model which was apparently adopted by some of the judges in Ashley in deciding that the defence of self-defence in tort operates differently from that defence in criminal law.14 As Lord Bingham recognised, ‘There is no reason in principle why it should be the same test as obtains in a criminal trial, since the ends of justice which the two rules respectively exist to serve are different’.15 But this assumption that the criminal and civil law are simply different can cause unnecessary confusion and complication in the law. This

13 See G Virgo, The Principles of Equity and Trusts (Oxford, Oxford University Press, 2012) 22–25. 14 Above n 5. 15 Ibid, at [3]. See also Lord Scott at [17].

A New Fusion Debate

99

is illustrated by R v Smith,16 where the defendant appealed against his conviction for robbery. He had used force to steal heroin from a drug dealer. The issue for the court was whether it was possible to be convicted of stealing something which it is unlawful for anyone to possess; specifically whether the possession of prohibited drugs which were unlawfully in the victim’s possession could constitute property belonging to another for the purposes of the Theft Act 1968. The basis of the defendant’s argument on appeal was that the owner or possessor of drugs could have no rights in them which were capable of being assumed by the person who deprived the victim of the drugs; in other words, the defendant had not infringed the victim’s property rights. It was held that for purposes of theft, property did not cease to belong to the victim because its possession or control was unlawful. It was recognised that: ‘The question of how litigation in the civil courts, brought by a person seeking damages for wrongful deprivation of drugs unlawfully in his possession from the person who removed the drugs from him, would operate is utterly irrelevant.’17 The key principle was considered to be, in the words of Smith’s Law of Theft, that: public policy which prevents the wrongdoer from enforcing a property right should have no application to criminal proceedings brought in the name of the Crown. The criminal law is concerned with keeping the Queen’s peace, not vindicating individual property rights.18

Consequently, the court held, the defendant was correctly convicted of stealing the drugs. But this produces a consequential inconsistency between criminal and tort law. For it was assumed that the defendant would not have been liable for conversion of the drugs for reasons of public policy, but he was still guilty of stealing them. Legitimate reasons need to be found for this inconsistency of approach, but the court did not attempt to find any. In fact, the inconsistency might have been justified because although the defendant had interfered with the victim’s property rights, the victim would not have been able to vindicate those rights as a result of his own participation in illegal activity. In other words, there was a potential claim on the facts but it would have been defeated by the illegality defence as against the victim because of his own participation in criminal activity. But the court did not consider it necessary to engage in such an analysis, it being sufficient to assert that the case involved the criminal law and not the law of tort.

16

R v Smith [2011] EWCA Crim 66, [2011] 1 Cr App R 379 [Smith]. Ibid, at [9]. 18 D Ormerod and DH Williams (eds), Smith’s Law of Theft, 9th edn (Oxford, Oxford University Press, 2007) 80, quoted ibid, at [10]. 17

100 Graham Virgo B. Hierarchy A second model recognises that whilst the functions of criminal and tort law are different, they can coexist in an elegant hierarchy with tortious liability being more generally applicable and the criminal law operating in specific cases where the defendant’s conduct can be considered to be particularly serious and blameworthy. This is especially well illustrated by liability for causing death by negligence. The crime of gross negligent manslaughter19 will have been committed where the defendant has caused death in the following circumstances: (i)

The defendant owes the victim a duty of care under the ordinary principles of the tort of negligence, which is a question of law for the judge.20 The law relating to the tort of negligence is applied to determine whether a duty of care was owed, although some of the implications of tortious liability are not relevant. So, for example, if the defendant and the victim are engaged in a joint criminal enterprise, such as illegal immigration or arson, the defendant might be guilty of gross negligent manslaughter irrespective of the fact that their joint participation in an illegal activity would preclude a successful tortious action in negligence by the victim’s estate against the defendant.21 (ii) The duty of care has been breached and this caused the death of the victim. (iii) The negligence of the defendant ‘went beyond a mere matter of compensation between subjects and showed such disregard for the life and safety of others, as to amount to a crime against the state and conduct deserving of punishment’.22 Where these three conditions are satisfied the otherwise tortious liability, which will still usually operate, will be supplemented by the imposition of criminal liability. So, for example, in R v Adomako the defendant anaesthetist had failed to notice that his patient was suffocating during a routine eye operation because a tube had become disconnected.23 The patient died. The defendant was convicted of manslaughter because his negligence was so gross that the imposition of criminal liability was considered to be appropriate. As Lord Atkin recognised in Andrews v Director of Public 19 See A Simester, J Spencer, G Sullivan and G Virgo, Simester and Sullivan’s Criminal Law: Theory and Doctrine, 4th edn (Oxford, Hart Publishing, 2010) 409–14. 20 R v Evans [2009] EWCA Crim 650, [2009] 1 WLR 1999. 21 This involves a similar relationship between tort and crime as in Smith, above n 16, at [4]. 22 R v Bateman (1925) 19 Cr App R 8, 11–12 (Lord Hewart CJ). See also Andrews v Director of Public Prosecutions [1937] UKHL 1, [1937] AC 576, 582–83 (Lord Atkin) and R v Adomako [1994] UKHL 6, [1995] AC 171 [Adomako]. 23 Adomako, above n 22.

A New Fusion Debate

101

Prosecutions, simple lack of care which is sufficient for the imposition of civil liability will not be sufficient for the criminal law. There are ‘degrees of negligence: and a very high degree of negligence is required to be proved’ before the crime is established.24 It is this additional culpability which justifies the expansion from tortious to criminal liability.

C. Consistency The third model, whilst acknowledging that criminal and tort law have different objectives, recognises that it is a legitimate aim of the legal system to seek consistency of interpretation and application of common rules and concepts where possible and appropriate. The validity of this principle of consistency was explicitly recognised by McLachlin J in the decision of the Supreme Court of Canada in Hall v Hebert:25 the law must aspire to be a unified institution, the parts of which—contract, tort, the criminal law—must be in essential harmony. For the courts to punish conduct with one hand while rewarding it with the other would be to ‘create an intolerable fissure in the law’s conceptually seamless web’.26

It follows logically that, as McLachlin J acknowledged, it is not appropriate to treat the defendant’s conduct as criminal but also to ‘reward’ him or her by recognising rights which are enforceable in the civil law. Similarly, and logically, it would not be appropriate to give the defendant a defence to potentially criminal conduct but to hold him or her civilly liable for that same conduct without very good reason for doing so. Where there is such inconsistency it can be resolved by amending either the criminal or tort law. But there is no general assumption that the criminal or tort law approach should prevail, as there is where there is a conflict between common law and equity.27 As Lord Steyn has recognised, ‘it would be wrong to assume on a priori grounds that the criminal law rather than the civil law is defective’.28 Rather, where there is unjustified inconsistency between tort and crime that inconsistency needs to be resolved one way or the other. If there are no other considerations which can assist, then it would be appropriate to remove the inconsistency by adopting the interpretation which negates liability. In fact, however, where the inconsistency has been removed by statute, it has typically resulted in the expansion of the criminal law. So, for example, trespass by squatters has historically triggered 24

Above n 22, at 583. Hall v Hebert [1993] 2 SCR 159, 165. 26 Ibid, quoting from EJ Weinrib, ‘Illegality as a Tort Defence’ (1976) 26 University of Toronto Law Journal 28, 42. 27 See above text accompanying nn 10–13. 28 R v Hinks [2000] UKHL 53, [2001] 2 AC 241, 252 [Hinks]. 25

102 Graham Virgo tortious but not criminal liability. But in 2012 a new offence of ‘squatting in a residential building’ was created,29 resulting in a significant expansion of criminal liability to encompass a person who enters a residential building as a trespasser with intent to live in the building. It does not follow from the recognition of the principle of consistency that differences between criminal and tort law must necessarily be removed, for that would be to ignore the different objectives of the two bodies of law. But the differences must be acknowledged and carefully justified by reference to the distinct principles which underpin criminal and tort law. As Lord Steyn noted in R v Hinks: ‘The purposes of the civil law and the criminal law are somewhat different. In theory the two systems should be in perfect harmony. In a practical world there will sometimes be some disharmony between the two systems’.30 One of the best examples of an inconsistency being justified is the decision of the House of Lords in Ashley,31 despite the dicta of Lords Bingham and Scott, which suggest that this case fell within the first model, separate bodies of law. In fact, it is a perfect example of this more subtle third model being adopted. Ashley concerned the application of self-defence in a tort claim for assault and battery. Although the House of Lords held that the defence in tort is interpreted differently from the way it is interpreted in the criminal law, it was acknowledged that most of the elements of the defence were the same. As Lord Rodger recognised: ‘the civil and criminal law would look at that situation in the same way: it would be absurd to say that the person under attack was justified in killing his assailant, but nevertheless potentially liable in damages’.32 Nevertheless, there were some differences in the operation of the defence and the House of Lords justified these differences carefully.33 Ashley arose from the fatal shooting of an unarmed victim by a police officer following a raid on the victim’s house. The police officer was charged with murder but pleaded self-defence and was acquitted. In the criminal law, self-defence will be available to the defendant if he or she honestly but mistakenly thought it was necessary to defend himself or herself against an imminent risk of attack.34 The victim’s relatives brought a tort claim against the police force. One of the key issues for the court was whether self-defence was available to this claim, where the police officer had made 29

Legal Aid, Sentencing and Punishment of Offenders Act 2012, s 144. Above n 28, at 252. This was a case where dissonance between criminal and tort law was not justified. See below text accompanying nn 76–82. 31 Above n 5. 32 Ibid, at [51]. 33 Lord Neuberger, ibid, at [87] recognised that the ‘criminal law and civil law of battery substantially march together’ (emphasis added). 34 R v Williams (Gladstone) [1987] 3 All ER 411 (CA); Beckford (Solomon) v R [1988] AC 130 (PC); Martin (Anthony) v R [2001] EWCA Crim 2245, [2003] QB 1. See also the Fourteenth Report of the Criminal Law Revision Committee, Offences Against the Person (1980) (Cmnd 7844), [281]–[283]. 30

A New Fusion Debate

103

an honest but unreasonable mistake as to the need to use force to defend himself from an imminent attack. Even though Lords Bingham and Scott emphasised that there was no reason in principle why the defence should be interpreted in the same way in the criminal and civil law, the court recognised that in both cases the defence will only be available if it was reasonable for the defendant to have acted as he or she did in the light of the facts as he or she believed them to be.35 So the defence would be available in both the criminal and civil law even if there was no imminent threat to the defendant but he or she was mistaken as to the need to use force. It was held, however, that in the civil law the mistake had to be reasonable,36 whereas in the criminal law an unreasonable mistake would suffice.37 The justification given for this difference38 was that nobody should be punished for a crime which he or she did not intend to commit or for the consequences of an honest mistake, especially because of the increasing emphasis in the criminal law on the principle of individual responsibility, which requires proof of subjective culpability.39 The function of tortious liability was considered to be different, namely to identify and protect rights and to strike a balance between conflicting rights.40 As regards the torts of assault and battery, it was recognised that everybody has the right not to be subjected to physical harm by the intentional acts of another, but they also have the right to protect themselves by using reasonable force to prevent an imminent attack. In striking the appropriate balance between these two conflicting rights, it was considered appropriate to qualify the right not to be subjected to physical harm to situations in which the defendant’s belief as to the need to use force to protect himself or herself was reasonably held.41 The approach of the House of Lords in Ashley is appropriate and defensible. Although the result of the decision is to create a dissonance between criminal and tort law, this is conscious and marginal, and it operates in favour of the defendant in that he or she is acquitted of a crime even though he or she remains civilly liable. This does not involve, to use Weinrib’s words quoted earlier in Hall v Hebert, an ‘intolerable fissure in the law’s 35

Ashley, above n 5, at [16] (Lord Scott), [53] (Lord Rodger). See also Gambriell v Caparelli (1974) 54 DLR (3d) 661 (Ont Co Ct). Lord Scott in Ashley did acknowledge, however, that he would have preferred to decide that mistake should not have been relevant at all, even if it had been reasonable, but this had not been argued in the appeal. Lord Rodger reserved his opinion on this point and Lord Carswell disagreed. 37 In fact, even in the criminal law unreasonable mistakes are not always relevant, such as where the defendant relies on the excuses of duress of threats or circumstances: R v Graham [1981] EWCA Crim 5, [1982] 1 WLR 294; R v Hasan [2005] UKHL 22, [2005] 2 AC 467, or where the mistake relates to the consent of the victim in the crime of rape: Sexual Offences Act 2003, s 1. At one stage even for self-defence in the criminal law any mistake had to be reasonable: R v Fennell [1971] 1 QB 428 (CA) 431 (Widgery LJ). 38 Ashley, above n 5, at [17] (Lord Scott). 39 Ibid, at [76] (Lord Carswell). 40 Ibid, at [18] (Lord Scott), [76] (Lord Carswell), [86] (Lord Neuberger). 41 Ibid, at [18] (Lord Scott). 36

104 Graham Virgo conceptually seamless web’, which would arise if the application of the rules were reversed so that the defendant was convicted of the crime if the mistake had been unreasonable but found not liable for the tort by virtue of such a belief. The state of the law on self-defence is also consistent with the hierarchical model of the relationship between criminal and tort law, since the liability in tort is established but the effect of the defendant’s mistake, albeit unreasonably held, is not sufficient to justify the imposition of additional criminal liability. This is coherent and sensible and also involves the application of the principle of consistency. This is because the defences are generally defined in the same way regardless of whether they are being deployed in criminal or tort law. Where they are different, that difference is justified by virtue of the divergent aims of punishment and compensation, but with the result that criminal liability is more restrictive than liability in tort. Procedural and structural differences between criminal and tort law may lead to other differences in result which do not undermine the principle of consistency because they can be justified on a principled basis. So, for example, the fact that the defendant has been acquitted of a crime does not mean that he or she cannot be held liable for a tort on the same facts, if only because the standard of proof for a criminal conviction is higher than for the imposition of civil liability. It was for this reason that in Ashley it was held not to be an abuse of process to hold the Chief Constable liable for the tort of battery even though the defendant had been acquitted of murder.42

D. Application of the Models As the subsequent parts of this chapter illustrate, the relationship between criminal and tort law can be examined in a wide variety of different contexts. In doing so it will be seen that the second and third models generally prevail, although the courts’ usual failure to recognise the principle of consistency often makes it appear that the judges only recognise the first model.

III. ILLEGALITY

Probably the most significant area where the definition of concepts in the criminal law has a direct impact on tort law relates to the operation of the

42 Above n 5, at [66] (Lord Rodger), [79] (Lord Carswell). See also Hunter v Chief Constable of the West Midlands Police [1981] UKHL 13, [1982] AC 529 and Raja v Van Hoogstraten [2005] EWHC 2890 (Ch) (where the defendant’s conviction for manslaughter had been quashed on appeal for technical reasons but he was still found civilly liable for the homicide in a trial where his previous convictions were admissible).

A New Fusion Debate

105

defence of illegality, otherwise known as the ex turpi causa rule.43 Although it is clear that the definition of illegality for the purposes of the law of tort is not coextensive with the criminal law, since an act can be considered to be illegal even though it is not criminal simply because it is considered to be contrary to public policy,44 nevertheless the most significant reason why conduct will be treated as illegal will be because it involves the commission of a crime. It is then necessary to determine whether the fact that conduct is criminal should have an impact on liability in tort, such that the illegality operates as a defence to the claim. In determining this, the principle of consistency is plainly important. The operation of the illegality defence in tort was considered by the House of Lords in Gray v Thames Trains Ltd,45 where Lord Hoffmann recognised that ex turpi causa was a policy which was implemented in a wide and a narrow form.46 In its wide form, the effect of the defence is that the claimant cannot recover for loss which was a consequence of his or her own criminal act.47 This version of the defence is not motivated by the principle of consistency,48 but can be justified instead on the ground that ‘it is offensive to public notions of fair distribution of resources that a claimant should be compensated (usually out of public funds) for the consequences of his own criminal conduct’.49 Further, this principle will only apply where the loss was caused by the claimant’s criminal conduct. Where the loss cannot be considered to have arisen from the commission of a crime, the illegality defence will not apply. So, for example, in Delaney v Pickett the claimant had been seriously injured in a car accident.50 His claim for damages against the driver for negligent driving was rejected by the trial judge because the car in which the claimant was being driven was being used to collect and transport illegal drugs for sale, so the illegality defence was considered to be applicable. However, the Court of Appeal held that the defence did not apply because the loss could not be considered to have

43 As recognised in Holman v Johnson (1775) 1 Cowp 341, 98 ER 1120; Tinlsey v Milligan [1994] AC 340 (HL). 44 See HG Beale (ed), Chitty on Contracts, 31st edn (Andover, Sweet & Maxwell, 2012) [16-03] et seq. Conduct which is illegal as being contrary to public policy but not criminal includes conduct relating to prostitution. See, eg, Girardy v Richardson (1793) 1 Esp 13, 170 ER 265. 45 Gray v Thames Trains Ltd [2009] UKHL 33, [2009] AC 1339 [Gray]. 46 Ibid, at [32]. 47 In Stone and Rolls Ltd v Moore Stephens [2009] UKHL 39, [2009] AC 1391, [20] [Stone and Rolls] Lord Phillips described the principle as preventing a claimant ‘from using the court to obtain benefits from his own illegal conduct’. This only applies where the claimant’s liability is primary rather than vicarious: ibid, at [28]. 48 Gray, above n 45, at [51] (Lord Hoffmann). 49 Ibid; Delaney v Pickett [2011] EWCA Civ 1532, [2012] 1 WLR 2149, [35] (Ward LJ). See Vellino v Chief Constable of the Greater Manchester Police [2001] EWCA Civ 1249, [2002] 1 WLR 218. 50 Ibid.

106 Graham Virgo been caused by the criminal activity; rather it only provided the occasion for the defendant’s tortious conduct. The narrower form of the defence means that the claimant cannot recover for loss which is the consequence of a punishment, such as loss of liberty or a fine, which is imposed on the defendant for a criminal act.51 It is this form of the defence which can be explained by reference to the principle of consistency, since it is the criminal law which causes the loss as a matter of penal policy and it would be inconsistent with that law for the civil law to allow the criminal to be compensated for that loss.52 As Denning LJ said in Askey v Golden Wine Co Ltd: In every criminal court the punishment is fixed having regard to the personal responsibility of the offender in respect of the offence, [and] to the necessity for deterring him and others from doing the same thing again … these objects would be nullified if the offender could recover the amount of the fine … from another by process of the civil courts.53

The operation of both forms of the defence is illustrated by Gray itself. The claimant had killed a pedestrian as the result of post-traumatic stress arising from his involvement in a train crash, which had been caused by the defendant train company’s negligence. The claimant was convicted of manslaughter on the grounds of diminished responsibility. Despite this, he was still considered to be sufficiently responsible for his actions so that the effect of the ex turpi causa defence was that he could not recover from the train company for loss of earnings whilst he was detained in a mental hospital. This was because the loss arose as a consequence of a sentence imposed for a crime, so the narrow version of the defence was applicable. The House of Lords explicitly justified this conclusion by the need to ensure consistency in the justice system.54 Neither could the claimant be indemnified by the train company against potential claims by the deceased’s dependents by virtue of the wider version of the defence, namely that it was not possible to recover damages for loss which arose from the claimant’s own criminal act.

51 Askey v Golden Wine Co Ltd [1948] 2 All ER 35 (KB) 38 (Denning J) [Askey]; Clunis v Camden and Islington Health Authority [1997] EWCA Civ 2918, [1998] QB 978. See J Goudkamp, ‘Can Tort Law be Used to Deflect the Impact of Criminal Sanctions? The Role of the Illegality Defence’ (2006) 14 Torts Law Journal 20. 52 Gray, above n 45, at [37] (Lord Hoffmann). See also Delaney v Pickett, above n 49, at [35] (Ward LJ); The Illegality Defence in Tort (Law Commission No 160, 2001) [4.100]; State Rail Authority of New South Wales v Wiegold (1991) 25 NSWLR 500 (SC) 514 (Samuels JA); British Columbia v Zastowny 2008 SCC 4, [2008] 1 SCR 27, [23] (Rothstein J). Weinrib has justified this as preventing the ‘stultification of the criminal law’: above n 26, at 52. 53 Above n 51, at 38. 54 Lords Phillips and Rodger recognised that it would have been different had the defendant been detained in a hospital by virtue of his mental condition rather than as a result of his criminal act, for then there would have been no inconsistency in allowing the tort claim to succeed.

A New Fusion Debate

107

The more general operation of the illegality defence was considered in the important decision of the Court of Appeal in Les Laboratoires Servier v Apotex Inc.55 In that case the holder of a patent had obtained an interim injunction prohibiting the appellant from selling products which infringed the patent. The patent was later held to be invalid and the injunction was discharged. The appellant brought a claim on a cross-undertaking in damages arising from the loss it had suffered by virtue of its not being able to sell goods in England because of the injunction. The problem was that these goods would have been manufactured in Canada, which would have been in breach of a patent that was, and remained, enforceable there. The question on appeal was whether the award of damages would infringe the principle of illegality, since comity required the English courts to have regard to the unlawfulness of the defendant’s conduct in manufacturing the goods in Canada. The nature of the unlawfulness in Canada is significant. The manufacture of the goods was not criminal there and did not involve any dishonest actions, but would have involved the commission of a strict liability tort. The appellant had argued that the illegality defence should only operate where there was culpability on the part of the claimant, which would be the case either where the wrongdoing was manifest (such as in murder) or the claimant had knowledge of all of the ingredients of the wrongdoing at the time the acts in question were committed. Consequently, it was argued, if the appellant honestly believed on reasonable grounds that the Canadian patent was invalid, it could not be considered to have acted with sufficient culpability to engage the illegality defence. Whilst the Court of Appeal agreed that the claim for damages was not defeated by the illegality defence, it adopted a more flexible and less dogmatic approach to the analysis of illegality than the appellant had advocated, albeit one that was explicitly not founded on vague notions of the public conscience.56 Etherton LJ sought to identify the policy considerations which underlie the illegality defence, as follows: ‘furthering the purpose of the rule which the illegal conduct has infringed; consistency; the claimant should not profit from his or her own wrong; deterrence; and maintaining the integrity of the legal system’.57 The court’s approach to determine the effect of illegality was to analyse the facts rigorously, to apply the policy considerations and to produce a just and proportionate response to the illegality. On the facts, the illegality 55 Les Laboratoires Servier v Apotex Inc [2012] EWCA Civ 593, [2012] WLR (D) 138 [Apotex]. 56 Ibid, at [63] (Etherton LJ). 57 Ibid, at [66], relying on the Law Commission’s report on The Illegality Defence (Law Commission No 320, 2010). For Laws LJ there were only three policies (at [98]): turpitude, consistency in the law and no general discretion exercisable by reference to the public conscience.

108 Graham Virgo defence was held to be inapplicable because: (i) although the unlawful act in Canada was the manufacture of goods there in breach of the patent, which was a statutory wrong in Canada regardless of the infringer’s state of mind, the applicant had acted honestly and had believed reasonably that the patent was invalid; (ii) since liability was strict, this was considered to be low on the scale of culpability; and (iii) the sale of the goods would not have been unlawful either in Canada or the United Kingdom. This approach is significant, particularly because of the explicit recognition of the policies of consistency and integrity of the legal system, the former of which Etherton LJ considered to be of considerable importance.58 The justification for these policies was explained by Laws LJ as follows: Self-evidently the law ascertains what is unlawful and condemns it. In the paradigm case it does so by the grant of a remedy (or penalty) against the wrongdoer. Likewise it withholds relief where the wrongdoer is the claimant; this is the Ex turpi rule. Were it otherwise, the law’s condemnation of what is unlawful would be partial and inconsistent.59

It follows that the fact that the relevant conduct is a crime is highly significant to the tort claim, although it does not follow automatically that this triggers the illegality defence in every case, for the crime may be trivial or the result of an inadvertent breach of some law or regulation.60 The illegal conduct must involve some degree of recognised turpitude or culpability before the defence should be triggered.61 The question then is what degree of culpability might trigger the defence. Subjective fault encompassing knowledge of, or suspicion about, the unlawfulness must be sufficient turpitude. The commission of a strict liability offence will not suffice. For example, in Osman v J Ralph Moss Ltd a motorist who had been convicted of driving without insurance was held to be entitled to sue the insurance brokers who had negligently recommended him to take out an insurance policy with a company whose shaky financial foundation was well known in insurance circles.62 The claimant recovered the value of the fine he had paid as damages, even though the penalty arose directly from a criminal conviction and so was caught by the narrow interpretation of the illegality defence, because the offence had been one of absolute liability which did

58 Apotex, above n 55, at [67]. See also Laws LJ at [92]. Both relied explicitly on the decision of McLachlin J in Hall v Hebert, above n 25, at 165. See also Cory J in Hall v Herbert, above n 25, at 152; Hewison v Meridian Shipping Pte [2002] EWCA Civ 1821, [82] (Ward LJ); Gray, above n 45, at [39] (Lord Hoffmann), [68]–[69] (Lord Rodger); Stone and Rolls, above n 47, at [226] (Lord Mance); Safeway Stores Ltd v Twigger [2010] EWCA Civ 1472, [2011] 2 All ER 841, [16] (Longmore LJ). 59 Apotex, above n 55, at [93]. 60 Ibid, at [74] (Etherton LJ), [94] (Laws LJ). 61 Ibid, at [92] (Laws LJ). See also Stone and Rolls, above n 47, at [24] (Lord Phillips). 62 Osman v J Ralph Moss Ltd [1970] 1 Lloyd’s Rep 313 (CA) [Osman].

A New Fusion Debate

109

not involve any fault on his part.63 In other words, the narrow version of the illegality defence did not apply because of the absence of culpability. So what if the commission of the crime involved objective fault: would this trigger the application of the illegality defence? This was considered by the Court of Appeal in Safeway Stores Ltd v Twigger.64 The claimant had infringed the provisions of the Competition Act 1998 relating to anti-competitive activity and was penalised by the Office of Fair Trading (OFT). It then sought to recover the value of the penalties and the costs of the investigation from its directors and employees who were responsible for the infringement, claiming breach of contract, breach of fiduciary duty and negligence. It was recognised that the OFT sanctions were penal and akin to a fine. Longmore LJ recognised that the rationale of the illegality defence: is the need for the criminal courts and the civil courts to speak with a consistent voice. It would be inconsistent for a claimant to be criminally and personally liable (or liable to pay penalties to a regulator such as the OFT) but for the same claimant to say to a civil court that he is not personally answerable for that conduct.65

Both the wide and narrow versions of the illegality defence were held to be applicable, so the claims failed.66 The narrow version potentially applied because recovery from the directors and employees of the value of the penalty imposed by the OFT could be interpreted as involving recovery in respect of the consequence of a sentence for the ‘quasi-criminal’ act of entering into an illegal agreement.67 The wider version of the defence potentially applied as regards the claim to recover the cost of the OFT investigation, since this could be considered to involve compensation for the consequences of making the illegal agreement. Whether this version of the defence applied turned on whether the cost of the investigation could be considered to have been caused by the directors and employees. The court noted that most of the cases where the narrow version of the defence had been applied involved crimes involving a mental element,68 which suggests that the defence will not be engaged where the crime was one of strict liability, but presumably it would be applicable where the crime involved objective fault, such as negligence,69 as was the case with the offence committed by the claimant in that case.

63 Ibid, at 318 (Edmund-Davies LJ), 319 (Phillimore LJ). See also R Leslie Ltd v Reliable Advertising and Addressing Agency Ltd [1915] 1 KB 652. 64 Above n 58. 65 Ibid, at [16]. 66 Ibid. 67 Ibid, at [29] (Longmore LJ). 68 Ibid, at [17]. 69 As in Askey, above n 51. See Osman, above n 62, at 318 (Edmund-Davies LJ).

110 Graham Virgo The law on the illegality defence is complex, but these cases show that the principle of consistency is acknowledged explicitly. This can operate in a variety of different ways, but for present purposes the most significant is that a defendant cannot deflect the consequences of a criminal sanction by obtaining relief from a third party. Otherwise the civil law would undermine the penal effect of the criminal law. But this is not an absolute rule. It only applies where the crime involves proof of culpability. Where the crime is one of strict liability, the nature of the liability can be characterised as quasi-criminal, so the need for consistency of approach between criminal and civil law is much weaker and so the claimant can recover in tort despite a conviction for the strict liability offence. In such cases the absence of criminal culpability justifies the exclusion of the principle of consistency.70

IV. PROPERTY AND PROPRIETARY RIGHTS

Probably the most notorious example of inconsistency in English law between criminal and tort law arises in the context of theft. Whilst the criminal law purports to adopt the civil law conception of property and proprietary rights,71 there is a fundamental tension between the approaches of the two systems such that the general reaction of the criminal courts is to eschew any attempt at consistency of interpretation. But equally tort law appears to eschew developments in the criminal law.

A. Proprietary Rights The criminal law draws on civil concepts to identify the victim’s proprietary rights. Theft in English law will be committed where the defendant has appropriated property belonging to another person, and ‘belonging to’ is widely defined to include rights of possession.72 It follows that the owner of property can be considered to have stolen it if he or she dishonestly interferes with another person’s possessory right, even if the owner has a right to take the property. So, for example, in R v Turner (No 2)73 the defendant was convicted of stealing his car from the garage which had repaired it, and which was in possession of it until the defendant had paid for the repairs, even though the garage owner was a bailee at will so that the defendant 70 A further justification for shifting the burden of the criminal sanction to a tortfeasor is that this may serve to allocate the true responsibility for breach of the criminal law. See Goudkamp, above n 51, at 45. 71 In Hinks, above n 28, at 263 Lord Hobhouse recognised that ‘theft is a crime which relates to civil property and … concepts from the civil law have to be used’. 72 Theft Act 1968, s 5(1). 73 R v Turner (No 2) [1971] 1 WLR 901 (CA).

A New Fusion Debate

111

could have taken the car perfectly lawfully whether or not any money was due in respect of the repairs.74 But this is inconsistent with a key objective of the principle of consistency, namely that a defendant should not be convicted of a crime where he or she was lawfully exercising a right.75 It has been recognised that a defendant can be convicted of theft even if the victim consents to the interference with his or her property rights.76 But this too creates a potential dissonance between criminal and civil law, since the victim would not have any claim in tort for the interference, even where the effect of the interference is that the defendant obtains enforceable rights in the property. This dissonance is particularly well illustrated by R v Hinks,77 where the defendant was convicted of theft even though the transaction was considered to be effective to transfer title to her. In that case the defendant had dishonestly obtained a significant gift from the victim, a person in a vulnerable mental state. Although the defendant was found guilty of theft, her proprietary rights were such that she would have been able to rely on the tort of conversion to protect those rights from interference by third parties and so benefit from the fruits of her crime. But this apparent dissonance between criminal and tort law is not as problematic as it might first appear. This is because a defendant can only be guilty of theft if he or she appropriated property which belonged to another person at the time of the appropriation. In Hinks the gift was voidable by virtue of the defendant’s undue influence of the victim, so the victim could be considered to have a continuing interest in the property. Whether this should be sufficient to establish the key components of the offence was considered by the Privy Council in Wheatley v Commissioner of Police of the British Virgin Islands.78 In that case the defendant was charged with theft.79 He was employed by the government as Financial Secretary. He had a direct interest in two business enterprises which contracted with the government to rebuild a wall. The defendant had acted dishonestly because, knowing of his conflict of interest, he signed the contracts on behalf of the government without disclosing his personal interest. The contracts were not, however, overpriced, so the conflict of interest had not affected the government adversely. The Privy Council recognised that the defendant was guilty of theft, since considerations of gain and loss were not relevant to

74

Ibid, at 904 (Lord Parker CJ). See Smith, above n 7. 76 Lawrence v Commissioner of Police of the Metropolis [1971] UKHL 2, [1972] AC 626; Director of Public Prosecutions v Gomez [1992] UKHL 4, [1993] AC 442 [Gomez]; Hinks, above n 28. 77 Above n 28. 78 Wheatley v Commissioner of Police of the British Virgin Islands [2006] UKPC 24, [2006] 1 WLR 1683 [Wheatley]. 79 Contrary to s 203 of the Criminal Code (Act No 1 of 1997) of the Laws of the British Virgin Islands, which is virtually identical to the English law of theft. 75

112 Graham Virgo the determination of whether there had been an appropriation of property, it being sufficient that the defendant had effected the disbursement of the government’s funds. Most importantly, the Privy Council recognised that there was: ‘no dissonance between the criminal and the civil law, since the contracts made by the [defendant], contrary to his authority and with the connivance of the second [defendant], were plainly voidable at the suit of the Government’.80 In other words, because of the defendant’s breach of fiduciary duty, namely the self-dealing rule, the transaction with the government was voidable. Although the Privy Council did not explain why this avoided any dissonance between the criminal and civil law, presumably this was because the government could have avoided the contracts in equity and this mere equity was sufficient to ensure that it had a sufficient continuing interest in the money, so that the defendant had interfered with the property rights of another person. The same would be true of Hinks. In other cases the transaction might be voidable because of fraud.81 The real concern with cases such as Hinks is not the result but the reasoning, which suggests indifference to the apparent clash between criminal and civil law.82 In fact, if the criminal courts were more concerned with the operation of the civil law, focusing on the validity of the underlying transaction, then they would see that where that transaction is void or voidable the conviction for theft is readily justifiable. Seeking consistency of interpretation can, therefore, be of significant benefit to the quality of judicial reasoning.

B. Property There are also significant differences in the way property is defined in the civil and criminal law. The most important difference concerns choses in action, which are capable of being property for the purposes of theft83 but not for the tort of conversion.84 Interference with a chose in action cannot ground a claim in conversion because, despite being identifiable and transferable, a chose in action cannot be possessed, whereas appropriation for the crime of theft does not depend on there being any interference with possessory rights. This consequently infringes the principle of consistency,

80

Wheatley, above n 78, at [11]. Gomez, above n 76. 82 See Hinks, above n 28, at 252–53 (Lord Steyn). Cf Lord Hobhouse (at 264). 83 But only under the Theft Act 1968 and not under the earlier offence of larceny. 84 See OBG v Allan [2007] UKHL 21, [2008] AC 1, [102]–[106] (Lord Hoffmann); Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch) [45], [2012] 3 WLR 835 (Stephen Morris QC). See generally A Goymour and S Watterson, ‘Testing the Boundaries of Conversion: Account-Holders, Intangible Property and Economic Harm’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 204, 215–18. 81

A New Fusion Debate

113

because it follows that the defendant can be convicted for interfering with a chose in action where he or she has acted lawfully in a civil sense in doing so. It does not follow from this that the identification of property in theft and in the tort of conversion must necessarily be the same, but the reasons for that difference of result need to be carefully articulated and they have not been. If the difference of approach cannot be justified, then ‘fusion’ would be appropriate. It is then necessary to consider which approach should be preferred. Whilst it has often been assumed that the tort of conversion should follow the approach of the criminal law, such that interference with choses in action should trigger tortious85 or some other form of liability,86 such a conclusion does not necessarily follow. The modern law of theft was actually intended to reflect the law of conversion. When the Theft Act 1968 was enacted, ‘appropriation’ was intended to mean the same thing as conversion but ‘appropriation’ was preferred because ‘conversion’ was considered to be a ‘lawyer’s word’.87 In any event, choses in action deserve the same protection as other forms of property and so interference with intangible property rights should trigger both criminal and civil liability. Consistency should prevail and here the civil law should follow the criminal law.

V. CONSPIRACY

Conspiracy is both a serious criminal offence88 and a tort. The principle of consistency has been acknowledged in this context as well. For example, in AAH Pharmaceuticals v Birdi Coulson J recognised that in the criminal law it is not possible for a person to conspire with a company of which he is the sole shareholder.89 He went on to acknowledge that the civil law is less clear, since a contract can be made between a company and the sole shareholder of the company90 and it is possible to identify a conspiracy between two or three wholly connected companies.91 But Coulson J acknowledged,

85

See S Green, ‘Theft and Conversion—Tangibly Different?’ (2012) 128 LQR 564. See A Goymour, ‘Conversion of Contractual Rights’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 67 who argues that conversion should not be extended to encompass ‘pure’ non-documentary contractual rights, but that the law of unjust enrichment should be used instead and, to the extent that this cannot provide appropriate protection, a new faultbased tort should be developed to protect intangibles. That fault-based tort would operate in a consistent way with the crime of theft, albeit that lesser fault would trigger the tort, which would be consistent with the hierarchical model of the relationship between crime and tort. 87 Criminal Law Revision Committee, Eighth Report (Cmnd 2977) 19. 88 Contrary to the Criminal Law Act 1977 in England. 89 AAH Pharmaceuticals v Birdi [2011] EWHC 1625 (QB) [31], citing R v McDonnell [1966] 1 QB 233. 90 Lee v Lee’s Air Farming Ltd [1961] AC 12 (PC). 91 Taylor v Smith [1991] IR 142. 86

114 Graham Virgo whilst not deciding the point, that it would be an unattractive result to distinguish between the civil and criminal jurisdictions in such a radical way as to conclude that a conspiracy between the sole shareholder and the company establishes liability in tort but not in criminal law. Since the essence of a conspiracy is agreement and there is no policy reason why the principle of consistency should not be applied in this context, it follows that the criminal and civil law should be interpreted consistently. Here, by virtue of the separate identity of the sole shareholder and the company, it would be appropriate for the criminal law to follow tort law so that a conspiracy between the sole shareholder and the company should result in liability in both regimes.

