Criminalization: The Political Morality of the Criminal Law [Hardcover ed.] 019872635X, 9780198726357

Citation preview

C R I M I N A L I Z AT I O N

Criminalization The Political Morality of the Criminal Law Edited by R A DUFF L I N D S AY FA R M E R S E MARSHALL M A S S I M O R E N ZO V I C TO R TA D RO S

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © The several contributors 2014 The moral rights of the authors‌have been asserted First Edition published in 2014 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Crown copyright material is reproduced under Class Licence Number C01P0000148 with the permission of OPSI and the Queen’s Printer for Scotland Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2014945176 ISBN 978–0–19–872635–7 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Acknowledgements This is the fourth volume to emerge from a research project on Criminalization, funded by a grant from the Arts and Humanities Research Council (Grant No 128737). We are grateful to the Arts and Humanities Research Council for the grant that made this project possible, and to our own universities for the further material and administrative support that they provided—the University of Stirling and the Stirling Department of Philosophy, the University of Glasgow and its School of Law, and the University of Warwick and its School of Law and Department of Philosophy. We are very grateful to the authors who have contributed to this volume, and to all of the participants at the conference from which this volume emerged. We are also especially grateful to Nikki Leavitt, Ambrose Lee, and Brian Ho for their invaluable work in helping to organize the conference. Antony Duff, Lindsay Farmer, Sandra Marshall, Massimo Renzo, Victor Tadros

Table of Contents List of Contributors

ix

1. Introduction: Towards a Theory of Criminalization? R A Duff, Lindsay Farmer, S E Marshall, Massimo Renzo, and Victor Tadros

1

2. Quantifying Criminalization James Chalmers and Fiona Leverick 3. Criminal Law as an Institution: Rethinking Theoretical Approaches to Criminalization Lindsay Farmer

54

80

4. Bureaucratic ‘Criminal’ Law: Too Much of a Bad Thing? Jeremy Horder

101

5. Criminalization in Republican Theory Philip Pettit

132

6. Contractarian Criminal Law Theory and Mala Prohibita Offences Susan Dimock

151

7. Liberty’s Constraints on What Should be Made Criminal Michael S Moore

182

8. Polygamy: A Novel Test for a Theory of Criminalization Douglas Husak

213

9. Civil Peace and Criminalization Anthony Bottoms

232

10. Marginality, Ethnicity, and Penality: A Bourdieusian Perspective on Criminalization Loïc Wacquant

270

11. ‘It Isn’t Just About You’: Victims of Crime, their Associated Duties, and Public Wrongs S E Marshall

291

Index

307

List of Contributors Anthony Bottoms is Emeritus Wolfson Professor of Criminology at the University of Cambridge and Honorary Professor of Criminology at the University of Sheffield. He is a Fellow of the British Academy. James Chalmers is Regius Professor of Law at the University of Glasgow. Susan Dimock is Professor of Philosophy at York University in Toronto Canada. R A Duff is a Professor Emeritus in the Philosophy Department at the University of Stirling, and holds the Russell and Elizabeth Bennett Chair at the University of Minnesota Law School. Lindsay Farmer is Professor of Law at the University of Glasgow. Jeremy Horder is Professor of Criminal Law at the London School of Economics and Political Science, and an Emeritus Fellow of Worcester College, Oxford. He was a Law Commissioner for England and Wales from 2005 to 2010. Douglas Husak is Professor of Philosophy and Law at Rutgers University and Editor-inChief of Criminal Law and Philosophy. Fiona Leverick is Professor of Criminal Law and Criminal Justice at the University of Glasgow. S E Marshall is a Professor Emeritus in the Department of Philosophy at the University of Stirling and a Visiting Research Scholar at the University of Minnesota Law School. Michael S Moore is currently the holder of the only University-wide chair at the University of Illinois, the Walgreen Chair; he is also Professor of Law, Professor of Philosophy, and Professor in the Center for Advanced Study at the Urbana-Champaign campus of that University. His most recent book is Causation and Responsibility: An Essay in Law, Morals, and Metaphysics (OUP, 2009; paperback edition, 2010; Spanish edition, 2011). Philip Pettit is L. S. Rockefeller University Professor of Politics and Human Values at Princeton University and Distinguished Professor of Philosophy at the Australian National University. Among his recent books are On the People’s Terms: A Republican Theory and Model of Democracy (CUP, 2012) and Just Freedom: A Moral Compass for a Complex World (Nortons, 2014). Massimo Renzo is an Associate Professor in the Department of Philosophy at the University of Warwick. Victor Tadros is Professor of Criminal Law and Legal Theory at the University of Warwick. Loïc Wacquant is Professor of Sociology at the University of California, Berkeley, and researcher at the Centre Européen de Sociologie et de Science Politique, Paris. His work deals with urban marginality, penality, carnality, and social theory. His books have been translated in two dozen languages and include Body and Soul: Notebooks of An Apprentice Boxer (2004, new expanded edition, 2014), The Two Faces of the Ghetto (2015), and Tracking the Penal State (2015).

1 Introduction Towards a Theory of Criminalization? R. A. Duff, Lindsay Farmer, S. E. Marshall, Massimo Renzo, and Victor Tadros

This is the fourth volume of papers arising from an AHRC funded project on Criminalization conducted by the five editors. The three previous volumes contained papers from three pairs of workshops held during the project’s first three years;1 most of the chapters in this volume were originally delivered as plenary papers at the project’s final conference in 2011;2 three monographs by members of the project team will complete the mini-series that the project has produced.3 We described the project’s overall aims in our Introduction to the first volume of papers. The Introduction to this fourth and final volume of papers gives us the opportunity to discuss what the project has and has not achieved, some of the ways in which debate about the issues of criminalization has developed during the last five years, and what we take to be the most promising lines of further enquiry.

I.  Where We Started—and Why We Didn’t Get Where We Intended ‘Our first objective’, we said, ‘is to develop a normative theory of criminalization: an account of the principles and values that should guide decisions about what to criminalize and about how to define offences’:4 this reflected, no doubt, the confidence that we felt after finishing our previous project on the criminal trial, 1  The Boundaries of the Criminal Law (2010), The Structures of the Criminal Law (2011), The Constitution of the Criminal Law (2013), all published by Oxford University Press. 2  The papers by James Chalmers and Fiona Leverick, by Lindsay Farmer, by Sandra Marshall, and by Loic Wacquant were not delivered at the conference; 54 papers were delivered and discussed at the conference. 3  R. A. Duff, The Realm of Criminal Law; Lindsay Farmer, The Institution of Criminal Law; Victor Tadros, Wrongs and Crimes; see Section VII. 4  From the ‘Objectives’ section of our original application to the Arts and Humanities Research Council (Grant No. AH/E007821/1).

2

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

when we thought that we had, as we had intended, developed at least the outlines of a normative theory of the criminal trial.5 Fortunately, we also spoke more cautiously about ‘working towards an overarching, theoretically informed, normative perspective’ on ‘the proper scope and structure of criminal law’;6 whilst we have not succeeded in our ‘first’, more ambitious, objective, we think that the project has achieved that second, more modest objective. A variety of reasons help to explain our failure to achieve that first objective—a failure that is also itself instructive and, we believe, productive. We will comment in more detail on some of the reasons in the following sections, but they are worth summarizing here. A simple reason for not producing even the outlines of a theory of criminalization is that it soon became clear that we would not be able to agree among ourselves on the structure or the content of such a theory. There is, as this Introduction should make clear, much on which we agreed or came to agree—in particular about what any such theory would need to include, about the questions that need to be asked and answered, and about the kinds of approach that are likely (or unlikely) to prove fruitful. But on matters of substantive content we found that our views were, and remained, too divergent to permit agreement even on a sketch of a theory of criminalization. That is hardly surprising—and in itself hardly of great interest: it is worth noting, however, that such disagreements have helped, rather than hindered, the more realistic (and still ambitious) project of developing a richer and more adequate overarching theoretical perspective on the problems of criminalization. A second, and theoretically more significant, obstacle to developing ‘a normative theory of criminalization’ lies in the very complexity of the phenomena—of the decisions, practices, and institutions—to be theorized. Theorists are prone to talk (as our talk of ‘the principles and values that should guide decisions about what to criminalize’ implied) as if criminalization is a matter of decision by an authoritative legal body—presumably a legislature. From this perspective, to criminalize a particular type of conduct is to pass a statute defining that conduct as criminal, and attaching a sanction to it. A normative theory of criminalization would then be a theory of the principles, values, and aims that should guide legislatures in making such decisions.7 To understand criminalization in this way is, however, too narrow. For one thing, to talk in such terms is to ignore the role still played in some legal systems by a non-statutory common law—or perhaps to wish it away, on the grounds that a rational, principled system of criminal law must be a codified system. It is also to ignore the role played by the courts and judges who have to interpret a criminal code or statute in deciding cases: whether we see the Herculean task that 5  R. A. Duff, L. Farmer, S. E. Marshall, and V. Tadros, The Trial on Trial (3): Towards a Normative Theory of the Criminal Trial (Oxford: Hart Publishing, 2007). 6  In the ‘Case for Support’ in our original application, p. 2. 7  See e.g. A. P. Simester and A. von Hirsch, Crimes, Harms and Wrongs (Oxford: Hart Publishing, 2011), 3, and 6 on ‘the act of criminalization’, which is clearly a legislative act.

Introduction

3

faces judges in hard cases as that of seeking ‘the right answer’, or as that of trying to arrive at the best, or at least an acceptable, decision, they play a crucial role in determining which kinds of conduct are actually treated as criminal by the legal system; so surely a theory of criminalization should have something to say to them, about the principles, values, and aims in the light of which they should approach their task.8 More importantly, though, to understand criminalization in this way is to overlook the role of enforcement of the law. As has been pointed out, from Oliver Wendell Holmes’s account of the ‘bad man’ onwards,9 the meaning of the law is less a matter of the law in the books than of the law in action. Criminal laws might be passed but not enforced; their enforcement and use will depend on interpretation by officials on the street; and so the content or meaning of the law, or at least its de facto effect on the lives of individuals, will ultimately depend on how it is interpreted by a range of enforcement officials—from police officers to prosecutors to those such as compliance officers or tax inspectors. From this perspective the question of who is criminalized, and how and for what, cannot be seen as a matter simply of legislative decision: it is a complex process, or set of processes, through which certain kinds of conduct come to be formally defined, to be understood (by officials and citizens), and to be treated (especially by various kinds of official) as criminal. A normative theory of criminalization will need to discuss the nature and structure, the proper operations and outcomes, of these processes. A further complication, making the task of theory development yet more demanding, is that we cannot treat the criminal law as an isolated institution or practice. The criminal law is, after all, just one amongst many kinds of law; its institutions and practices function within the overall institutional structure of the law. A theory of criminalization must therefore have something to say about the ­relationships, and the proper differences in function and in scope, between the criminal law and other modes of legal regulation. We will return to this point in Section III, but it is worth indicating here three ways in which this point raises complex issues for a theory of criminalization. First, it is common to draw a contrast between ‘real’ criminal offences and ­‘regulatory’ criminal offences.10 There is more than one contrast here, and the different contrasts are sometimes conflated. One contrast is between offences that crimina­ lize conduct that is wrong independently of its being regulated and offences that criminalize conduct that is wrong only in virtue of its being regulated (the contrast between crimes that are mala in se and crimes that are mala prohibita). Another is the contrast between offences that are deemed punishable and those to which mere penalties are attached (and there is more than one way to understand the 8  We should also note that this perspective implies a somewhat dated view of the law-making process. First, it ignores the impact of membership of international and transnational bodies on lawmaking, given which the source of law might not always be a sovereign national parliament. Second, it overlooks the role of secondary or delegated legislation as a source of law. Both of these points are discussed further in Section II. 9  Oliver Wendell Holmes, ‘The Path of the Law’, Harvard Law Review, 10 (1897), 457, at 459. 10  See Horder, and Chalmers and Leverick, in this volume, and Section II in this chapter, text at nn. 29–31.