VI. CULPABILITY

The key characteristic of a serious criminal offence is that subjective fault must be established, since a defendant should only be punished for a crime where he or she can be considered to have voluntarily assumed responsibility for the commission of the offence.92 There has been a general tendency in the criminal law to move away from objective to subjective notions of culpability,93 requiring proof that the defendant intended the prohibited harm or foresaw it as a possibility. This subjective approach to the identification of fault has been incorporated in the definition of dishonesty, which is the key fault requirement for offences of theft and fraud. Dishonesty for purposes of the criminal law was defined in R v Ghosh94 by reference to whether the reasonable person would consider the defendant’s conduct to be dishonest and, if so, whether the defendant realised that this would be how the reasonable person would characterise the conduct, even if the defendant did not consider the conduct to be dishonest by his or her own standards. This hybrid interpretation of dishonesty has been extended to other crimes, such as the common law offence of misconduct in public office where, at least where the misconduct involves the acquisition of property by theft or fraud, it is an essential ingredient of the crime that the defendant acted dishonestly in this subjective sense.95 The hybrid interpretation of dishonesty has been recognised in the civil law as well. For example, to establish liability for the equitable wrong of

92

See Ashley, above n 5, at [76] (Lord Carswell). Most significantly in R v G [2003] UKHL 50, [2004] AC 1034 where Lord Bingham recognised, at [33], that it is ‘neither moral nor just to convict a defendant … on the strength of what someone else would have apprehended if the defendant himself had no such apprehension’. 94 R v Ghosh [1982] QB 1053 [Ghosh]. 95 R v W(M) [2010] EWCA Crim 372, [2010] QB 787, [11] (Lord Judge CJ). 93

A New Fusion Debate

115

dishonestly assisting a breach of trust or fiduciary duty,96 the House of Lords in Twinsectra Ltd v Yardley97 adopted the Ghosh test, so that the defendant would not be liable as an accessory unless he or she was aware that the reasonable person would have characterised the conduct as dishonest.98 This appears, therefore, to be a situation where the principle of consistency has been recognised. But decisions of the Privy Council99 and the Court of Appeal100 have rejected the Ghosh test of dishonesty in respect of establishing accessorial liability. Instead, a primarily objective test has been adopted whereby honesty is assessed with reference to the ordinary standards of reasonable and honest people, even if the defendant did not appreciate that such people would consider the conduct to be dishonest. This objective test does appear to have subjective components, since it is necessary to determine what the defendant knew or suspected about the relevant transactions, but that is only because this assists the court to determine whether the defendant’s conduct involved sufficient want of probity to enable it to be characterised as objectively dishonest. But the courts have not clearly identified any convincing reason as to why the principle of consistency should not be applied in this context, and some judges have even expressed concern about whether ‘the concept of dishonesty for the purposes of civil liability differed to any marked extent from the concept of dishonesty as understood in the criminal law’.101 But there is no reason for such concern. Dishonesty in the criminal law operates to establish moral culpability for conduct which will result in the imposition of punishment. In the civil law of accessorial liability, dishonesty operates simply to characterise the defendant’s conduct, rather than his or her mind, as wrongful, so that the claimant’s rights should prevail over those of the defendant. This focus on dishonesty of conduct rather than dishonesty of mind was specifically recognised by Lord Millett in his dissenting judgment in Twinsectra.102 In Barlow Clowes International Ltd v Eurotrust International Ltd, Lord Hoffmann considered dishonesty to be a subjective state of mind, but the standard by which the law determined this was an 96

A wrong which can be characterised as tortious in effect if not in origin. Twinsectra Ltd v Yardley [2002] UKHL 12, [2002] 2 AC 164 [Twinsectra]. 98 This test was applied by the Court of Appeal in Bultitude v The Law Society [2004] EWCA Civ 1853. See also Ultraframe (UK) Ltd v Fielding [2005] EWHC 1638 (Ch), [2006] FSR 16, [1481] (Lewison J). 99 Royal Brunei Airlines Sdn Bhd v Tan Kok Ming [1995] UKPC 4, 2 AC 378, 392; Barlow Clowes International Ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 WLR 1476, [10] (Lord Hoffmann). 100 Abou-Rahmah v Al-Haji Abdul Kadir Abacha [2006] EWCA Civ 1492, [2007] 1 All ER (Comm) 827, [65] (Arden LJ); Starglade Properties Ltd v Nash [2010] EWCA Civ 1314. See also Fattal v Walbrook Trustees [2010] EWHC 2767 (Ch), [2012] Bus LR D7, [81] (Lewison J). 101 Starglade Properties Ltd v Nash [2010] EWCA Civ 1314, [2011] Lloyd’s Rep FC 102, [42] (Leveson LJ). See also Hughes LJ at [41]. Leveson LJ went on to assert that the Court of Appeal (Criminal Division) should revisit the test of dishonesty at the earliest opportunity. 102 Twinsectra, above n 97, at [121]. 97

116 Graham Virgo objective one.103 But this unnecessarily confuses the criminal and civil law objectives, which can properly be explained by distinguishing between dishonest states of mind and dishonest conduct. Only the latter is relevant to civil law claims because such claims do not depend on proof of culpability and the identification of individual responsibility. Consequently the principle of consistency legitimately need not be applied in this context.

VII. SELF-HELP

In both criminal and civil law there is a defence to liability where the claimant acted reasonably in self-help. This defence was interpreted restrictively by the Court of Appeal in the criminal law in R v Burns,104 where the defendant had picked up a prostitute and, having taken her in his car to a more secluded area, decided that he did not wish to engage in sexual activities with her and asked her to leave the car. She refused to do so and the defendant then removed her by force and prevented her from re-entering, causing her to sustain scratches and bruises. The defendant was charged with causing her actual bodily harm by assault and pleaded the defence of self-help, on the ground that he needed to use force to recover exclusive possession and occupation of his car. It was held that although the defence was available in principle, it was not available on the facts because the exercise of the right to self-help by the defendant was not considered to be reasonable since there had been a mutual understanding that the victim would be returned to where the defendant had picked her up. It was acknowledged that self-help could only justify the use of force where there were no reasonably practicable alternatives105 to enable the defendant to acquire exclusive possession of the car, and only then where a person had entered the vehicle without permission and refused to leave it, not where he or she had entered the car as a passenger at the invitation of the car owner. The court also briefly considered the operation of the defence in the civil law, where there is a right of recaption of property whereby the defendant can use force to recover exclusive possession of property. Even here the defence is interpreted restrictively and is circumscribed by the reasonableness of the force exercised, assessed by reference to the identification of any reasonable alternatives available to the property owner.106 The Court of Appeal in Burns noted that the civil law did recognise the right to use

103

Above n 99. R v Burns [2010] EWCA Crim 1023, [2010] 1 WLR 2694 [Burns]. 105 Ibid, at [13], citing Lloyd v Director of Public Prosecutions [1992] 1 All ER 982, 992 (Nolan LJ). 106 Southwark London Borough Council v Williams [1971] Ch 734 (CA) 745–46 (EdmundDavies LJ). 104

A New Fusion Debate

117

reasonable force to eject an individual from any property, such as a house, after he or she had refused to leave on request, even if the individual had originally been invited into the property, for this permission could subsequently be revoked. Liability in tort would not be imposed. But the court did not consider the implications of any potential dissonance between the interpretation of the defence in the criminal and civil law. Although this dissonance did not arise on the facts because the defendant’s behaviour would be characterised as unreasonable in both regimes, such dissonance would have arisen if the force used had been reasonable. This is because the defence is available in the civil law where permission is given to enter the property and that permission is subsequently revoked, but not in the criminal law where the permission cannot be revoked. It follows that the defendant would have a private law right to remove the occupant from his vehicle and prevent her return without being liable for the tort of battery but he could be convicted of the crime of battery for exercising that right. Here the principle of consistency is infringed and this infringement cannot be justified. The defence of self-help should be available in all cases to recover possession of property using reasonable force, even if permission was originally given to enter or use the property, since it should be possible to revoke the permission to enable the defence to apply, regardless of whether the defence is considered in the criminal law or the law of tort. By focusing on the principle of consistency, absurdity of result can be avoided.

VIII. CONCLUSIONS

The boundary between criminal and tort law is a frontier for legal analysis which has not received the attention it deserves. There is a tendency amongst the judiciary and commentators to assume that things are done differently in criminal and tort law and that nothing can be gained by even considering what might be happening on the other side of the boundary. The aim of this chapter has been to challenge this orthodoxy by arguing that it is appropriate for those working on different sides of the boundary to have regard to what is being done on the other side. It does not follow from this exercise that fusion of concepts and policies will necessarily occur; the functions of criminal and tort law are such that difference will be common and can be justified. But the fundamental point is that where difference cannot be justified, the principle of consistency should apply to the interpretation of common concepts such that it becomes necessary to consider carefully, by reference to relevant principles and policies, whether the criminal law should follow the law of tort or vice versa. Careful analysis of disparate areas of the law reveals that the principle of consistency is actually firmly embedded in the law, although it is rarely articulated. It can be identified in the interpretation of defences, such as

118 Graham Virgo self-defence, and the operation of the defence of illegality in tort claims. But consistency of approach is not inevitable in every case. There is certainly no obvious inconsistency in the imposition of civil liability without imposing criminal liability, since this may arise from the general need for greater culpability for criminal liability and different standards of proof. But a situation in which criminal liability is imposed in circumstances where the defendant is legitimately exercising a right recognised in the civil law will infringe the principle of consistency and should be avoided if it cannot be justified, such as where a defendant is guilty of theft but not liable for conversion or through the differential operation of the defence of self-help. The crucial lesson to be drawn from this chapter is that there are real advantages to those who characterise themselves as private lawyers having regard to what occurs in the criminal law and vice versa. The web of the law should, as McLachlin J recognised, be conceptually seamless and a unified institution which works in essential harmony.107 That harmony can tolerate difference; it is only the ‘intolerable fissures’ which are unacceptable.

107

Hall v Hebert, above n 25, at 165, citing Weinrib, above n 26, at 42.

5 Challenging the Orthodoxy of Crime’s Precedence over Tort: Suspending a Tort Claim Where a Crime May Exist MATTHEW DYSON

I. INTRODUCTION: ORTHODOXIES AND THE RELATIONSHIP BETWEEN TORT AND CRIME

A. Orthodoxy: Sticking to the Straight and Narrow

W

HAT IS THE orthodox understanding of the relationship between tort and crime, and is it right?

The first step is to decide what ‘orthodox’ and hence, ‘orthodoxy’, mean. The Oxford English Dictionary gives the original meaning of orthodoxy as: ‘The body of opinions, doctrines, or beliefs held to be orthodox by a particular religion, society, or group’.1 First, this comes from ρθο´δοξος, orthódoxos, from orthós, straight, right and dóxa, opinion.2 Thus at its earliest root there is a clear link between the physical euphemism, straight or right, and the background metaphysical quality of the object, an opinion. Second, it thereafter comes to mean not just views and beliefs, but also, according to the OED, ‘practices prevalent or established in a particular society, field, profession, party, etc.; conventional’. Hence it comes to designate any branch of learning or discipline whose ‘principles and methods are regarded as authoritative or founded in established theory’. It covers not

1

Oxford English Dictionary Online, www.oed.com (last accessed July 2013). CT Onions, Oxford Dictionary of English Etymology (Oxford, Oxford University Press, 1966) 634. 2

120 Matthew Dyson only opinions and beliefs, but habits, techniques, forms of reasoning and processes of thought and action. Third, an orthodoxy is likely to become, in one way or other, a simplification. The orthodox view is commonly a reduction of a difficult idea, favouring at best the majority position. With the counter-positions stripped out for simplicity’s sake, we see only an unqualified generalisation. Fourth, orthodoxy might, in extended use, not only refer to ‘right’ practices and established disciplines, but to the maintenance of them; ‘maintaining [them] in accordance with those prevailing or officially sanctioned in one’s profession, discipline, party, etc.; conventional, conservative’.3 Therefore, orthodoxy has gone beyond just established practice, to extend to the practice of establishment. Our thinking too easily focuses on bifurcating discourses into the accepted and the doubted, the norm and the abnormal, the default and the challenger. We shift onto the new idea the burden of proving itself, often within a framework which favours the established reasoning. This preference for establishment can lead to undesirable consequences. In particular, it can lead to difficulties in seeing discrete situations where the orthodoxy may not actually be true: a discrepancy might be missed, or dismissed, in the face of the established position. Finally, the underlying belief that orthodoxies sometimes exist in some areas can lead to the assumption that there must always be an orthodoxy. In other words, the default belief of legal actors is often in the existence of an established position that can be expressed simply enough for those with only minimal time or interest to find it out. Even more than defaulting to the orthodox being correct, we default to the existence of an orthodoxy in the first place. This implication of orthodoxy is particularly problematic in two areas: where an established rule is on the cusp of changing and where there is a new and developing area of law. In both these areas, the presumption of an established position can easily lead to an oversimplification of what the law is in order to make it expressible as an orthodoxy, thus stifling the development of that area of law. The relationship between tort and crime is just such an area: the law is often in flux and developing but too often gets oversimplified, hindering detailed analysis. This oversimplification is not being sufficiently challenged and tested. B. Orthodoxies in Tort and Crime: The Three Ps What then is the orthodoxy about the relationship between tort and crime? The significant discourses of the modern common law, particularly within England, suggest that three statements of orthodoxy could be made. These 3

Oxford English Dictionary Online, above n 1.

Challenging the Orthodoxy of Crime’s Precedence over Tort 121 three separate but related orthodoxies are the ‘three Ps’ of tort and crime: partition, priority and permanence. (i) The Orthodoxy of Partition The orthodoxy of partition holds that a legal system should not co-ordinate tort and crime together explicitly. It represents a default position that there is no benefit in co-ordination and that a divide should be used to keep each area separate and simple. Such a barrier between tort and crime has three key dimensions: location, direction and who is building it. There are four locations where tort and crime might meet, such that a partition could be placed there to prevent co-ordination: institutions, norms, substantive law and procedure. Institutionally we might ask whether the same legal actors or offices are able to perform both civil and criminal functions, or liaise with their partners on the other side of that divide. Normatively, a sample question would be whether the same theories are used in tort law as in criminal law. This includes, at least for present purposes, whether the terminology and conceptual understanding is the same across tort and crime. Substantively, the focus is on a specific rule and whether that rule, for instance in tort law, should be co-ordinated with a parallel rule in, say, criminal law. Finally, procedurally, a partition prevents the processes of crime and tort from being linked up and made compatible. One example of a procedural location is the timing rules between tort and crime, discussed in detail later. In terms of direction, the partition could be put up by the criminal law in the face of the civil law, or by the civil law in the face of the criminal law. Some examples of each will be given, but as the paper later focuses on an example from the civil law side, more examples from the criminal law will be given below. Finally, we could focus on who is building the partition, for instance, those in practice or those who research and teach. To focus on the law in practice we would look most closely at legislation and the role of courts in interpreting it as well as their expression of the common law. To focus on teaching and research we would look more to the work of academics and commentators. There are such a host of instances of interaction across these locations, directions and personnel that any examples given here could only give the barest sense of the whole picture. Nonetheless, four partition examples will be given. An instance of where a criminal court has responded to the normative and substantive links between tort and crime is R v Sheppard and Whittle.4 The Court of Appeal held that people who use a web server in another 4

R v Sheppard and Whittle [2010] EWCA Crim 65, [2010] 1 WLR 2779.

122 Matthew Dyson country to incite racial hatred in England and Wales can be liable under the Public Order Act 1986. The material, described as ‘truly pernicious’,5 doubted the existence of the Holocaust and was intensely derogatory of Jewish and black people. Does such material require a publishee? The only evidence of a publishee proven at trial was that a single police officer had downloaded the material in the course of the criminal investigation. Defence counsel tried to argue that libel concepts, both civil and criminal, were helpful in exploring the idea of ‘publication’. The Court of Appeal was unmoved: ‘The point that there cannot be publication without a publishee is in our judgment fundamentally misconceived. It is based on an irrelevant comparison with the law of libel’.6 Surely similar techniques of legal reasoning apply in either case, so there may well be something to gain by at least being aware of the range of possible meanings of the word ‘publication’, in civil and criminal law. Indeed, in the course of argument in court, Cranston J had drawn on examples of publication from the world of financial prospectuses in just this way.7 An even more powerful example, where a statute directly referred to civil law but nonetheless a barrier was put up, can be seen in the controversial case of R v Hinks.8 The defendant had acquired property from an infirm patron she claimed to have been looking after. The litigation was shaped by the inability of the prosecution to prove that the ‘gifts’ had been obtained by deception, undue influence or in circumstances where the donor did not have capacity to make them. The defendant was charged with theft and the case went to the House of Lords on whether the defendant appropriated property when taking up a valid gift. The relevant legislation, the Theft Act 1968, clearly referred to civil law norms, particularly the ‘owner’ of property,9 proprietary interests at common law and in equity10 and the belief in a civil right to property being conclusive evidence of a lack of dishonesty.11 It might therefore be assumed that no partition could be drawn between the civil and the criminal law. The House of Lords disagreed, the leading speech being given by Lord Steyn: The purposes of the civil law and the criminal law are somewhat different. In theory the two systems should be in perfect harmony. In a practical world there will

5

Ibid, at [39]. Ibid, at [35]. The Court of Appeal noted the need for damage in civil libel as if this was the only relevant matter; they did not explain why a publishee would not be necessary for inciting or encouraging others. 7 The author was present in court. 8 R v Hinks [2000] UKHL 53, [2001] 2 AC 241 [Hinks]. The case is about criminal protection of civil law interests broadly, not only those recognised by tort law but also by the law of property. 9 Theft Act 1968, s 3(1). 10 Ibid, s 5(1)–(5). 11 Ibid, s 2(1)(a). 6

Challenging the Orthodoxy of Crime’s Precedence over Tort 123 sometimes be some disharmony between the two systems. In any event, it would be wrong to assume on a priori grounds that the criminal law rather than the civil law is defective ... The tension between the civil and the criminal law is therefore not in my view a factor which justifies a departure from [earlier case law]. Moreover, these decisions of the House have a marked beneficial consequence. While in some contexts of the law of theft a judge cannot avoid explaining civil law concepts to a jury (eg in respect of section 2(1)(a)), the decisions of the House of Lords eliminate the need for such explanations in respect of appropriation. That is a great advantage in an overly complex corner of the law.12

Lord Steyn posits a normative ‘harmony’ between the civil and criminal law, but ‘in a practical world’ they might be in disharmony, which in that particular context, means that they might have substantively different rules. Lord Steyn grasps for some benefit, citing the potential complexity of the law, such that the jury might be confused by what was ‘mine’ and ‘yours’ in practice. The irony is that this was the very distinction that the Criminal Law Revision Committee had been relying on when it drafted the Bill.13 The difficulty behind this partition is made even more obvious by Lord Hobhouse, in dissent: Another point which has arisen from the general intention of the Act and its drafting is the assumption that all questions arising in connection with the law of theft should now be capable of answer without involving any concept or rule derived from the civil law or using any technical legal terminology. Whilst there can be no doubt about the general intention of the Act, to proceed from such a general intention to that assumption is simplistic and erroneous ... The truth is that theft is a crime which relates to civil property and, inevitably, property concepts from the civil law have to be used and questions answered by reference to that law.14

Lord Hobhouse continued, giving examples of famous instances where appellate courts had had to sort out earlier failures to apply civil law concepts properly.15 A third example of partition can be seen where a criminal power is tracking a paradigmatic civil law concern, such as providing compensation. Here the criminal courts have been keen to be free to do ‘justice’ rather than be tied too tightly to the civil law. Section 1(1) of the Criminal Justice Act 1972 first gave criminal courts a general power to award compensation for damage

12

Hinks, above n 8, at 252–53. ‘The important element of them all is undoubtedly the dishonest appropriation of another person’s property—the treating of ‘tuum’ as ‘meum’; and we think it not only logical, but right in principle, to make this the central element of the offence’ Criminal Law Revision Committee, Eighth Report, Theft and Related Offences (1967) (Cmnd 2977), [33]. 14 Hinks, above n 8, at 263–64. 15 One, R v Preddy [1996] UKHL 13, [1996] AC 815, being particularly famous and leading not only to a number of quashed convictions, but also to a rapid legislative response in the form of the Theft (Amendment) Act 1996. 13

124 Matthew Dyson caused by the commission of a crime.16 That provision was re-enacted in section 1(1) of the Powers of Criminal Courts Act 1973. The criminal law’s approach to causation under this provision has become markedly different from that in the civil law. A classic example is a conviction under the Trade Descriptions Act 1968, R v Thomson Holidays Ltd.17 There the Court of Appeal held that the civil rules of causation were not to be imported into the criminal law. In response to the defence that the damage suffered by two holidaymakers stemmed from a failure to fulfil a contractual obligation, not from the misleading statements contrary to section 14(1)(b) of the Trade Descriptions Act 1968, the court held that: Parliament, we are sure, never intended to introduce into the criminal law the concepts of causation which apply to the assessment of damages under the law of contract and tort ... It must do what it can to make a just order on such information as it has ... the court must ask itself whether loss or damage can fairly be said to have resulted to anyone from the offence for which the accused has been convicted or which has been taken into consideration.18

For academics, an even more telling partition example is how much we write our articles and textbooks on tort law without exploring the reality of the criminal law’s influence. We can pick up the same example: compensation orders. Patrick Atiyah was opposed to criminal compensation without a parallel civil liability. In 1979 he devoted an article in the Criminal Law Review to arguing this, particularly against any right to compensation which flits in and out of existence dependent on the discretion of a sentencer.19 The article concluded: ‘writers of textbooks on the civil law may need to draw their readers’ attention to the same possibility [of compensation orders] in a wide variety of circumstances’.20 Strangely enough, the 1980 and 1987 editions of his seminal work, Accidents, Compensation and the Law, the latter edited by Peter Cane, do not do this.21 Of course, they are in good company: even today many of the leading texts on tort law only briefly note any relevant criminal law material.22

16 See now Powers of Criminal Courts (Sentencing) Act 2000, ss 130–134. For more detail on powers to compensate see M Dyson, ‘Connecting Tort and Crime: Comparative Legal History in England and Spain since 1850’ [2009] Cambridge Yearbook of European Legal Studies 247. 17 R v Thomson Holidays Ltd [1974] QB 592 (CA). 18 Ibid, at 599. 19 PS Atiyah, ‘Compensation Orders and Civil Liability’ [1979] Criminal Law Review 504, 505. 20 Ibid, at 509. 21 PS Atiyah, Accidents, Compensation and the Law, 3rd edn (London, Weidenfield and Nicolson, 1980); P Cane (ed), Atiyah’s Accidents, Compensation and the Law, 4th edn (London, Weidenfield and Nicolson, 1987). 22 Eg, there are two entries under the heading ‘Crime’ in WVH Rogers, Winfield and Jolowicz on Tort, 18th edn (London, Sweet & Maxwell, 2010) 1213; A Beever, Rediscovering the Law of Negligence (Oxford, Hart Publishing, 2007) 527–28 refers those looking for

Challenging the Orthodoxy of Crime’s Precedence over Tort 125 Common lawyers tend to see the places where tort and crime grate against each other as isolated incidents which do not impinge on the overall position that tort and crime should not be co-ordinated. This is certainly the perception in England. Where such incidents do arise, the typical way to approach them, and attempt to ‘resolve’ them, leads to our second orthodoxy on tort and crime. (ii) The Orthodoxy of Priority The second orthodoxy is that, if a situation of conflict ever arises, tort law is logically subservient to crime. This could be called the orthodoxy of priority. It too could be looked at in terms of location, direction and who is making it, but just two examples will be given to show the point. Since 2003 an English statute has sought to reduce trespass to the person claims brought concerning the events which led to the claimant’s conviction for an imprisonable offence.23 Such claims are thought to denigrate the criminal justice process and so are restricted: the plaintiff must obtain leave to bring the trespass claim and leave is only granted where the defendants’ act was ‘grossly disproportionate’, presumably to the wrong being prevented. In 2009 the Court of Appeal retrospectively granted leave to bring such a battery action, but when the matter actually came to court on the merits, the defendants’ acts were held not to have been grossly disproportionate and the claim failed.24 This is a particularly interesting example because the rule, originally designed to protect ‘have-a-go heroes’ or those defending their homes, has instead become a shield for the police to use excessive force, so long as they can muster a conviction on a related charge. In the same year, the House of Lords held that the defence of ex turpi causa can defeat a claim when a tort had caused the victim to lose full mental responsibility before he killed someone: the plaintiff sought damages for the income lost while he was serving time in prison for the killings and the House of Lords rejected this ‘shift’ of the criminal law’s sanction to the tortfeasor.25 It could be argued that making the railway company pay for ‘criminal law’ to ‘illegality’, which in fact deals only with the defence of ex turpi causa in negligence. 23

Criminal Justice Act 2003, s 329(2). Adorian v Commissioner of Police of the Metropolis [2009] EWCA Civ 18, [2009] 1 WLR 1859 (granting leave), [2010] EWHC 3861 (QB) (dismissing the claim). See JR Spencer, ‘Legislate in Haste, Repent at Leisure’ [2010] CLJ 19. See also Ashley v Chief Constable of Sussex Police [2008] UKHL 25, [2008] 1 AC 962, where the House of Lords refused to strike out a claim in trespass for a police force killing an unarmed suspect based on a defence of self-defence. 25 The maxim ex turpi causa and a line of reasoning based on maintaining the dignity of the criminal conviction would apply to defeat a claim for loss of earnings and general damages where a train accident victim, suffering from post traumatic stress disorder, committed manslaughter by diminished responsibility: Gray v Thames Trains [2009] UKHL 33, [2009] 24

126 Matthew Dyson foreseeable harms resulting from their negligence did not produce sufficient conflict with the criminal law; certainly tort law does not regard all criminal acts as unforeseeable26 so all criminal sanctions should not automatically be outside the scope of civil liability either. There are further recent examples of intersections between tort and crime, many being difficulties or ambiguities in nineteenth-century statutes which prioritise crime over tort.27 There has been little analysis of the underlying tensions between tort and crime and their true impact on both areas. Only recently has scholarship started to develop.28 For example, at the Fifth Biennial Conference on the Law of Obligations, held at Oxford in July 2010 and entitled ‘Rights and Private Law’, 18 out of 31 papers on tort law discussed aspects of the relationship with criminal law, with four contributors basing their papers on it entirely. Even so, rarely does such scholarship go deeply into the developments of the rules as they are now, focusing only on what they are or should be in modern law and assuming the permanence or worse, the irrelevance, of the past. (iii) The Orthodoxy of Permanence The third orthodoxy is that the relationship between tort and crime has not changed: what we do now in tort and crime, we have always done. This, the orthodoxy of permanence, is part of the prevailing low priority given to understanding why we have the rules that we have now. For instance, in Hinks, mentioned above, Lord Steyn referred to a letter from Parliamentary Counsel drafting the Bill that became the Theft Act 1968, noting Dr Glanville Williams’s view that ‘appropriation’ would occur where you accept a gift or buy an apple. The letter said that, if this view were shared by the rest of the Criminal Law Revision Committee, the whole language should be changed lest only the requirement of dishonesty stop

1 AC 1339, on which see J Goudkamp, ‘The Defence of Illegality: Gray v Thames Trains Ltd’ (2009) 17 Torts Law Journal 1; cf Pitts v Hunt [1990] EWCA Civ 17, [1991] 1 QB 24, 39 (Beldam LJ). 26 Among many examples, Home Office v Dorset Yacht [1970] UKHL 2, [1970] AC 1004 rather stands out. 27 See Wong v Parkside Health NHS Trust [2001] EWCA Civ 1721, [2003] 3 All ER 932 concerning the bar to a later civil action or prosecution after a summary conviction for assault contained in ss 44 and 45 of the Offences Against the Person Act 1861. This was last tinkered with by the Courts Act 2003 c 39 Sch 10, [1]; but cf the contemporaneous Criminal Justice Act 2003, s 329(2). For a very recent decision on an old problem, see Crawford Adjusters v Sagicor General Insurance (Cayman Islands) Ltd [2013] UKPC 17, holding that there can be an action for malicious prosecution even in relation to civil proceedings. 28 See, eg, D Mead, ‘When the Police come Knocking—The Case for Review of Public Law Trespass and Consensual Entries onto Property’ [2012] Criminal Law Review 97; S Green, ‘Conversion and Property Torts’ (2012) 128 LQR 564; N Cobb, ‘Property’s Outlaws: Squatting, Land Use and Criminal Trespass’ [2012] Criminal Law Review 114.

Challenging the Orthodoxy of Crime’s Precedence over Tort 127 everyone from being liable for every acquisition of property. Lord Steyn responded: While this anecdote is an interesting bit of legal history, it is not relevant to the question before the House ... How far it in fact takes one is a matter considered in Gomez [1993] AC 442 ... to delve into the intentions of individual members of the Committee, and their communications, would be to rely on material which cannot conceivably be relevant. If statutory interpretation is to be a rational and coherent process a line has to be drawn somewhere. And [the note] was demonstrably on the wrong side of the line.29

On a point of statutory interpretation, Lord Steyn’s application might be arguable. What is more interesting is his dismissal of any different history than the one he wished to enlarge. The majority in Hinks missed lessons from the past and thereby made mistakes for the future.

C. Challenging the Orthodoxies Having set up these three orthodoxies, and noted the difficulty of challenging orthodoxy, let us turn to a test. Imagine reading the following summary of the law: Since 1607, civil courts in England have prohibited civil claims based on events which were both torts and crimes where the interests of the criminal justice system required it.

Does this sound as if it supports each of the three orthodoxies? There is partition save where there is priority and it has been the law since 1607, which certainly suggests permanence. The next section of this chapter will analyse this rule, putting flesh to the bones of the summary just given. The rule co-ordinating the question of whether a civil case can precede a criminal one will be called a ‘timing rule’. In the common law, this has often been described as the rule that ‘a trespass merges in a felony’, even though, as will be seen later, this phrasing was only accurate for a short while. Such timing rules are significant. They have had a practical role in solving civil disputes and in shaping possible claims in tort. While it is true that criminals are often not worth suing, the case law shows that plaintiffs clearly saw practical value in civil claims preceding criminal prosecutions.30

29

Hinks, above n 8, at 247–48 (Lord Steyn). Sometimes such actions are against peripheral participants in the crime, such as banks whose employees are involved in fraud. The actions might also seek to take advantage of insurance, more beneficial civil rules of evidence and procedure or to force a defendant to return or pay the equivalent in value of specific property. 30

128 Matthew Dyson II. THE EVOLUTION OF THE TIMING RULES

The history of timing rules in English law can be divided into three phases. The first is from 1607 to 1914: here a merger rule became established, largely to protect the state’s interests in forfeited property and in promoting private prosecutions. However, the courts had not developed a way to enforce it. The second period is from 1914 to 1967, when the courts enforced a timing rule by a mandatory stay of proceedings. This temporarily stabilised that timing rule but ultimately its value was sufficiently doubted that it was removed by legislation. From 1967 English law has been in its third phase, with the removal of the rule followed by its regrowth as a discretion to suspend a parallel civil claim but only where the interests of the defendant, not the state, are prejudiced. In the early common law, ‘crime’ and ‘tort’, as we call them now, were equally valid ways for a victim to pursue justice for a wrongful act.31 The most ‘civil’ remedy, the writ of trespass, could be brought or a more penal action (indictment or appeal of felony) could be used. The choice seems to have been between compensation and vengeance, and this choice was one for the victim. This position continued from around the 1200s to the end of the 1500s.32 The law’s indifference was then replaced by precedence for the criminal action. From 1607 until 1967, a civil action based on facts which constitute a felony could not take place before the prosecution for the felony.

A. 1607–1914 The analysis of Watkin Williams J in the 1881 case of Midland Insurance Company v Smith is an accurate summary of the law in the first period, from 1607 to 1914: The history of the question shows that it has at different times and by different authorities been resolved in three distinct ways. First, it has been considered that the private wrong and injury has been entirely merged and drowned in the public wrong, and therefore no cause of action ever arose or could arise. Secondly, it was thought that, although there was no actual merger, it was a condition precedent to the accruing of the cause of action that the public right should have been

31 See generally, DJ Seipp, ‘The Distinction between Crime and Tort in the Early Common Law’ (1996) 17 Boston University Law Review 59; JB Ames, Lectures on Legal History and Miscellaneous Legal Essays (Cambridge Mass, Harvard University Press, 1913). 32 In this period there is strong evidence that the two actions were independent: PR Glazebrook, ‘The Merging of Misdemeanours’ (1962) 78 LQR 560, 561; JB Ames, ‘Substantive Law before the time of Bracton’ in Lectures on Legal History, above n 31, at 45–46; Hudson v Lee (1589) 4 Co Rep 43 a, 76 ER 989, 989–90, where it appears that an action at law had been maintained for what must have been a felony.

Challenging the Orthodoxy of Crime’s Precedence over Tort 129 vindicated by the prosecution of the felon. Thirdly, it has been said that the true principle of the common law is that there is neither a merger of the civil right nor is it a strict condition precedent to such right that there shall have been a prosecution of the felon, but that there is a duty imposed upon the injured person, not to resort to the prosecution of his private suit to the neglect and exclusion of the vindication of the public law.33

These three phases, or logics, represent increasing sophistication in the analysis of the relationship between the two actions.34 The first idea, drowning, was evident in what was arguably the earliest case discussing the relationship between a felony and trespass arising on the same facts: Higgins v Butcher in 1607.35 As is typical of the early cases, there is doubt over what it actually decided. Most likely it decided a husband could not maintain an action for the harm suffered by his wife when she was killed. However, in addition to the idea that the right of action died with her, her death was also said to constitute a wrong to the Crown. That wrong was ‘converted into felony, and that drowns the particular offence and private wrong offered to the master before; his action is thereby lost’.36 The reasoning in respect of the felony was unsupported by authority and arguably not necessary to determine the case.37 The second idea was that the prosecution of the felon was a condition precedent for the actionability of the civil law claim. Cases in this phase began in 1652 with Dawkes v Coveneigh and expressed the fear that the felony prosecution will be ‘smothered’ if the plaintiff could elect to bring a civil suit.38 None of the eighteenth-century cases offers significant clarification as to why there should be a condition precedent in some cases and not

33 The Midland Insurance Co v Smith (1881) 6 QBD 561, 568. A number of cases are missed out, but the basic pattern is valid. 34 See further: M Dyson, ‘The Timing of Tortious and Criminal Actions for the Same Wrong’ (2012) 71 CLJ 85. 35 Higgins v Butcher (1607) Yelv 89, 80 ER 61, 61 that the felony ‘drowns ... the private wrong’. On this claim see, eg, WS Holdsworth, History of English Law (London, Methuen, 1923) vol iii, 331–33. 36 See CK Burdick, Burdick’s Law of Torts (Albany NY, Banks & Company, 1926) [36]; PH Winfield, A Text-Book of the Law of Tort (London, Sweet and Maxwell Ltd, 1937) 168, 170. 37 In the same line are Markham v Cobb (1675) Jones W 147, 82 ER 79; cf Wells v Abrahams (1872) LR 7 QB 554, 560–61 [Wells]. See also the general dicta in Cooper v Witham (1669) 1 Sid 375, 82 ER 1166. 38 Dawkes v Coveneigh (1652) Style 346, 82 ER 765, 766. See Sir Matthew Hale, The History of the Pleas of the Crown (London, R Nutt and R Gosling, 1736) 546–47; cf William Blackstone, Commentaries on the Law of England (Oxford, Clarendon, 1765–1769) vol 4, 6, who did not give authority but stated that ‘the private wrong is swallowed up in the public: we seldom hear any mention made of satisfaction to the individual; the satisfaction to the community being so very great’, then noting that forfeiture would make it impossible. See also Lutterell v Reynell (1670) 1 Mod 282, 283, 86 ER 887.

130 Matthew Dyson others, when the condition precedent was satisfied, and why in particular the condition precedent was the best approach.39 The third phase in this period, promoting viable criminal prosecutions, began in 1810 with Crosby v Leng.40 Leng had been acquitted on a charge of felonious assault and the court held that a civil action could now proceed. Lord Ellenborough CJ expressed this as a question of timing: while public justice must not be prejudiced, private actions could be brought on the same facts once public justice was secured.41 This was perhaps the first case to ‘freeze’ a civil claim while a criminal action was thought necessary.42 Other than occasional returns to the condition precedent analysis,43 the courts turned to exploring and delimiting what would prejudice public justice. There was some sophistication here, defining persons who were sufficiently within the scope of the state’s interests to be required to prosecute.44 A similar desire to specify the relevant interests of justice would later be picked up in the cases deciding exactly who was bound to prosecute.45 The most significant difficulty with all three of these phases was that there were no settled means to enforce the suspension rule. From about the middle of the nineteenth century, once suspension of the civil action was the agreed outcome, the courts began to consider the mechanism to achieve it.46 There had been some inconclusive but influential discussion of the matter in the 1880s Irish case of A v B, where in fact neither party wanted the rule to apply.47 Holmes J thought there was a summary power to stay civil proceedings, one the court could invoke of its own motion where it appeared proper, for example when a criminal case was ‘actually

39 Cf Gibson v Minet (1791) 1 H Bl 569, 126 ER 326, 336 (Perryn B) who restated the merger rule, with Master v Miller (1791) 4 TR 320, 100 ER 1042, 1048–59, where Buller J expressly declined to rule on a future civil action’s validity. Finally there was Whitfield v Lord Despencer (1778) 2 Cowper 754, 758, 761, 98 ER 1344, where counsel on both sides agreed on a merger rule, but counsel for the plaintiff argued that the servant’s felony should not suspend the action against the master. 40 Crosby v Leng (1810) 12 East 409, 104 ER 160. 41 Ibid, at 161. 42 TN Talfourd (ed), Dickinson’s Practical Guide to the Quarter Sessions, 5th edn (London, Baldwin and Cradock et al, 1841) 302. 43 See, eg, Gimson v Woodfull (1825) 2 Car & P 41, 42–43, 172 ER 19, 21; even in 1889 Bigelow wrote that there seemed to be a condition precedent, but he questioned how it could be enforced: MM Bigelow, The Law of Torts (Cambridge, Cambridge University Press, 1889) 129–30. 44 See, eg, Stone v Marsh (1827) 6 B & C 551, 108 ER 554; Marsh v Keating (1834) 1 Bing NC 198, 131 ER 1094, 1101–2 (Park J); White v Spettigue (1845) 1 Car & K 673, 174 ER 986 [Spettigue]; Dudley & West Bromwich Banking Co v Spittle (1860) 1 J & H 14, 70 ER 642, 643 [Dudley]. 45 Osborn v Gillett (1873) LR 8 Ex 88, 93; Ex parte Ball, In re Shepherd (1879) 10 Ch D 667, 673–74. 46 Gimson v Woodfull (1825) 2 Car & P 41, 172 ER 19; Spettigue, above n 44, at 679 (Rolfe B); Wellock v Constantine (1863) 2 H & C 146, 159 ER 61; Wells, above n 37. 47 A v B (1889) 24 LR Ir 235.

Challenging the Orthodoxy of Crime’s Precedence over Tort 131 pending’.48 However, as he thought the facts before them did not suggest this, he held that they should not interfere. This was also the opinion of Johnson J and Sir Michael Morris CJ, though the Chief Justice expressly did not decide what sanction was available.49 Only O’Brien J approved both the court’s discretion to stay on its own motion, and a stay on the facts before him.50 Despite these and other discussions, no solution had been conclusively adopted by the close of the nineteenth century.51 Commentary on the timing rules had been sparse but by the end of the nineteenth century, most commentators criticised the suspension rule. Commentators also seemed to believe that judges doubted it. Stephen’s Commentaries moved from a position of ‘gross and atrocious’ injuries causing the private action to be ‘swallowed up in the public’ in the first edition in 1845,52 to acknowledging the mere suspension of the private wrong in the eighth edition in 1880;53 but by 1890 he thought the rule ‘practically impossible to enforce’.54 Addison, having been clear on the duty in the first edition in 1860,55 moved to a position of doubt about the rule, at least by 1893.56 A writer in the Justice of the Peace journal in 1883 finished a summary of the law without enthusiasm: ‘although the rule may still be said to have some shadowy existence, it is so honeycombed by exceptions, and so difficult, if not impossible, to bring into application, that it merits very little consideration’.57 Ames, lecturing around 1890, said that idea ‘has been much criticized, and it is doubtful if it is still law’.58 From the first edition of his work on tort law in 1887, Pollock thought that the tide had turned against the rule: But so much doubt has been thrown upon the supposed rule in several recent cases, that it seems if not altogether exploded, to be only awaiting a decisive

48 Ibid, at 237. The case appears to be identical with that reported in S v S (1882) 16 Cox 566 (Queen’s Bench of Ireland). 49 A v B, above n 47, at 238–39. 50 Ibid, at 239–40. 51 Appleby v Franklin (1885) 17 QBD 93 (Dist Ct) 94 (Huddleston B), 95 (Wills J). 52 James Fitzjames Stephen, Stephen’s New Commentaries on the Laws of England (London, Henry Butterworth, 1845) 56. 53 James Fitzjames Stephen, Stephen’s New Commentaries on the Law of England, 8th edn (London, Butterworths, 1880) 5. 54 James Fitzjames Stephen, A General View of the Criminal Law, 2nd edn (London, Macmillan & Co,1890) 502–3. 55 Though he located discussion in the ‘recapture and restitution for the wrongful conversion of chattels’ section, thus linking to the second of Watkin Williams J’s categories. 56 Cf CG Addison, On Wrongs and their Remedies (London, V and R Stevens and Sons, 1860) 219–20, without citations, with H Smith and AP Percival Keep, Addison on Wrongs and their Remedies, 7th edn (London, Stevens and Sons Ltd, 1893) 76: ‘The duty [to prosecute], if it exists, is confined to felonies’. 57 ‘The Merger of Tort in Felony’ (1883) 47 Justice of the Peace 291, 292, though he thought some discretion might remain for serious cases. Compare with the somewhat derivative ‘Merger of Trespass in Felony’ (1888) 52 Justice of the Peace 803 and the highly critical ‘The Merger of Trespass in Felony’ (1898) 104 Law Times 498 (2 April 1898). 58 Ames, above n 32, at 46. The lectures were given at intervals between 1886 and 1895.

132 Matthew Dyson abrogation. The result of the cases in question is that, although it is difficult to deny that some such rule exists, the precise extent of the rule, and the reasons of policy on which it is founded, are uncertain, and it is not known what is the proper mode of applying it ... On the whole there is apparent in quarters of high authority a strong though not unanimous disposition to discredit the rule as a mere cantilena of text-writers founded on ambiguous or misapprehended cases, or on dicta which themselves were open to the same objections.59

However, while authors were busy attacking the rule particularly during a lull in cases, the courts were about to select their means to implement it.

B. 1914–1967 In Smith v Selwyn, the female plaintiff brought a civil action alleging deception, drugging and actual or attempted non-consensual sexual intercourse.60 The plaintiff argued that these facts constituted a misdemeanour contrary to section 3(3) of the Criminal Law Amendment Act 1885.61 The defendant argued that the claim disclosed a felony, contrary to section 22 of the Offences Against the Person Act 1861, so suspending the civil action until after the plaintiff prosecuted. Counsel for the defendant argued from Cave J in Roope v D’Avigdor,62 Cockburn CJ in Wells v Abrahams63 and the Irish case of A v B,64 that a stay of action was the correct course.65 Counsel for the plaintiff argued that Wells should be followed: the case should proceed but the judge may decline to enter judgment for the plaintiff. Somewhat surprisingly, the court accepted the arguments of the defendant: It is not easy to find a statement in any case as to what is the course which the Court ought to adopt in a matter of this kind. Some of the decisions are not easy to reconcile. This, however, is certain, that the Court has a right, if not an imperative duty, to stay the proceedings in a civil action for damages, if it is clear that that which is the basis of the claim in the action is a felony committed by the defendant against the plaintiff.66

The court had received relatively detailed citations, including to the ninth edition of Pollock in 1912 (still bearing his forecast of the impending abrogation

59

F Pollock, The Law of Torts (London, Stevens and Sons, 1887) 172–73. Smith v Selwyn [1914] 3 KB 98 (CA) [Selwyn]. 61 The husband was also claiming in his own name for the harm of having lost the consortium of his wife, but the Court of Appeal treated the plaintiffs together: ibid, at 104 (Kennedy LJ). 62 Roope v D’Avigdor (1883) 10 QBD 412 (Div Ct). 63 Above, n 37. 64 Above, n 47. 65 Selwyn, above n 60, at 100–1. 66 Ibid, at 103 (Kennedy LJ); see also 106 (Swinfen Eady LJ), 106–7 (Phillimore LJ). 60

Challenging the Orthodoxy of Crime’s Precedence over Tort 133 of the rule).67 However, judgment was not reserved and it is unclear how much time was taken in consideration. If the timing rule was to be followed, the case before the court could not easily have been resolved without a decision on how to implement the rule. Given that fact, a stay was probably the result more supported by authority. The proceedings were stayed but leave granted to amend their claim to better avoid relying on facts suggesting a felony. The certainty which Smith v Selwyn finally provided may have helped to ensure the survival of the rule to which it gave effect, at least for a while.68 Thereafter judges turned to criticising the rule: it was artificial,69 confusing70 and out of date.71 The judiciary were echoing the doubts expressed by commentators at the end of the nineteenth century, even though the timing rule’s content and existence were now more certain. Peter Glazebrook summed up the position in the 1960s thus: ‘In its present form the rule can find favour only with those who believe that litigation should resemble as closely as possible a game of snakes and ladders’.72

C. 1967–Present These doubts led to reform, but by means that were indirect and ultimately unsuccessful. The Criminal Law Revision Committee’s Report on Felonies and Misdemeanours in 1965 [CLRC Report] had recommended the removal of the distinction between misdemeanour and felony and this was brought about by section 1 of the Criminal Law Act 1967 [CLA 1967]. The Report had dealt with the timing rule in only one paragraph, finding that the rule had ‘become unnecessary, especially as it does not apply to misdemeanours, and should be allowed to lapse’.73 However, while the Report argued that removing distinctions between felonies and misdemeanours would remove

67 Ibid, at 102. F Pollock, The Law of Torts, 9th edn (London, Stevens and Sons Ltd, 1912) 205–8. Pollock is somewhat sullen in the 10th edition, noting that after Smith v Selwyn, ‘Discussion of the earlier authorities is therefore no longer useful’: F Pollock, The Law of Torts, 10th edn (London, Stevens and Sons, 1916) 210. 68 See, eg, H Street, The Law of Torts, 3rd edn (London, Butterworths, 1963) 97, n 5. 69 Jack Clark (Rainham), Ltd v Clark [1946] 2 All ER 683, 685 (Tuckey LJ). The court also stressed that the stay was not automatic: ibid, at 684 (Morton LJ); noted (1947) 63 LQR 6. 70 Fowler v Lanning [1959] 1 QB 426, 440. See also, on malicious prosecution, Abbott v Refuge [1962] 1 QB 432, 452–53, 460–62, 469–70. 71 See, eg, Oloro v Ali [1965] 3 All ER 829 (QB), 830–31 (Milmo J) [Oloro]; Gouldbourne v Magnus [1959] CLY 2661 (County Court). 72 See Glazebrook, above n 32, at 562. 73 Criminal Law Revision Committee, Seventh Report, Felonies and Misdemeanours (1965) (Cmnd 2659), [80]; Winn, Personal Injuries Litigation, (1968) (Cmnd 3691), [389]; Sir Derek Hodgson, Profits of Crime and Their Recovery (London, Heinemann, 1984) 13.

134 Matthew Dyson the timing rule, no section in the Act was devoted to that task. There was no new approach, and no guidance on how such cases should be resolved.74 The first case post-CLA 1967 was Jefferson v Bhetcha in 1979.75 Forbes J, at trial, had adjourned an application for summary judgment in respect of misappropriated cheques, apparently because he believed that a defendant should not be forced to disclose a defence in a civil case while a criminal one on the same matter was pending.76 The Court of Appeal was not referred to the CLRC Report nor the CLA 1967, and it did not refer to the cases of Smith v Selwyn and Wells v Abrahams, which had been cited in argument.77 Indeed, the only case cited by Megaw LJ in the sole judgment from the two-man court was doubted by him. That was the Supreme Court of Victoria case of Wonder Heat v Bishop where the English timing rule was applied to adjourn a claim of money had and received because the defendant was already committed for trial.78 That case was relevant because it had persuaded Forbes J at first instance in Jefferson to adjourn the case pending the criminal prosecution in order to prevent disclosure of the defendant’s defence. Megaw LJ doubted the outcome of Wonder Heat and Forbes J’s belief that a criminal prosecution must be protected from parallel civil proceedings. The judgment in Jefferson, handed down only a day after argument was heard, set out a new approach and has become the leading case.79 First, the court having control of the civil proceedings could, in the exercise of its discretion under section 41 of the Supreme Court of Judicature (Consolidation) Act 1925, stay those proceedings if it appeared to the court that justice so required. To say that a vexatious civil action could be stayed was orthodox reasoning and had been used in purely civil cases for some years.80 However, the innovation was that the discretion could be exercised where the civil courts were dealing with a matter that could or should come before the criminal courts first. It was also the first time statutory authority had been used for the timing rule.

74 This is in contrast to the other issues dependent on there being a felony, such as the power of the arrest, penalties for assisting offenders or concealing crimes, where the CLA created new rules using the term ‘arrestable offence’ in place of ‘felony’: ss 2–3, 4 and 5 respectively. 75 Jefferson Ltd v Bhetcha [1979] 2 All ER 1108 (CA) [Jefferson]. 76 Ibid, at 1111–12. 77 Ibid, at 1109. 78 Wonder Heat Pty Ltd v Bishop [1960] VR 489. The distinction between felony and misdemeanour had been abolished in Victoria by the Crimes Act 1958, s 322B, though Pape J did not seem to think this relevant, even though this was probably the first merger case since the Crimes Act came into force. 79 See, eg, Panton v Financial Institutions Services Ltd [2003] UKPC 86 (Jamaica) [7]. 80 See, eg, St Pierre v South American Stores Ltd [1936] 1 KB 382 (CA), an action relating to a lease in a foreign jurisdiction, especially at 398 (Scott LJ), on the principles affecting the grant of a stay. The section has been used by about 13 cases in total, two of which are merger cases.

Challenging the Orthodoxy of Crime’s Precedence over Tort 135 Second, in deciding whether to exercise this discretion, the judge would have to balance the justice between the parties: There is, I say again, in my judgment, no principle of law that a plaintiff in a civil action is to be debarred from pursuing that action in accordance with the normal rules for the conduct of civil actions merely because so to do would, or might, result in the defendant, if he wished to defend the action, having to disclose, by an affidavit ... or in the pleading of his defence, or by way of discovery or otherwise, what his defence is or may be ... in the contemporaneous criminal proceedings ... By way of example, a relevant factor telling in favour of a defendant might well be the fact that the civil action, or some step in it, would be likely to obtain such publicity as might sensibly be expected to reach, and to influence, persons who would or might be jurors in criminal proceedings ... [or] for example, enabling prosecution witnesses to prepare a fabrication of evidence or by leading to interference with witnesses or in some other way.81

No stay was given in Jefferson. Courts have continued to apply the test strictly and stays have not been given easily.82 The law has continued to develop but not fundamentally change. Thus, for example, when criminal law downgraded the protection of the ‘right to silence’ the civil law took this as validation of its reluctance to consider aspects of criminal procedure.83 The new Civil Procedure Rules also changed the rule’s form though not especially its substance.84 A Practice Direction became the governing provision for the timing rule in 2001.85 It made clear that, in an application for the stay of civil proceedings pending the determination of related criminal prosecution: any party to the civil or criminal proceedings may make the application; every other civil party must be made a respondent in the application; the application must give grounds for and estimate the duration of the stay; and finally, that it was not necessary for the prosecutor or defendant in the criminal proceedings to be joined as a party to the civil case.86 In V v C, the first case after the Civil Procedure Rules came into force, the court added some examples of factors to balance when considering a stay: for example, that adverse inferences could be drawn from 81 Jefferson, above n 75, at 113 (Megaw LJ). Thus the so-called ‘right to silence’ was not a civil law rule. The discretion had therefore been exercised poorly on the facts: ibid, at 114. 82 For a rare example see In Re DPR Futures Ltd [1989] 1 WLR 778, 790–92. There, according to Millett J, the real risk to the defendant’s chance of a fair trial if a civil case were to precede a criminal one was outweighed by the potential losses to the large number of small investors. 83 Surrey Oaklands NHS Trust v Hurley (unreported, QBD 25 June 1999), where Sullivan J also held that if one of multiple defendants had no answer to the civil claim, fairness required that to be established at the earliest possible stage. 84 Before these reforms the statutory power had been preserved in s 49(3) of the Supreme Court Act 1981. 85 Practice Direction—Applications (2001) PD 23, [11A.1]–[11A.4]. 86 On occasion judges have even taken the Civil Procedure Rules as a reason to be aware of criminal law issues in the cases before them: eg, Secretary of State for Trade and Industry v Crane [2001] 2 BCLC 222 (HC) 226–27 (Ferris J).

136 Matthew Dyson silence in certain circumstances in both the civil and criminal law and that a positive defence is more likely to exculpate than incriminate.87 The court also held that where the defendant makes the application, the onus is on him or her to demonstrate why that discretion should be exercised, and that the stronger the case against the defendant in the civil context, the higher the onus on the defendant should be.88

III. CHALLENGING THE ORTHODOXIES

Having explored in detail one interface between tort and crime, we are in a position to assess the ideas of partition, priority and permanence. The history of the timing rules suggests that these orthodoxies can and should be challenged, at the very least as being too simplified descriptions of the complexity of the law in practice.

A. Partition The timing rules show clearly that the partition orthodoxy is inaccurate: English law does connect some procedural aspects of the relationship between tort and crime, and this connection has been reinforced so often that legal actors clearly believe in some benefit from doing so. Further support for the rejection of partition can be seen in the availability of compensation in criminal courts,89 in the admissibility of convictions as evidence in later civil cases90 and in the willingness of civil courts to consider any serious criminal penalties already imposed in deciding to award punitive damages.91 What is more, this connection makes sense. It seems likely that parties to civil claims would avail themselves of any means to win or delay their actions, making a partition put up by the civil law against the criminal law less appealing to them and thereby bringing such comparisons before the courts. Such pleadings would in turn force judges to deal with procedural links between civil and criminal law. Aggregating all the places where tort and crime interact together therefore jeopardises understanding how the law is actually working. It fails to show the context of when and why such a reference will be useful. Similarly it skates over the intricate points of

87

V v C [2001] EWCA Civ 1509, [2002] CP Rep 8. Ibid, at [24]–[44]. 89 See Dyson, above n 16. 90 See, for instance, M Dyson, ‘Civil Law Responses to Criminal Judgments in England and Spain’ (2012) 3 Journal of European Tort Law 308. 91 Rookes v Barnard [1964] UKHL 1, [1964] AC 1129, 1226 (Lord Devlin); Borders (UK) v Commissioner of Police of the Metropolis [2005] EWCA Civ 197, [2005] Po LR 1, [20], [41], [46]; cf Gray v Motor Accident Commission [1998] HCA 70, 196 CLR 1, 14. 88

Challenging the Orthodoxy of Crime’s Precedence over Tort 137 application and even encourages us not to notice such points, let alone deal with them properly. The timing rules clearly suggest that the orthodoxy of partition is inaccurate if generalised. It may be that partition is inherently more viable in two of the other locations: norms and substantive law.92 Further research on this is highly desirable. Similarly, it may be that any partitions are more usually erected by the criminal law against the influence of the civil law. It is as yet unclear whether such a partition is more commonly erected in practice, or in academic work. Nonetheless, it has been shown that an overarching orthodoxy of partition without sensitivity to its location is inaccurate. It is also harmful, as it hides the reality of the interactions between tort and crime.

B. Priority If the timing rules do not support partition, what about priority? Let us first take priority to mean that if a situation of conflict arises, tort law is logically subservient to crime. In which case, much turns on how ‘conflict’ is understood and then whether tort giving way really is a benefit. An obligation to stay proceedings, without any express evidence of an actual conflict with the criminal law, supports the priority understanding only if we use an unsatisfying definition of ‘conflict’. A different approach to ‘conflict’ and ‘benefit’ can be seen in the three primary reasons for the timing rules: (i) the Crown’s interest in forfeiture of the felon’s property; (ii) that the state’s interest in punishment outweighs the civil interest; and (iii) that otherwise the rights of the defendant in the criminal trial would be jeopardised. These rationales do not map neatly onto the phases of the timing rules. The timing rules began with a first phase where the civil claim drowned in the felony, moving to prosecution being a condition precedent before focusing instead on the encouragement of prosecutions. In their second phase, there was a mandatory stay of parallel proceedings. We are now in the third phase where a stay will be given in the rare circumstances of a real risk to the defendant’s rights. The first two phases at least can only be seen as expressions of an imagined or assumed conflict with the criminal law leading to a priority rule. The rules have steadily become more sophisticated but it is only in the third phase that we see sufficient sophistication to justify an orthodoxy of priority, and only then in specific circumstances.

92 In fact, the legal actors involved in civil and criminal cases are likely not always to see the benefit in partition. See, eg, Dyson, above n 16, at 260.

138 Matthew Dyson (i) Forfeiture Forfeiture presents an obvious and unsophisticated example of priority: that the state would lose money if the civil claim proceeded, therefore there was a conflict and the civil claim was not allowed to proceed. Until 1870 a convicted felon forfeited all his property to the Crown, including that not related to the crime. The rationale was mentioned in case law93 and in commentaries.94 If one assumes the validity of this rule, then a civil claim risks diminishing the felon’s property to satisfy the claim and/or costs, to the detriment of the Crown. This is only a risk, not a certainty, since arguably not all successful civil claims would be able to be proven to the criminal burden of proof. As such the idea of conflict, and hence any ‘priority’ being assigned, was questionable. Nonetheless, if a real conflict can be proven, the forfeiture rationale supports some form of priority reasoning. (ii) The State’s Non-financial Interests in Punishment The state’s non-pecuniary interests in punishment may provide some further evidence for the priority of criminal law over tort law. Such interests have sometimes been lumped together to form a rule of public policy.95 Some expression of the state’s interest in punishment was probably the dominant basis for the rule after 1870: Probably [the rule’s] origin is not unconnected with the interest of the Crown in the forfeiture of goods of a convicted felon; and certainly since the abolition of such forfeiture, it can only be justified, if it is to be justified at all, on the grounds of public policy—namely, that there is a public interest in ensuring that the prosecution of an offender in the criminal courts is not subordinated to the enforcement of civil rights by private individuals.96

If this is founded on the subjective idea that the individual is putting himself before the state,97 then this surely fades at some point in the early twentieth century once the state has taken over the responsibility of prosecuting.98 There are also a couple of examples of this reasoning which cropped up in the nineteenth century, justifying it as a way to prevent potential private litigants from colluding to settle a civil claim rather than prosecute the wrong.99

93

Chowne v Baylis (1862) 31 Beav 351, 54 ER 1174. Stephen, above n 54, at 107. 95 See, eg, Gibson v Minet (1791) 1 H Bl 569, 126 ER 326, 336 (Perryn B): ‘for the sake of the public’. 96 Oloro, above n 71, at 830 (Milmo J). 97 As was suggested in, eg, Spettigue, above n 44, at 988. 98 This has been judicially recognised, eg, Rose v Ford [1936] 1 KB 90, (1935) Ll L Rep 1 847–48; Oloro, above n 71, at 831. See also Winfield, above n 36, 170–71. 99 Dudley, above n 44, at 643; Wallace v Hardacre (1807) 1 Camp 45, 170 ER 870, 871. 94

Challenging the Orthodoxy of Crime’s Precedence over Tort 139 This non-financial interest reasoning can support a priority approach only so far. If forfeiture to the Crown must be protected, that is one thing, but the general need for punishment can be protected by the certainty of a criminal action rather than its chronological precedence. The civil action could have been allowed to proceed, but the prosecution still guaranteed. Thus the plaintiff would have had the chance for compensation for the tort to put him in a better position to mount a prosecution for the felony. On the other hand, if there were doubts about him carrying through, enforcement of a successful civil action could have been subject to the bringing of a prosecution, or costs could have been used as a motivator. This rationale does not easily support a conflict between civil and criminal law. (iii) Protection of Defendant’s Rights in the Criminal Trial The final justification for a timing rule is perhaps the best example of an evolved or detailed priority approach. The CLA 1967 expunged the merger rule and its previous rationales;100 a new discretion to stay where the defendant’s interests in the criminal trial were at risk later took its place. That discretion to stay, rather than an obligation, forced the defendant in the civil case to show how his or her rights in the criminal trial were being infringed. It therefore shows a much more detailed and flexible priority rule, focused on the defendant in the civil action showing actual conflict with his or her rights in a pending criminal trial. The use of this discretion highlights the point made in the introduction, that like most or perhaps even all orthodoxies, the three Ps of tort and crime are apt to become generalisations, which lack the necessary precision to be accurate.

C. Permanence The timing rules have been shown to change greatly over time and as such, the orthodoxy of permanence too cannot always be true. The rule has moved from the automatic destruction of the civil claim to its suspension in rare cases. Yet there is a thread running through it, the same thread that disproves the orthodoxy of partition: through the last 400 years there has been an acceptance that the criminal law and civil law should not be disconnected, but rather, their unity should be intellectually appreciated and practically monitored. There was a steady liberalising trend but legal actors apparently believed in a need for some sort of rule, even if they did 100 There is one early mention during counsel’s argument in Peddle v Rutter (1837) 8 Car & P 337, 173 ER 521, 522–23 (Lord Denman CJ): a criminal action had been ended because the common serjeant thought the civil action was enough. However, this was not referred to in post-1967 cases.

140 Matthew Dyson not create a clear sanction for its breach. The underlying problem was not solved even by statutory assassination of a timing rule, so a rule regrew in a similar (but more liberal) shape.101 What can the timing rules tell us about challenging orthodoxies about permanence? There are a number of interesting points in the development of the timing rules but they cannot be explored in detail here.102 Let us take just one example: that there can be lags between the law in practice and its underlying rationales. The state’s interest in the finances of the defendant has been shown to be the main reason for the merger timing rule. However, enforcement of forfeiture was on the wane well before the Felony Act 1870 which formally removed it.103 This slow decline makes it hard to trace any impact the removal of this rationale might have had on what was by then a ‘suspension’ rule. It is clear that the rule itself survived after 1870. The lack of cases between 1889 and 1914 might have been a time of transition away from the forfeiture rationale. This in turn may have led to the greater resolution in Smith v Selwyn in 1914. Perhaps the simplest lesson to draw is the resilience of the rule concerning parallel criminal proceedings. For example, the leading criminal law textbook by Smith and Hogan argued in 1965: ‘Smith v Selwyn is important, then, because it shows that the rule survives the erosion of its foundations [in forfeiture]’.104 In other words, did orthodoxy remain even after the reasoning did? Or, more troublingly, are we challenging the orthodoxies of today, or those of yesterday?

IV. CONCLUSION

The complex and important relationship between tort and crime merits a better understanding than the three orthodox Ps. Three lessons should be learned.

101 The English legal system has tended to solve practical problems as they arise in the courts. These ad hoc solutions have not had particular guidelines but have been guided by fact patterns and the conservative but equitable traits of the judiciary. On the other hand, this lack of guiding principle has also left the legislature free to tinker with the system, though it has only done so rarely. 102 See, eg, Dyson, above n 34, at 103–12. 103 Ibid, at 108. 104 JC Smith and B Hogan, Criminal Law (London, Butterworth & Co, 1965) 22, referring to Street, above n 68, 97. See also ‘Damages for Assault’ (1935) 80 Law Journal 376, 376: the rule was founded in forfeiture, and ‘therefore, existed originally for the benefit of the Revenue, and, as this reason no longer exists, it has been argued that the rule should be abolished. The cases show, however, that the rule is still useful as a test of the bona fides of the plaintiff in a civil action. As the rule is far from being an anachronism, it may, therefore, be expected to survive the scrutiny of the law reformers.’ By comparison, the legislative reboot of the rule in 1967 effected a very neat change in the timing rule, and it coincided with a brand new rationale. Yet this shift still took 12 years, from 1967 to 1979, to bear fruit.

Challenging the Orthodoxy of Crime’s Precedence over Tort 141 First, that orthodoxies can easily be too simplistic, robbing them of value. Typically, statements about the law can operate on a number of levels: this chapter has noted institutional, normative, substantive and procedural. It has focused on one procedural connection to show that the law has had to co-ordinate the relationship between tort and crime on at least that level. The other levels merit further study. Similarly, we have identified the fact that the purposes of legal actors may differ: that the civil law and the criminal law might look at each other, as well as themselves, differently. Indeed, so might an academic have a different conception of the law than a practitioner and judge. A proper understanding of the law, and its implications, cannot exist without this combination of structure and nuance. Second, not only are locations within the law important, but so is an understanding of the way the law is developing. For instance, we have learned that an orthodoxy can exist while being incomplete: the existence of early timing rules was not doubted even though they had no clear method of enforcement for the first 300 years of their existence. We have also learned that legal actors can so value a rule that they can resurrect it, with a completely different purpose, only 12 years after killing it off. That this can be done surreptitiously by a court, sidestepping the intention of Parliament, is all the more interesting. These events emphasise the risks in uncritically believing in orthodoxies expressed as simple rules: the reality can be much more complex. We can also learn about legal development from other legal systems, both the similar and the different. In Germany, § 403(1) of the Code of Criminal Procedure states that, in proceedings before the Local Court, the aggrieved person may, without the consent of the accused, also assert a property claim arising out of the criminal offence. Nonetheless, less than 1% of Local Court cases have civil claims for harm appended to the criminal process, despite 97% of victims saying they would like that to happen.105 Why is this? A familiar line of reasoning was used by one German commentator: The reason for this is possibly that the strict separation between criminal and civil law instilled during legal studies, and the mental distinction between punishment and reparation, run deep; also criminal judges might fear that they will be put under pressure and overburdened by having to deal with complex issues of civil law.106

On the other hand, in Spain, not only will a criminal prosecution usually deal with any compensation for the victim but, since 1848, it has done so

105 See, eg, E Siegismund, ‘Ancillary (Adhesion) Proceedings in Germany as Shaped by the First Victim Protection Law: An Attempt To Take Stock’ in H Iitsuka and R Findlay-Debeck (eds), Resource Material Series No 56 (Tokyo, United Nations Asia and Far East Institute for the Prevention of Crime and Treatment of Offenders, 2000) 102, 108–9. 106 Ibid, at 108–9.

142 Matthew Dyson using rules in the Spanish Criminal Code, not the Civil Code.107 Similarly France has long perceived some kind of unity of civil and criminal fault and precedence for a criminal prosecution over parallel civil claims.108 In addition, the French partie civile links the procedures for civil wrongs and criminal actions, with the victim being a key player in a criminal action.109 We can learn a lot from understanding their outright rejection of partition and why nonetheless they adopt priority in certain situations. Even in other commonwealth jurisdictions a greater willingness to look behind the apparent force of rules can be seen. It has been stated that American courts have taken ‘every possible view’ of how to answer the question of concurrent civil and criminal proceedings.110 Even before the end of the nineteenth century, the merger rule was being rejected across much of the USA: Although the English rule was adopted in early American cases in some states, it has been swept away by statute in those and other states ... The majority of courts in the United States regard the English rule as not adapted to our situation and needs, and of little practical use, and hold that there is no merger, and that a party who has been injured by the felonious conduct of another may maintain his civil action for the injury without waiting for the institution of a criminal prosecution against the offender.111

Third, it might even be better to reject the urge to reach for an orthodoxy in the first place. The practice of establishing orthodoxies can encourage stagnation and laziness. To shift the orthodoxy takes time and effort, so inertia sets in: challenging the orthodoxy becomes harder over time. Ultimately, even being consciously aware of practices which breach the orthodoxy becomes hard. All this can be accepted in many areas of law: the pressures and difficulties of the day-to-day operation of the law and the demands on academic time do not allow us to do better. But it cannot be justified where the picture of the law we get is so fundamentally skewed. This is particularly a risk in areas of the law when they are in a state of flux, or developing as new outcroppings. The timing rules have been one instance of a larger dispute that has surfaced from time to time in the courts and legislature over the centuries. 107 See, for a general introduction to the present position, G Quintero Olivares and JM Tamarit Sumalla, ‘De la responsibilidad civil derivada de los delitos y faltas y de las costas procesales’ in Gonzalo Quintero Olivares (ed), Comentarios al Nuevo Código Penal (Cizur Menor, Navarra, Editorial Aranzadi, 2005). 108 See, eg, R Demogue, De la réparation civile des délits (Paris, Arthur Rousseau, 1898) and G Viney, Introduction à la responsabilité, 3rd ed (Paris, LGDJ, 2008) titre II. 109 See, eg, PO Lapie, ‘The Partie Civile in the Criminal Law of France’ (1928) 10 Journal of Comparative Legislation and International Law (3d Ser) 33; J Larguier, ‘Civil Action for Damages in French Criminal Procedure’ (1964–1965) 39 Tulane Law Review 687. 110 JP Bishop, Criminal Law, 1st edn (Boston, Little, Brown & Co, 1856) vol i, §§ 333–35. 111 DS Garland and LP McGehee (eds), The American and English Encyclopedia of Law, 2nd edn (New York, Edward Thompson, 1902) vol XX, 601–2. See also JB Ames, Select Cases on Tort (Cambridge, Cambridge University Press, 1874) 354–90.

Challenging the Orthodoxy of Crime’s Precedence over Tort 143 There are great practical implications depending on how this kind of dispute is resolved and it is time to face up to them. This chapter has argued that a logical first step to understanding the relationship between tort and crime is to challenge the established, often unsophisticated and vague orthodoxies of their relationship. From there, new and nuanced understandings can be developed, for instance by building up a picture of how they have been co-ordinated until now. Many of what are thought to be novel arguments now can be traced back to other points in the history of the relationship and there is a wealth of material on the outcomes of those arguments waiting to be explored. For instance, take one of the examples given at the start of this chapter: the requirement, created in 2003, for leave to bring a trespass to the person claim based on events that had led to the defendant’s conviction for an imprisonable offence. It is premised on the assumption that the criminal prosecution will precede the civil claim. Since 1979 there has no longer been a rule requiring the criminal proceedings to come first.112 Yet Parliament proceeded as if the civil claim would always, not just often, come second. The message is clear. If we fail to challenge orthodoxy, both as established practice and as the practice of establishment, we will constantly struggle to regulate an ever-changing world.

112 See Adorian v MPC [2009] EWCA Civ 18, [2009] 1 WLR 1859 on s 329(2) of the Criminal Justice Act 2003. The existence of s 329(2) could be a reason for the stay to be exercised under Practice Direction 23, above n 85, at [11A.1]–[11A.4]. Otherwise the civil court would decide on the liability between the parties without reference to the criteria in the statute, eg, that the defendant’s actions were grossly disproportionate. On the other hand, as a matter of statutory interpretation, if the civil case comes first, there may simply be no force to the 2003 provision. No reported cases on this have been found since Adorian v MPC. Of course, in the majority of situations a criminal conviction will still precede a civil claim.

6 Legislative Challenges to Orthodoxy JAMES LEE*

I. INTRODUCTION

A

NY ACCOUNT OF the law of tort is incomplete without an examination of the role of legislation in the development of the law. In this chapter, my engagement with the theme of this book is twofold. First, I aim to understand the ways in which legislation may ‘challenge orthodoxy’ in tort; second, I shall challenge the manner in which it does so. Whether seeking to reverse an individual decision,1 reaffirm an existing tenet, introduce limitation statutes or enact wholesale reform, legislative interventions can have a significant impact upon fundamental principles. Some ‘statutes avowedly [reform] the common law on its own terms [while] other statutory initiatives can appear to be puzzling and retrograde steps’.2 As examples of such interventions, ‘legislative affirmation’ is examined in Part II and issues relating to the law of limitation are examined in Part III. We shall see that these various instances may pose significant challenges to orthodoxy, albeit that some of the challenges are not always deliberate. Problems arise, I shall argue, when such interventions do not engage sufficiently faithfully with the relevant principles. And yet, as Burrows has recently remarked, the role of legislation within the law of obligations, at least from a theoretical perspective, has traditionally been ‘woefully

* Much of the research for this chapter was undertaken while I was a Visiting Scholar at the University of Melbourne, with the benefit of a grant from the Society of Legal Scholars: I am grateful to these institutions for their support. I also appreciate the comments of the editors, those who attended the presentation of an earlier draft of this article at the Obligations VI Conference, and my fellow presenters in that session, Roderick Bagshaw and Neil Foster. This article draws upon material in my book, Legislation and Reform in the Law of Obligations (Hart Publishing, forthcoming in 2014). Any errors are my own. 1 J Lee, ‘“Inconsiderate Alterations in our Laws”: Legislative Reversal of Supreme Court Decisions’ in J Lee (ed), From House of Lords to Supreme Court: Judges, Jurists and the Process of Judging (Oxford, Hart Publishing, 2011). 2 P Mitchell, ‘Patterns of Legal Change’ (2012) 65 Current Legal Problems 177, 188.