4

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

distinction between punishments and penalties). A crucial question for any theory of criminalization is whether the latter range of offences really belong within the criminal law: or should they be formally separated off, into a distinct realm of noncriminal ‘regulatory’ or ‘administrative’ violations—as in the German system of Ordnungswidrigkeiten?11 Second, victimizing crimes are also very often torts: the perpetrator could, in principle if not often in practice, face not only criminal prosecution, but a civil case brought by the victim (or the victim’s relatives or dependants). So how should we understand the respective aims of these two kinds of legal process—if we do have good reason to maintain them both? If tort law already provides victims with legal recourse against those who harmed them, why should we also criminalize the harmful conduct? If only some kinds of tort should also be criminal, which should be—and why? Third, practices of ‘preventive justice’, which are increasingly common and significant as governments look for effective ways of preventing various kinds of harm or annoyance, also challenge theorists of criminal law, since they seem to blur the boundaries between criminal law and other modes of legal regulation and control.12 This is especially true of measures that impose constraints on the conduct of individuals judged to be dangerous, or likely to offend, and criminalize any violation of those constraints: anti-social behaviour orders and control orders13 are the two most familiar examples of this phenomenon; such measures restrict liberty through civil orders, the breach of which is sanctioned by the criminal law. We must ask both whether it is appropriate to use the criminal law in this way; and whether these kinds of restrictive order are an appropriate way of addressing kinds of harmful or wrongful conduct—rather than simply leaving it to the criminal law to deal with such conduct once it is committed. The larger question that these points raise is whether there is something distinctive about the character or function of criminal law. Is the criminal law just another mode of regulation? If it is not, as is implied by the suggestion that we can distinguish ‘proper’ criminal law from regulatory offences, then what are the distinctive features or characteristics of the criminal law? This also then raises the question of when it is appropriate to resort to the criminal law. Should it be used as a matter of first, or of last, resort; or are there other considerations that might come into play? The task of building a theory of criminalization is already, one might think, challenging enough given the complex range of institutional practices and processes that it must capture and the need to explicate the criminal law’s proper place

11  Gesetzüber Ordnungswidrigkeiten (1968; consolidated in 1975); for a useful introductory (and critical) discussion, see T. Weigend, ‘The Legal and Practical Problems Posed by the Difference between Criminal Law and Administrative Penal Law’, Revue Internationale de Droit Pénal, 59 (1988), 67. 12 See especially A. J. Ashworth and L. Zedner, ‘Preventive Orders: A Problem of UnderCriminalization?’, in The Boundaries of the Criminal Law, and Section V in this chapter. 13  See further (and on the new provisions that are replacing anti-social behaviour orders and control orders) at nn. 137–41.

Introduction

5

among the range of other kinds of legal institution that make up the legal system as a whole. But three further challenges should be noted here. First, although philosophical theorists of criminal law have often discussed the problems of punishment, of criminal liability, and of criminalization as if they are essentially problems in moral philosophy, they are also problems in political philosophy. The criminal law is not an institutional embodiment of the moral law, addressing and binding us as moral agents; it is a political institution, part of the state, and must address us as citizens—members of the polity whose law it is. A theory of criminalization must therefore include or depend on a political theory of state and society: it must be a theory of the role that criminal law should play within a particular kind of polity. Ambitiously universalist theorists might hope to offer theories of criminal law that will apply to every kind of polity—theories of what criminal law should be and mean in any and every kind of society, whatever its political structure. Given the criminal law’s essentially political structure and foundations, however, it is arguable that any such universal theory of the criminal law would have to be grounded in a universal political theory of society and the state: failing such a political theory, we can only ask what kind of criminal law, serving what kinds of aim and structured by what principles, would be appropriate for this or that more particular kind of polity. Second, and relatedly, the criminal law has a history; more precisely, every contemporary system of criminal law has a long and complex history—and different systems have different histories. What kind of attention should would-be normative theorists pay to those histories? Should they recognize that what they are theorizing is and cannot but be a particular system, with a history from which it cannot be detached? Or can they hope, to at least some degree, to transcend that history? To the extent that normative theorizing must attend to, or depends on a grasp of, the contingent, historical actualities of existing systems of criminal law, it must also therefore draw not only on the resources of legal theory and of normative philosophy (moral and political), but also on the disciplines that deal with those actualities: on history, on criminology, on sociology, on political science. The extent and character of such attention to historical contingencies is of course a matter of continuing debate (and one of the matters on which the editors have rather different views): is it, modestly, a matter of seeing how such historical contingencies make a difference to the practical application and implications of a set of ahistorical, non-contingent, normative principles; or is it, more radically, a matter of grounding the normative principles themselves in particular historical settings? On either kind of view, however, a substantive normative theory of criminalization must attend to the particular histories of the systems of criminal law that are to be theorized. Third, further questions need to be addressed about the shape and structure, as well as the starting points, of a theory of criminalization. Such a theory must, we have noted, include, or be able to draw on, a larger political theory of the state; we have also noted that any aspiration to an ahistorical universality is at least controversial. But whether the theory is to be universal or not, we must also ask how neat and systematic we should expect it to be. It is tempting, in this as in other

6

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

contexts, to search for a grand unifying theory: an account of the proper aims of the criminal law (which must be part of any normative theory of its scope), of its proper scope and limits, and of how decisions about criminalization should be made, which appeals if not to a single master principle from which all else can be derived, then at least to a coherent set of principles which either do not conflict or include meta-principles by which any conflict can be resolved and from which we can derive specific and substantive conclusions about what should or should not be criminal. Prominent examples of this type of approach are the republican theory of criminalization defended by John Braithwaite and Philip Pettit,14 the legal moralist theory defended by Michael Moore,15 and the view recently proposed by Larry Alexander and Kimberley Ferzan, according to which the only purpose of the criminal law is to prevent the imposition of unjustifiable risks on legally protected interests.16 Perhaps, however, the grail of grand theory is illusory: all that theorists should hope for or aspire to is a much messier, more piecemeal account that can do justice, as grand theories could not, to moral and/or social complexity. In the following sections we will explore some of these challenges to the construction of a theory of criminalization in more detail, and use them as a framework in which to discuss some of the main developments in public and scholarly debate since 2007 (when we applied for our grant). We hope in this way, if not to sketch or even gesture towards a theory of criminalization, at least to indicate the directions in which future work can fruitfully proceed, the issues that need to be addressed, and some of the ways in which they can be addressed. This will contribute to our aim of helping to develop an ‘overarching, theoretically informed, normative perspective’ on criminalization: that perspective will need to be rather different from the more limited perspectives suggested by current debates, which too often take the basic divisions to be those between consequentialist and non-consequentialist theories, or between versions of legal moralism and harm-based theories.

II.  The Complexity of the Phenomena We have already noted the distorting tendency among theorists to see criminalization as a matter essentially of legislation. One effect of this is that the perceived problem of ‘over-criminalization’ is then understood as essentially a problem of legislative overactivity: that politicians are too prone to pass a new criminal law when faced with a perceived social mischief or problem; that criminal provisions are too routinely tacked onto pieces of legislation without proper scrutiny of the appropriateness of such uses of the criminal law or of their consistency with existing norms 14  J. Braithwaite and P. Pettit, Not Just Deserts: A Republican Theory of Criminal Justice (Oxford: Oxford University Press, 1990). See also Pettit’s chapter in this volume. 15  M. S. Moore, Placing Blame: A General Theory of the Criminal Law (Oxford: Oxford University Press, 1997). See also Moore’s chapter in this volume. 16  L. Alexander and K. Ferzan (with contributions by S. J. Morse), Crime and Culpability: A Theory of Criminal Law (Cambridge: Cambridge University Press, 2009).

Introduction

7

and principles. This understanding of over-criminalization explains the salience of claims such as the one which has acquired a certain currency in Britain: that the New Labour government created over 3,000 new offences between 1997 and 2006.17 There is indeed clear evidence of ‘penal populism’ in many Western countries: politicians are too ready to seek electoral popularity by demanding more punitive sentences, building more prisons, appeasing victims’ movements, passing new laws in response to moral panics, and so on.18 It is not clear, however, that such penal populism is closely related to over-criminalization in the narrow sense of excessive legislation. One difficulty with such claims about the increasing number of criminal laws or offences is that they are rarely accurate—and one reason for this is that quantifying criminal laws or offences is far from easy. As Chalmers and Leverick show, identifying and individuating crimes is fraught with difficulties.19 The question of whether a particular piece of legislation has created several offences or only one might be a matter of drafting style or might be down to the judgment of the researcher—and in either case, while this might lead to a high or a low headline figure for the number of ‘new’ crimes, it might be better evidence of the level and forms of parliamentary activity than of the actual criminalization of conduct. A second problem, identified both by Chalmers and Leverick and by Horder,20 is that it is hard to know what counts as legislation. While many offences are created by primary legislation, a great number are created as adjuncts to other statutes regulating a wide range of activities, by statutory instruments, by local authorities’ by-laws, and by regulatory bodies. This might lead to confusion for citizens—and perhaps point to the need for a requirement for legislators clearly to identify criminal legislation and to explain the reasons for it;21 but this again points to concerns with the drafting and enactment of legislation rather than with criminalization as such. Even if we can satisfactorily resolve these kinds of issues of quantification—and Chalmers and Leverick have come closest to developing a satisfactory measure—there is no clear standard against which we can measure what would count as too many criminal laws or criminal offences; to which it might be replied that we do not need to know the appropriate amount of criminalization to know when there is too much.22 Another way of framing the claim about over-criminalization is to argue that we have too much of the wrong sort of criminal law. Theorists may distinguish the 17  N. Morris, ‘Blair’s Frenzied Law Making: A New Offence for Every Day Spent in Office’, The Independent, 16 August 2006, ; discussed by Chalmers and Leverick (in this volume) pp. 58–9. 18  See J. V. Roberts, Penal Populism and Public Opinion: Lessons from Five Countries (Oxford: Oxford University Press, 2003). See also J. Simon, Governing through Crime (Oxford: Oxford University Press, 2007). 19  Chalmers and Leverick, in this volume, pp. 62–3. 20  Both in this volume. See also Law Commission, Criminal Law in Regulatory Contexts (Law Com. CP No. 195; London: HMSO, 2010). 21  See e.g. Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), ch. 1 suggesting that there is a need for reasons to be given. 22  See also D. N. Husak, Overcriminalization: The Limits of the Criminal Law (Oxford: Oxford University Press, 2007), ch. 1.