146 James Lee underexplored’ in England,3 and Duxbury has noted that ‘academic lawyers generally … can often seem less inspired by statute law than they are by case law’.4 If legislation and the common law do ‘exist in a symbiotic relationship’,5 we must investigate whether that relationship is mutually beneficial or parasitic. Both the United Kingdom Supreme Court and the High Court of Australia have been called upon to examine aspects of this relationship on many occasions in the past decade. In Australia, this is in no small part the result of the increased willingness at a state level to legislate.6 In response to a perceived ‘insurance crisis’ in the early years of the twenty-first century, an inquiry was launched into reforms of the law of tort, resulting in Civil Liability Acts in various states which have dramatically affected the legal landscape.7 There has also been the implementation of uniform defamation legislation across the states.8 A consequence has been that scholars in Australia have been required to re-examine the roles of legislation within the field of civil liability.9 English scholars, on the other hand, have not had quite the same cause to assess the influence of statutes. We shall therefore draw upon themes from the Australian literature and apply them to 3 A Burrows, ‘The Relationship Between Common Law and Statute in the Law of Obligations’ (2012) 128 LQR 232, 232. Burrows does identify some exceptions, to which we might now add a recently published collection: J Steele and TT Arvind (eds), Tort Law and the Legislature: Common Law, Statute, and the Dynamics of Legal Change (Oxford, Hart Publishing, 2012). 4 N Duxbury, Elements of Legislation (Cambridge, Cambridge University Press, 2012) xii. 5 As Gleeson CJ put it in Brodie v Singleton Shire Council [2001] HCA 29, 206 CLR 512, [31]. 6 As noted by Gleeson CJ, Gummow, Hayne and Heydon JJ in D’Orta-Ekenaike v Victoria Legal Aid [2005] HCA 12, 223 CLR 1, [53] (on solicitor/advocate’s immunity): ‘Since 1999, State legislatures have given close attention to what has been called “tort law reform”. In particular, close attention has been paid to the law of negligence, and a number of statutes have been passed since 2000 which have dealt with that general subject’. 7 See, eg, Hon R Debus MP, ‘Tort Law Reform in New South Wales: State and Federal Interactions’ (2002) 25 University of New South Wales Law Journal 825; B McDonald, ‘Legislative Intervention in the Law of Negligence: The Common Law, Statutory Interpretation and Tort Reform in Australia’ (2005) 27 Sydney Law Review 443. The Acts have been met with considerable criticism: ‘the Australian Acts, in as far as they pretend to capture and clarify the law of negligence while in fact altering it significantly, are insidious’: A Beever, Rediscovering the Law of Negligence (Oxford, Hart Publishing, 2007) 509 fn 11. 8 In force from 2006: Defamation Act 2005 (NSW); Defamation Act 2005 (Qld); Defamation Act 2005 (SA); Defamation Act 2005 (Tas); Defamation Act 2005 (Vic); Defamation Act 2005 (WA); Civil Law (Wrongs) Act 2002 (ACT) ch 9; Defamation Act 2006 (NT). See D Rolph, ‘A Critique of the National, Uniform Defamation Laws’ (2008) 16 Torts Law Journal 207 and A Kenyon, ‘Six Years of Australian Uniform Defamation Law: Damages, Opinion and Defence Meanings’ (2012) 35 University of New South Wales Law Journal 31. 9 A particular critic of the civil liability reforms has been Barbara McDonald. See, eg, McDonald, above n 7; B McDonald, ‘The Impact of the Civil Liability Legislation on the Fundamental Policies and Principles of the Common Law of Negligence’ (2006) 14 Torts Law Journal 268; B McDonald, ‘Teaching Torts: Where to Start in an Age of Statutes?’ (2010) 18 Torts Law Journal 173. See also J Dietrich, ‘Teaching Torts in the Age of Statutes and Globalisation’ (2010) 18 Torts Law Journal 141.

Legislative Challenges to Orthodoxy 147 the jurisdictions within the United Kingdom.10 In order to consider the ways in which legislation may have an impact on the development of tort doctrine,11 we shall use some recent decisions from the Supreme Court and Privy Council: Spread Trustee Company Ltd v Hutcheson (Guernsey),12 Ministry of Defence v AB,13 and AXA General Insurance Ltd v Lord Advocate.14 Legislative challenges to orthodoxy raise the spectre of ossification:15 that the law may become fixed by the statute, thus limiting the scope for continuing development of a common law principle by the courts.16 It is important for the law to retain its incremental dynamism;17 yet, as I have remarked elsewhere, ‘when it legislates, Parliament can only take a snapshot of the state of the law, a still frame from the moving picture of the common law’.18 This feature can lead to a disconnect between the common law and legislation. Indeed, in one of the cases considered below, Lady Hale observed that: Law reformers have to take the law as it has currently been decided to be, and those whom they consult will be divided amongst those who consider it to their advantage and those who do not. It is by no means unknown for a higher court subsequently to take a different view of the law.19

Following on from that observation, my second challenge to orthodoxy is broader: it is to question the manner in which Parliament does intervene, in order to understand what Mitchell has called the ‘patterns of legal change’.20 To that end, the constitutional jurisprudence from the United

10

And the British Crown Dependency of Guernsey (see below n 12). In this chapter, I shall generally use examples from the law of tort but also some from the law of equitable wrongs since, as I have argued elsewhere, the problems of legislative interventions apply to both common law and equitable wrongs: J Lee, ‘A Civil Law for the Age of Statutes’ in Steele and Arvind, above n 3. See also P Birks, ‘Equity in the Modern Law: An Exercise in Taxonomy’ (1996) 26 University of Western Australia Law Review 1, 25–26; J Edelman, ‘Equitable Torts’ (2002) 10 Torts Law Journal 64. The dicta in these equitable cases are of general application. 12 Spread Trustee Company Ltd v Hutcheson (Guernsey) [2011] UKPC 13, [2012] 2 AC 194, rev’g 2009–10 GLR 403 (CA) [Spread Trustee]. 13 Ministry of Defence v AB [2012] UKSC 9, [2012] 2 WLR 643 [AB]. 14 AXA General Insurance Ltd v Lord Advocate [2011] UKSC 46, [2012] 1 AC 868, aff’g [2011] CSIH 31, 2011 SLT 439 [AXA]. 15 As Lord Millett observed in Dubai Aluminium Company Ltd v Salaam [2002] UKHL 48, [2003] 2 AC 366, [108]. 16 See also McDonald, above n 7, at 461. 17 On the limits of such dynamism, see Lord Dyson MR, ‘Where the Common Law Fears to Tread’, Annual Lecture for the Constitutional and Administrative Law Bar Association, 6 November 2012 www.judiciary.gov.uk/Resources/JCO/Documents/Speeches/mr-speech-wherecommon-law-fears-to-tread-06112012.pdf (last accessed January 2013). 18 Lee, above n 11, at 102. 19 Spread Trustee, above n 12, at [136] (PC). 20 Mitchell, above n 2. 11

148 James Lee Kingdom and Australia on legislative interference with existing causes of action will be explored in Part IV. The argument throughout the various themes considered here is for the necessity of a coherent approach to legislation: even where the legislature seeks to challenge orthodoxy, such legislation should nevertheless be implemented with an understanding of the common law. Respecting coherence and understanding orthodoxy can lead to more effective legislation. Thus, I shall challenge the idea that, in the field of tort, we should accept that the law ‘is a chaotic jumble of (surviving) common law rules and piecemeal, inconsistent and unsystematic statutory incursions, whether we like it or not’.21

II. LEGISLATIVE AFFIRMATION

A. General One way in which Parliament may intervene is through ‘legislative affirmation’,22 where legislation expressly endorses relevant principles.23 That might be seen to be a case of supporting orthodoxy, but I shall suggest that such interventions may pose just as much of a challenge to orthodoxy as a deliberate attempt to change the law. A straightforward example is to be found in section 1 of the Compensation Act 2006.24 The Act has three purposes, as set out in its title: ‘to specify certain factors that may be taken into account by a court determining a claim in negligence or breach of statutory duty; to make provision about damages for mesothelioma; and to make provision for the regulation of claims management services’. The majority of the Act’s provisions relate to the introduction of a regulatory framework for claims management services:25 it was indicative of the general trend in England and Wales towards reforming tort procedure and process in response to concerns about the ‘compensation

21

J Dietrich, ‘What is “Lawyering”? The Challenge of Taxonomy’ (2006) 65 CLJ 549,

573. 22 Transport for London (London Underground Ltd) v Spirerose Ltd [2009] UKHL 44, [2009] 1 WLR 1797, [18] (Lord Walker of Gestingthorpe), quoting Lord Nicholls of Birkenhead in Waters v Welsh Development Agency [2004] UKHL 19, [2004] WLR 1304, [28]: both were referring to the ‘affirmation’ of the approach of Fletcher Moulton LJ in Re Lucas and Chesterfield Gas and Water Board [1909] 1 KB 16 (CA) in rule 3 of s 2 of the Acquisition of Land (Assessment of Compensation) Act 1919. 23 McDonald, above n 7, at 460. 24 Perhaps also s 2, which provides that ‘An apology, an offer of treatment or other redress, shall not of itself amount to an admission of negligence or breach of statutory duty’. 25 A Morris, ‘The “Compensation Culture”’ in Steele and Arvind, above n 3.

Legislative Challenges to Orthodoxy 149 culture’.26 Section 1 of the Act concerns the ‘deterrent effect of potential liability’ and provides: A court considering a claim in negligence or breach of statutory duty may, in determining whether the defendant should have taken particular steps to meet a standard of care (whether by taking precautions against a risk or otherwise), have regard to whether a requirement to take those steps might— (a)

prevent a desirable activity from being undertaken at all, to a particular extent or in a particular way, or (b) discourage persons from undertaking functions in connection with a desirable activity.

Given its similarity to the principles as clarified by the House of Lords in Tomlinson v Congleton Borough Council,27 it is questionable whether this section effects any change in the law. This suspicion may be confirmed by the fact that the section has not been frequently cited in the courts. When it has been mentioned, judicial views have been unanimous. The Scout Association v Barnes involved a claim for a personal injury suffered during a game played amongst a Scout troop.28 The first instance judge had found the Scout Association to be in breach of duty, a view upheld by a majority of the Court of Appeal. In giving judgment, Jackson LJ stated that: It is not the function of the law of tort to eliminate every iota of risk or to stamp out socially desirable activities … This principle is now enshrined in section 1 of the Compensation Act 2006. That provision was not in force at the time of the claimant’s accident. However, the principle has always been part of the common law.29

Jackson LJ dissented on the application of that principle to the facts, but the court was not divided on the principle itself.30 This view of continuity enshrined in section 1 has been reaffirmed by the Court of Appeal in subsequent cases.31 But what is the point of a provision such as section 1 if it is not changing anything? The Explanatory Notes to the Act state it was intended: to contribute to improving awareness of this aspect of the law; providing reassurance to the people and organisations who are concerned about possible litigation; and to ensuring that normal activities are not prevented because of the fear

26 See Morris, ibid; A Morris, ‘Spiralling or Stabilising? The Compensation Culture and Our Propensity to Claim Damages for Personal Injury’ (2007) 70 MLR 349; A Higgins, ‘Referral Fees—the Business of Access to Justice’ (2012) 32 Legal Studies 109. 27 Tomlinson v Congleton Borough Council [2003] UKHL 47, [2004] AC 46, [34]–[36], [41]–[42], [46] (Lord Hoffmann) [Tomlinson]. 28 The Scout Association v Barnes [2010] EWCA Civ 1476. 29 Ibid, at [34]. 30 See ibid, at [49] (Smith LJ). 31 Uren v Corporate Leisure (UK) Ltd [2011] EWCA Civ 66, [13] (Smith LJ); Sutton v Syston Rugby Football Club Ltd [2011] EWCA Civ 1182, [13] (Longmore LJ).

150 James Lee of litigation and excessively risk-averse behaviour … This provision reflects the existing law and approach of the courts as expressed in recent judgments of the higher courts.32

So the section was intended to be a legislative affirmation, and it has been interpreted as such: Owen J found it ‘somewhat difficult to see why it was felt necessary to enact’,33 as it did not seem to him ‘that section 1 adds anything to the common law’.34 Now there are often grounds to be suspicious of such statutory statements of the obvious: that they are unnecessary and that they are really enacted to allow the legislature to look ‘busy’.35 But what are the other possible reasons for legislative affirmation? How are the courts to make sense of such provisions? The Chief Justice of Australia, Robert French, has previously observed the potential utility of these legislative interventions: ‘Statutory principles of liability promise coherency, order and democratic legitimacy in the development of the law without the need to codify the entire field’.36 It is that point of ‘democratic legitimacy’ which might be particularly relevant here. McDonald has also suggested that affirmations are enacted ‘to demonstrate legislative support for or give the legislature’s imprimatur to a particular principle, perhaps to dispel doubts about the principle’s validity’.37 A related but distinct reason for legislative affirmation may be, as in section 1, to act as ‘political rhetoric’38 in order to increase publicity about the existing principles and to counter a perception (even if in truth a misconception) about the principles employed by the courts. There may be more of a case for doing so where the principle is thought to be latent or immanent in the law than when there has been a clear recent decision from our highest court on the very point in issue.39 Indeed, the strength of this ‘publicity’ argument may be less significant now that there is wider media coverage

32 Explanatory Notes to the Compensation Act 2006, [10], [17]. Despite s 1 being in Part 1 of the Act, which is labelled ‘Standard of Care’, the Notes state that the section ‘is not concerned with and does not alter the standard of care’: instead, it is solely focused on what steps should or could have been taken—although the difference is not quite clear. Cf Uren v Corporate Leisure (UK) Ltd [2013] EWHC 353 (QB), [75] (Foskett J). 33 Wilkin-Shaw v Fuller [2012] EWHC 1777 (QB), [42]: the claim failed on the basis that the defendants were not in breach of their duty of care. 34 Ibid, at [46]. 35 McDonald, above n 7, at 457, 460. 36 The Hon Justice R French, ‘Statutory Modelling of Torts’ in N Mullany (ed), Torts in the Nineties (Sydney, LBC Information Services, 1997) 211, quoted in Dietrich, above n 9, at 156. 37 McDonald, above n 7, at 460. 38 I am grateful to an anonymous reviewer for a comment along similar lines. 39 The contemporary decision in Tomlinson, above n 27, would suggest that s 1 of the Act falls into the latter category and is redundant. See on transparency and court visibility generally A Kavanagh, ‘From Appellate Committee to United Kingdom Supreme Court: Independence, Activism and Transparency’ in Lee, above n 1.

Legislative Challenges to Orthodoxy 151 of decisions of the Supreme Court than was often the case with House of Lords decisions.40 But even if there are good reasons for ‘affirming’, it is responsible and sensible for the legislature, when doing so, to be clear as to its intentions as to the related principles of law.

B. Legislative Toleration? An express and clear example of legislative affirmation, or perhaps toleration, of existing principles can be found in the treatment of defences to defamation claims under the uniform legislation in Australia.41 Division 2 of the Defamation Act (NSW) includes a list of various statutory defences,42 but section 24(1) of the Act provides that: A defence under this Division is additional to any other defence or exclusion of liability available to the defendant apart from this Act (including under the general law) and does not of itself vitiate, limit or abrogate any other defence or exclusion of liability.43

The intention of the legislation is clearly to extend the range of defences available.44 Thus, despite Kirby J’s assertion in his 2007 prophecy of ‘the dying hours of the common law of defamation in Australia’45 the common law defences live on.46 The significance of this preservation can be seen in Papaconstuntinos v Holmes a Court,47 in which the defendant was able to rely upon the common law defence of qualified privilege, so avoiding the statutory requirement of reasonableness.48 The appellant claimant argued that there was a requirement that the defendant must show a ‘pressing need’ to protect its interest, judged on the basis of ‘reasonable necessity’. The majority (French CJ, Crennan, Kiefel and Bell JJ) rejected that argument. But Heydon J in dissent observed that in declining to rely upon the 40 The Supreme Court’s decisions in AB, above n 13, www.bbc.co.uk/news/uk-17364359 (last accessed January 2013) and AXA, above n 14, www.bbc.co.uk/news/uk-scotlandglasgow-west-15264262 (last accessed January 2013) were both covered extensively. 41 Kenyon, above n 8. 42 In ss 25–33. 43 Emphasis added. For ease of reference, I refer only to the Defamation Act 2005 (NSW) but the equivalent provisions are Defamation Act 2005 (Qld), s 24; Defamation Act 2005 (SA), s 22; Defamation Act 2005 (Tas), s 24; Defamation Act 2005 (Vic), s 24; Defamation Act 2005 (WA), s 24; Civil Law (Wrongs) Act 2002 (ACT) ch 9, s 118(2); Defamation Act 2006 (NT), s 21. 44 For example, s 30 of the Act does not require reciprocity of interest for the defence of qualified privilege to be available: Aktas v Westpac Banking Corporation Limited [2010] HCA 25, [14]. 45 Channel Seven Adelaide Pty Ltd v Manock [2007] HCA 60, [105]. 46 On clauses which preserve the common law, see Burrows, above n 3, at 234–36. 47 Papaconstuntinos v Holmes a Court [2012] HCA 53 [Papaconstuntinos]. 48 Ibid, at [7]. Section 30(1)(c) of the Act requires that ‘the conduct of the defendant in publishing … is reasonable in the circumstances’.

152 James Lee statutory defence, the defendant had ‘conceded that his conduct in doing so could not be defended as reasonable’,49 a factor which he viewed as relevant. Informed by this view, he went on to divine the reasonable necessity requirement from the words of Parke B in Toogood v Spyring.50 And the majority, notwithstanding their dismissal of the appellant’s argument in the particular appeal, did seem to leave open whether the law might develop to include a reasonableness requirement.51 It may then be that the existence of a related statutory defence, though it ‘does not of itself vitiate, limit or abrogate’ a common law defence, can still have a gravitational effect upon its future development. We may contrast this Australian approach with the recently enacted Defamation Act 2013 in England and Wales.52 The Act takes the approach of introducing new defences of ‘truth’,53 ‘honest opinion’54 and ‘[reasonable] publication on [a] matter of public interest’.55 In doing so, it expressly abolishes their respective common law predecessors of justification,56 fair comment57 and the ‘Reynolds defence’.58 The Act, in the interests of certainty, thus mounts a clear legislative challenge to orthodoxy: the common law of these defences is thus not to have scope to continue to develop alongside the statutory frameworks.

C. Spread Trustee Company Ltd v Hutcheson (Guernsey) A more complicated example of the legislative affirmation phenomenon was the source of disagreement in the Privy Council in an appeal from Guernsey. Spread Trustee Company Ltd v Hutcheson (Guernsey) involved a claim for breach of trust against a professional trustee company which was alleged to have failed to investigate prior breaches by previous trustees of

49

Ibid, at [58]. Toogood v Spyring (1834) 1 Cr M & R 183, 149 ER 1044. 51 Papaconstuntinos, above n 47, at [48]. 52 A limited number of provisions also apply to Scotland, as provided by s 17(2). The case for reform had been judicially recognised: eg Spiller v Joseph [2010] UKSC 53, [2011] AC 852, [105]–[117] (Lord Phillips PSC). 53 Defamation Act 2013, s 2. 54 Ibid, s 3. 55 Ibid, s 4. 56 Ibid, s 2(4). 57 Ibid, s 3(8). The Act refers to the defence of ‘fair comment’ notwithstanding the defence being renamed ‘honest comment’ by the Supreme Court in Spiller v Joseph, above n 52, at [117]. 58 Ibid, s 4(6). The ‘Reynolds defence’ was a species of qualified privilege as developed in Reynolds v Times Newspapers Ltd [1999] UKHL 45, [2001] 2 AC 127, applying a standard of ‘responsible journalism’: see Jameel v Wall Street Journal Europe Sprl [2006] UKHL 44, [2007] AC 359. 50

Legislative Challenges to Orthodoxy 153 the relevant settlements.59 Amongst the allegations were that these failures amounted to gross negligence. The trustee company sought to rely upon an exemption clause in the trust settlements.60 The Privy Council was required to decide whether at the relevant times Guernsey law permitted the trustee company to rely upon such a clause. The Trusts (Guernsey) Law 1989 provided a statutory footing for trusts in Guernsey. The States (of Guernsey) Advisory and Finance Committee had recommended legislation to clarify the recognition of trusts in the law of Guernsey because ‘there was a need to dispel the present uncertainty concerning trusts in Guernsey’.61 The proposed legislation ‘would seek to set out a basic infrastructure of legal principles on the authority of which trustees, beneficiaries and settlors could operate with certainty and confidence’. Section 34(7) originally provided that ‘Nothing in the terms of a trust shall relieve a trustee of liability for a breach of trust arising from his own fraud or wilful misconduct’. It was subsequently amended by the Trusts (Amendment) (Guernsey) Law 1990, which added ‘or gross negligence’ to the breaches for which liability could not be excluded.62 It was therefore necessary for the Board to determine the effect of the 1989 Law. It was common ground between the parties that, as a matter of principle, a trustee could not exclude liability for fraud or wilful wrongdoing, but that he or she could exclude liability for negligence. But was it, in 1989, possible for a trustee’s liability for gross negligence to be excluded? There was no dispute between the parties that, before 1989, no statutory provision or Guernsey case law addressed the question. It was therefore necessary to work out what the customary law of Guernsey would have been: given Guernsey’s legal heritage, it would have looked to either English or Scots law. The position in English law was uncertain, but the position in Scots law was clear in that liability could not be excluded for culpa lata (the equivalent of ‘gross negligence’). The Privy Council, by a majority, held that it was ‘implicit’ in section 34(7) that ‘a trust could relieve a trustee for a breach of trust arising from other causes. Otherwise there would have been no point in expressly prohibiting the exclusion of liability for fraud and wilful misconduct’.63 Since ‘the purpose of the Law was to replace the existing customary law and to

59

Above n 12. The clause was framed as follows: ‘In the execution of the trusts and powers hereof no trustee shall be liable for any loss to the Trust Fund arising in consequence of the failure depreciation or loss of any investments made in good faith or by reason of any mistake or omission made in good faith or of any other matter or thing except wilful and individual fraud and wrongdoing on the part of the trustee who is sought to be made liable’. 61 President of the States Advisory Committee, ‘Trusts Law’ Letter to the President of the States of Guernsey: Spread Trustee, above n 12, at [15] (PC). 62 See s 1(f), which came into force on 19 February 1991. 63 Spread Trustee, above n 12, at [21] (PC). 60

154 James Lee clarify the rights and obligations of trustees in Guernsey’, it followed that, in the absence of any reference to gross negligence, a term to exclude such liability was lawful.64 The Court of Appeal had held that the failure to mention liability for gross negligence was likely the ‘product of a mistake’.65 The majority in the Privy Council inferred, on the contrary, that the omission was deliberate,66 stating that ‘It appears to the Board to be more likely than not that [Guernsey law] would have looked to the law of England’.67 As noted above, the position in England was not clear at the relevant time. It was subsequently authoritatively established by the Court of Appeal in Armitage v Nurse,68 which held that liability could be excluded for everything other than a breach of the trustee’s ‘core’ duty ‘to perform the trusts honestly and in good faith for the benefit of the beneficiaries’ which ‘is the minimum necessary to give substance to the trusts’.69 Thus, liability for gross negligence could be excluded, but liability for fraud or dishonesty could not. The majority also rejected the notion that gross negligence was a distinct category of fault in this area of the law.70 The next question was whether the Law as amended had retrospective effect,71 so as to prevent the trustee from relying upon the exclusion clause. The Board unanimously held, in agreement with the Court of Appeal, that, in the absence of any expression of contrary intention, the Law did not apply retrospectively. Lord Mance noted that, beyond fraud and dishonesty, ‘courts do best to leave the nature and extent of any intervention in parties’ own arrangements to legislators’.72 Sir Robin Auld framed the question differently, but supported the conclusion of Lords Clarke and Mance: What matters is what a pre-1991 Guernsey Court should have decided as a matter of Guernsey law on a logical and otherwise legally correct process of reasoning—an outcome that might also have required examination by the Board at the time. It is an issue on which the present Board—given the extensive material and submissions put before it—is as well placed as the Lieutenant Bailiff and Court of Appeal, then or now, to determine.73

64

Ibid, at [22] (PC). Ibid, at [36] (CA). 66 Ibid, at [26] (PC). 67 Ibid, at [45] (PC). 68 Armitage v Nurse [1997] EWCA Civ 1279, [1998] Ch 241 [Armitage]. 69 Ibid, at 253–54. 70 Ibid. See the clarification by the Privy Council on the contexts in which gross negligence is recognised in English law: Spread Trustee, above n 12, at [50]–[51] (PC). 71 Another argument concerning a subsequent Trusts (Guernsey) Law 2007 was also decided in favour of the trustee: Spread Trustee, above n 12, at [75]–[79] (PC). 72 Spread Trustee, above n 12, at [111] (PC). 73 Ibid, at [127] (PC). 65

Legislative Challenges to Orthodoxy 155 Lady Hale and Lord Kerr dissented. The problem for the minority was that they doubted whether English law was reasonably clear on the point in 1988 and, even if so, whether the courts in Guernsey would have followed English law. Lady Hale’s view was that the addition of ‘or gross negligence’ by the amendment could be seen as simply a clarificatory ‘minor change’.74 Lady Hale’s reasoning is marked by a desire to keep the correctness of the position stated in Armitage open, pending consideration by the Supreme Court in the future.75 Lord Kerr believed that English law would probably have given effect to such a clause in 1989.76 The peculiarity of the legislative affirmation in the 1989 Law was that it was not clear whether there was any ‘orthodox’ view of the relevant principles in the first place. It is therefore difficult for any principle to be ‘affirmed’, and that renders the interpretation of the statute a somewhat counterintuitive intellectual exercise. Two dicta from Lord Reid are relevant here. The first is from Rookes v Barnard:77 ‘One assumes that Parliament knows the law, but if the law is notoriously uncertain we must not attribute to Parliament prescience of what the law will ultimately be held to be’.78 The second is from a less well-known case: ‘the mere fact that an enactment shows that Parliament must have thought that the law was one thing does not preclude the courts from deciding that the law was in fact something different’.79 That approach is the one taken by the minority in Spread Trustee. Not only is the common law not to be ossified for evermore in the state in which it was at the time of a legislative enactment; it is also open to the court to conclude that the common law was not what it was assumed to be.

III. THE LIMITATIONS OF THE LIMITATION ACT 1980

As noted in AB, ‘Limitation of actions is a creature of statute, not of the common law’.80 And so the problem discussed at the end of the previous part, namely that the common law may turn out to have been different from the basis upon which legislation was enacted, is also found in the context of the Limitation Act 1980. It is compounded by the language in the statute, which does not necessarily reflect the way in which the relevant common 74

Ibid, at [139] (PC). Ibid, at [129] (PC). 76 As the Court of Appeal subsequently did in Armitage, above n 68. 77 Rookes v Barnard [1964] UKHL 1, [1964] AC 1129. On the presumption against the alteration of the common law more than expressly provided, see Duxbury, above n 4, at 36–39. 78 Ibid, at 1174 (Lord Reid). 79 Birmingham Corporation v West Midland Baptist (Trust) Association (Inc) [1970] AC 874, 898. 80 Above n 13, at [162]. 75

156 James Lee law claims are framed.81 We shall take two cases in order to examine the issues: AB on negligence and Central Bank of Nigeria v Williams82 on dishonest assistance. The relationship between legislation and the common law in the context of limitation is a reflexive one: the tactical advantage of limitation (which may be to the claimant or the defendant) leads to contortions, whether in the interpretation of the language of the statute or in the development of cause of action itself or even in both. Thus the mere fact of the enactment of limitation statutes may present challenges to orthodoxy. A. AB v Ministry of Defence: The Test Veterans Case In AB the Supreme Court considered whether veterans of British nuclear tests from the 1950s could bring claims in respect of their alleged exposure to radiation. The veterans, and veteran groups of which they were members, had a long-standing belief that various illnesses from which they suffered were caused by this exposure, yet they did not commence any claim until after they had seen, in 2007, the results of the ‘Rowland Report’, a study of New Zealand test veterans.83 The decision involved the analysis of sections 11 and 14 of the Limitation Act 1980. Section 11 provides that the limitation period for actions in respect of personal injuries is three years from either ‘the date on which the cause of action accrued’ or ‘the date of knowledge (if later) of the person injured’.84 The definition of the ‘date of knowledge’ in section 11 is provided by section 14, which lists various facts of which knowledge is required: (a) that the injury in question was significant; and (b) that the injury was attributable in whole or in part to the act or omission which is alleged to constitute negligence, nuisance or breach of duty; and (c) the identity of the defendant; and (d) if it is alleged that the act or omission was that of a person other than the defendant, the identity of that person and the additional facts supporting the bringing of an action against the defendant; and knowledge that any acts or omissions did or did not, as a matter of law, involve negligence, nuisance or breach of duty is irrelevant.

81 I examine the problems concerning the limitation periods for equitable wrongs in another essay: Lee, above n 11. 82 Central Bank of Nigeria v Williams [2012] EWCA Civ 415, [2012] WLR (D) 108 [Williams]. 83 The report is available at: www.llrc.org/epidemiology/subtopic/nzvetsrept.pdf (last accessed January 2013). 84 Section 11(4)(a) and (b).

Legislative Challenges to Orthodoxy 157 The Supreme Court split 4:3 in holding that the veterans’ claims were timebarred. Lord Wilson set out the central policies behind such statutes: The statutes of limitation, which stretch back to 1540, have been in place for two main reasons. One is to protect defendants from being vexed by stale claims. They are Acts of peace: … The other is to require claims to be put before the court at a time when the evidence necessary for their fair adjudication is likely to remain available, or, in the words of the preamble to the 1540 Act, at a time before it becomes ‘above the Remembrance of any living Man … to … know the perfect Certainty of such Things’.85

Those two reasons militated against permitting recovery on the facts in AB. For the majority, led by Lord Wilson, there was an incongruity in accepting the claimants’ arguments (which the minority did), as ‘it is a legal impossibility for a claimant to lack knowledge of attributability for the purpose of section 14(1) at a time after the date of issue of his claim. By that date he must in law have had knowledge of it’.86 The minority (Lord Phillips, Lady Hale and Lord Kerr) dissented on the basis that, notwithstanding the case law, which had recognised the relevance of the claimants’ belief, the statutory language was clearly focused on ‘knowledge’ and the claimants could not ‘know’ something of which there was no evidence. No matter how strong the claimants’ suspicions, beliefs or convictions that the Ministry of Defence was to blame for their subsequent illnesses, that could not equate to knowledge. Lady Hale seemed to accept that this conclusion was problematic, but viewed it as inevitable: ‘There is a very good case for the law being different. But I do not think that we should translate our view of what a sensible law of limitation would say into our view of what it does say’.87 The case is chiefly of interest in the context of this chapter because of the court’s more general observations about the state and operation of the law of limitation. Lord Walker explained that the ‘words of the 1980 Act themselves must be the starting-point, illuminated where necessary by judicial exposition’.88 Yet those words must then be applied to the relevant principles of the common law. Lady Hale, a former academic and Law Commissioner, offered more strident views of the role of legislative intervention in this area: ‘this is a field in which statute has intervened for policy reasons. But in policy terms the current regime for personal injury claims, combining discoverability with discretion, might be thought to have the worst of all possible worlds.’89

85 86 87 88 89

AB, above n 13, at [6]. Ibid, at [3]. Ibid, at [174]. Ibid, at [68]. Ibid, at [164].

158 James Lee AB primarily concerned the construction of the words of the relevant sections, rather than trying to fit the cause of action into particular provisions: section 11 deliberately covers every possible claim for personal injury regardless of the nature of the action. But the mismatch can even lead to more fundamental questions of tort law orthodoxy, such as the musing by Lady Hale that, rather than straining to construe the limitation provisions, it ‘might be better still if the cause of action accrued at the date of the wrongful act or omission rather than at the date of damage’.90 In her view, the majority instead sought to circumvent the apparent undesirability of allowing the instant claims to proceed by endorsing an interpretation of the statute which was inconsistent with the language.

B. Central Bank of Nigeria v Williams The applicable limitation period for equitable wrongs is unclear, having ‘a somewhat chequered history’,91 or if one prefers the more frank view of Swadling, ‘the current limitation regime … is a mess’.92 I have dealt in more detail with the issues elsewhere,93 but one example suffices to make a broader observation about the role of legislation. In Williams the claimant alleged that he had been the victim of a sting operation by the Nigerian Security Services and sought to reclaim some $6 million that he had paid pursuant to a trust arrangement. The relevant facts took place in 1986, but the claim was only brought in 2010. The general rule for actions by a beneficiary in respect of a breach of trust (or to recover trust property no longer in the possession of the trustee) is contained in section 21(3) of the Limitation Act 1980: six years from the date on which the cause of action accrued. So if the general period applied, then the claim was statute-barred. However, the claimant relied upon section 21(1)(a) which provides that no limitation period applies to an action by a beneficiary under a trust if the action is ‘in respect of any fraud or fraudulent breach of trust to which the trustee was a party or privy’. Can a claim in dishonest assistance be said to be ‘in respect of any fraud’ to which the trustee was a party?94 The difficulty with such a conclusion is that, since the Privy Council’s decision in Royal Brunei Airlines v Tan,95 it 90

Ibid, at [166]. Peconic Industrial Development Ltd v Lau Kwok Fai [2009] HKCFA 16, [2009] 2 HKLRD 537, [17], on the Hong Kong Limitation Ordinance, which borrows from the English provisions. 92 W Swadling, ‘Limitation’ in P Birks and A Pretto (eds), Breach of Trust (Oxford, Hart Publishing, 2002) 344. 93 Lee, above n 11. 94 There is also an argument as to whether references to dishonest assistants as ‘constructive trustees’ mean that they may fall within the meaning of ‘trustee’ for the purposes of the section, but I leave that to one side here: see ibid. 95 Royal Brunei Airlines v Tan [1995] UKPC 4, [1995] 2 AC 378. 91

Legislative Challenges to Orthodoxy 159 is not necessary to show that the trustee was dishonest: our focus is on the accessory’s role, and so there can be liability for dishonestly assisting in an innocent breach of trust. The Court of Appeal in Williams recognised the anomaly, but nevertheless held that the claim fell within section 21(1)(a) and was thus not out of time. The point for present purposes is that Black LJ identified that the problem arose precisely because the common law would continue to develop independently of the statute: The intervention of Royal Brunei Airlines v Tan in the jurisprudence gives rise to the anomaly identified by CBN which must, I think, be regarded as unforeseen … Whilst the anomaly cannot be ignored, one must, I think, remember that it is an accident of the developing law and would not have been in the contemplation of the pre-Tan draftsman.96

C. Reform As Lady Hale has noted, ‘The current law of limitation is complicated and incoherent’.97 The English law of limitation suffers from its history of incremental reform, particularly responding to individual decisions.98 The Law Commission did propose reform of the Act,99 which would have created a general three-year limitation period100 applying to ‘civil claims’, defined as ‘a claim made in civil proceedings in which the claimant seeks (a) a remedy for a wrong, (b) restitution, or (c) the enforcement of a right’.101 That would seem to cover all claims within the law of obligations, and this approach at least also has the advantage of clarity and certainty. These proposals were accepted by the Labour government in July 2002, subject to further consultation.102 Many years later, the government announced that it was abandoning the proposals on the basis that ‘consultation with key stakeholders has demonstrated that there are insufficient benefits and potentially large-scale costs associated with the reform. In addition, the courts have remedied some of the most significant difficulties with the law that the Law Commission identified’.103 The delay in responding to Law Commission proposals in areas of civil law reform has been identified as a matter of ‘deep regret’ by both the former Chairman of the

96

Williams, above n 82, at [58]–[59]. AB, above n 13, at [163]. 98 See generally Lady Hale’s opinion in AB. 99 Law Commission, ‘Limitation of Actions’ (Law Comm No 270, 2001). 100 With a ‘long stop’ of 10 years: ibid. 101 Ibid, Appendix A, cl 1(4) of the Draft Bill. The Law Commission explains its reasoning as to trusts at [4.94]–[4.106]. 102 Lord Irvine, HL Deb 16 July 2002, col WAc127, in response to question HL5294. 103 Parliamentary Under-Secretary of State for Justice (Bridget Prentice), 19 Nov 2009, col 13WS. The principal case in the interim period was the decision of the House of Lords in A v Hoare [2008] UKHL 6, [2008] 1 AC 844. 97

160 James Lee Commission104 and the House of Commons Justice Committee105 and has been ‘deplored’ by Sir Henry Brooke.106 But the problems with the limitation framework are not merely that some claimants miss out on, or benefit from, a more favourable limitation period when we think they should not. Any regime may result in some unfortunate instances. What the example of Williams demonstrates is that the difficulties encountered are partly born out of the inability of legislation to keep pace with the common law, let alone anticipate future developments. So equitable wrongs, for example, are claims which have taken on a greater commercial (and, for that matter, academic) significance in the three decades since the Limitation Act 1980 was enacted.107 Of course, that is partly the point: statutory ambivalence regarding certain actions creates an incentive to resort to them, so as to be able to argue for the more favourable option. The best the judges can do is to squeeze these claims into the wording of provisions which do not directly cover them. Such an approach in turn affects the more general orthodox principles of the relevant obligations.

IV. CHALLENGING LEGISLATIVE CHALLENGES TO ORTHODOXY

This part considers some of the constitutional constraints that may operate on the legislative variation of orthodox principles of private law, particularly in the light of provisions which protect possessions and property.