8

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

‘core’ of criminal law from criminal laws lying outside the core.23 It is natural then to see these ‘outside the core’ crimes as ‘peripheral’—as not being part of the ‘real’, legitimate, criminal law. It is hard to know what to make of such contrasts between the core and the periphery without being given a clearer idea of how the core is to be identified. If it is taken, for instance, to consist in that set of familiar ‘mala in se’ which is salient in so much theoretical writing about criminal law, the bulk of existing criminal law falls outside the ‘core’: but a normative theory of criminal law cannot simply exclude so much of existing criminal law by definitional fiat. Two relevant suggestions are, first, that crimes ‘outside the core’ are those that do not (obviously) involve the kind of pre-legal moral wrongfulness that might be said to characterize ‘core’ crimes—which implies that criminal law, properly speaking, is concerned only with pre-legal wrongs; and second, that crimes ‘outside the core’ are those that do not respect some supposedly general principles of criminal law—for instance that they are offences of strict liability,24 or impose reverse burdens of proof. These suggested criteria for identifying the core of criminal law might then be treated as normative criteria of legitimate criminalization: statutes should criminalize conduct only if it is (pre-legally) wrongful, and should conform to such general principles as ‘actus non facit reum nisi mens sit rea’ and the presumption of innocence. This kind of approach, which has received some support recently both from courts and from academics,25 might not by itself provide much of a brake on the overenthusiastic legislation of criminal offences, but might at least help to ensure that criminal statutes meet minimal standards of justice. However, these principles can become problematic when we consider the vast array of ‘regulatory’ offences that constitute a large proportion of ‘crimes outside the core’.26 One problem is that there is no close relationship between the mala in se/mala prohibita distinction and the serious/non-serious distinction. Some offences might be regulatory in the sense that the wrongness of the conduct prohibited depends on the existence of a regulation governing it, but this need not imply that the conduct is not very seriously wrongful; the fact that the conduct’s wrongfulness is contingent upon its being regulated has no implications for its seriousness.27 Even theft and criminal damage might be argued to be regulatory in this sense, in virtue of the fact that the wrongness of the conduct, at least in some instances, 23  See e.g. W. Stuntz, ‘The Pathological Politics of Criminal Law’, Michigan Law Review, 100 (2001), 506; for a more critical discussion of the distinction between ‘core’ and ‘periphery’, see D. N. Husak, ‘Crimes Outside the Core’, Tulsa Law Review, 39 (2004), 755. 24  This connects with the first suggestion insofar as it is supposed that the criminal law is concerned with culpable wrongdoing, and that strict liability offences permit conviction without proof of fault. 25  See e.g. Clingham (formerly C (a minor)) v Royal Borough of Kensington & Chelsea, R v Manchester Crown Court ex parte McCann [2002] UKHL 39; [2003] 1 AC 787; for academic support, see e.g. A. J. Ashworth, ‘Is the Criminal Law a Lost Cause?’, Law Quarterly Review, 116 (2000), 225, and A. J. Ashworth and J. Horder, Principles of Criminal Law (7th edn.; Oxford: Oxford University Press, 2013), chs 2–3; and Husak, Overcriminalization (n. 22), ch. 2, on ‘internal constraints’. 26 But not all of them: see Husak, Overcriminalization (n. 22), 36–45, on ‘overlapping’ and ‘ancillary’ offences, and offences of ‘risk creation’; see also the Introduction to The Boundaries of the Criminal Law (n. 1), 3–5. 27  See e.g. Bottoms’ discussion of the criminalization of drink driving in this volume.

Introduction

9

depends on property law. This fact hardly makes theft and criminal damage either non-serious or ‘outside the core’ of criminal law. So if there is a proper distinction to be drawn between ‘real’ and ‘regulatory’ criminal law, it is to be found elsewhere, perhaps in the form of the regulation, the penalties that attach to it, and the conditions of liability that are specified. Furthermore, worries about overcriminalization are not restricted to regulatory offences. For example, lying is wrong independently of its being criminalized. Very broad dishonesty offences, such as the offence of fraud in English criminal law,28 are nevertheless problematic. The fact that lying is typically wrong in itself, and not in virtue of state regulation, does not make its criminalization unproblematic. Lying might typically not involve wrongfulness that is serious enough, or of the right kind, to warrant its criminalization. Despite such difficulties, it might still be claimed that some offences, in particular some kinds of regulatory offence, are not ‘really’ criminal offences: they are ‘quasi-criminal’, since conviction for them does not involve the kind of stigma that conviction for a ‘real’ criminal offence brings; and for that very reason we need not be so concerned about the requirements of mens rea and burdens of proof that are appropriate for ‘real’ criminal offences.29 Theorists who want to preserve the principled purity of the criminal law are likely to argue in response that such offences have no place in the criminal law: we might be able to justify them as part of a separate, non-criminal, system of ‘administrative’ or ‘regulatory’ law, which prohibits and penalizes conduct but does not criminalize it;30 but any criminal law must conform to the wrongfulness requirement and to the general principles of criminal liability. This, they might argue, is where the problem of over-criminalization is most acute: the problem is not that we have too much criminal law in the ‘core’, but that we have too much criminal law outside the core—too many mala prohibita, too much regulatory criminal law; the criminal law’s reach is overextended, into areas where its use seems inappropriate.31 One problem with this line of argument is that, as Horder forcefully argues in his chapter in this volume, the line between ‘regulatory’ and ‘proper’ criminal offences 28  Fraud Act 2006, s. 1; for discussion, see D. Ormerod, ‘Criminalising Lying’, Criminal Law Review (2007), 193. 29 See e.g. the judicial dicta cited in A. P. Simester, ‘Is Strict Liability Always Wrong?’, in A. P. Simester (ed.), Appraising Strict Liability (Oxford: Oxford University Press, 2005), 21, at 23–4; but Simester adds some appropriately cautionary remarks about the relationships between the ideas of ‘quasi-criminal’ laws, of ‘regulatory’ offences, and of ‘mala prohibita’. 30  See J. Feinberg, ‘The Expressive Function of Punishment’, in Doing and Deserving (Princeton: Princeton University Press, 1970), 95, at 96–8, on the difference between ‘punishments’ (which have a ‘reprobative’ meaning) and ‘penalties’ (which do not). But see Weigend, ‘The Legal and Practical Problems’ (n. 11); Duff et al., The Trial on Trial (3) (n. 5), 189–98; V. Tadros, ‘Criminalization and Regulation’, in The Boundaries of the Criminal Law (n. 1), 163. 31  This claim taps into a broader argument about an over-regulation by the modern state which is at least implicit in many recent analyses of criminalization. See e.g. J. Habermas, The Theory of Communicative Action (Boston: Beacon Press, 1985), ii. 357–73 on the colonization of the lifeworld; G. Teubner, ‘Juridification: Concepts, Aspects, Limits, Solutions’, in G. Teubner (ed.), Juridifcation of Social Spheres (Berlin: de Gruyter, 1987). See also S. Veitch et al., Jurisprudence: Themes and Concepts (2nd edn.; London: Routledge, 2012), 255–64.

10

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

has become blurred, so that it is hard to draw any sharp distinction between them. A more substantive objection is found both in Horder’s chapter and in Chalmers and Leverick’s: that whilst it is easy to object that there are too many regulatory offences, a closer and deeper examination suggests that many might be justifiable. We must at least be open to the possibility that different values or goals are properly relevant in different contexts of criminalization; that principles concerning moral culpability, and concerns with efficiency, might weigh differently in different contexts; and thus that not every legitimate criminal offence need involve the requirements of fault that characterize offences in the ‘core’. Dimock, in her chapter in this volume, offers a contractarian rationale for a range of so-called mala prohibita, focused especially on market offences and offences against the state:32 this is the kind of discussion in which theorists of criminalization need to engage more thoroughly than they often do, to tackle difficult questions about the precise form, the substantive content, and the normative grounding of ‘regulatory’ offences. Such discussion will also, as Dimock’s chapter makes clear, involve important questions in political theory about the nature and aims of the state, about the scope of liberty, and about the legitimate grounds and forms of state coercion. We touch on these in Section III. A further problem with the focus on legislation, which will turn our attention to other key players in the processes of criminalization, is that when a legislature passes a statute defining certain types of conduct as criminal, it might not intend that every instance of conduct fitting that definition should actually be treated as criminal—that all such conduct should, ideally, be detected, prosecuted, and punished.33 This quite common phenomenon is sometimes made helpfully explicit. When what became the Sexual Offences Act 2003 was passing through the House of Commons, concern was expressed that ss. 9 and 14 would criminalize any kind of sexual activity (however consensual) between two young people aged 15. Paul Goggins, a Home Office Minister, assured the House: That is not the intention of the Bill; nor will it be its effect in practice. . . . There have . . . been no prosecutions simply for kissing; nor will there be in future. [I]‌f we find no other way to deal with this question, . . . we shall be able to trust the Crown Prosecution Service to ensure that that intention is followed.34

Legislatures often pass statutes that they know to be, if read literally, too broad, relying upon police or prosecutorial discretion to ensure that only kinds of conduct involving the mischief at which the statute is ‘really’ aimed are prosecuted: they delegate the task of criminalization, the task of deciding which kinds of conduct are actually to be treated as criminal, to others. A normative theory of criminalization 32  See also S. P. Green, ‘Why it’s a Crime to Tear the Tag off a Mattress: Over-Criminalization and the Moral Content of Regulatory Offenses’, Emory Law Journal, 46 (1997), 1533; R. A. Duff, Answering for Crime: Responsibility and Liability in the Criminal Law (Oxford: Hart Publishing, 2007), 89–94, 166–74. 33  We can leave aside here some familiar questions about what it means to say that a legislature intends this or that in passing a statute, or how such intentions can be discerned. 34  Hansard vol. 409, 15 July 2003, col. 248.