A. European Convention on Human Rights and Fundamental Freedoms In AXA, the United Kingdom Supreme Court had to consider whether the Damages (Asbestos-Related Conditions) (Scotland) Act 2009 was compatible with the human rights of the petitioners.108 Under section 29 of the Scotland Act 1998, which defines the legislative competence of the Scottish 104

Law Commission, ‘Annual Report 2010–11’ (Law Comm No 328) 2. House of Commons Justice Committee, Draft Civil Law Reform Bill: Pre-Legislative Scrutiny (Sixth Report) (HC 2009–10, 300–II) 55–56. 106 In his response to the Consultation on the Civil Law Reform Bill: www.judiciary.gov.uk/ NR/rdonlyres/36B981B526AC46B0A9642ADCED109055/0/henrybrookeresponsecivillawre formbill.pdf (last accessed January 2013). 107 So too the law of unjust enrichment: see J Lee, ‘Unjust Enrichment in a Parallel Universe? Restitution of Tax, Limitation and European Law’ [2013] Lloyd’s Maritime and Commercial Law Quarterly 11. 108 There was also a public law challenge, which turned on the status of Acts of the Scottish Parliament, but that is beyond the scope of the present chapter: see, eg, C Himsworth, ‘The Supreme Court Reviews the Review of Acts of the Scottish Parliament’ [2012] PL 205; M Elliott, ‘Holyrood, Westminster and Judicial Review of Legislation’ (2012) 71 CLJ 9; A McHarg, ‘AXA General Insurance Ltd v Lord Advocate’ (2012) 16 Edinburgh Law Review 224. 105

Legislative Challenges to Orthodoxy 161 Parliament, any legislation of the Scottish Parliament is not law insofar as it is incompatible with any of the rights under the European Convention on Human Rights and Fundamental Freedoms (‘ECHR’).109 The Supreme Court unanimously upheld the Act’s validity as being within the Parliament’s legislative competence.110 Although the validity of Westminster legislation is not affected by incompatibility with human rights,111 parties may nevertheless seek either a compatible interpretation of such statutes112 or a declaration of such incompatibility.113 The issues raised in AXA are therefore relevant to all United Kingdom legislation. The Act was a response to the decision of the House of Lords in Rothwell v Chemical and Insulating Co,114 in which the House of Lords had held that pleural plaques, which are asymptomatic changes in the pleural membrane which serve as evidence of asbestos exposure, did not amount to actionable damage. Thus the orthodoxy in issue was the principle (or ‘truism’)115 that a claim in negligence requires actionable damage. The Act was a legislative response by the Scottish Parliament in order to reverse the outcome of Rothwell. Similar legislation was enacted in Northern Ireland, but not in England and Wales (where a system of ex gratia payments was established by the government instead). Section 1 of the Act provides as follows: (1) Asbestos-related pleural plaques are a personal injury which is not negligible. (2) Accordingly, they constitute actionable harm for the purposes of an action of damages for personal injuries. (3) Any rule of law the effect of which is that asbestos-related pleural plaques do not constitute actionable harm ceases to apply to the extent it has that effect.116

However, the section expressly purported not to affect ‘any enactment or rule of law which determines whether and in what circumstances a person may be liable’ beyond the change effected by its other sections.117 The Act thus appears to be ‘affirming’ all the other relevant principles of law and merely changing the legal judgment on whether pleural plaques amount to actionable damage. 109

Specifically s 29(1) and (2)(d). See J Lee, ‘Legislative Interventions, Human Rights and Insurance’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 9. 111 Human Rights Act 1998, s 3(2)(b) and s 4(6)(a). 112 Ibid, s 3. 113 Ibid, s 4. 114 Rothwell v Chemical and Insulating Co [2007] UKHL 39, [2008] AC 281 [Rothwell]. 115 J Stapleton, ‘The Gist of Negligence: Part 1: Minimal Actionable Damage’ (1988) 104 LQR 213, 213. 116 Section 2 provides, in similar words, that asymptomatic pleural thickening and asbestosis are also injuries which are not negligible. It is unnecessary here to distinguish between these three injuries. Section 3 stops the limitation period from running for the time between the decision in Rothwell and the coming into force of the statute. 117 Section 1(4). 110

162 James Lee In the Supreme Court, the argument focused on a challenge under Article 1 of the First Protocol (‘A1P1’), the first paragraph of which provides that: Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.

A different type of challenge, not pursued before the Supreme Court in AXA, focuses on the right under Article 6(1), which provides that a claimant is entitled to a fair hearing in ‘the determination of his civil rights and obligations’ by ‘an independent and impartial tribunal established by law’. In the courts below, there had been some discussion on the question whether, even though a claimant’s chose in action counts as a ‘possession’, the same could be said of the situation where the defendant does not have any obligation to compensate a claimant for ‘harm’ which is not actionable. The Supreme Court, however, was satisfied with a more general proposition as to the engagement of Article 1, as Lord Hope concluded: A person’s financial resources … are capable of being possessions within the meaning of A1 P1 … [and therefore] the amount of money that they would be required to pay to satisfy their obligations under the insurance policies is a possession for the purposes of A1 P1.118

Similarly, Lord Reed was in no doubt that: In the case of an insurance company, the fund out of which it meets claims must therefore constitute a possession within the meaning of the article. Legislation which has the object and effect of establishing a new category of claims, and which in consequence diminishes the fund, can accordingly be regarded as an interference with that possession.119

Having concluded that the petitioners’ possessions were the subject of interference by the legislation, it was then necessary to decide whether that interference was proportionate. On proportionality, it could be argued that the legislation does not amount to a challenge to orthodoxy. Rather, it is merely the result of the legislature’s specific disagreement on the question of whether pleural plaques satisfy the legal test for actionable damage. So, on this argument, the Act does not affect the orthodox principle. That said, it does amount to the imposition of liability in circumstances where there previously was none. Furthermore, the pertinent point which is implicit in the court’s analysis is that there might be legislative reform of such a fundamental principle

118 119

AXA, above n 14, at [26]–[28] (SC). Ibid, at [114] (SC).

Legislative Challenges to Orthodoxy 163 so as to render the interference disproportionate.120 Given the tenor of the ECHR’s jurisprudence, that proposition must be correct. However, what we do not know after AXA is where judges would draw the line, either conceptually or on the specific facts: how great a departure from principle was the Act?121 If we look again at Rothwell, Lord Hope had started his judgment by describing the requirement of actionable damage as the ‘most basic … principle of the law of negligence’,122 while the Ministry of Justice, in deciding not to legislate for England and Wales, had expressly cited the fact that Rothwell was ‘a unanimous decision based on fundamental principles of the law of negligence’ as a reason not to intervene.123 In AXA itself, Lord Brown noted that it ‘is established law, however, that neither the risk of injury nor the apprehension of its happening are actionable’.124 To decide otherwise would have been an ‘admittedly novel’ development.125 One hypothetical challenge to orthodoxy which might well be disproportionate was offered during argument: what if Parliament had legislated that any substantial exposure to asbestos should constitute actionable damage?126 For Lord Brown, whilst the basic principles were clear, their application to pleural plaques was not so. Thus, Parliament could not ‘be regarded as having completely overturned a body of established law unambiguously supporting the appellants’ position so as to destroy what they could properly characterise as a legitimate expectation of being permanently immune from such claims’.127 Had Parliament so overturned a body of established law, however—and I suggest that such a decision is not that far away—then it would appear that the legislation would have been held to be invalid. The awkwardness of the intervention can further be seen from this purported affirmation of (or at least neutrality as to) the other applicable principles of law. The Act says nothing about causation or the extent of liability. It is therefore eminently possible that recovery for pleural plaques would be proportionate to each former employer’s material contribution to the risk of contracting the plaques,128 applying the Fairchild principle129 as

120

Lee, above n 110, at 13–14. AXA, above n 14, at [80] (SC) (Lord Brown). 122 Above n 114, at [36]. See also the decision of the Inner House in AXA, above n 14. 123 Ministry of Justice, Pleural Plaques Consultation Paper CP 14/08, [35]. 124 Above n 14, at [66] (SC). 125 Ibid, at [77] (SC). Of course cases may be made for such developments: see, eg, D Nolan, ‘New Forms of Damage in Negligence’ (2007) 70 MLR 59, 88. 126 AXA, above n 14, at [82] (SC). 127 Ibid, at [81] (SC) and see also Lord Mance at [96] (SC). 128 I made this point in Lee, above n 1, at 92–93 and in Lee, above n 110, at 14–15. See also more recently P Reid, ‘Damages for Pleural Plaques in Scotland: Only the Beginning?’ (2012) 28 Professional Negligence 52, 54–57. 129 Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, [2003] AC 32 [Fairchild]. 121

164 James Lee applied in Barker.130 Again, here we find an attempt to affirm a common law principle simply causing more problems. Although the pursuers in AXA were unsuccessful, the Supreme Court’s approach does reveal the potential for future challenges to be based upon interference with rights under A1P1. And, in appropriate cases, the Article 6 challenge on the basis of interference with pending claims may also afford an avenue for redress. Significantly, in Article 6 cases the state is subjected to a stricter test. In AXA, the legislature’s judgment as to the public interest was upheld because it could not be said to be ‘manifestly without reasonable foundation’. Under Article 6, the state must demonstrate that there are ‘compelling grounds of the general interest’,131 and the European Court of Human Rights requires that ‘any reasons adduced to justify such measures be treated with the greatest possible degree of circumspection’.132 Although it was not argued before the Supreme Court in AXA, at least Lord Brown would have held that the Act did not satisfy this standard,133 which underlines how finely balanced the case was.

B. Australia Legislative deprivation of a valuable chose in action was raised before the High Court of Australia in Haskins v The Commonwealth,134 in which a naval serviceman who had been detained for breaches of discipline alleged that certain provisions of the Military Justice (Interim Measures) Act (No 2) 2009 (Cth) interfered with his asserted cause of action against the Commonwealth for false imprisonment. On the facts, the High Court held that no action lay against the Commonwealth because he had been detained by an officer acting on the basis of what appeared on their face to be lawful orders.135 The claimant thus had no cause of action of which he could be deprived. But it is still worthwhile to explore the Australian position on such deprivations. Although not identical to the provisions under the ECHR, the Australian Constitution also places limits on the scope of legislative interference. However, the Australian Constitution also governs

130 Barker v Corus (UK) Plc [2006] UKHL 20, [2006] 2 AC 572 [Barker]. Although s 3 of the Compensation Act 2006 was enacted to reverse the Barker interpretation of Fairchild, the section can only apply to cases of asbestos-related mesothelioma: s 3(1)(b). See Mitchell, above n 2, at 188–94. 131 Zielinski, Pradal and Gonzalez v France (2001) 31 EHRR 19, [57]. 132 The National & Provincial Building Society, The Leeds Permanent Building Society and The Yorkshire Building Society v United Kingdom (1998) 25 EHRR 127, [112]. 133 AXA, above n 14, at [83] (SC). 134 Haskins v The Commonwealth [2011] HCA 28. 135 Ibid, at [67] (French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ).

Legislative Challenges to Orthodoxy 165 the relationship between the Commonwealth and the states, and so its provisions must be understood in that context. Section 51 of the Australian Constitution provides for the legislative powers of the Parliament. In section 51(xxxi) the Parliament is empowered to legislate in respect of ‘the acquisition of property on just terms from any State or person for any purpose in respect of which the Parliament has power to make laws’. The relevance of this wording for this chapter is the extent to which it may apply to the legislative removal or modification of causes of action. In Victoria v The Commonwealth (Industrial Relations Act Case), the High Court of Australia regarded it as ‘well established’ that: the guarantee effected by s 51(xxxi) of the Constitution extends to protect against the acquisition, other than on just terms, of ‘every species of valuable right and interest including ... choses in action’.136

Similarly, in a concurring opinion in Georgiadis v Australian & Overseas Telecommunications Corporation, Brennan J (as he then was) remarked that ‘it needs no extension of the meaning of “property” in s 51(xxxi) to comprehend a chose in action for damages for negligence causing personal injury’.137 McHugh J was more cautious as to whether such changes could relevantly engage section 51(xxxi).138 Certain opinions in the cases on section 51(xxxi) have suggested that whether the Commonwealth has the power to vary existing causes of action without compensating on ‘just terms’ depends upon the source of the right, since statutory rights are ‘of [their] nature … susceptible of modification or extinguishment’.139 Recent cases appear to have limited the scope of this proposition: ‘It is too broad a proposition … that the contingency of subsequent legislative modification or extinguishment removes all statutory rights and interests from the scope of s 51(xxxi)’.140 Instead, the inquiry must be directed to the specific context.141 The breadth of the Supreme Court’s approach in AXA, on the other hand, bypasses many of the nuances of this debate. Instead, the focus is on whether the interference can be justified. That is not the Australian approach.

136 Victoria v The Commonwealth (Industrial Relations Act Case) [1996] HCA 56, 187 CLR 416, 559 (Brennan CJ, Toohey, Gaudron, McHugh and Gummow JJ). 137 Georgiadis v Australian & Overseas Telecommunications Corporation [1994] HCA 6, 179 CLR 297, 312. 138 Ibid, at 328–29. 139 Ibid (Mason CJ, Deane and Gaudron JJ); Health Insurance Commission v Peverill [1994] HCA 8, 179 CLR 226 (Mason CJ, Deane and Gaudron JJ). 140 Attorney-General for the Northern Territory v Chaffey [2007] HCA 34, [24] (Gleeson CJ, Gummow, Hayne and Crennan JJ). 141 Telstra Corporation Limited v The Commonwealth [2008] HCA 7, [49] (Gleeson CJ, Gummow, Kirby, Hayne, Heydon, Crennan and Kiefel JJ).

166 James Lee There are limits to the comparison between the jurisdictions, as in Australia reforms of private law principles are largely the realm of the states, as demonstrated in the Civil Liability Acts,142 notwithstanding that the High Court of Australia has insisted upon ‘a unified common law which applies in each state but is not itself the creature of any State’.143 Section 51(xxxi) only applies to the powers of the Commonwealth Parliament. That said, comparing the Australian and British case law on legislative interference with or removal of causes of action reveals a contrast between the literal focus of the High Court and the more purposive approach of the ECHR and European Union lines of authority which shape the British analysis. There is therefore greater scope within the British framework for challenging such interventions. Allen has explored the High Court’s development of the jurisprudence under section 51(xxxi) and noted that whilst there are shared themes between the provisions there is some scepticism over whether it is appropriate for the High Court to borrow the proportionality-based approach from ECHR jurisprudence.144 In both countries, the challenge for the courts in such cases is that virtually any reform of tort law will involve providing either that there is liability where there previously was not or that there is no liability where there previously was. A balance must be struck between protecting proprietary interests and respecting that certain law reforms fall within the province of the legislature.

V. CONCLUSIONS: OSSIFYING ORTHODOXY?

As Fuller noted, ‘If every time a man relied on existing law in arranging his affairs, he were made secure against any change in legal rules, the whole body of our law would be ossified forever’.145 What emerges from the foregoing discussion is that legislative challenges to orthodoxy can cause significant difficulties and encounter the law of unintended consequences. The consequences are often more extensive than the legislature seems to realise, but that is a function of the process of legal change when it is ‘accomplished by ignoring the sensibilities of lawyers, resulting in content that strikes a dissonant note when placed next to more general, common law principles’.146 Indeed, questions arise even when the intention of the

142

See above n 7. Kable v Director of Public Prosecutions (NSW) [1996] HCA 24, 189 CLR 51, 112 (McHugh J). See R French, ‘State of the Australian Judicature’ (2010) 84 Australian Law Journal 310. 144 T Allen, ‘The Acquisition of Property on Just Terms’ (2000) 22 Sydney Law Review 351, 368. 145 L Fuller, The Morality of Law, revised edn (New Haven, Yale University Press, 1969) 60. 146 Mitchell, above n 2, at 191. 143

Legislative Challenges to Orthodoxy 167 legislation is not to alter the common law principle at all. And that is a challenging paradox. We have also seen that, in both England and Australia, there are checks on the extent to which legislation may interfere with existing private law rights and obligations (or non-obligations, where liability is imposed). Although such constraints are examined in narrow and specific contexts, the jurisprudence is in principle equally applicable (or even more so) to more wide-ranging reforms of fundamental principles. However, there are and must be limits to the effect of this jurisprudence. In AXA, Lord Reed quoted Fuller’s observation above, using ‘ossification’ in a different sense from how I have used it previously in this chapter.147 Although, as I have argued, there is considerable potential for challenging legislative interventions in the ECHR jurisprudence applied in AXA, it must be recognised that the legislature is to be afforded a great degree of latitude in its determination of matters of social policy, even if it is not entirely clear what that means or how great the latitude is. A broader conclusion is that we should require more of our legislators: both a greater effort to engage faithfully with the relevant principles of common law and the recognition that any intervention may have wider implications (even when legislation is narrowly intended). There must be deeper understanding of ‘the continuing inter-relationship between the common law and statute and the policy choices that courts and legislatures … must make in their respective, but not mutually exclusive, roles of fixing and distributing the burden of liability’.148 That will lead not only to a more coherent law of tort but also, on a practical level, to more effective legislation, achieving whatever policy aims Parliament should identify. As we have seen, it may also ensure compliance with important constitutional requirements. Beatson has described the archetypal common lawyer’s view of the relationship between judge-made law and legislation as ‘oil and water’.149 That view can be avoided by a constructive engagement between the courts and the legislature and by requiring ‘law reformers and legislative draftsmen to think carefully about how their solutions fit into the overall structure of the legal system and to be as helpful as they can about the reasons for their enactment’.150 And the judiciary have their role too, for, as MacCormick emphasised: everybody knows that the job of legislation is never completed when the text of a statute leaves the legislature. In due course problems, foreseen or unforeseen by the legislature, will arise concerning the exact scope of a legislative text, or its

147

Above n 14, at [120]. McDonald, ‘Teaching Torts’, above n 9, at 186. 149 J Beatson, ‘Has the Common Law a Future?’ (1997) 56 CLJ 291; J Beatson, ‘The Role of Statute in the Development of Common Law Doctrine’ (2001) 117 LQR 247. 150 Beatson, ‘Has the Common Law a Future?’, above n 149, at 313. 148

168 James Lee exact application to some situation that arises and that has significance enough for a party affected to litigate the question. The final process of concretization or determination … will still have to take place through judicial decision.151

This inevitability of judicial interpretation of the statute, and parallel judicial development of the common law independent of the context of a given statute, must be accepted. But it is also inevitable that statute and common law will be forced to meet again. Burrows has made the convincing argument that ‘statutes and common law must be seen as integrated parts of the whole law of obligations’.152 I have further urged that we must also endeavour to understand the manner of such integration; the ‘complex and myriad ways in which statute and common law rules interact’.153 Arguments from coherence have a tendency to seem excessively deferential to the ideals of tort orthodoxy,154 but regard for coherence has the potential to contribute to effectiveness and efficiency. Although legislating more coherently is important, my argument has not been that our common law orthodoxies should be regarded as ossified. As Lord Dyson MR has recently remarked, ‘history has shown that the product of today’s buccaneer sometimes becomes tomorrow’s orthodoxy’.155 If we are to have an integrated understanding of both common law and statutory law as part of the law of tort, we may need to rethink our ideas as to what amounts to ‘orthodoxy’ in the first place.

151 N MacCormick, Rhetoric and the Rule of Law (Oxford, Oxford University Press, 2005) 10. 152 Burrows, above n 3, at 258. 153 Dietrich, above n 9, at 150. 154 Eg J Conaghan, ‘Tort Law and Feminist Critique’ (2003) 56 Current Legal Problems 175, 182: ‘much of British tort scholarship betrays a wariness of ends-oriented analyses of tort, preferring instead to focus on questions of internal coherence, structure, and classification’. See too the work of Steve Hedley, such as S Hedley, ‘Rival Taxonomies within Obligations: Is There a Problem?’ in S Degeling and J Edelman (eds), Equity in Commercial Law (Sydney, Law Book Co, 2005) 77. 155 Dyson, above n 17, at 10.

7 Statutes and Civil Liability in the Commonwealth and the United States: A Comparative Critique NEIL FOSTER*

T

HIS CHAPTER CHALLENGES current orthodoxy on the question of how statutes can be used as a source of private law liability. In the United States the accepted view seems to be (at the risk of oversimplification) that: 1. At a state level, breach of many statutes will provide a basis for civil damages; but whether the courts will so find in a particular case is decided on broad ‘policy’ grounds, and an outcome is almost impossible to predict; 2. At a federal level, hardly any federal statutes will these days create civil liability. I want to suggest a radical revision of these views. It is ‘radical’ in the sense that it goes back (as the etymology of the word suggests)1 to the ‘roots’ of the development of the doctrine. The roots of both US state and federal jurisprudence in this area, as I hope to show, will be found in the classic United Kingdom decisions relating to the tort of ‘breach of statutory duty’. I want to suggest that if courts in the United States took into account the contours of that action as it has been, and still is, applied in the Commonwealth, there may be a way forward to deal with both of the major problems (that is, on the one hand unconstrained use of ‘policy-based’ considerations at state level in applying statutes; on the other hand, a narrow refusal at the

* Research for this article was mainly conducted at Villanova Law School, Pennsylvania. I would like to thank my colleague Professor Chaim Saiman and other colleagues at VLS for their warm hospitality and support during my visit in 2011. All errors in US or other law are, of course, my responsibility. 1 The first entry in the Oxford English Dictionary for the word is: ‘Of or relating to a root or to roots’ (from the Latin radix).

170 Neil Foster federal level to read the implications of statutes) which currently beset statute-based civil liability in US private law jurisprudence.

I. OVERVIEW OF THE ISSUES

The contours of the common law tort of negligence have been reasonably stable in the Commonwealth since Donoghue v Stevenson,2 and in the United States of America since MacPherson v Buick.3 These two legal systems require a court to find a duty of care owed by the defendant to the plaintiff, a breach of that duty, damage caused by that breach which is not too remote, of a type for which the law will provide compensation. In contrast, there is a fairly wide apparent divergence of approach when the question arises as to how standards of behaviour set by statute are relevant to civil liability. In the majority of Commonwealth countries,4 there is a specific tort action called ‘breach of statutory duty’ (‘BSD’), which provides that once certain preconditions are met,5 the defendant’s breach of a statute, which causes harm to the plaintiff, provides in itself a ground of civil liability.6 The situation (like much else in the law of torts) is much more complex in the United States. The majority of states adopt the view that breach of a statute by a defendant is not a separate tort action, but will provide the answer to the question of whether there has been a breach of duty in a negligence claim: it is ‘negligence per se’. However, there are some cases where breach of the statute will not automatically resolve the issue of breach, providing instead simply evidence that may be taken into account by the factfinder (perhaps by way of a presumption of breach). In both of these cases a wide range of policy-based judicial ‘excuses’ may be applied at the discretion of the court. In other cases, particularly (though not solely) when federal statutes are involved, the courts may still sometimes find that

2

Donoghue v Stevenson [1932] AC 562 (HL). MacPherson v Buick Motor Co 217 NY 382, 111 NE 1050 (1916), cited of course in the later decision of Donoghue v Stevenson, above n 2. See John CP Goldberg and Benjamin Zipursky, ‘The Moral of MacPherson’ (1998) 146 University of Pennsylvania Law Review 1733. 4 The exception being Canada, where the Supreme Court abolished this separate action in R v Saskatchewan Wheat Pool [1983] 1 SCR 205. 5 See Section II B below. 6 For an overview of this Commonwealth tort action, see C Sappideen and P Vines (eds), Fleming’s The Law of Torts, 10th edn (Pyrmont, Law Book Co, 2011) ch 18. I was the revising author for this latest edition of the new ch 18 and relevant parts of ch 24, ‘Employers’, dealing with the statutory tort. More detailed analysis of the action can be found in KM Stanton, Breach of Statutory Duty in Tort (London, Sweet & Maxwell, 1986) and KM Stanton, P Skidmore and M Harris, Statutory Torts (London, Sweet & Maxwell, 2003). For a defence of the continuing validity of the BSD action against some suggestions it should be abolished, see N Foster, ‘The Merits of the Civil Action for Breach of Statutory Duty’ (2011) 33 Sydney Law Review 67. 3

Statutes and Civil Liability in the Commonwealth and US 171 there is an ‘implied right of action’ which looks and sounds very similar to the Commonwealth BSD action. The aim of this chapter is not to resolve all the problems presented by the jurisprudence and commentary on these various forms of action, but to provide a starting point for comparison between the approaches that the two systems have adopted on this issue of implied statutory civil liability. It is hoped that a proper understanding of the development of the law, and the current differences between the two systems, may illuminate the choices that courts have made, illustrate some of the major differences that exist between the Commonwealth and the United States in the area of tort law, and suggest some ways forward for courts in the United States. A better understanding of both the similarities and the differences may also help commentators and courts to avoid the problems that can be created by assuming that the law in one system is the same as the other. There are, of course, many dangers in ‘cross-system’ comparisons, especially between the Commonwealth and the United States, some of which are helpfully illuminated by Jane Stapleton in an important article.7 In the United States tort law is usually regarded as a state matter, whereas in Commonwealth jurisdictions usually there is a unitary approach (even in a federation like Australia, where there is a ‘unified’ common law). In particular the role of the jury in US tort litigation must be taken into account in explaining the formulation of legal rules: ‘a covert concern with jury decision-making in the U.S. generates a pronounced tendency to crystallize rules of law with which the trial judge can govern access to the jury’.8 As will also become apparent in this chapter, the complexities created by the US federal division of powers are significant: Stapleton notes that ‘important aspects of U.S. tort law … have no close parallels elsewhere, such as the specific constitutional constraints on it recognized by the U.S. Supreme Court’.9 Stapleton also makes a number of invaluable comments about the phenomenon (as to which there is no real parallel in the Commonwealth system) of the Restatements of the law, especially relevant in the area of torts. The Restatements are not issued with any formal legal authority—they are not adopted by any legislature, nor does any judicial body formally approve them. They are intended in effect to be a summary or ‘harmonisation’ of the law of torts as it is found in the various state jurisdictions.10 Since they 7 J Stapleton, ‘Benefits of Comparative Tort Reasoning: Lost in Translation’ (2007) 3:1 Journal of Tort Law Art 6. For another helpful general overview of differences between the US and the European systems, see U Magnus, ‘Why is US Tort Law so Different?’ (2010) 1 Journal of European Tort Law 102. 8 Stapleton, above n 7, at 28. 9 Ibid, at 29. 10 See JCP Goldberg, AJ Sebok and BC Zipursky, Tort Law: Responsibilities and Redress, 2nd edn (Alphen aan den Rijn, Wolters Kluwer, 2008) 15: ‘Specifically, the Restatement aims

172 Neil Foster come with the imprimatur of respected academics and judicial officers (acting non-judicially), the law as summarised in a Restatement may be influential in further development of the law in court decisions. However, Stapleton notes that there is always going to be some debate as to whether the Restatement exercise is a ‘neutral’ attempt to summarise the ‘best’ or ‘most common’ law, or a ‘political’ exercise designed by at least some of the participants to push the law in a particular direction.11 In addition, a state court can always decline to follow the guidance given by a Restatement where it conflicts with judicial authority in that state. And, of course, different state courts, even if agreeing to follow a Restatement rule, may develop their own interpretation of the rule.12 Keeping these matters in mind, we turn to the tricky business of crosssystem comparisons. In Part II of the chapter, I briefly outline the ways that statutes create civil liability in the Commonwealth legal system. In Part III I discuss the rules developed for implied civil liability based on statutes in the United States. After outlining the two main options (‘negligence per se’ and ‘implied rights’), Part IIIA describes and critiques the ‘negligence per se’ (‘NPS’) analysis, which squeezes statutory obligations into the broad common law action for negligence, starting with a classic decision of Cardozo J. I outline the similarities and differences between the NPS approach and Commonwealth BSD approach, demonstrating the historical links between the two actions which are often not noticed, and pointing out where I think the NPS analysis goes wrong by ignoring the fundamental preliminary question of legislative intention. Ignoring this question has led to the development of an unsatisfactory policy-based set of ‘excuses’ which operate at a trial level, and a confusion of the questions of duty and breach, as well as incoherence in dealing with federal/state interactions. Part IIIB deals with the other stream of cases, creating ‘implied rights of action’, mostly based in federal statutes, which are shown to have the same roots as the NPS cases in the BSD jurisprudence. Yet the increasingly narrow approach taken to statutory interpretation in the federal sphere has the result that federal rights impliedly created by Congress go unenforced. In Part IV I bring out some general themes in the area and suggest that a better way forward may

to gather together and interpret decisional law coming out of all US jurisdictions in an effort to identify “black letter” law: rules and standards on which there is a wide degree of consensus among judges’. 11 For example, she quotes Frank J Vandall, ‘Constructing a Roof Before the Foundation is Prepared: The Restatement (Third) of Torts: Products Liability Section 2 (b) Design Defect’ (1997) 30 University of Michigan Journal of Law Reform 261, 279: ‘the ALI’s mission is no longer to restate the law, but rather to issue pro-manufacturer political documents’. A similar comment was made to me in private discussions with a colleague who teaches Torts in a US Law School. 12 See Stapleton, above n 7, at 40.

Statutes and Civil Liability in the Commonwealth and US 173 be for US courts to return to the roots of statutory civil liability claims in the Commonwealth BSD model.

II. STATUTORY CIVIL LIABILITY IN THE COMMONWEALTH

How do statutes create civil liability in the Commonwealth? Not all statutory obligations lead to civil liability; many of course may only be enforced by the criminal law. Statutes can have an impact on civil liability in many ways. This chapter will not be concerned with all of them.

A. Express Statutory Civil Liability in the Commonwealth A statute may, of course, explicitly create civil liability for breach. The best example of explicit statutory civil liability in the Australian context was the former section 52 of the Trade Practices Act 1974 (Cth). Under that section, a company was bound not to engage, in trade and commerce, in misleading or deceptive behaviour. Persons who could show that this had happened, and that as a result they had suffered loss, could under section 82 of the Act recover damages from the company.13 This action under section 82 was held by the High Court not to be an action in tort, but a special action created by the Act: see Marks v GIO Australia Holdings Ltd.14 The court held that, in determining what rule of ‘remoteness’ should apply to section 82 actions, neither the ‘contractual’ rule nor the ‘tort’ rule should automatically be applied; the rule to be applied was to be derived from the language of the statute itself, although analogies from the other areas could be considered. Similarly, Henville v Walker held that the doctrine of contributory negligence did not apply to section 82 actions, since it was not present in the Act, and the section 82 action could not be classified as a ‘tort’ action.15 Thus, not all statutory civil

13 This provision has now been replaced by the Australian Consumer Law (the ‘ACL’) in Sch 2 of the Competition and Consumer Act 2010 (Cth) from 1 January 2011. An equivalent prohibition on misleading and deceptive conduct is now contained in s 18 of the ACL (and damages for breach awarded under s 236 of the ACL). The new provision now applies not only to corporations but also, because of a co-operative agreement with the states, to individuals. See, eg, the Fair Trading Amendment (Australian Consumer Law) Act 2010 (NSW), inserting new s 28 into the Fair Trading Act 1987 (NSW). It seems likely that the status of the new action (as a non-tortious, statutory civil liability provision) will be the same as that of former s 52, but no occasion has yet arisen for the courts to comment on this aspect. 14 Marks v GIO Australia Holdings Ltd [1998] HCA 69, 196 CLR 494. 15 Henville v Walker [2001] HCA 52, 206 CLR 459, [140] (McHugh J).

174 Neil Foster liability is ‘tortious’. Parliament may even decide to abolish a tort action, and then substitute for it a specific statutory procedure.16 On the other hand, Parliament may choose to declare that a particular form of civil action is to be treated as a tort, either generically, or by ‘inserting’ it into the pre-existing common law in some way. In New South Wales, a statutory cause of action for damages for removal of support from a neighbouring building, which was not actionable in many cases under common law, was created by amendment to section 177 of the Conveyancing Act 1919 (NSW). Under section 177(1), this was accomplished by declaring that: ‘For the purposes of the common law of negligence, a duty of care exists in relation to the right of support for land’. So in this case the statutory right is treated as if it were part of the common law of negligence, presumably to avoid specifically defining issues like remoteness, limitation and contributory negligence.17 As we will see, while this is a rare expedient in the Commonwealth, the notion of ‘incorporating’ a statutory provision into the pre-existing law of negligence has become a significant aspect of US tort law, in the doctrine of NPS.

B. Implied Statutory Civil Liability in the Commonwealth The more interesting area, for the purposes of this chapter, however, is the question of whether a statute can create implied civil liability, without expressly so providing.18 In the Commonwealth, this is generally dealt with under the rubric of the tort of BSD.19 The modern Commonwealth view of the criteria for determining whether a statutory obligation creates a civil remedy is usually seen as well summed up in the judgment of Lord Browne-Wilkinson in X (Minors) v Bedfordshire County Council: a private law cause of action will arise if it can be shown, as a matter of construction of the statute, that the statutory duty was imposed for the protection of a

16 In New South Wales this was accomplished by the Trees (Disputes Between Neighbours) Act 2006 (NSW) s 5, which abolished the tort action for nuisance ‘as a result of damage caused by a tree to which this Act applies’, vesting jurisdiction to decide such disputes in the specialist Land and Environment Court. 17 For a recent case discussing this provision, see Lym International Pty Ltd v Marcolongo [2011] NSWCA 303. To illustrate the operation of this ‘hybrid’ statutory/ common law provision, Campbell JA held at [198] that a provision of the legislation which required something ‘not to be done’ had to be read, in light of the characterisation of the action as a part of the law of negligence, as an obligation to ‘take reasonable care’ to see that the action not be done. 18 For discussion of these issues in the context of UK legislation, see KM Stanton, ‘New Forms of the Tort of Breach of Statutory Duty’ (2004) 120 LQR 324. 19 With the exception of Canada. See above n 4.

Statutes and Civil Liability in the Commonwealth and US 175 limited class of the public and that Parliament intended to confer on members of that class a private action for breach of duty.20

Similarly, the High Court of Australia said in Byrne and Frew v Australian Airlines Ltd: A cause of action for damages for breach of statutory duty arises where a statute which imposes an obligation for the protection or benefit of a particular class of persons is, upon its proper construction, intended to provide a ground of civil liability when the breach of obligation causes injury or damage of a kind against which the statute was designed to afford protection.21

The Tenth Edition of Fleming’s The Law of Torts summarises the elements of the BSD action as follows: The elements of the civil action for breach of statutory duty … can be identified as: (a) the intention of Parliament to allow an action; (b) the plaintiff must fall within the ‘limited class’ of the public for whose benefit the statutory provision was enacted; (c) the damage suffered must also fall within the intended scope of the statute; (d) the obligation under the statute was imposed on the defendant; (e) the defendant must have breached the statute; and (f) that breach must have caused actual damage of some sort to the plaintiff.22

The general details of the action are dealt with in Fleming’s and in other standard tort texts.23 For those who are more familiar with the American context than the Commonwealth, it seems worth stressing a number of important features of the action. First, while there may have been some wavering on the question over the years, the orthodox view that is now accepted in all jurisdictions that recognise the action is that it is a separate action from that of the tort of negligence. It is not a sub-set of the law of negligence. So wrote Lord Wright in London Passenger Transport Board v Upson: I think the authorities … show clearly that a claim for damages for breach of statutory duty intended to protect a person in the position of the particular plaintiff is a specific common law right which is not to be confused in essence with a claim for negligence.24

20

X (Minors) v Bedfordshire County Council [1995] UKHL 9, [1995] 3 All ER 353, 364. Byrne and Frew v Australian Airlines Ltd [1995] HCA 24; 185 CLR 410, 424. 22 Fleming, above n 6, at 424. This list of elements was cited with approval in Alcoa of Australia Ltd v Apache Energy Ltd [2012] WASC 209, [80]. 23 See above n 6, and also more generally H Luntz, D Hambly, K Burns et al, Torts: Cases and Commentary, 7th edn (Sydney, LexisNexis Butterworths, 2013) ch 10; M Jones and AM Dugdale (eds), Clerk & Lindsell on Torts, 20th edn (London, Sweet & Maxwell, 2010); J Murphy and C Witting, Street on Torts, 13th edn (Oxford, Oxford University Press, 2012); RP Balkin and JLR Davis, Law of Torts, 4th edn (Chatswood, LexisNexis Butterworths, 2009); K Barker, P Cane, M Lunney et al, The Law of Torts in Australia, 5th edn (South Melbourne, Oxford University Press, 2012). 24 London Passenger Transport Board v Upson [1949] AC 155, at 168 (HL). 21

176 Neil Foster More recently, the comments of Crennan and Kiefel JJ in the High Court of Australia decision in Stuart v Kirkland-Veenstra also support this wellestablished view: ‘The action for breach of statutory duty, although itself a tort, is regarded as distinct from the tort of negligence’.25 Second, there is still an orthodox role in the law of negligence for statutes. Essentially, breach of a statutory standard may be used as evidence of (though not conclusive evidence of) a breach of a duty of reasonable care. The judgment of Ipp JA in Talbot-Price v Jacobs sums up the modern Australian approach: It has long been the law that breach of a statute or regulation may be evidence of negligence but is not irrefutable proof of negligence. Every case has to be decided according to its own circumstances. The breach of a statute or regulation is not definitive of a duty of care, or the performance of that duty.26

III. IMPLIED STATUTORY CIVIL LIABILITY IN THE UNITED STATES

What, then, is the law in relation to implied statutory civil liability in the United States? Probably the most important thing to note at the outset is that there is no general law of torts in the United States. Under the division of powers between the federal Congress and the various states, the area of tort law was generally left to the states. While this could also be said of the Australian situation, one key difference between the US and Australian Constitutions is that the High Court of Australia is the final court of appeal from state Supreme Courts, not simply on ‘federal’ issues, but on all matters of law.27 The result that has been reached in Australia is that the common law of torts (as with the common law of contracts, or unjust enrichment, or indeed criminal law where it remains uncodified) is a unitary body of law that is interpreted authoritatively by the one final court. In other words, there is only one common law of torts in Australia.28 But in the United States, the received doctrine since the decision of the US Supreme Court in Erie Railroad v Tompkins29 is that the federal Congress

25

Stuart v Kirkland-Veenstra [2009] HCA 15, 237 CLR 215, [130]. Talbot-Price v Jacobs [2008] NSWCA 189, [56]. Other recent comments on this issue can be found in Leighton Contractors Pty Ltd v Fox [2009] HCA 35, [49] and Caltex Refineries (Qld) Pty Limited v Stavar [2009] NSWCA 258, [220]–[222]. 27 See the Australian Constitution, s 73(2). 28 John Pfeiffer Pty Ltd v Rogerson [2000] HCA 36, 203 CLR 503, [15]: ‘there is a single common law of Australia’. 29 Erie Railroad v Tompkins 304 US 64, 78 (1938). 26

Statutes and Civil Liability in the Commonwealth and US 177 has no specific power over tort law in the states. ‘Tort law is state law’.30 So anyone attempting to summarise ‘the’ US law on statutes and torts (or any other tort doctrine, for that matter) has a difficult task. Nevertheless, the Restatement tries to provide a general summary of majority views in the various states. In addition, despite the finding in Erie that tort law generally is a matter for the states, it is clear that the federal courts have had to develop a version of ‘federal tort law’ for various purposes legitimately within the legislative power of the federal Congress. And whether or not most US lawyers would regard it as ‘tort law’, US federal courts regularly deal with the question of whether civil actions may be implied from statutes. According to Dobbs, The Law of Torts,31 commonly regarded as one of the main authoritative torts texts in the United States, statutes may create civil liability in the following ways: 1. By being adopted by a court as the definitive standard of care, in the classic NPS analysis (§§ 314 ff). 2. By providing explicitly for civil recovery, although then perhaps ‘picking up’ the common law of negligence with specific adjustments. Dobbs refers particularly to the Federal Employers Liability Act (FELA), which applies to railroad workers.32 3. By impliedly creating a civil action, as in Bivens v Six Unknown Named Agents of the federal Bureau of Narcotics,33 where the US Constitution was held to create an implied civil remedy.34 In a later note Dobbs suggests that the implication of a right of civil recovery may be easier under a federal statute than that of a state, where there is no general body of tort law in the background: ‘federal statutes differ from state statutes in that there is no federal common law of tort, only law emanating from the Constitution and the federal statutes. Conceivably, one could more readily find that Congress implicitly created a tort claim in such a system than in the ordinary state-law system’.35 One might argue that, there being no ‘back-up’ system to allow wrongs to be compensated for, 30 B Kritchevsky, ‘Tort Law is State Law: Why Courts Should Distinguish State and Federal Law in Negligence-Per-Se Litigation’ (2010) 60 American University Law Review 71, 75. 31 Dan B Dobbs, The Law of Torts (St Paul, Minn, West Publishing, 2000) § 133. For other treatments of the area see WL Prosser, WP Keeton, DB Dobbs et al, Prosser and Keaton on Torts, 5th edn (St Paul, Minn, West Publishing, 1984) § 36; FJ Vandall, E Wertheimer and MC Rahdert, Torts: Cases and Problems, 2nd edn (Los Angeles, LexisNexis, 2003) 231–45; VR Johnson and A Gunn, Studies in American Tort Law, 3rd edn (Durham, NC, Carolina Academic Press, 2005) 304–27. 32 45 USCA § 51. For a recent example of the continuing relevance of FELA, see the Supreme Court decision in CSX Transportation Inc v McBride 131 S Ct 2630 (2011). 33 Bivens v Six Unknown Named Agents of the federal Bureau of Narcotics 403 US 388 (1971) [Bivens]. 34 For comment on recent developments see CM Vazquez and SI Vladeck, ‘State Law, the Westfall Act, and the Nature of the Bivens Question’ (2013) 161 University of Pennsylvania Law Review 509. 35 Dobbs, above n 31, at § 135.