Introduction

11

will need to have something to say about such legislative tactics. Could they and should they be avoided, for instance through more careful legislative drafting and a principle of compulsory prosecution; or must such quasi-legislative discretion inevitably be delegated to police and prosecutors? If it is inevitable, criminalization is already shown to be a much more complex process than statutory legislation: we must then ask both about the conditions under which, and the considerations in the light of which, a legislature can properly pass what it believes to be (on its face) an overly broad statute, and about the considerations that should then guide other officials, especially police and prosecutors, in the exercise of the quasi-legislative power that is thus delegated to them.35 More generally, and even apart from such delegation of quasi-legislative power, police and prosecutors (as well as courts) play important roles in determining what kinds of conduct are in practice defined and treated as criminal. The police must decide where to focus their investigative resources—which kinds of case to investigate as (actual or potential) crimes. Whether this is a matter of developing policies or can be done on a more ad hoc, case-by-case basis, they must decide which cases to investigate further and to pass on to prosecutors, and which to ignore, or to deal with by formal or informal diversionary processes. Sometimes policing policies might amount to the de facto decriminalization of types of conduct that the law formally defines as criminal: if a police force institutes a policy of not prosecuting or reporting those found in possession of small amounts of a prohibited drug, it might be argued that such possession has, within that force’s area, been decriminalized.36 Prosecutors must also decide which cases to prosecute: in England and Wales, this requires them to decide not only whether there is sufficient evidence to ground a realistic prospect of conviction, but also whether prosecution would be ‘in the public interest’.37 Here too, prosecutorial policies might bring about the de facto decriminalization of conduct that is formally defined as criminal. For example, in 2010 the English Director of Public Prosecutions issued a formal Policy specifying the factors that would guide decisions about the prosecution of those who assist others’ suicides, in particular by helping them to travel to the 35  Compare Stuntz, ‘The Pathological Politics of Criminal Law’ (n. 23), on the uneasy equilibrium between over-criminalization and under-enforcement in contemporary criminal law. 36 See e.g. , and , on such policies in the UK; also the Obama administration’s instruction on federal prosecution of possession of medical marijuana: see . But we should bear in mind that decriminalization is not the same as legalization: if, for instance, those small amounts of the prohibited drug are liable to be confiscated, such possession has been de facto decriminalized, but not legalized. 37 See Code for Crown Prosecutors (), s. 4; A. J. Ashworth and M. Redmayne, The Criminal Process (4th edn.; Oxford: Oxford University Press, 2010), 199–206; also S. Moody and J. Tombs, Prosecution in the Public Interest (Edinburgh: Scottish Academic Press, 1982). For further discussion see J. Rogers, ‘The Role of the Public Prosecutor in Applying and Developing the Substantive Criminal Law’, in The Constitution of the Criminal Law (n. 1), 53. They must also, of course, decide just what charges to pursue—a decision of particular significance in systems that allow widespread plea bargaining and ‘charge stacking’: see Husak, Overcriminalization (n. 22), 22–3.

12

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

Dignitas clinic in Switzerland:38 the document specifies sixteen ‘public interest factors tending in favour of prosecution’ (para. 43), and six ‘tending against prosecution’ (para. 45). It declares that ‘only Parliament can change the law on encouraging or assisting suicide’, and that ‘[t]‌his policy does not in any way “decriminalise” . . . assisting suicide’, or give an ‘assurance that a person will be immune from prosecution if he or she . . . assists’ another’s suicide (paras 5–6). However, the court’s reasons for requiring the DPP to issue these guidelines had to do with the need to enable those whose contemplated conduct would fall within the statutory definition of assisting suicide to predict whether they would face prosecution, and plan their actions accordingly.39 It is therefore hard not to read the document as, in effect, assuring anyone to whose case most or all of the factors ‘tending against prosecution’ (and none of those ‘tending in favour of prosecution’) apply that they will not face prosecution; indeed, it is at least arguable that if such a person was nonetheless prosecuted, the court should dismiss the case as an abuse of process. It is therefore tempting to say that, although such conduct is still formally criminal as a matter of statute, and although there is no evidence of a legislative intent or desire that it should not be treated as criminal, it has now been in effect decriminalized by the DPP. A theory of criminalization must have something to say about the roles of such officials in the practice of criminalization: what kind of power should they have to determine the effective scope of the law, and what kinds of consideration should guide their exercise of that power?40 The case of assisted suicide is perhaps more complicated than this, in ways that also bear on the project of developing a theory of criminalization, since we need to ask why the factors that the DPP specifies as tending against prosecution should do so. If the answer is that when these factors obtain, the assister’s conduct is not (sufficiently) wrongful to merit conviction as criminal, or does not perpetrate the kind of mischief against which the statute can be taken to be aimed,41 the policy amounts to a de facto decriminalization: it makes clear to citizens that such conduct, although it fits the law’s formal definition of a crime, is not a kind of conduct from which they ought to desist as a criminal wrong. We might instead, however, 38  Policy for Prosecutors in Respect of Cases of Encouraging or Assisting Suicide, issued by the CPS in 2010 (): the DPP was required to issue these guidelines by the court in R (ex parte Purdy) v DPP [2009] 3 WLR 403, [2009] UKHL 45. 39  See e.g. Purdy (n. 38), paras 40–3 (Lord Hope), 84–6 (Lord Brown), and 96 (Lord Neuberger). The court was concerned with the requirements of ‘accessibility and foreseeability’ implicit in art. 8(2) of the European Convention on Human Rights: could a person know ‘what acts and omissions will make him criminally liable’, and foresee ‘the consequences which a given action may entail’ (para. 40; Lord Hope)? 40  See also C. Steiker, ‘Criminalization and the Criminal Process: Prudential Mercy as a Limit on Penal Sanctions in an Era of Mass Incarceration’, in The Boundaries of the Criminal Law (n. 1), 27, arguing that at various stages in the criminal process officials should have, and use, a greater discretionary power to exercise mercy, to mitigate the systemic tendency to over-punish. 41  Compare Model Penal Code § 2.1, on ‘De Minimis Infractions’: one ground for dismissing a prosecution is that ‘the defendant’s conduct . . . did not actually cause or threaten the harm or evil sought to be prevented by the law defining the offense’; see D. Husak, ‘The De Minimis “Defence” to Criminal Liability’, in R. A. Duff and S. P. Green (eds.), Philosophical Foundations of Criminal Law (Oxford: Oxford University Press, 2010), 328.