178 Neil Foster it should be easier to imply the existence of civil liability in federal statutes. This area of ‘implied rights of action’ is discussed in much more detail in section B below. 4. By explicitly limiting or abolishing tort liability. In the US context one important question is whether a federal statute setting a particular course of behaviour can preclude the operation of (or ‘pre-empt’) state tort law.36 Broadly speaking, we may say for the purposes of this chapter that implied statutory civil liability arises under US law in the following ways: 1. In ‘ordinary’ tort actions under state law, courts will apply a statute in some cases as ‘NPS’, that is, they will read a statutory standard into the ‘breach’ stage of a negligence action. This represents, according to the Restatement, a majority approach, discussed below in section A(i). In some states, however, breach of a statute may be regarded as providing a ‘presumption’ of breach that needs to be rebutted by the defendant, or in other states simply as ‘evidence’ of breach in a negligence claim, as discussed in section A(ii). 2. In some cases the courts may rule that a statute itself creates an ‘implied right of action’. Perhaps because of the strength of the NPS jurisprudence at the state level, this argument seems to be raised very rarely in relation to state statutes (though there seems no doubt that, in theory, it could be).37 But because of the lack of involvement of federal courts in general tort law following Erie, and because for a number of reasons federal courts may provide an attractive venue for plaintiffs, claims that a federal statute has created an implied right of action have often been made. As we will see, as these matters are clearly within federal jurisdiction they have often been considered by the US Supreme Court.

A. As Part of the Law of Negligence Application of statutory standards regularly comes before US state courts as part of negligence cases. (i) Negligence Per Se The majority view is that a statute, where it is applicable, will conclusively determine the ‘breach’ question in a negligence action. The Restatement

36 Ibid, at § 133. See Geier v American Honda Motor Co, Inc 529 US 861, 120 S Ct 1913 (2000), holding that where a federal authority had specified that cars made in a particular year need not be fitted with airbags, a tort claim under state law alleging negligence due to lack of an airbag could not succeed. 37 See the Illinois litigation in Citizens Opposing Pollution v ExxonMobil Coal USA 404 Ill App 3d 543, 556, 936 NE 2d 181, 193 (2010), discussed briefly below.

Statutes and Civil Liability in the Commonwealth and US 179 of the Law (Third) Torts: Liability for Physical and Emotional Harm (Tentative Draft No 5) puts it this way: § 14. Statutory Violations as Negligence Per Se An actor is negligent if, without excuse, the actor violates a statute that is designed to protect against the type of accident the actor’s conduct causes, and if the accident victim is within the class of persons the statute is designed to protect.38

It is also worth noting how the Restatement (Second) treated the issue: § 286. When Standard Of Conduct Defined By Legislation Or Regulation Will Be Adopted The court may adopt as the standard of conduct of a reasonable man the requirements of a legislative enactment or an administrative regulation whose purpose is found to be exclusively or in part (a) (b) (c) (d)

to protect a class of persons which includes the one whose interest is invaded, and to protect the particular interest which is invaded, and to protect that interest against the kind of harm which has resulted, and to protect that interest against the particular hazard from which the harm results.39

The contrast between the two formulations is interesting, but probably not as great as it appears. The Restatement (Second) seems to give more discretion to a court to adopt a statute or not (‘may adopt’), whereas the Third seems more directive (the actor ‘is negligent’). But the words ‘without excuse’ in the Restatement (Third) give the clue to the fact that, while apparently a strict rule, there is still a wide scope for ‘excuses’ to be accepted even under the latest formulation.40 The origin of the NPS doctrine is usually found in the judgment of Cardozo J in Martin v Herzog.41 The case was interesting because it arguably involved both plaintiffs and defendant being in breach of a statutory provision. The plaintiffs, riding in a carriage, were injured when struck by the defendant’s motor vehicle near dusk. The issue that became important was whether, by failing to have a lantern displayed on the carriage contrary to a local statute, the plaintiffs were guilty of contributory negligence. 38 Restatement of the Law (Third) Torts: Liability for Physical and Emotional Harm (Tentative Draft No 5) (2005) § 14. 39 Restatement (Second) of Torts (1965) § 286. 40 So much so that the next section of the Restatement (Third) is entitled ‘Excused Violations’. 41 Martin v Herzog 126 NE 814 (NY 1920) [Martin]. No doubt this judgment, as well as the future development of the area, was also heavily influenced by the important article by E Thayer, ‘Public Wrong and Private Action’ (1913–14) 27 Harvard Law Review 317. Professor Thayer argued strongly that when evidence of statutory breach emerged in a negligence case (dealing with prohibitory legislation), the breach must be seen as negligence per se and not merely as evidence of breach (at 322).

180 Neil Foster However, the decision of the court has subsequently been taken to be authoritative on the question of liability for negligence in general. The trial judge had allowed the jury to decide whether breach of the statute was careless in the circumstances. But the majority of the New York Court of Appeals, concurring in a judgment written by Cardozo J, agreed with his comments that ‘the unexcused violation of the statutory signals is more than some evidence of negligence. It is negligence in itself’. The possibility that the defendant’s vehicle had crossed the centre line of the road before the collision (also in breach of a statute) was not in the end taken into account; the plaintiffs’ breach of the law, however, was to preclude them from recovery (at a time when ‘contributory negligence’, as in the UK and elsewhere in the Commonwealth, was a complete defence rather than a matter for apportionment of damages). Other elements of the judgment would find their way into the later law. One aspect of the finding was that this was a ‘statute intended for the protection of travellers on the highway, of whom the defendant at the time was one’.42 This requirement is reflected, of course, in later formulations in the Restatement. To the Commonwealth lawyer a question that arises is: why choose to incorporate the statute into the law of negligence, as opposed to treating it as a separate tort? Much more work needs to be done before a serious answer can be offered. But it is interesting to note that Cardozo J himself was conscious of the debate. He commented: ‘There may be times when, if jural niceties are to be preserved, the two wrongs, negligence and breach of statutory duty, must be kept distinct in speech and thought … In the conditions here present they come together and coalesce’.43 He cited English texts for this proposition in the form of Clark & Lindsell and Salmond, as well as referring to the (then reasonably recent) decision of the US Supreme Court in Texas & Pacific Railway Company v Rigsby,44 which is usually cited as a major landmark in the development of ‘implied rights of action’. If Cardozo J himself was sufficiently clear in thought to realise the distinction, his comments on this issue seem to have been forgotten by later US state courts, which have continued the pattern set in Martin of regularly incorporating the statutory standard into the law of negligence, rather than discussing in detail whether statutes of the type concerned should give rise to stand-alone civil liability. It is instructive to compare the development of the law of the Commonwealth on this question of whether a breach of traffic rules would automatically create (or, as in Martin, remove) civil liability. Only a few 42

Martin, above n 41, at 815. Ibid. 44 Texas & Pacific Railway Company v Rigsby 241 US 33, 39–40 (1916) [Rigsby], discussed in more detail in section B below. 43

Statutes and Civil Liability in the Commonwealth and US 181 years later the English Court of Appeal considered the question in Phillips v Britannia Hygienic Laundry Co Ltd.45 The issue was approached in what might be called the orthodox way, by discussing whether the particular statute (effectively requiring vehicles on the road to be in a safe condition) was intended to create liability. All the judges agreed that it was not. There was a difference of opinion between Bankes LJ and Atkin LJ as to the precise reason for the finding. Bankes LJ concluded that where a statute is designed for the protection of the ‘public as a whole’ it would not be civilly actionable. Atkin LJ differed on that point, noting the oddness of a rule that would allow enforcement of less important statutes but preclude those designed to protect a larger group of people.46 But he agreed in the end that Parliament did not intend to allow a civil action in the case of this sort of provision (a general ‘roadworthiness’ requirement). Whatever the precise rationale adopted, Phillips has been taken as standing for the proposition that, in general, highway and traffic regulations will not be civilly actionable in the Commonwealth.47 It has also been consistently cited for the proposition that only duties owed to a ‘limited class’ of the public will be recognised as civilly actionable.48 But the acceptance in the United States of the reasoning in Martin seems to have ‘opened the floodgates’ in the US to NPS cases based on traffic laws. To what extent, then, was the NPS doctrine influenced by the previous history of the BSD action in the UK? It becomes apparent in the early cases that a number of the features of the NPS analysis are directly related to the early UK BSD jurisprudence. It may be helpful to demonstrate some of the general similarities first. The following table arranges the issues usually thought to be relevant to Commonwealth BSD claims and notes some of the main NPS cases that raise the same or similar issues.

45

Phillips v Britannia Hygienic Laundry Co Ltd [1923] 2 KB 832 (CA). Interestingly, two members of the High Court of Australia, Evatt and McTiernan JJ, agreed with these comments in one of the major Australian cases affirming the availability of BSD actions in workplace safety cases: O’Connor v S P Bray Ltd [1937] HCA 18, 56 CLR 464. Still, the main judgment in that case is usually regarded as that of Dixon J, who did not cast any doubt on the ‘special class’ rule. 47 See, eg, in Australia Hopewell v Baranyay [1962] VR 311; Tucker v McCann [1948] VLR 222; Abela v Giew (1965) 65 SR (NSW) 485; Bowling v Weinert [1978] 2 NSWLR 282 (applying the previous case to hold that no BSD action arose for breach of water traffic regulations governing boats); in New Zealand Gardiner v McManus [1971] NZLR 475 (noted in BH Giles, ‘Town Planning and the Tort of Breach of Statutory Duty’ (1972) 2 Auckland University Law Review 39, 42). 48 See, eg, the comments of the UK Supreme Court in Morrison Sports Ltd v Scottish Power Plc [2010] UKSC 37, [2010] 1 WLR 1934, [40]: ‘one of the necessary preconditions of the existence of a private law cause of action is that the statutory duty in question was imposed for the protection of a limited class of the public’. The dicta to the contrary of Atkin LJ in Phillips were specifically disapproved. 46

‘A judge must first examine the statute to determine if it is the sort of legislative pronouncement appropriate to set the standard of care in a negligence case.b One factor: whether there is an ‘explicit’ or ‘specific’ rule rather than a general principle. ‘Whether the class of persons designed to be protected by the statute includes the plaintiff’.d See specifically Dobbs §142 ‘Statutes Creating Duties Only to the Public’ for some decisions refusing liability on this ground. ‘Whether the statute was designed to protect against the type of harm suffered by the plaintiff’.f This seems to be a clearly obvious requirement. ‘All the jury needs to find is that the statute was violated’.g ‘When courts conclude that the actor’s conduct is outside the scope of the statutory command, the actor is not in violation and cannot be guilty of negligence per se’.h In the most recent cases US courts are willing to ‘excuse’ statutory violation where fault is not involved.i Still, there are some examples of strict liability.j

(a) The intention of Parliament to allow an action.a

(b) The plaintiff must fall within the ‘limited class’ of the public for whose benefit the statutory provision was enacted. Hence a statute that imposes duties for the benefit of the ‘public at large’ may not be actionable.c

(c) The damage suffered must also fall within the intended scope of the statute.e

(d) The obligation under the statute was imposed on the defendant.

(e) The defendant must have breached the statute.

Hence there can be ‘strict’ liability, as no ‘fault’ needs to be shown.

US Negligence Per Se cases

Commonwealth BSD

Table 7.1 Elements of Commonwealth BSD and US NPS Compared

182 Neil Foster

‘She must still establish the other elements of a negligence cause of action: cause-in-fact, proximate cause and damages’.k

eg, Atkinson v Newcastle and Gateshead Waterworks Co (1877) LR 2 Ex D 441 (whether an action is available for breach of statute ‘must, to a great extent, depend on the purview of the legislature in the particular statute, and the language which they have there employed’). b JL Diamond, LC Levine and A Bernstein, Understanding Torts, 4th edn (New Providence, LexisNexis, 2010) 86. c See Phillips v Britannia Hygienic Laundry Co [1923] 2 KB 832 (traffic violations). d Ibid, at 87–88. See also Tentative Draft (No 5), above n 38, at § 14; Dobbs, above n 31, at §§ 137–39. e Gorris v Scott (1874) LR 9 Ex 125. f Diamond et al, above fn b, at 87, citing (as many cases in this area do), Gorris, above fn e. See also Tentative Draft (No 5), above n 38, at § 14 (‘a statute that is designed to protect against the type of accident the actor’s conduct causes’); Dobbs, above n 31, at §§ 137–39. g Diamond et al, above fn b, at 89. h Dobbs, above n 31, at § 140. i Diamond et al, above fn b, at 90. See also Tentative Draft No 5, above n 38, at § 15 (‘Excused violations’). j See Dobbs, above n 31, at § 141, citing Spalding v Waxler, 205 NE 2d 890 (Ohio 1965) (brakes failed, no evidence of carelessness, defendant in breach and liable). k Diamond et al, above fn b, at 89.

a See,

(f) That breach must have caused actual damage of some sort to the plaintiff.

Statutes and Civil Liability in the Commonwealth and US 183

184 Neil Foster Because of the similarity between elements of the BSD tort and the NPS approach, and their shared historical roots, it is not surprising that similar issues have been raised by the courts in the different systems. Some further examples may illustrate the point. An interesting example of the ‘limited class of persons’ rule can be seen in a case cited by Dobbs, DiCaprio v New York Central Railroad.49 There a statute requiring a railroad to be fenced was held to have been designed to stop cattle wandering onto the tracks, and not to protect a small child who wandered onto the track.50 There is a very good example of the ‘type of harm’ rule in another case discussed by Dobbs, Stafford v Borden.51 A statute prohibited supply of petrol in containers smaller than a certain size. Allen, an arsonist, purchased fuel in a smaller container, and then used it to light a fire. The victims of the arson sued the petrol station for breaching the statute, arguing that the only way that the arson had been possible was that the fuel was transported in a small container. The case is interesting because it seems clear that the statute was designed to prevent harm by fire, which of course was the type of harm suffered. But it seems fair to say that the statute had intended to guard against accidental harm by fire, not the sort of harm that actually ensued. As the court said: Unquestionably, the protection of human life and property is inherent in this purpose; however, the specific dangers these rules address are those attending the possible leakage of flammable liquids from their containers, or the potential for disaster if persons inadvertently come in contact with such liquids unaware of their nature. There is nothing in the language of the statutes or the rules promulgated by the fire marshal that indicates that the prevention of arson was the intent behind their enactment.52

Dobbs points out that exclusionary rules of this sort may be described as dealing with the ‘type of risk’ (describing how the harm comes about) or the ‘type of harm’ (classifying what interests the law will protect, eg, bodily integrity, personal property, economic loss). The cases reveal that different analyses can be used in different cases. As he notes, Gorris v Scott (discussed below) could be analysed as a ‘risk’ case (the statute was not aimed at injury caused by water washing over the boat) or as a ‘harm’ case (aimed at preventing disease of the animals, not their drowning).53

49 DiCaprio v New York Central Railroad 231 NY 94 (1921) [DiCaprio]; Dobbs, above n 31, at § 138. 50 To be fair, the statute did explicitly say that ‘a railroad corporation shall erect and maintain fences on both sides of its right of way sufficiently high and strong to prevent horses, cattle, sheep, and hogs from going upon its road from the adjacent lands’, ibid, at 96. 51 Stafford v Borden 252 Ill App 3d 254 (1993) [Stafford]; Dobbs, above n 31, at § 137. 52 Stafford, above n 51, at 259 (citations omitted). 53 Gorris v Scott (1874) LR 9 Ex 125 [Gorris].

Statutes and Civil Liability in the Commonwealth and US 185 The decision of the US Supreme Court in Kernan v American Dredging Co54 illustrates the difficulty in this process of analysis and classification. Kernan was a passenger on a barge navigating the Schuylkill River in Philadelphia and was injured by an explosion. The cause of the explosion was an open lamp, which was located only three feet above the water level, in a situation where there was an ‘extensive accumulation of petroleum products’ on the river. Federal regulations required that lights on such barges be elevated at a height of eight feet. The lower courts found that if the light had been at eight feet, the explosion would not have occurred. Despite the fact that it seemed fairly clear that the ‘eight foot’ requirement was enacted for the purposes of better navigation, not to avoid explosions, the majority of the Supreme Court held that the worker had a valid civil action based on breach of the statute. However, Dobbs gives a misleading impression of the decision. Dobbs discusses the facts of Kernan and gives the impression that the court found that it fell within an NPS analysis by a strained interpretation. In fact Brennan J, writing for the majority, based his decision explicitly not on the traditional NPS reasons, but on the particular statute involved, the ‘Jones Act’,55 which itself adopted the provisions of FELA and applied them to sailors. That legislation gave a civil remedy to injured workers (and their families in the case of wrongful death, as here); previous decisions held that in relation to two specific federal safety Acts, ‘a defect resulting from a violation of either statute which causes the injury or death of an employee created liability without regard to negligence’.56 The issue in the decision was simply whether the navigation regulations could be added to the list of statutes that created this form of ‘strict’ liability, and according to Brennan J the interpretation of the relevant clause in FELA meant that it could. Hence it was not a common law issue of NPS. Brennan J commented: The courts, in developing the FELA with a view to adjusting equitably between the worker and his corporate employer the risks inherent in the railroad industry, have plainly rejected many of the refined distinctions necessary in common law tort doctrine for the purpose of allocating risks between persons who are more nearly on an equal footing as to financial capacity and ability to avoid the hazards involved. Among the refinements developed by the common law for the purpose of limiting the risk of liability arising from wrongful conduct is the rule that violation of a statutory duty creates liability only when the statute was intended to protect those in the position of the plaintiff from the type of injury in fact incurred. This limiting approach has long been discarded from the FELA.57

54 55 56 57

Kernan v American Dredging Co 355 US 424 (1958) [Kernan]. 46 USCA § 688. Kernan, above n 54, at 431. Ibid, at 438 (emphasis added).

186 Neil Foster It is true that the dissenters in the case objected on the basis that they would have applied the decision in Gorris and ruled out the navigation statute as applicable. But no member of the Supreme Court thought that he was simply deciding an issue of common law NPS. The similarity of issues raised in the US NPS and UK BSD cases is no accident. It seems clear that the US courts were conscious of the developing BSD tort in the UK when developing the NPS doctrine. Some of the material about to be noted is also relevant to the later issue of ‘implied rights of action’—but both streams of US law seem to find their fons et origo in the common law BSD action. We find the following passage in what is usually regarded as the key early US Supreme Court decision on implied rights of action, Texas & Pacific Railway Co v Rigsby: A disregard of the command of the statute is a wrongful act, and where it results in damage to one of the class for whose especial benefit the statute was enacted, the right to recover the damages from the party in default is implied, according to a doctrine of the common law expressed in 1 Comyn’s Dig. title, ‘Action upon Statute’ (f), in these words: ‘So, in every case, where a statute enacts or prohibits a thing for the benefit of a person, he shall have a remedy upon the same statute for the thing enacted for his advantage, or for the recompense of a wrong done to him contrary to the said law.’ (Per Holt, Ch. J., Anonymous, 6 Mod. 26, 27.) This is but an application of the maxim, Ubi jus ibi remedium.58

Further, Bellia notes that: ‘The Supreme Court of the United states has cited Couch [v Steel]59 at various times for the proposition that courts have broad authority to order remedies for statutory violations’.60 Gorris,61 which seems to have struck law teachers in the USA as well as elsewhere around the Commonwealth as a stark example of the ‘type of harm’ rules, is regularly cited. (In that decision it was held that a statute requiring sheep being carried by ship to be penned was designed to further veterinary health issues, not to protect the wider physical well-being of the sheep. When a number of sheep were washed overboard in a heavy storm,

58

Rigsby, above n 44, at 39–40. Couch v Steel 3 E & B 402, 23 LJ (QB) 121. 60 AJ Bellia Jr, ‘Article III and the Cause of Action’ (2004) 89 Iowa Law Review 777, 844. However, Bellia seems to be wrong when on the next page he concludes that ‘Couch was thus not a case about implying rights of action from a statute’. While it is true that it involved personal injury, Lord Campbell, in a passage that Bellia himself quotes on that page, said: ‘the simple enactment requiring the supply of medicines would have entitled the plaintiff to an action, in the same manner as if the obligation had been imposed by the common law’ (emphasis added). The point of the case was that the common law at the time (following Priestley v Fowler (1837) 3 M & W 1, 150 ER 1030) did not give the plaintiff an action, and so it had to be based on the statute. See the discussion of Couch in N Foster, ‘Breach of Statutory Duty and Risk Management in Occupational Health and Safety Law: New Wine in Old Wineskins?’ (2006) 14 Tort Law Review 79, 80–81. 61 Gorris (n 53). 59

Statutes and Civil Liability in the Commonwealth and US 187 the fact that there were no pens as required by the statute was held not to ground a BSD action.) In the US Supreme Court decision in Kernan we read about: the familiar principle in the common law of negligence that injuries resulting from violations of a statutory duty do not give rise to liability unless of the kind the statute was designed to prevent. Indeed that principle, which is but an aspect of the general rule of negligence law that injuries in order to be actionable must be within the risk of harm which a defendant’s conduct has created … was established as long ago as 1874 by a leading English case, Gorris v. Scott, and has been followed in this country almost without exception.62

More recently the Seventh Circuit has continued to cite the case with approval, noting that ‘the old tort cases are often the most illuminating’.63 The case was called a ‘hardy perennial’ in another decision.64 A Westlaw search reveals that it is cited on more than 40 occasions in US decisions ranging from 1895 to 2011. Yet, as noted previously, despite the influence of the BSD cases, following Martin state tort decisions continued to ‘insert’ statutory duty into the negligence analysis rather than to deal with it under the BSD rules. Clearly one factor may have been the immense regard in which Cardozo J was held by US courts and academics on questions of tort law.65 Another may have been the fact that approaching the matter in this way allowed the courts to respond to a wider group of claims from plaintiffs. In other words, opening up the use of traffic laws in negligence claims bluntly meant that more plaintiffs were likely to succeed. To some extent it seems that NPS liability is also subject to the same critiques of inconsistency and judicial ‘policy making’ that have been directed at the BSD action. Dobbs notes, for example, that for many years courts rejected any NPS liability based on breaches of laws directed at those selling alcohol—the courts ‘simply rejected the statutory standard as inappropriate’.66 But later the courts began finding such liability ‘[a]s public and judicial attitudes changed’.67 (In the Commonwealth, an Australian

62

Kernan, above n 54, at 442 (citations omitted). Aguirre v Turner Const Co 582 F 3d 808, 815 (7th Cir 2009). 64 Gauger v Hendle 349 F 3d 354, 363 (7th Cir 2003), overruled by Wallace v City of Chicago 440 F 3d 421 (7th Cir 2006) (though not with any relevance to the authority of Gorris). 65 Cardozo J, of course, was the principal architect of the decision in MacPherson v Buick Motor Co, above n 3, only four years previously, which even at the time was regarded as a landmark in the law of negligence: see JCP Goldberg and BC Zipursky, Torts (Oxford Introductions to US Law) (Oxford, Oxford University Press, 2010) 80: ‘MacPherson … established Benjamin Cardozo’s reputation as a leading jurist’. So perhaps it is not surprising that his approach to this question of the relationship between statute and civil liability became quickly established as the prevailing view in US tort law. 66 Dobbs, above n 31, at § 135. 67 Ibid, citing Rong Yao Zhou v Jennifer Mall Restaurant Inc 534 A 2d 1268 (DC 1987). 63

188 Neil Foster BSD decision rejected the liability of a hotel by holding that the relevant statute was passed for the benefit of the ‘public at large’.68 This involved an application of a specific, if slightly illogical, rule in the BSD analysis, rather than broad public policy considerations.) Despite the similarities, a fundamental difference between the BSD cases and the current NPS doctrine is worth noting. The ‘orthodox’ BSD analysis in the Commonwealth retains a strong emphasis on the fact that the finding of civil liability in a statute is very much a question of legislative intention. Of course this does not require an explicit statement by Parliament, but the way that the issue is to be resolved is through the tools of statutory interpretation. Later courts have regularly cited the comments of Kitto J in the High Court of Australia decision of Sovar v Henry Lane Pty Ltd: [T]he question whether a contravention of a statutory requirement of the kind in question here is actionable at the suit of a person injured thereby is one of statutory interpretation. The intention that such a private right shall exist is not, as some observations made in the Supreme Court in this case may be thought to suggest, conjured up by judges to give effect to their own ideas of policy and then ‘imputed’ to the legislature. The legitimate endeavour of the courts is to determine what inference really arises, on a balance of considerations, from the nature, scope and terms of the statute, including the nature of the evil against which it is directed, the nature of the conduct prescribed, the pre-existing state of the law, and, generally, the whole range of circumstances relevant upon a question of statutory interpretation.69

By contrast, in dealing with NPS from § 314, Dobbs treats the issue under US law as one where the courts are ‘free to adopt’ a statutory standard as part of the law of negligence. The issue is framed as one for the court, not for the legislature. From a Commonwealth perspective, this is odd. Of course, in the end it is a court that has to answer the question of interpretation, but most common law courts would reject Dobbs’ characterisation as a way of approaching statutory civil liability. The US cases, while paying ‘lip service’ to legislative intention, regularly continue to leave a wide discretion to courts. Yowell discusses some Texas decisions on NPS and raises the question whether a court is bound to apply a statute once it finds that it is intended to protect a class including the plaintiff from harm of the type that occurred.70 His view is that the courts ought to retain a residual discretion to decline to apply a statutory standard in negligence even if these criteria are satisfied. 68 See Re Laszlo Joseph Chordas v Bryant (Wellington) Pty Limited [1988] FCA 462, [32], a decision of the Federal Court of Australia, on appeal from the ACT Supreme Court: ‘Section 79 was enacted in the general public interest, not for the protection of persons who may be injured by the conduct of intoxicated persons’. 69 Sovar v Henry Lane Pty Ltd (1967) 116 CLR 397, 405 (HCA). 70 P Yowell, ‘Judicial Discretion in Adopting Legislative Standards: Texas’s Solution to the Problem of Negligence Per Se?’ (1997) 49 Baylor Law Review 109.

Statutes and Civil Liability in the Commonwealth and US 189 Arguably he is right, but not for the reasons he puts forward for addressing this question: adapted from the Restatement (Second), he suggests the questions should be about obsolescence, unreasonableness or obscurity. In my view this is a key area where the Commonwealth BSD analysis would be more helpful. The preliminary question there is always, ‘was it the intention of the legislator to allow a civil action’? That, with respect, is the better way of addressing the issues. If a law is clearly penalising trivial behaviour and with a very minor penalty, it will be unlikely that one could say that the lawmaker intended an unlimited civil action. The absence of the ‘legislative intention’ inquiry in NPS cases can be seen in the original decision in Martin. No part of Cardozo J’s judgment seems directed to whether Parliament intended to imply a civil remedy for breach of the legislation. The questions of parliamentary intention as to whom to protect are impliedly discussed, when it is noted that the legislation in question was prescribed ‘for the benefit of another that he may be preserved in life or limb’, and was intended to protect ‘travelers on the highway, of whom the defendant at the time was one’.71 But the question of the legislature’s intention as to the creation of civil liability is not discussed. Indeed, at one telling point Cardozo J simply said that the jury should not be able to relax ‘the duty that one traveler on the highway owes under the statute to another’.72 Here is an implied finding that the statutory provision itself has created a bilateral relationship between travellers on the road, each owing a duty ‘to another’ not to breach the law.73 This failure to raise the question of legislative intention about creation of private duties, then, becomes a theme that allows the US state expansion of the law of NPS. But of course, while the door is opened to a wider group of claimants, once the size of the potential group becomes clear, the courts are then left to wrestle with ways of ‘controlling’ the flood. In the US jurisprudence it seems that this is done by the introduction of a series of ‘excuses’ (authorised by the telling word in the judgment in Martin, ‘unexcused’) that can be used when it becomes apparent that the automatic assumption of civil liability based on statutes can no longer be sustained.

71

Martin, above n 41, at 815. Ibid. 73 Ironically, in yet another famous decision handed down by Cardozo J, four years after Martin, Palsgraf v Long Island Railroad 162 NE 99 (NY 1928), he was very conscious of the question whether the law created not just a general ‘obligation’, but a duty to the specific plaintiff. As Goldberg and Zipursky, above n 65, at 102, note, the key concept was that the victim in an action for negligence ‘sues for breach of a duty owed to himself’ and not anyone else. Nevertheless, the implication of the comments in Martin are that when a statute aims to protect a certain class against particular harm, members of that class are owed individual duties by others who are addressed by the statute. 72

190 Neil Foster ‘Excuses’ that are discussed in the Restatement (Third) include: — — — — —

whether a violation was ‘reasonable’ in light of personal characteristics of the actor (such as age or capacity); whether ‘reasonable care’ was taken to attempt to comply with the statute; subjective knowledge of the facts rendering the statute applicable; manner of presentation of statutory obligations; or whether compliance would have involved greater risk than noncompliance.74

Blomquist gives an interesting overview of the historical development of the doctrine, and a strong critique of how it operates at the moment.75 He points out that the early cases involving the phrase ‘negligence per se’ (the earliest he cites is Simpson v Hand76) seem to use it not in reference to statutory rules, but as a way of emphasising how clearly negligent some behaviour was. In that context it seems to have functioned something like the Latin tag res ipsa loquitur: the carelessness ‘speaks for itself’. But gradually the term came to be applied more specifically to carelessness involving breach of statutes designed for safety. Blomquist sees the decision of Cardozo J in Martin as ‘crystallizing’ this sort of approach. He criticises the decision itself, as inconsistent with Cardozo’s own extrajudicial writings, and as internally incoherent. As he points out, in Martin both the buggy driver (in failing to have lights) and the car driver (in crossing the middle line) were in breach of statute. Why then was the buggy driver’s breach focused on as the relevant one?77 Nevertheless, courts later regularly referred to Martin, and Blomquist reports that from 1921 to 2000 there were over 10,000 opinions mentioning ‘negligence per se’. He reviews a number of these in his article, and then offers a critique of the jurisprudence through application of a theoretical model offered by Summers.78 Whether or not this particular model is persuasive, Blomquist correctly brings out a number of defects in NPS doctrine and practice. He points out that courts wrestle with the need to interpret legislation as creating a civil right when none is spelled out. In a significant critique, he notes that a wide-ranging definition of ‘broad excuses’ for statutory violation means

74

See Tentative Draft No 5, above n 38, at § 15. R Blomquist, ‘The Trouble with Negligence Per Se’ (2009) 61 South Carolina Law Review 221. 76 Simpson v Hand 6 Whart 311, 323 (Pa 1841). 77 Blomquist, above n 75 at 250–51. 78 RS Summers, Form and Function in a Legal System: A General Study (Cambridge, Cambridge University Press, 2006), noted in Blomquist, above n 75, at 272, and applied in section IV of Blomquist’s article. 75

Statutes and Civil Liability in the Commonwealth and US 191 that, far from simplifying a standard negligence analysis, cases may be prolonged: [T]he very conceptual foundation of the rationale for the negligence per se doctrine—‘the doctrine’s ability to provide greater certainty than the usual reasonable person standard’—is undermined by the broad and far ranging excuses that may be considered as reasons for why a tort litigant violated a particular nonprescriptive standard.79

Finally, he considers the doctrine arguably an inappropriate usurpation of legislative power, as the court is reading something in which the legislative body failed to deal with. In my view some of Blomquist’s analysis is not convincing. If the role of the court is actually interpretation of legislative intent, as it should be, rather than policy-based creation of liability, the separation of powers argument is not strong.80 For this reason I would also not support his final proposal, when he concludes, after arguing for limitation of cases where the doctrine should be applied, that it should simply be seen as based on ‘policy analysis’ by the judge. However, I think his point about the undermining effect of the ‘excuses’ that are accepted is correct, and in my view US NPS jurisprudence would be clearer and conceptually more satisfying if the only ‘excuses’ that were accepted were those which can be found from the statute itself, or those which are generally applicable to all torts. The area of ‘excused violations’ is one of those, then, which reveals fundamental problems with the NPS doctrine, and why a more straightforward BSD doctrine would be better. Leonard is another author who has pointed out the conceptual difficulties with allowing a wide range of excuses in NPS claims.81 He comments that of course there is no problem allowing excuses that would be allowed under a criminal prosecution.82 But logically the court ought not to allow other excuses, as that would undermine the statutory purpose that is meant to be implemented by the NPS doctrine. But Leonard argues against this result, as this would turn what is meant to be ‘negligence’ litigation into ‘strict liability’.83 To the Commonwealth lawyer, given that a BSD claim is a separate tort, this is not a problem— strict liability has long been seen to be a key characteristic of much of the BSD litigation, and indeed one of its main advantages over negligence actions. However, it seems clear that this ‘dislike’ of strict liability by US courts may be part of what drives the artificial introduction of excuses.

79

Above n 75, at 279. Presumably the words ‘tort litigant’ here mean primarily ‘defendant’. This point is developed in more detail in section B below. 81 DP Leonard, ‘The Application of Criminal Legislation to Negligence Cases: A Reexamination’ (1983) 23 Santa Clara Law Review 427. 82 Ibid, at 467. 83 Ibid, at 469. 80

192 Neil Foster Yet the NPS litigation reveals many instances in which ‘excuses’ have been accepted. Leonard argues this is necessary. He uses the example of a defendant who has breached a statute forbidding driving with a prescribed concentration of alcohol in the blood. Suppose a defendant in that condition injures the plaintiff. But suppose the defendant can show that he was only driving because, unexpectedly, his child became ill and he needed to seek medical assistance. In that case a court might excuse the driving. Leonard says that if the court would not immunise the defendant against civil liability here, ‘something seems wrong with that result’.84 But one may beg to differ. Suppose it to be true that in a criminal prosecution a court might find the defendant not guilty (perhaps with a defence of necessity?), or else impose a nominal fine. Suppose it also to be true that in a civil action the conclusion that there was a statutory breach would have to lead to a large award of damages.85 It is not self-evident that this is wrong. There is a difference in kind between a criminal conviction and a civil action. One might want to allow a criminal court not to impose a conviction (with all the attendant reputational consequences as well as a possible prison term), where it is simply a question of how to punish the defendant alone. But in a case where the defendant, while breaking a law put in place by the community to avoid just this kind of incident, has caused immense personal injury to an otherwise blameless plaintiff, why should the plaintiff’s interests be put below those of the defendant? Does it not seem equally wrong to leave the plaintiff without civil remedy?86 Leonard is perfectly correct, however, to criticise the inconsistency of allowing these excuses if a consistent NPS doctrine is to be applied: ‘If a criminal statute imposes responsibility without fault, courts deciding negligence cases based on the statute are departing from the basis of the theory when they permit excuses which would not be available in criminal prosecutions.’87 It is suggested, however, that this critique demonstrates the problems with the NPS theory, and points back to a better view being the traditional BSD analysis. The many examples of cases where a court has had to introduce an extraneous excuse, or strain to read a statute oddly, point, for example, to the wisdom of the Commonwealth view that in general traffic laws are not suitable for the automatic creation of civil liability.88 84

Ibid, at 471. In fact, of course, in the Commonwealth, driving offences of this sort will almost invariably be held not to give rise to actionable civil claims. But we will assume for the present purposes that the BSD analysis would in some cases allow reference to driving statutes of this sort. 86 And, of course, though this should not be relevant to the liability issue, almost always there will be an insurance policy to which the driver will have access. 87 Leonard, above n 81, at 472. 88 Whether this should be justified on the conceptually unsatisfactory grounds that such legislation is enacted for the ‘public benefit’ and not to protect a limited class of the public is another question. It would be possible to maintain the exclusion of such laws as a matter of the intention of the legislature in general, taking into account the sort of issues that Leonard raises 85

Statutes and Civil Liability in the Commonwealth and US 193 In short, Leonard’s critique of the majority US version of NPS, where in theory the case remains a negligence one but the ‘breach’ is irrefutably established by the statutory breach, is persuasive, and is another reason to suggest that the classic BSD analysis is preferable. The more fundamental problem revealed by the ‘excuses’ jurisprudence is that which has been noted before: that once the relevance of the statute to civil liability has been automatically assumed, without paying attention to the likely intention of the lawmaker, then it becomes necessary to artificially constrain the application of the statute by a series of ad hoc devices. It seems that the better approach would be to introduce a specific stage in the NPS analysis addressing the question of legislative intention as to relevance of the statute to civil liability. Alternatively, it is suggested that the Commonwealth BSD analysis provides a better overall model, rather than attempting to ‘shoe-horn’ a statutory obligation into the common law action for negligence. It will be recalled that the very first hurdle that needs to be crossed in establishing a viable BSD claim is that of legislative intention. A series of decisions may clarify the general class of legislation which will, or will not, give rise to such liability.89 Of course this will leave some claims as uncertain at the outset—but it is suggested that there is less uncertainty where the proper question is being asked, rather than the trackless wilderness of ‘excuses’ which justify behaviour seen to be contrary to the legislation.90 Another factor tending to show that the prevailing NPS theory is unsatisfactory is the way that the cases deal with the fundamental issue of ‘duty of care’. If the common NPS view is adopted, then in theory all that the statute does is to determine the issue of breach, and the question of whether the defendant owed the plaintiff a duty of care ought to be decided on the other common law principles governing this question. But in most of the NPS cases the duty issue seems to be completely elided. Indeed, if the Restatement (Third)’s § 14 view is adopted, while in theory the question to be resolved is whether ‘an actor is negligent’, the questions used to resolve the question include whether the statute was ‘designed to protect against the type of accident the actor’s conduct causes’, and whether the ‘victim is within the class of persons the statute is designed to protect’. If a

in his critique, referring to ‘the frequency with which such laws are violated, or the seeming unfairness of saddling defendants with huge liability for violating admittedly safety-oriented, but often minor, provisions’, ibid, at 481. 89 Such as the line of cases in the Commonwealth, previously noted, declining to make traffic regulations civilly actionable. 90 See also H Miles Foy III, ‘Some Reflections on Legislation, Adjudication, and Implied Private Actions in the State and Federal Courts’ (1986) 71 Cornell Law Review 501, which undertakes an excellent review of the history of the action for negligence per se (as well as the area of implied federal rights of action) and makes a very persuasive case for a return to something very similar to the Commonwealth BSD approach.

194 Neil Foster court finds in favour of a plaintiff on these points, can it be doubted that they would find a duty of care owed to the plaintiff by the defendant? It seems that the most common understanding of ‘duty’ is simply to base it on foreseeability, in which case, if there is a statute providing protection for a class of victims, it would be impossible to claim that such harm was unforeseeable. Note that Dobbs, in his analysis of the law of negligence, deals with the issue of duty of care in a brief paragraph before turning at length to breach: It sounds surprising, but there are cases in which the defendant literally owes the plaintiff no duty that bears on the harm suffered. In the ordinary case, however, the defendant does owe a duty of care. The only question about the duty in such cases is whether the care owed is some especially high care or whether it is more modest.91

The Restatement (Third) looks at first glance as though it contains a substantial discussion of ‘Duty’ in § 7, but in effect it is very minimal: (a) (b)

An actor ordinarily has a duty to exercise reasonable care when the actor’s conduct creates a risk of physical harm. In exceptional cases, when an articulated countervailing principle or policy warrants denying or limiting liability in a particular class of cases, a court may decide that the defendant has no duty or that the ordinary duty of reasonable care requires modification.92

While in many ways this resembles the Commonwealth approach, it provides a much stronger presumption and ends up ignoring the duty question in all but the most unusual cases. As the Comment to § 6 notes: Ordinarily, an actor whose conduct creates risks of physical harm to others has a duty to exercise reasonable care. Except in unusual categories of cases in which courts have developed no-duty rules, an actor’s duty to exercise reasonable care does not require attention from the court.

Some discussion of this duty question is offered in an old, but still valuable, article by Linden.93 In particular, he discusses a series of cases where statutes requiring a driver to stop and provide assistance to a pedestrian hit by his or her vehicle were breached, and civil liability was found. The main

91 Dobbs, above n 31, at § 115. Similarly, in a more modest, student-oriented text, JL Diamond, LC Levine and A Bernstein, Understanding Torts, 4th edn (New Providence, LexisNexis, 2010), the discussion is opened in ch 3 by defining ‘duty’ as ‘a legally recognized relationship between the parties’ but almost immediately moves on to discussion of the ‘standard of care’ and other breach issues. 92 Tentative Draft No 5, above n 38, at § 7. 93 AM Linden, ‘Tort Liability for Criminal Nonfeasance’ (1966) 44 Canadian Bar Review 25, which discusses the question under the heading ‘Breach of Legislation which Creates New Duties Where None Existed at Common Law’ from 41 ff.

Statutes and Civil Liability in the Commonwealth and US 195 case he cites is Brooks v E J Willig Transport Co.94 The court held that the statute would be breached even if the driver concerned were not careless in any way in causing the collision, and ruled that this would create civil liability. In effect, as Linden notes, this is not simply crystallising a standard of care, but creating a new duty of care to those who have been, without one’s own carelessness, hit by one’s vehicle. (Incidentally, the case arguably illustrates again that the Commonwealth BSD analysis is preferable here— rather than forcing this situation into the law of negligence, it would be better to conduct an analysis of the statute under the BSD approach and, if the view were taken that the legislative intention was to give a civil right of suit to those injured by the breach, to say so directly.) Apparent confusion between the duty and breach questions can be seen in the most careful of discussions. In an oft-cited article reviewing the NPS area, Leonard comments that the effect of applying the majority NPS view is that the duty of care which would have governed the case at common law (if indeed the common law would have prescribed any duty at all) is completely supplanted by the standard of care implicit in the criminal statute adopted by the legislature.95

To the Commonwealth eye, this seems at first hopelessly confused. Surely NPS deals with the breach question, not the duty? But when it is noticed that in the second half of the sentence what ‘supplants’ the duty at common law is ‘the standard of care’ prescribed by the statute, it seems that the phrase ‘duty of care’, when first occurring, must be a reference to something like ‘the standard of the reasonable person implied by the common law duty of care’. In other words, the word ‘duty’ is being used as a shorthand way of referring to what a more formal analysis would call ‘breach’. In contrast to some of the other approaches noted above, what seems to be a more principled analysis can be found in Sorensen v State Farm Automobile Insurance Co.96 There the question arose whether an insurance company could sue a driver who had failed to maintain insurance on her vehicle. The company had been unable to recover (by way of subrogation) damages it had to pay out to clients who had been injured by the uninsured driver. Kite CJ for the Supreme Court of Wyoming, in overturning a lower court decision that breach of the statute requiring vehicle owners to take out insurance was civilly actionable, said: The difficulty with the district court’s approach is that it determined the statute established the standard of care without first determining that a duty existed. The concepts of standard of care and duty are not synonymous. Before a statute can be said to establish a standard of care, there must be a legal duty to which the

94 95 96

Brooks v E J Willig Transport Co 40 Cal 2d 669, 255 P 2d 802 (1953). Leonard, above n 81, at 448 (emphasis added). Sorensen v State Farm Automobile Insurance Co 2010 WY 101 (2010).

196 Neil Foster statutory standard of care can be applied. Thus, in deciding the issue presented, we address first whether § 31-4-103 should be construed to establish a common law duty actionable in negligence for failure to maintain liability insurance. Stated more broadly, we consider whether such a relationship exists between members of the public and a vehicle owner that the law will impose a duty actionable in negligence on the latter to maintain insurance for the protection of the former.97

The court ruled that no duty of care existed, and hence that the statutory standard was not relevant.98 A further difficulty that faces US state courts in applying the classic NPS view (that the statutory standard provides conclusive evidence of breach) is the fact that statutes in a federal system emanate from more than one source. In particular, how is a state court in a common law negligence action to treat a statutory standard laid down by the federal Congress? Dobbs notes that state courts are ‘free to adopt’ a standard prescribed by federal law in deciding a state tort case. If the question were one of statutory interpretation, of course, then it would better be framed as, ‘did Congress assume or expect that a state court in a tort case would apply this as a standard for breach?’99 In an important article, Kritchevsky argues at length that the practice of state courts adopting standards set by federal law for NPS actions is wrong.100 The first premise is, as noted previously, that there is no US federal common law of tort. This flows from the decision noted above in Erie Railroad v Tompskins,101 that the federal Congress has no specific power over tort law in the states. Kritchevsky then argues that the generally accepted rationale for NPS actions today is ‘institutional comity’—that a state court should not ‘second guess’ the state legislature in a negligence action by allowing a jury to conclude that behaviour which is contrary to state law is ‘reasonable’ behaviour. But, she says, this rationale does not exist where the federal legislature had no direct power to pass laws on the question of tort liability in the state; and hence state courts should not apply those federal rules where they have not otherwise been made the law of the state.102 97

Ibid, at 234 (emphasis added, citations omitted). There was also a discussion whether the statutory prohibition itself created a civil cause of action, and this was also rejected. Interestingly the old UK decision of Monk v Warby [1935] 1 KB 75 held that similar legislation requiring insurance did create an actionable duty. 99 A Commonwealth court would probably not feel free to ‘expand the scope’ of a statute beyond persons who were actually covered, as Dobbs suggests a US court would be able to do, above n 31, at § 136, under the heading ‘Expanding a Statute’s Scope’. Conceptually the Commonwealth courts still insist they are following Parliament, not simply choosing a standard that the judge thinks appropriate. 100 Kritchevsky, above n 30. 101 Above n 29. 102 Kritchevsky recognises, of course, that federal law may ‘pre-empt’ state law on an area if valid, or a state may have chosen to adopt federal standards voluntarily—see 128–29, nn 379–83. But outside these circumstances state courts should not adopt federal laws if the state legislature has not done so. 98

Statutes and Civil Liability in the Commonwealth and US 197 Dobbs refers as an example of the application of federal law to Coker v Wal-Mart Stores, Inc, where violation of a federal statute prohibiting sale of guns to a minor was considered NPS.103 There was no real debate in that decision on the application of the statute. However, Congressional intent was taken into account in discussion of what sort of ‘proximity’ of connection was required between the illegal sale and the harm (occasioned by subsequent use of the illegally purchased gun in a robbery): Relying on legislative intent articulated by the United States Supreme Court in Huddleston v. United States, the third district concluded that the ‘risk of harm’ Congress meant to prevent by enacting 18 U.S.C. § 922 was just the ‘type’ of conduct which occurred in the case before it. The court noted the Supreme Court’s extensive quotation in Huddleston from the legislative history of the Gun Control Act and, in particular, the fact that ‘Congress determined that the ease with which firearms could be obtained contributed significantly to the prevalence of lawlessness and violent crime in the United States.’ As the third district observed, the Huddleston ruling underscores the principal purpose of the legislation: to prevent those deemed too dangerous or irresponsible due to age, criminal background, or incompetency from obtaining firearms and ammunition. To accomplish that purpose, Congress chose to control the initial dissemination of firearms and ammunition, and not simply to prohibit the subsequent possession of them. The argument that legality of possession is pivotal to the issue of proximate causation under these circumstances entirely defeats the congressional purpose of the Act and renders its provisions a nullity. For the same reason, the proposition that criminal acts resulting from the unlawful sale of firearms or ammunition should be treated differently under a proximate causation analysis from negligent acts is equally unavailing.104

Interestingly, in a similar Commonwealth decision, the Queensland Court of Appeal held in Pask v Owen that a provision making it illegal to supply a firearm to a minor created civil liability.105 The plaintiff, a 13-year-old boy, had accidentally shot himself after being given a gun by the defendant’s son (aged 15), who had been given it by the defendant. The Court of Appeal all agreed that the legislation prohibiting supply of a firearm to a minor created civil liability. As Thomas J put it: it is a provision which sets out to prescribe certain precautions for the safety of others, and … no contrary intention appears. I therefore conclude (as did the learned trial judge) that the legislature has prescribed certain duties which add to the general duties imposed by the law.106

103 104 105 106

Coker v Wal-Mart Stores, Inc, 642 So 2d 774, 775 (Fla Dist Ct App 1994). Ibid, at 777–78. Pask v Owen [1987] 2 Qd R 421. Ibid, at 434.

198 Neil Foster The legislation in question was state, not federal, law, but the facts of the cases are obviously similar. In the Queensland case, of course, the statute was given direct effect as a BSD, rather than as part of the law of negligence. (ii) Statutory Breach as Evidence of Negligence of Presumptive Breach To return to the US situation, Dobbs notes that there is some variation among the states, some treating the statutory standard as simply creating a ‘rebuttable presumption’ of breach in a negligence action, or as evidence of the matter.107 He argues that the difference in outcome between these states and those that apply the NPS analysis is minimal, especially given that NPS includes as part of its definition the word ‘unexcused’. But on general principles it would seem to be more conceptually satisfactory to provide from the outset that, when approached as part of the law of negligence, the statutory breach is only one of a number of factors that may lead to a finding of breach (a position, it should be noted, shared across the common law world, including Canada),108 rather than to impose what often look like fairly arbitrary ‘excuses’. A description of how the doctrine operates in a state where a presumption applies can be found in a recent decision on the law of California: In California, negligence per se is ‘a presumption of negligence [that] arises from the violation of a statute which was enacted to protect a class of persons of which the plaintiff is a member against the type of harm which the plaintiff suffered as a result of the violation of the statute.’ However, ‘it is the tort of negligence, and not the violation of the statute itself, which entitles a plaintiff to recover civil damages.’

To ensure that the presumption of negligence arises when appropriate, ‘the Legislature ... codified this presumption with the adoption of Evidence Code 669’. Accordingly, because NPS is simply a codified evidentiary doctrine, ‘the doctrine of negligence per se does not establish tort liability’. In other words, even when the requirements of Evidence Code section 669 are satisfied, ‘this alone does not entitle a plaintiff to a presumption of negligence in the absence of an underlying negligence action’. Thus, ‘to apply negligence per se is not to state an independent cause of action’ because ‘[t]he doctrine does not provide a private right of action for violation of a statute’.109 107

Dobbs, above n 31, at § 134(1)(b). In Saskatchewan Wheat Pool, above n 4, at [38], Dickson J for the court noted that, even though the tort of BSD would be abolished in Canada, ‘the violation of the statute should be evidence of negligence on the part of the defendant’. For an application of this approach, see Street v Ontario (Racing Commission) 2008 ONCA 10, 88 OR (3d) 563. 109 People of California v Kinder Morgan Energy Partners LP 569 F Supp 2d 1073, 1087 (SD Cal 2008) (citations omitted). 108

Statutes and Civil Liability in the Commonwealth and US 199 B. Implied Rights of Action110 Rather than incorporating statutory obligations into the law of negligence, statutes can create ‘implied rights of action’ simply by virtue of the statutory scheme. The similarity between the principles operating in this area, and those applied in the NPS cases, seems to be disguised to the US lawyer because there is a strong tendency to deny that civil liability created by implication from a federal statute can be called a ‘tort’. Yet on most general definitions of the term adopted in the Commonwealth literature, as a civil wrong, classifying the action in this way seems unexceptionable. Indeed, it is interesting that some of the more recent US writers are moving in this direction.111 The Restatement (Third) recognises that a statute may create an ‘implied cause of action’ in addition to the operation of the NPS doctrine: The court, relying on ordinary principles of legislative interpretation, may in appropriate cases infer from the statute a cause of action for damages against the violator. In cases involving conduct that causes physical or emotional harm, courts have not often exercised the authority referred to in this Comment; no doubt this is because the longstanding recognition of the common-law rule affirmed in this Section reduces the significance of an implied statutory cause of action.112

The ‘common-law rule’ referred to is the NPS doctrine. Sherman notes that, in the early days of the Supreme Court, the decision in Rigsby adopted a rule which seemed to favour implied liability.113 Indeed, as noted previously, the rule sounds very similar to late nineteenthcentury formulations in the BSD area in the UK, from which it is explicitly supported.114 Stabile correctly points out that, while Rigsby is often regarded as the main early decision in favour of an ‘implied right of action’, a number of previous cases (including the seminal Marbury v Madison115 itself) had relied on the general principle that where there is a right, there must be a

110 While there seems, surprisingly, to be no monograph devoted to this topic, there are many discussions in journal articles (a number of which are cited below) and some reviews of the area in texts devoted to legislative interpretation generally. One of the most helpful recent overviews is the material in C Nelson, Statutory Interpretation (New York, Foundation Press, 2011) 653–89, under the heading ‘The Case of federal Causes of Action’. 111 See Goldberg and Zipursky, above n 65, at 30, discussing the scope of their introductory review of tort law: ‘we will also have occasion to consider interests and invasions … that are recognized and protected through modern statutory schemes … These statutorily defined legal wrongs are torts in structure and functions.’ 112 Tentative Draft (No 5), above n 38, at § 14. 113 P Sherman, ‘Use of Federal Statutes in State Negligence Per Se Actions’ (1992) 13 Whittier Law Review 831, 864. 114 See above n 44. 115 Marbury v Madison 5 US 137 (1803).

200 Neil Foster remedy.116 Further, Noyes analyses the Rigsby decision in some detail to determine its impact on more recent decisions concerning implied rights of action.117 He concludes with some justification that it was a ‘common law’ decision, rather than one based on the federal issues, following the UK precedents. In general the story that is commonly told in the literature about the implication of rights of action from federal statutes is that in the nineteenth and early part of the twentieth centuries implication was possible applying the broad Rigsby analysis. But following the New Deal, there was a ‘rush’ of federal legislation, which led to the Supreme Court ‘tightening up’ the criteria for implication. The case that is seen as starting this trend was Cort v Ash.118 There the law in question was a federal law prohibiting companies from making political donations. A shareholder wanted to seek damages against the company directors for a breach of the statute, constituted by press advertisements paid for during the 1972 Presidential election. By the time the case came to the Supreme Court, of course, the election was over, but the questions remained whether injunctive relief could be sought for the future, and whether damages could be awarded for any previous breach. In fact the action for an injunction was doomed because, although no enforcement mechanism had been provided when the Act came into force, by the time the matter reached the Supreme Court an elaborate scheme for enforcement had been put into place by statutory amendment (by creation of a federal Election Commission, and detailed rules governing complaints). Hence Brennan J ruled that, since the court had to decide the matter in accordance with the circumstances prevailing at the time of the decision, there was clearly no case for saying that Congress intended to allow a private civil action.119 Nevertheless, the court went on to discuss the question whether, as enacted in 1972, the statute allowed a private action for damages. Brennan J summarised the considerations that the court would take into account in ‘determining whether a private remedy is implicit in a statute not expressly providing one’ (effectively the common law BSD issue, though as we shall see in the unique US federal context). These factors were said to be:

116 SJ Stabile, ‘The Role of Congressional Intent in Determining the Existence of Implied Rights of Action’ (1995–96) 71 Notre Dame Law Review 861, 864. 117 JE Noyes, ‘Implied Rights of Action and the Use and Misuse of Precedent’ (1987–88) 56 University of Cincinnati Law Review 145. 118 Cort v Ash 422 US 66 (1975) [Cort]. See the general review of the development of the doctrine in JA Marcantel, ‘Abolishing Implied Private Rights of Action Pursuant to Federal Statutes’ (27 January 2013), available at SSRN: http://ssrn.com/abstract=2207656. 119 Ibid, at 76.

Statutes and Civil Liability in the Commonwealth and US 201 1. Whether the plaintiff is a member of a limited class for whose ‘especial benefit’ the statute was enacted. This was very interestingly paraphrased as ‘does the statute create a federal right in favor of the plaintiff?’. 2. Is there any indication of legislative intent either way? 3. Would it be consistent with the purposes of the legislative scheme to imply such a remedy? 4. Is the cause of action one ‘traditionally relegated to state law’, so that it would be inappropriate to infer a cause of action based on federal law?120 To the Commonwealth lawyer’s eye this is a tantalisingly similar set of criteria to those adopted in BSD cases, yet oddly arranged. Item 2, ‘legislative intent’, would seem to logically be the first on the list—yet here it comes in second place to the interests of the plaintiff. Item 3 looks like simply a part of item 2. And item 4 has a particular federal flavour that would be called in Australian terms a ‘state reserved powers’ approach. This was a way of reading the Australian Constitution which was abandoned by the High Court of Australia in the constitutional sphere many years ago121— the question in Australia now not being, ‘is this a matter for the states?’, but ‘can it fairly be seen to fall within the Commonwealth legislative power in section 51?’ If the answer is ‘yes’ to the second question, then the fact that ‘traditionally’ it was a matter of state power should not be a barrier to federal lawmaking. Clearly the US federal context was quite different on this point, at least in 1975. Other elements of this decision are similar to common law BSD jurisprudence. So Brennan J went on to say: ‘provision of a criminal penalty does not necessarily preclude implication of a private cause of action for damages’.122 This is precisely the rule adopted by the common law courts in BSD cases.123 But in the end an examination of this statute led to the view that it was not intended to provide a civil right to shareholders as: —

the statute was aimed at combating corruption in the election process as its primary purpose, not protecting the interests of shareholders; — there was no explicit discussion of private remedies in the ‘legislative history’; — a private remedy after the fact would not assist the purpose of preventing corruption;

120

Ibid, at 78. Amalgamated Society of Engineers v Adelaide Steamship Co Ltd (1920) 28 CLR 129 (HCA). 122 Cort, above n 118, at 79. 123 See Groves v Lord Wimborne [1898] 2 QB 402 (CA), where the statute which was found actionable not only provided a criminal penalty but also gave a discretion to an officer to pay some of the fine to the injured worker. But since the fine was so small the court still implied a separate civil action. See comment on the decision in Foster, above n 60, at 81. 121

202 Neil Foster — in general corporations are governed by state law and in an appropriate case the shareholders may still have a state remedy. The subsequent history following Cort reveals what can probably best be described as a continued narrowing of the grounds for legitimate implication of a civil action. In Cannon v University of Chicago the court found that federal legislation prohibiting discrimination on the basis of sex in admission to federally funded education implied a personal civil right of action to and applicant so excluded.124 Powell J delivered a strong dissent, claiming that applying the Cort principles was a usurpation of legislative power by the court. In Transamerica Mortgage Advisors, Inc (TAMA) v Lewis a differently constituted Supreme Court ruled against civil liability for breach of a federal statute, and applied the Cort criteria.125 But the majority suggested that not all the criteria were always relevant, the most important question being Congressional intent. In dissent in that case, White J (joined by Brennan, Marshall and Stevens JJ) applied the Cort questions and found in favour of a tort action. The authority of Cort itself was not directly questioned, but clearly there could still be a difference of opinion about intent. In Middlesex County Sewerage Authority v National Sea Clammers the court as a whole ruled against a private civil action being available in relation to particular provisions of legislation aimed at preventing pollution.126 But Powell J, now in the majority, stressed that Congressional intent alone was the test; whereas Stevens J, who joined in the result, adhered to the need to consider a range of factors above any explicit statements by Congress. He referred to the ‘traditional common law analysis’ and conducted a general application of the Cort factors. He ruled against actionability on the basis of a lack of intent to benefit a specific section of the public (a part of the classic BSD reasoning). In Merrill Lynch, Pierce, Fenner & Smith Inc v Curran the majority of the court upheld a private action related to securities fraud.127 Even Stevens J now affirmed that the focus must be on legislative intent, but found such an intent in the fact that private actions in relation to the statute had been upheld by federal courts for many years, and that Congress must be taken to have been aware of this. Powell J dissented on the basis, effectively, that Congress must be taken only to have said what it actually said. Within the court there was a division growing between those who took the view that Congress, except in very rare cases, can only have expressed an ‘intent’

124 125 126 127

Cannon v University of Chicago 441 US 677 (1979). Transamerica Mortgage Advisors, Inc (TAMA) v Lewis 444 US 11, 100 S Ct 242 (1979). Middlesex County Sewerage Authority v National Sea Clammers 453 US 1 (1981). Merrill Lynch, Pierce, Fenner & Smith Inc v Curran 456 US 353 (1982).

Statutes and Civil Liability in the Commonwealth and US 203 about an action in plain words, and those who were prepared to address the issue of Congressional intent more broadly. In Central Bank of Denver v First Interstate Bank of Denver,128 the majority upheld the continued availability of a civil action in relation to s 10(b)-5 of the Securities Exchange Act 1934 (on the basis of the prior recognition of such an action) but refused to extend the action against those who aided and abetted the breach. Stevens J in dissent objected, partly on the basis that in 1934 courts were much more willing to imply causes of action, and hence the court ought to be prepared to find an intent based on that view. By 1999, in a trial decision, Kramer v Secretary of Defense, Kennedy J held that the effect of the later treatment of Cort was that ‘the Court has continued to find all four Cort factors relevant, but has shifted the emphasis among these factors to focus on legislative intent’.129 However, he noted that another view had been taken of Cort by Scalia J, concurring in the later decision of Thompson v. Thompson: It could not be plainer that we effectively overruled the Cort v. Ash analysis in Touche Ross & Co. v. Redington, and Transamerica Mortgage Advisors, Inc. v. Lewis, converting one of its four factors (congressional intent) into the determinative factor, with the other three merely indicative of its presence or absence.130

Still, the majority position represented in Thompson was that the fourfactor test is still to be applied.131 Perhaps the most recent authoritative summary of this area comes from the following comments of Ginsburg J, writing for the Supreme Court in 2011: ‘“[R]ecognition of any private right of action for violating a federal statute,” currently governing decisions instruct, “must ultimately rest on congressional intent to provide a private remedy”.’132 Earlier in this chapter I argued that an important part of the BSD approach is the question of ‘legislative intent’. But intent can be found in a variety of ways, and what we find in the US Supreme Court jurisprudence is a shift towards an intense focus on the precise terms of the statute. The shift towards a greater focus on the exact words of the text itself can also be illustrated in cases dealing with the question of what might be called ‘ancillary’ issues that arise even if an intention to create civil liability has been 128

Central Bank of Denver v First Interstate Bank of Denver 511 US 164 (1994). Kramer v Secretary of Defense 39 F Supp 2d 54, 57 (DDC 1999). 130 Thompson v Thompson, 484 US 174, 189, 108 S Ct 513, 521 (1988) (citations omitted). Note that this was the judgment quoted in the High Court of Australia by Gleeson CJ and Gummow and Hayne JJ in Slivak v Lurgi (Australia) Pty Ltd [2001] HCA 6, 205 CLR 304, [28]. 131 See the comments of Molloy DJ in First Pac Bancorp, Inc v Helfer 224 F 3d 1117, 1122 (9th Cir 2000): ‘Nevertheless, we still find the four-factor test helpful in determining whether a statute provides a private right of action’. 132 Astra USA, Inc v Santa Clara County, Cal, 131 S Ct 1342, 1347 (2011) (citations omitted). 129

204 Neil Foster found. That is, even if there is civil liability, issues arise as to the relevant parties (to whom does the liability attach?), the extent of damages available, statutes of limitations, and myriad other issues. Popkin notes that one approach to the question was found in Franklin v Gwinnett County Public Schools,133 where it was held that, once the court has decided that a civil remedy is available, the question of which civil remedy is completely at large and does not itself need to be justified by the legislation concerned.134 Hence an order for damages may be made even if Congress has not addressed its mind to the question. However, in the later decision in Gebser v Lago Vista Independent School District the majority (Stevens J dissenting) held that there was no vicarious liability for breach of the implied civil action.135 The legislation in question was the sex discrimination legislation found actionable in Cannon.136 In Gebser a student had been the victim of sexual harassment (actually seduction and sexual activity) by a teacher, and the question was whether the school district could be held vicariously liable. It is interesting that the facts of the case are so similar to those of the major Commonwealth common law cases dealing with (and ruling in favour of) vicarious liability for sexual assault by caregivers: Bazley v Curry,137 Lister v Hesley Hall138 and Lepore v NSW.139 Yet here the decision was based on principles relating to federal funding of schools, and the majority found that there was no liability.140 The ‘shrinking’ of implied federal civil liability can also be illustrated from the interpretation of another provision. Popkin notes how the ‘implied right of action’ becomes relevant under 42 USC §1983.141 That legislation gives a civil action to someone: 1. deprived of a ‘right’ secured by the Constitution or the laws of the United States (ie a right given by federal statute), 2. by the action of someone purporting to act under a state law. There was initially some doubt as to whether a ‘right’ secured by federal law had to be a fully blown implied civil right, or whether something less

133

Franklin v Gwinnett County Public Schools 503 US 60 (1992). See WD Popkin, Materials on Legislation, 5th edn (St Paul, Minn, Foundation Press, 2009) 793–95. 135 Gebser v Lago Vista Independent School District 524 US 274 (1998). 136 Above, n 124. 137 Bazley v Curry [1999] 2 SCR 534. 138 Lister v Hesley Hall [2001] UKHL 22, [2002] 1 AC 215. 139 Lepore v NSW [2003] HCA 4, 212 CLR 511. 140 Dobbs, above n 31, at § 335 cites opposing views among state courts as to whether there can be vicarious liability for sexual assault of children in care: Stropes v Heritage House Chidrens Ctr of Shelbyville Inc 547 NE 2d 244 (Ind 1989) (holding that there can be), Niece v Elmview Group Home 131 Wash 2d 39, 929 P 2d 420 (1997) (saying not). 141 Above n 134, at 797–801. 134

Statutes and Civil Liability in the Commonwealth and US 205 than that could be protected under § 1983. However, the decision of the Supreme Court in Gonzaga University v Doe,142 now holds that the ‘right’ under federal statute must be one which would satisfy the rules for existence of an ‘implied right of action’. There are then comments by Rehnquist CJ suggesting that the test is whether Congress has created rights ‘in clear and unambiguous terms’.143 All this means that the remedy provided by the provision has been made much harder to access. Similarly, the recent jurisprudence on implied constitutional rights of action also demonstrates this trend. The decision in Bivens held that there would be a civil action against a federal government official who violates a constitutional right of the plaintiff.144 But Popkin notes that in recent years this action has been interpreted narrowly, no action being available, for example, against a private corporation contracted to perform services for the federal government.145 The more recent approach of the Supreme Court, then, to implied rights of action is summed up as follows by Wright and Miller:146 Long ago, it might have been appropriate to assume that Congress intended that courts would routinely provide other remedies to persons injured by a violation of the substantive rule. As complex regulatory and social engineering schemes have proliferated, however, this assumption is no longer appropriate.

All the above discussion relates to federal statutes. It is clear that a state court may also grant a remedy under a state statute without fitting it into the NPS framework. See, for example, the following comments from a recent decision of the Illinois Appellate Court: Four factors are to be considered in determining whether a private right of action may be implied from a statute: ‘Implication of a private right of action is appropriate if: (1) the plaintiff is a member of the class for whose benefit the statute was enacted; (2) the plaintiff’s injury is one the statute was designed to prevent; (3) a private right of action is consistent with the underlying purpose of the statute; and (4) implying a private right of action is necessary to provide an adequate remedy for violations of the statute.’ Fisher v. Lexington Health Care, Inc.

142

Gonzaga University v Doe 536 US 273 (2002). Perhaps not surprisingly, though, where there has been a clear finding that an implied right of action exists, a § 1983 action will often be available as a supplementary remedy: see Fitzgerald v Barnstable School Committee 555 US 246 (2009). 144 Above n 33. 145 Above n 134, at 802, citing Correctional Services Corp v Malesko, 534 US 61 (2001). 146 CA Wright, AR Miller et al, Federal Practice and Procedure, 3rd edn (St Paul, Minn, West, 2011) vol 13A, § 3531.6. 143

206 Neil Foster Courts will imply a private remedy where a clear need to effectuate the purpose of a statute exists.147

Nevertheless, where the doctrine of NPS is available, litigants at the state level may prefer to rely on that, which (as noted above) does not at the moment involve an investigation of legislative intent.

IV. GENERAL ISSUES

There are a number of general matters that arise from the preceding review.

A. Methods of Statutory Interpretation One set of issues has to do with statutory interpretation. The question to what extent courts may find a Congressional intent when it has not been explicitly spelled out is fundamentally linked to the controversial issues that continue to arise in US constitutional jurisprudence about the ‘original meaning’ of the text of the Constitution, and the appropriate role of judges in statutory interpretation. Eskridge et al, for example, describe in some detail a shift in techniques of statutory interpretation during the twentieth century.148 The ‘Legal Process Era’ from 1940 to 1973 was a time when the legal system as a whole was seen as being shaped by each of the participants, and where courts were justified in furthering what they saw as legislative purpose in the way they read statutes.149 However, a shift seemed to take place in the 1970s and 1980s into an era of what has been called the ‘New Textualism’, stressing the need, on ‘separation of powers’ and ‘institutional competency’ grounds, for courts to pay close attention to what statutes actually say, and to move away from the temptation to imply rights and duties.150 The most prominent judicial voice in this movement is, of course, Justice Scalia.151 While there is an ongoing debate about whether the theory of ‘textualism’ is

147 Citizens Opposing Pollution v ExxonMobil Coal USA, above n 37, at 556 (citations omitted). 148 WN Eskridge Jr, PP Frickey and E Garrett, Cases and Materials on Legislation: Statutes and the Creation of Public Policy, 4th edn (St Paul, Minn, Thomson, 2007). 149 They cite as an example Moragne v States Marine Lines, Inc 398 US 375 (1970), where the Supreme Court was prepared to develop the law of compensation to seamen to provide an action for wrongful death which had not been explicitly given in the Jones Act, 46 USC § 688. 150 See Eskridge et al, above n 148, at 765 ff. 151 See A Scalia, A Matter of Interpretation (Princeton, Princeton University Press, 2007).

Statutes and Civil Liability in the Commonwealth and US 207 still authoritative,152 there is no doubt that it has had a significant influence on the way that US courts approach statutes. The intense debates about statutory interpretation techniques conducted in the United States have not really been replicated in the Commonwealth. Of course there are differences of opinion over interpretation, and sometimes those differences have been very stark. For example, the decision of the High Court of Australia in Commissioner of Taxation v Ryan153 involved the expiry of a ‘limitation period’ after which a taxpayer’s return could not be revised by the Taxation Office. Section 170(3) of the Income Tax Assessment Act 1936 (Cth) provided: Where a taxpayer has made to the Commissioner a full and true disclosure of all the material facts necessary for his assessment, and an assessment is made after that disclosure, no amendment of the assessment increasing the liability of the taxpayer in any particular shall be made after the expiration of 3 years from the date upon which the tax became due and payable under that assessment.

The majority of the High Court found that a notice (despite being labelled ‘assessment’) that specified a zero amount of money due was not within the terms of the Act an ‘assessment’. Kirby J in dissent registered outrage at what he saw as ‘literalistic’ reading of the legislation that paid no attention to the overall purpose of the provision. He commented that [T]here is an alternative construction of the relevant provisions of the Act. That construction is to be preferred because it promotes the purpose for which the Parliament enacted s 170(3). It avoids the bizarre and clearly unjust consequences which would flow from the construction urged for the Commissioner.154

Despite this occasional tension, overall it seems that the High Court of Australia today is happy to describe the authoritative approach to interpretation as one which both pays high respect to the words that Parliament has used, but also is prepared where necessary to apply the ‘purposive’ approach of considering what Parliament intended.155 152 Eskridge et al, above n 148, at 793, citing eg J Molot, ‘The Rise and Fall of Textualism’ (2006) 106 Columbia Law Review 1. 153 Commissioner of Taxation v Ryan [2000] HCA 4, 201 CLR 109. 154 Ibid, at [61]. 155 For example, in one of the most controversial decisions of the court in recent years, Plaintiff M70/2011 v Minister for Immigration and Citizenship; Plaintiff M106 of 2011 v Minister for Immigration and Citizenship [2011] HCA 32, the court ruled by majority that a ministerial decision to expel unauthorised entrants to Australia to Malaysia, pursuant to an agreement with that country, was invalid. The issues in the case were fundamentally to do with how the legislation was to be interpreted. Yet all members of the court were generally agreed as to how to approach the matter: as French CJ put it at [13]: ‘The Court, however, must look to the text, context and purpose of the relevant statutory provision’. This included, as one of the other majority judgments said at [85] ‘close attention to the relevant statutory text’. But it also permitted, as a supplement to this, consideration of the broader purposes of the legislation (to allow Australia to accede to international conventions dealing with refugees) and discussion of the ‘legislative history’ (at [96]).

208 Neil Foster B. Negligence Per Se and Implied Rights of Action—Comparison How does the US law on ‘implied rights’, then, relate to the law on NPS? Sherman’s perceptive article points out how similar the two questions of ‘implied right of action’ (from federal statutes) and NPS (as an issue of state law) are in fact.156 One may note, as has been done above, that both doctrines have their roots in the common law BSD jurisprudence. But as Sherman notes, in states where the ‘majority’ NPS approach is in place (a breach of the statutory standard is automatically a breach for the law of negligence) the courts have ‘retained, in large part, the doctrine of implied remedies which existed at common law’.157 He goes on to suggest that they have ‘replac[ed] the duty and breach elements of negligence with the concept of statutory violation’. This remark brings out something that was noted previously. The classic statements of the majority position simply refer to the breach of the standard being replaced. But when it is noticed that a precondition for this happening is that the statute be designed to protect a person in the situation of the plaintiff, what we have in effect is also a replacement of the duty issue. If this judgment is made about the statute, then we are in effect finding that a duty to obey the statute (the equivalent of a general ‘duty of care’) is owed to the plaintiff.158 Once this is realised, then, as Sherman says, ‘it is difficult to distinguish the cause of action for negligence per se from an implied cause of action’.159 If the statute in question is a federal statute, ‘[t]he effect is exactly the same as it would be if a state court determined that an implied right of action was created by a federal statute’.160 One may note, however, that in light of the discussion above of the NPS jurisprudence there will be one major difference: that in finding that there is liability, the state court will have bypassed the question of legislative intention to create civil liability. Sherman seems to sum the situation up well when he notes: There has been a radical shift in the approach taken by the Supreme Court toward implied rights of action that has not been mirrored in the state courts. State rules of negligence per se and implied causes of action are descended from the old common law concept of implied remedies. However, the federal approach as reflected in Cort v Ash, has concentrated on a mixture of legislative intent and federalism

156

Above n 113. Ibid, at 889. 158 See the quote from Cardozo J in Martin, above n 41, at 815: ‘the duty that one traveler on the highway owes under the statute to another’ (emphasis added). 159 Above n 113, at 889. 160 Perhaps the only difference would arise in states where a large menu of ‘excused violation’ rules were in place, which would not be drawn from the statute. See also Stabile, above n 116, at 865. 157

Statutes and Civil Liability in the Commonwealth and US 209 issues; the state approach has run the gamut of approaches within the broad rubric of negligence.161

The complexity of the whole area can be illustrated by the problem Sherman wrestles with in the final part of his article. Suppose a federal court has ruled that a particular federal statute does not create an implied cause of action. A state court is then asked to decide whether a breach of this statute can be used as the basis for an NPS suit. Should the state court automatically defer to the decision of the federal court? The problem is not simply one of judicial comity, because on the prevailing analysis, the federal court was asked a different question to that facing the state court. On current theory, the state court should presumably be free to make up its own mind on the issue. The obvious solution adopted by the court whose decision is discussed in RBJ Apartments v Gate City Savings & Loan Assn162 is criticised by Sherman, as he is correct to do on the current analysis. The RBJ court, in holding simply that where there is no implied federal right of action, there can be no state action in NPS, does indeed ‘equate implied remedies with negligence per se’.163 One response would be: so they should, and US courts should align the two bodies of law.

V. CONCLUSIONS

It is hoped that this review of the US law in comparison to, and its development in the context of, the ‘traditional’ common law tort of BSD has provided some illumination about features that are common to both systems. In both contexts, where a question arises as to whether a provision of a statute provides a civil remedy, the courts will address important questions as to whether the statute in question was designed to protect the interests of the person concerned (rather than simply to be addressed to everyone in the community), and whether the type of harm that it aims to avoid was the type of harm suffered by the plaintiff. Cases from each system on specific statutes may illuminate decision making in the other system where similar issues arise. 161 Above n 113, at 888–89. A question which is beyond the remit of this chapter is also raised of the appropriate venue for determination of civil liability under federal law. In the US federal system this matters because there is a limited jurisdiction enjoyed by the federal courts, and it becomes important to know whether determination of the issue is a ‘federal issue’. Sherman, ibid, at 884 ff discusses Merrell Dow Pharmaceuticals Inc v Thompson 478 US 804 (1986), where the ultimate finding was that it was not a matter of ‘federal jurisdiction’ for a state court to be asked to apply a federal statute in a negligence per se case. (Hence the case could not be removed from a state court into a federal court.) The majority concluded that, because there was no direct implied remedy under the Act, it was not a matter of ‘federal jurisdiction’, and hence the case could continue in a state court. 162 RBJ Apartments v Gate City Savings & Loan Assn 315 NW 2d 284 (ND 1982). 163 Sherman, above n 113, at 900.

210 Neil Foster But it has also been concluded that there are some genuine difficulties and conceptual problems with the US law on these issues at the moment, and with some diffidence it is suggested that US lawmakers should consider seriously whether a better approach is possible, taking into account the solutions reached by common law courts around the Commonwealth on the BSD action. In the area of NPS actions, the approach that is taken in the majority of states follows the guidance given by the influential Cardozo J in Martin v Herzog of treating the statutory breach as part and parcel of a negligence action, rather than a separate civil action. While this approach opens up the possibility of a wide range of claims by victims of harm, it has been the subject of a number of convincing critiques.164 Perhaps the key issue is the fact that nowhere in the judgment in Martin is there any consideration of whether the legislator intended the statutory standard to be applied as part of a civil action in this way. The fact that application of statutes to civil claims is then opened up so widely has led to both the appearance of injustice in those few cases where a defendant has been held liable for a large award of damages based on breach of a fairly ‘trivial’ provision, or more commonly to the fairly unprincipled development of a wide range of judicially crafted ‘excuses’. This has in turn cut away any hope that the blanket application of statutes as part of the negligence action would create more certainty in comparison to the open-ended ‘breach’ question at common law. Ironically, it has also led to increased institutional tension between the courts and the legislature, as it appears that courts are taking it upon themselves not merely to apply a standard which the legislature has not directed to be applied, but then to arbitrarily dispense with that standard when it seems it will not achieve justice in a particular case. It is suggested that a return to the model of the Commonwealth BSD action may be a better option.165 That model requires a decision to be taken at the outset as to whether a particular statute, when considered in light of the range of legislative construction techniques usually applied, ‘intends’ to allow a civil action.166 The difficulty of making this decision on a statuteby-statute basis is alleviated to a large extent by a course of decisions of the courts where, in effect, interpretative ‘presumptions’ are used to deal with general classes of statute.167 164 See Leonard, above n 81; Sherman, above n 113; Blomquist, above n 75, and Dobbs, above n 31. 165 See Foy, above n 90. 166 See the quote from Kitto J in Sovar, above n 69. 167 So that, for example, workplace safety statutes are almost always interpreted as giving rise to civil liability unless this is clearly excluded. See NJ Foster, ‘Private Rights and Public Regulation: Breach of Statutory Duty and Workplace Safety’ (Conference on ‘Private and Public Law—Intersections in Law and Method, TC Beirne School of Law, University of Queensland, Brisbane, July 2011). On the other hand, as previously noted, there is a strong Commonwealth tradition of not allowing traffic regulations to be used as the basis for BSD claims. See above n 47.

Statutes and Civil Liability in the Commonwealth and US 211 If the US state NPS decisions seem from the Commonwealth perspective to be too ready to create civil liability on the basis of statutes, it has to be said that it seems that the (mostly federal) US jurisprudence dealing with ‘implied rights of action’ has become far too narrow. The line of decisions noted above, from Rigsby through Cort and into more recent years, seems correct in holding that the primary question should be that of Congressional intent. But those members of the Supreme Court who are committed to confining the issue of legislative intent to a myopic examination of only the words used in the confines of the one piece of legislative text are arguably engaged in an undemocratic process under the guise of furthering democracy and the separation of powers. The broader Commonwealth approach to legislative interpretation takes into account the fact that every text has a context—not simply the context of its immediate paragraph, or the Act within which it appears, but a context which involves a wider range of factors: the broad reasons for which the Act was passed, the shared community understanding of various concepts, the legal system into which the legislation has been placed (including the ‘ancillary’ rules about liability of other parties, types of remedies, limits on damages, etc), and of course the previous judicial interpretation of this legislation, or of similar legislation. Even the current Supreme Court, of course, has acknowledged this in being prepared to allow implied civil enforcement of legislation that has previously been directly held to be enforceable by the Supreme Court itself.168 It is suggested that there are many reasons for the Court to return to a slightly wider vision of what legitimate statutory interpretation involves; to see the judicial task as that of furthering, not cramping by narrow reading, the purposes of Congress in conferring certain rights. Foy puts it in this way: But courts do provide remedies for the wrongs that the law defines. Courts do not have to be told to provide these remedies. They do it because they are courts. The idea that courts cannot provide remedies for wrongs defined by law, absent proof of an affirmative legislative intention that they should do so, is at odds with this old-fashioned conception of the judicial function.169

It may be hoped that reference to the way courts in the Commonwealth are approaching the issue of statutory civil liability may contribute to an ongoing dialogue to allow this to happen. VI. CONCLUDING EXCURSUS: FEDERAL ISSUES IN BSD LITIGATION IN AUSTRALIA

This chapter raises an interesting question for Australia, which, while a Commonwealth country, of course has a federal constitution. Should an 168 169

See Janus Capital Group Inc v First Derivative Traders 131 S Ct 2296 (2011). Foy, above n 90, at 582.

212 Neil Foster Australian state court find that a breach of federal legislation is actionable under a BSD analysis? So far the question seems not to have arisen. Perhaps one reason for this is that, as noted, there is ‘one common law’ in Australia, as opposed to in the United States.170 So in theory a court of whatever stripe (federal or state) should probably be applying the same law wherever located in Australia. Hence it would make sense for a state court to hold that, if other elements of the action were satisfied, a BSD claim in relation to federal law would be available. However, the state court would still have to address the issue of the ‘intention’ of Parliament—could the federal Parliament have ‘intended’ that its prohibition create liability actionable in state courts? Arguably so—the relevant intention is to prohibit certain behaviour for the benefit of a member of a relevant class; the right to have this duty enforceable should probably be seen to be the issue, not the forum. State courts are generally invested with federal jurisdiction, so would have the power to enforce a federal right created in this way. Perhaps the trickier question is that raised by a state court adjudicating on a breach of legislation that was passed by another state. While the issue, again, does not seem to have arisen, presumably the court would need to find that on ‘conflict of laws’ principles, the victim was entitled to the benefit of an obligation which actually bound the alleged tortfeasor. The ‘choice of law’ rule for tort actions in Australia since John Pfeiffer Pty Ltd v Rogerson is that the ‘place of the tort’, the lex loci delicti, governs substantive rights and obligations.171 So it would seem to make sense that a court in one state would be willing (once appropriately seized of the issue) to apply a statutory obligation that was imposed on a wrongdoer under the law of another state where the wrongdoer was present. The ‘full faith and credit’ provision in section 118 of the Constitution would seem to support such a view. Another ‘federal’ issue in Australia is that which seems to have been hinted at in Byrne and Frew172—if there is a constitutional separation of powers, then how can a court in the judicial branch create liability based on a statute passed by the legislative branch (noting that in Australia it is not a breach of separation of powers for the Executive to pass regulations)? Gummow and McHugh JJ commented in that case: The result would have to be that there was ‘arising under’ a law ‘made by the Parliament’, in the sense of s 76(ii) of the Constitution, a new species of ‘matter’. However, where a question arises as to the creation of new rights and liabilities which will engage Ch III of the Constitution, it is to be expected that the Parliament will clearly state its will.173

170 171 172 173

John Pfeiffer Pty Ltd v Rogerson [2000] HCA 36, 203 CLR 503, [15]. Ibid. Above, n 21. Above, n 21, at 458 (some citations omitted).

Statutes and Civil Liability in the Commonwealth and US 213 The short answer to this is that common law courts, in ruling on an actionable BSD, are in theory not at all legislating—they are merely spelling out the necessary implication of what Parliament has said, that a right given by Parliament can be acted on in civil proceedings. (This would preclude the view that is sometimes expressed in US academic writing, however, that the courts have ‘authority’ to decide what statutory standard should be applied separate from any question of legislative intent.)174

174 It should be noted that comments about the federal complications of the BSD analysis were made by the High Court of Australia in Byrne v Australian Airlines Ltd [1995] HCA 24, 185 CLR 410, [104], fn 146, citing comments of Scalia J in dissent in Pennsylvania v Union Gas Co [1989] USSC 116, 491 US 1 about ‘plumbing the intent’ of Congress, and in Slivak v Lurgi (Australia) Pty Ltd [2001] HCA 6, 205 CLR 304, [28], fn 12, citing comments of Scalia J in Thompson v Thompson, above n 130. Some members of the High Court at this stage seem to have been heavily influenced by the approach to statutory interpretation adopted by Justice Scalia. Whether this approach will be pursued in the future is not clear.

8 The ‘Doubles the Risk’ Test for Causation and Other Related Judicial Misconceptions about Epidemiology CLAIRE McIVOR

I. INTRODUCTION

I

N RECENT YEARS, a significant amount of judicial time and effort has been spent discussing the applicability of the ‘doubles the risk’ test to questions of probabilistic causation in tort law. Unfortunately, this discussion has resulted in more questions being raised than answered. Given the frequency with which difficult cases of probabilistic causation come before the courts, most notably in the context of claims involving negligent occupational exposures to carcinogenic substances, the lack of clear guidance on the ‘doubles the risk’ test is a matter of concern. Through an analysis of recent case law, this chapter will demonstrate that there are serious problems with the test. It will show that at a theoretical level the test belies a series of misconceptions about the science of epidemiology and its potential relevance to tort law. It will further show that at a practical level the purportedly scientific manner in which the test is applied by the courts is both unscientific and internally inconsistent. Focusing largely on Lord Phillips’s speech in Sienkiewicz v Greif (UK) Ltd,1 the first part of this chapter will expose the errors underlying the judicial association of the ‘doubling of the risk’ approach with epidemiology. It will address: (i) the mistaken connection made by the courts between the doubling of the risk rule and the epidemiological concept of relative risk; and (ii) the common misconception that epidemiologists deal only in naked statistics. At a more fundamental level, the analysis of Sienkiewicz will also demonstrate that there are serious legal problems underlying the current 1 Sienkiewicz v Greif (UK) Ltd [2011] UKSC 10, [2011] 2 AC 229, aff’g [2009] EWCA Civ 1159, [2010] 1 QB 370 [Sienkiewicz].

216 Claire McIvor judicial approach to assessing issues of probabilistic causation in tort. These problems relate primarily to a misapplication of the balance of probabilities standard of proof and may be said to stem from a judicial tendency, when faced with statistical evidence, to lose sight of the nature and purpose of the factual causation stage of the negligence enquiry. Addressing the practical problems associated with recent applications of the ‘doubles the risk’ test in tort law, the second part of the chapter will turn to the recent High Court decision in Jones v Secretary of State for the Department of Energy & Climate Change (commonly described as ‘the Phurnacite Litigation’).2 It will demonstrate that current disagreements as to the remit of the ‘doubles the risk’ test are bound up in differing interpretations of the ‘material contribution to harm’ test for causation stemming from Bonnington Castings Ltd v Wardlaw3 and, in particular, the significance of the distinction between ‘divisible’ and ‘indivisible’ injuries. Such differences of opinion are long-standing and unlikely to be resolved any time soon. It will be argued that Lord Phillips’s views on these issues are at least partially influenced by his overt scepticism of the legal value of epidemiological evidence. Moving on to look at the manner in which the ‘doubles the risk’ test was applied by the Court of Appeal in Novartis Grimsby Ltd v Cookson,4 and comparing this with the application of the test at first instance in Sienkiewicz, the discussion will further demonstrate that there are inherent problems with the manner in which relevant risks are calculated and compared.

II. ‘DOUBLES THE RISK’ AND MISCONCEPTIONS ABOUT EPIDEMIOLOGY

The ‘doubles the risk’ test for causation has been discussed recently by the United Kingdom Supreme Court in Sienkiewicz, the latest in a line of highprofile mesothelioma cases.5 The claimant in Sienkiewicz was the daughter and administratrix of Mrs Enid Costello, who died of mesothelioma in 2006 at the age of 74. Mrs Costello had worked for the defendant, Greif (UK) Ltd, from 1966 until 1984 at its factory in Ellesmere Port. The factory manufactured steel drums and this involved the release of asbestos dust into the factory atmosphere. Although Mrs Costello worked in one of the factory offices rather than on the factory floor, it was accepted at trial that her work took her all over the factory and that she would have come into contact with 2 Jones v Secretary of State for the Department of Energy & Climate Change [2012] EWHC 2936 (QB) [Jones]. 3 Bonnington Castings Ltd v Wardlaw [1956] AC 613 [Bonnington]. 4 Novartis Grimsby Ltd v Cookson [2007] EWCA Civ 1261 [Novartis]. 5 Above n 1. See Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, [2003] AC 32 [Fairchild]; Barker v Corus UK Ltd [2006] UKHL 20, [2006] 2 AC 572.

The ‘Doubles the Risk’ Test for Causation 217 asbestos dust from time to time. In addition to this occupational exposure, Mrs Costello had also been exposed to low levels of asbestos in the general atmosphere of the area where she lived, Ellesmere Port. At first instance, Judge Main QC held that, though the defendant company had been in breach of its duty of care by exposing Mrs Costello to asbestos at work, the claim failed for lack of causation. He held that to establish causation the claimant had to show that the occupational exposure had at least doubled the risk from that due to environmental exposure alone. On the basis of his own quantitative assessments of Mrs Costello’s cumulative occupational and environmental exposures, he found that the occupational exposure had increased the total exposure by only 18%.6 The claimant appealed to the Court of Appeal on the primary ground that the judge had erred in his approach to causation by failing to apply the exceptional test for causation set out in Fairchild v Glenhaven Funeral Services Ltd.7 In accordance with this test, the claimant could establish causation merely by showing that the defendant had made a material contribution to the risk of the disease being contracted. Given that the occupational exposure had materially contributed to the risk of mesothelioma, the claimant ought to have succeeded and recovered in full. The Court of Appeal accepted this argument and reversed the decision of the trial judge. On appeal to the Supreme Court, Greif (UK) Ltd contested the decision of the Court of Appeal on two alternative but interrelated grounds: (i) that the Fairchild test did not apply to single exposure cases and the default test was the ‘doubles the risk’ test; or (ii) that if the Fairchild test did apply, the occupational exposure would still need to at least double the risk posed by the environmental exposure before it could be labelled as a ‘material’ contribution to the risk of contracting mesothelioma. Rejecting both arguments, the Supreme Court found that the Fairchild test for causation applied to all mesothelioma cases, regardless of whether the relevant exposures were single or multiple and irrespective of the involvement of non-tortious elements. No member of the Supreme Court had any doubt about the applicability of the Fairchild test to the claim in Sienkiewicz. Nevertheless, considerable time was spent obiter discussing the general applicability of the ‘doubles the risk’ test. This was motivated largely by a desire to take issue with a suggestion made by Lady Justice Smith in the Court of Appeal to the effect that the ‘doubles the risk’ test should operate as the default test for causation in so-called ‘evidential gap’ cases (cases in which it is not scientifically possible to determine exactly how a particular harm has been brought about).8 Significantly, to cast doubt on the general applicability of the doubling of 6 7 8

Sienkiewicz, above n 1, at [9] (CA). Above n 5. Sienkiewicz, above n 1, at [23] (CA).

218 Claire McIvor the risk approach, the Supreme Court thought that it was necessary to take issue with epidemiology. This is due to a common judicial misconception that the ‘doubles the risk’ test is a specialist epidemiological principle. As the following discussion will demonstrate, the misconception that epidemiologists treat evidence of a mere doubling of the risk as adequate proof of causation is just one of a series of judicial errors about epidemiology as a discipline.

A. Confusing ‘Doubles the Risk’ with the Epidemiological Concept of Relative Risk At the very beginning of his speech in Sienkiewicz, Lord Phillips states that the ‘doubles the risk’ test is usually applied to epidemiological evidence and that it operates by attributing causative effect to any factor that more than doubles a risk that would otherwise have been present.9 Later in his speech, he proceeds to put forward a seriously flawed account of epidemiology generally and of the concept of relative risk (hereafter abbreviated as RR) in particular. Lord Phillips takes the view that epidemiological evidence cannot provide a satisfactory basis upon which to make findings of factual causation in law. However, this is primarily due to his mistaken belief that epidemiology is largely concerned with the bare calculation of incidence rates.10 He begins by describing epidemiology as ‘statistical analysis’.11 After explaining that epidemiology is ‘the study of the occurrence and distribution of events (such as disease) over human populations’ and that it ‘seeks to determine whether statistical associations between these events and the supposed determinants can be demonstrated’,12 he proceeds to claim that epidemiology is not 9 Above n 1, at [4] (SC). See also at [73] (SC), where Lord Phillips refers to the ‘doubles the risk’ test as an ‘epidemiological principle’. 10 Also telling in this respect is the discussion by both Lords Phillips and Rodger in Sienkiewicz of what they describe (at [28], [149]) as the ‘epidemiological example’ used by Lord Mackay in Hotson v East Berkshire Health Authority [1988] UKHL 1, [1987] AC 750. This example draws on the facts of McGhee v National Coal Board [1972] UKHL 7, [1973] 1 WLR 1, in which the causal issue for determination was whether the provision of washing facilities at the end of the shift for employees who had been exposed to brick dust during the course of their employment would have prevented the victim in question, one of those employees, from developing dermatitis. Lord Mackay (at 786) hypothesised the following sets of figures as applying to the claim in McGhee: ‘One might, for example, say that it was established that of 100 people working under the same conditions as the pursuer and without facilities for washing at the end of their shift 70 contracted dermatitis: of 100 people working in the same conditions as the pursuer when washing facilities were provided for them at the end of the shift 30 contracted dermatitis’. By presenting this as an ‘epidemiological example’, their Lordships create the misleading impression that such bare statistical information about incidence rates represents the sum total of what epidemiology can provide. 11 Sienkiewicz, above n 1, at [11] (SC). 12 Ibid, at [80] (SC).

The ‘Doubles the Risk’ Test for Causation 219 concerned with the further question of whether these statistical associations demonstrate an underlying biological causal relationship. With respect, this is simply untrue. Statistics are certainly an important tool used by epidemiologists. However, they are no more than a tool and epidemiologists are more than just statisticians. In addition to designing and executing studies to collect relevant data, epidemiologists use various sophisticated techniques for interpreting the data precisely in order to determine whether any indicated statistical associations provide evidence of an underlying biological relationship of cause and effect.13 What Lord Phillips does not appear to understand is that epidemiologists are trained in both statistics and research methodologies and that many also have training in medicine. The highly specialised techniques used by epidemiologists in drawing causal inferences from empirical data are informed by all three disciplines.14 It is in the course of pressing his concerns about the reliability and adequacy of epidemiological evidence as a legal tool that Lord Phillips purports to explain the epidemiologic concept of RR. He explains that RR compares the relevant experience of a group that is subject to a particular exposure with the experience of a group that is not. This comparison measures the risk of a disease relative to exposure. He states that where RR exceeds 2, the statistical likelihood is that the particular exposure was the cause of the disease.15 Having just stated that epidemiology is concerned with the question of statistical associations and not with the separate question of whether such associations demonstrate an underlying biological cause, he thereby implies that epidemiologists would not look any further than the statistical likelihood calculated by the RR when formulating their conclusions about causation. Thus he concludes that RR>2 is a ‘tenuous basis for concluding that the statistical cause of a disease was also the probable biological cause, or cause in fact’,16 with the term ‘cause in fact’ being used in its bespoke legal sense. In actual fact, epidemiologists do not treat evidence of a doubling of a statistical risk as proof of causation. As explained above, epidemiology involves much more than the calculation of incidence rates. Epidemiologists

13 One technique involves the application of a set of causal inference principles known as the Bradford Hill Criteria. Developed by an epidemiologist called Austin Bradford Hill, these principles act as a checklist for determining whether an observed statistical association is likely to be a causal one. See AB Hill, ‘The Environment and Disease: Association or Causation?’ (1965) 58 Proceedings of the Royal Society of Medicine 295, 295–300. 14 Moreover, epidemiologists can also stratify data to take account of personal characteristics of claimants and produce evidence that speaks directly to the individual circumstances of the legal claim. For an example of how statistical evidence can be personalised using a technique known as Bayes’ Theorem, see R Goldberg, ‘Using Scientific Evidence to Resolve Causation Problems in Product Liability: UK, US and French Experiences’ in R Goldberg (ed), Perspectives on Causation (Oxford, Hart Publishing, 2011) 163. 15 Sienkiewicz, above n 1, at [82] (SC). 16 Ibid, at [83] (SC).

220 Claire McIvor do not base any conclusions about causation on RR results alone, and certainly not on a mere result of RR>2. Moreover, contrary to the impression given by Lord Phillips, RR>2 holds no intrinsic significance or value in epidemiology. While RR>2 can sometimes be used as a floor value for deciding whether further epidemiological investigations into the relationship between a particular agent and a disease outcome are warranted, such investigations will often proceed even with results showing the RR to be less than 2. Furthermore, the RR calculation is just one of many different measurements of disease occurrence used by epidemiologists. When analysing the results of a study, an epidemiologist will employ a variety of complicated mathematical equations, each designed to measure a specific aspect of disease occurrence.17 Lord Phillips’s doubts about the reliability of epidemiological evidence are similarly founded on a lack of knowledge of epidemiology. He suggests that such evidence is unreliable as a guide to causation in law for the following reasons: (i) the difficulty in collating sound epidemiological evidence; (ii) the difficulty of obtaining reliable evidence as to the relevant experience of the victim; and (iii) uncertainty as to the adequacy of the epidemiological evidence that is available. He gives the impression that such difficulties are insurmountable. However, these concerns can be easily addressed through the use of epidemiological expert witnesses. There are numerous ways of assessing the scientific robustness of a piece of epidemiological evidence and its particular pertinence (or lack thereof) to the specific legal issue being determined in an individual claim. Such assessments can be carried out by any suitably qualified and experienced epidemiologist who would look first at the design and execution of the study that produced the data in question. Some types of study (such as clinical trials or cohort studies) are better suited to certain research questions than others and much will depend, for example, on how the research question is framed.18 The epidemiological expert witness will be able to assess how well the authors of the study have accounted for, inter alia, selection bias, information bias, effect modification and confounding.19 Further reliability measures which will be considered are the ‘p-value’ and the ‘confidence interval’ associated with the study. Put very simply, these measure the degree of variation in the 17 See further KJ Rothman, Epidemiology: An Introduction (New York, Oxford University Press, 2002) 24–56. 18 See further S Loue, Forensic Epidemiology: A Comprehensive Guide for Legal and Epidemiology Professionals (Carbondale, Southern Illinois University Press, 1999) 59. 19 See, eg, Rothman, above n 17, at 94–112. Moreover, Lord Phillips is entirely wrong to suggest, at [84] (SC), that epidemiological evidence can only focus on one specific causal agent or a number of causal agents and that no account is taken of other relevant factors such as the age or genetic susceptibility of the victim. Any good epidemiologist will account for such obvious confounders. Goldberg, above n 14, at 162, identifies a similar misconception on the part of the trial judge, Lord Nimmo Smith, in the Scottish case of McTear v Imperial Tobacco Ltd [2005] CSOH 69.

The ‘Doubles the Risk’ Test for Causation 221 results produced by data against the size of the sample of individuals used for the study. Their function is to measure the likelihood that the results produced by the study are due to the unrepresentativeness of the sample rather than a biological association. The higher the p-value, for example, the higher the likelihood that the results have been skewed by the atypical nature of the individuals making up the study sample and that those results are therefore not reliable.20 In short, through the use of epidemiological expert witnesses it is possible to weed out unreliable and irrelevant epidemiological evidence. It is clear that the ‘doubles the risk’ test is based on a misinterpretation of the epidemiological concept of RR. The above discussion has also revealed a general lack of judicial understanding of the science of epidemiology. This lack of understanding has in turn engendered an attitude of scepticism towards epidemiological evidence. This has meant that the benefits of epidemiological evidence are not being properly explored by lawyers. This is a matter of regret since epidemiology arguably constitutes a valuable source of information about causation for the purposes of deciding some tort claims, particularly in cases where there is a gap in the relevant biological knowledge. One of the main arguments against the use of epidemiological evidence is that it can only ever indicate what might have happened in an individual case and not what actually happened.21 Thus Wright argues that such evidence can only be used to establish general causation and not to establish specific causation. As such, he rejects the use of such evidence at the factual causation stage of the negligence enquiry.22 However, it must be remembered that the balance of probabilities standard of proof does not require conclusive proof of causation. Under this standard, the claimant is only required to show that it is more likely than not that the relevant breach of duty caused the pleaded harm. The balance of probabilities standard thus allows for a significant margin of error. B. Misapplication of the Balance of Probabilities Standard of Proof In Novartis, where the issue was whether the claimant’s bladder cancer had been caused by his occupational exposure to chemicals or to his own

20

See further Rothman, above n 17, at 116–29. In Sienkiewicz, above n 1, at [190] (SC) Lord Mance expresses reservations about epidemiological evidence on the basis that ‘the law is concerned with the rights or wrongs of an individual situation and should not treat people and even companies as statistics’. For Lord Dyson, at [218] (SC), the problem with epidemiology is that it cannot conclusively prove causation in an individual case: ‘At best, it can only establish a certain probability that a randomly selected case of disease was one that would not have occurred absent exposure’. 22 R Wright, ‘Liability for Possible Wrongs: Causation, Statistical Probability and the Burden of Proof’ (2008) 41 Loyola of Los Angeles Law Review 1295. 21

222 Claire McIvor smoking, Lady Justice Smith treated the ‘doubles the risk’ test as an application of the standard ‘but for’ test on the balance of probabilities. In her view, ‘if occupational exposure more than doubles the risk due to smoking, it must, as a matter of logic, be probable that the disease was caused by the former’.23 In Sienkiewicz Lord Phillips expressly endorses this apparent logic.24 However, this reasoning is based on a flawed understanding of the relationship between the probability value of a piece of statistical evidence and the probability concept that applies to the balance of probabilities standard of proof.25 Contrary to the suggestion of Lady Justice Smith and Lord Phillips, the factual causation requirement is not automatically satisfied by a statistical probability of more than 50%. To satisfy the balance of probabilities standard of proof, the court must hold a minimum degree of belief in the evidence indicating a causal connection. The numerical value associated with a piece of statistical evidence indicating a causal connection does not correlate directly with the separate issue, explained in detail below, of belief probability. The main source of the confusion here is the tendency of the courts, when faced with statistical evidence, to conflate the substantive and evidential aspects of the test for factual causation and then reduce that conflation to a nonsensical numerical test. As emphasised by Steel and Ibbetson,26 there are two separate parts to factual causation in tort: (i) the conceptual/substantive part, which requires the claimant (as the party carrying the burden of proof) to establish that there is a necessary connection between the defendant’s breach of duty and the pleaded actionable damage (the ‘but for’ test); and (ii) the evidential part, which requires the claimant to satisfy the substantive part in accordance with the relevant standard of proof, which in the tort context is the balance of probabilities standard.27 According to this standard, the claimant need only convince the court that it is at least more likely than not that his or her substantive arguments about causation, when judged against the corresponding arguments put forward by the defendant, represent what actually happened. In the context of statistical evidence, the problem is due to the common practice of describing the balance of probabilities standard as requiring a minimum of 51% proof (with everything above 50% equating with

23

Above n 4, at [74]. Above n 1, at [78] (SC): ‘I agree with her that, as a matter of logic, if the defendant is responsible for a tortious exposure that has more than doubled the risk of the defendant’s disease, it follows on the balance of probability that he has caused the disease’. 25 This reasoning is also based on the implausible premise that every case of disease has only one cause. 26 S Steel and D Ibbetson, ‘More Grief on Uncertain Causation in Tort’ (2011) 70 CLJ 451, 459–60. 27 Note that in Sienkiewicz, above n 1, Lord Phillips repeatedly refers to the test for causation as the ‘test of balance of probability’: see, eg, at [4], [9], [16] and [26] (SC). 24

The ‘Doubles the Risk’ Test for Causation 223 100% proof and everything below equating with 0%) and then mapping this figure of 51+% directly onto the relevant piece of statistical evidence. Thus, if the claimant were to produce a piece of evidence which indicated that there was a 64% likelihood that the pleaded harm was caused by the defendant’s faulty conduct, this would be taken to mean that the claimant has duly satisfied the test for causation on the basis that 64% is more than 50%. Such reasoning is completely nonsensical, yet it is employed even at the highest judicial levels.28 To assist lawyers in understanding the significance of the distinctions between the probability notion associated with the balance of probabilities standard in law and the various probabilities associated with a piece of statistical evidence, Barnes has advocated a helpful conceptual framework.29 This framework involves three probabilistic categories: fact probabilities, belief probabilities and sampling error probabilities. Under the Barnes model, the term ‘fact probability’ is used to refer to a piece of probabilistic evidence which speaks to the existence (or non-existence) of a causal connection between the relevant tortious agent, X, and the pleaded damage, Y. A piece of statistical evidence indicating, for example, a 64% likelihood that Y was caused by X is therefore relevant to the question of whether the ‘but for’ test has been satisfied. As a relatively strong likelihood value, this piece of statistical evidence may be said to speak moderately in favour of the existence of a ‘but for’ connection between X and Y. As indicated already, fact probabilities are never determinative in their own right of the legal causal enquiry, regardless of the strength of the probability value that they indicate. As such, the statistical value of 64% holds no intrinsic legal significance. This is because the legal weight of the fact probability (the result of 64%) will depend entirely on the level of confidence that the court holds in it as a piece of evidence about causation. This is where the notions of ‘belief probability’ and ‘sampling error probability’ come in. The term ‘belief probability’ relates to the overall strength or credibility of the fact probability evidence. In assessing its level of confidence in the fact probability, the court will look at a variety of factors which speak to the credibility not just of the study that produced the fact probability, but 28 Note, eg, Lord Nicholls’s suggestion in Gregg v Scott [2005] UKHL 2, [2005] 2 AC 176, [46] that ‘a patient with a 60% chance of recovery reduced to a 40% prospect by medical negligence can obtain compensation. But he can obtain nothing if his prospects were reduced from 40% to nil’. 29 D Barnes, ‘Too Many Probabilities: Statistical Evidence of Tort Causation’ (2001) 64 Law and Contemporary Problems 192. As explained by Miller, Barnes’s model builds on a model previously developed by Gold: C Miller, ‘Causation in Personal Injury: Legal or Epidemiological Common Sense?’ (2006) 26 Legal Studies 544, 550, referring to S Gold, ‘Causation in Toxic Torts: Burdens of Proof, Standards of Persuasion, and Statistical Evidence’ (1986) 96 Yale Law Journal 376. Interestingly, in Sienkiewicz, Lords Rodger, Mance and Dyson all refer to the Gold article. Of the three, only Lord Dyson demonstrates any real engagement with its central points: see above n 1, at [217], [218], [221] (SC).

224 Claire McIvor also the credibility of the person who carried out the study. For instance, evidence submitted to the court indicating that the author of the study had previously been exposed as having falsified data would significantly reduce the court’s confidence in the reliability of the result of 64%. As regards the quality of the study itself, in determining belief probability the court would look for information about the quality of the study’s design and the robustness of its execution. For this purpose, the use of appropriately qualified expert witnesses is essential. Where the fact probability is presented as a piece of epidemiological evidence, the relevant expert witness should be an epidemiologist. When assessing the scientific robustness of a piece of epidemiological evidence, an epidemiologist would look at factors such as the size of the study, the manner in which the data was collected, the way in which potential sources of bias were accounted for, the control of relevant variables and adjustments for confounders and effect modifiers. As an expert witness, the epidemiologist would be able to provide valuable information to the court about the pertinence of the study’s results to the individual circumstances of the claimant. Additional indicators of scientific reliability noted earlier are the p-values and the confidence intervals for the studies in question. While these measurements also speak to the scientific robustness of the study, Barnes advocates placing them under a separate probability category, that of the ‘sampling error probability’. These measures are expressed as numerical values and so dealing with them separately reduces the potential for confusion with the other numerical evidence and helps to clarify the nature of the relationship between general belief probabilities and specific sampling error ones. As discussed already, p-values and confidence intervals measure the degree of variation in the results produced by data against the size of the sample of individuals used for the study. Their function is to measure the likelihood that the results produced by the study are due to the unrepresentativeness of the sample rather than a biological association. Miller explains that medical scientists customarily treat a 5% (p