Introduction

13

read the policy as implying that when these factors are present the assister’s criminal culpability is mitigated to such an extent that, even though the law provides no formal excuse, it should be treated as excused - as not meriting conviction and punishment. If we also agree that while justificatory defences are addressed ex ante to citizens as guides for their conduct, excuses and excusatory mitigations are relevant only ex post, and do not provide ex ante guidance on which citizens may rely,42 we might read the policy as declaring that, whilst assisting suicide is still a criminal wrong even when the factors tending against prosecution are present, it is one for which citizens can expect not to be held criminally liable when those factors obtain.43 We do not intend to try to settle this question here:44 but it highlights a further complexity in the phenomena of criminalization, to do with the familiar distinction between two faces of the criminal law. On the one hand, it speaks to the citizens ex ante, specifying the kinds of conduct that are to count as public wrongs, and from which they ought to refrain. From that perspective, to criminalize some type of conduct is to define it as such a not-to-be-committed wrong, whilst to issue guidelines of the kind that the DPP has issued for assisted suicide is to suggest that some conduct which is formally criminal does not constitute such a not-tobe-committed wrong—which is in effect to decriminalize it. On the other hand, the criminal law also speaks ex post to those whose responsibility it is to administer the law, in particular but not only to courts, about the procedures through which, and the conditions under which, those who commit such wrongs ought to be prosecuted and punished. It might be argued that from that perspective, a ‘public interest’ policy of not prosecuting certain types of conduct that fall within the law’s formal definition of an offence need not be understood as decriminalization, since it could instead be read as a policy of exempting from prosecution (for reasons either of compassion or of public policy) conduct that is still to be seen as criminally wrongful. If we reject (as we should) any policy of ‘acoustic separation’,45 42  As is implied by the familiar distinction between ‘rules for citizens’ (among which justifications belong) and ‘rules for courts’ (which include excuse doctrines): see e.g. P. Alldridge, ‘Rules for Courts and Rules for Citizens’, Oxford Journal of Legal Studies, 10 (1990), 487; P. H. Robinson, Structure and Function in Criminal Law (Oxford: Oxford University Press, 1997); J. Gardner, ‘The Gist of Excuses’, Buffalo Criminal Law Review, 1 (1998), 575. 43  It is not entirely clear which of these readings the Law Lords favoured in Purdy (n. 38). Lord Hope insisted (para. 26) ‘that it is no part of our function to change the law in order to decriminalise assisted suicide’, but some of his colleagues’ comments suggest the former view: see e.g. Baroness Hale (para. 59: ‘People need and are entitled to be warned in advance so that, if they are of a law-abiding persuasion, they can behave accordingly’); also Lord Brown (para. 83: ‘I seriously question whether one should always deprecate conduct criminalised by section 2(1)’). 44  It might be clarified, or further complicated, by thinking about the implications of such guidelines for other officials, such as the police. If the guidelines amount to an effective decriminalization of conduct to which the ‘factors tending against prosecution’ apply, it would presumably be inappropriate for a police officer to try to prevent such conduct—to prevent the would-be assister from providing the assistance; but if they serve only to indicate the conditions under which what is still a crime will (for reasons of mercy or public policy) not be prosecuted, the police could still properly intervene to prevent the commission of the crime. 45  See M. Dan-Cohen, ‘Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law’, Harvard Law Review, 97 (1984), 625; for criticism, see R. Singer, ‘On Classism and Dissonance

14

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

and insist that ‘rules for courts’ (and other legal officials) must also be accessible to citizens, such a policy should still be published, so that citizens can predict what kinds of conduct might result in prosecution; but what they are enabled to predict is when they can hope to escape prosecution for committing a crime. What this review suggests is that, while legislation is clearly an important dimension of criminalization, a theoretical approach to criminalization must have a broader focus. In this context, Lacey offers a useful typology of different forms or dimensions of criminalization.46 She distinguishes between criminalization as an outcome and criminalization as a practice—each understood as having both formal and substantive aspects.47 Criminalization as outcome is a matter of what is or should be criminalized, either formally in legislation or through judicial decisions, or substantively in the actual implementation of those formal norms. Criminalization as social practice is a matter of who does the criminalizing (formal or substantive), through what procedures, and according to what principles. As Lacey points out, while both outcome and practice have normative dimensions, normative theorizing has tended to focus on criminalization as outcome, with the practice of criminalization being treated too often as an empirical question for criminology or socio-legal studies. This also highlights the need for an appropriate theoretical language in which to discuss the relations between these areas. This is a challenge which is taken up in a distinctive way by Wacquant, both in his contribution to this volume and his work more generally.48 Drawing on the work of Pierre Bourdieu, Wacquant suggests that developments in crime and punishment must be understood in terms of an ambitious social theory, which can incorporate an analysis of class, race, urban transformation, and the rise of the neoliberal state. On this account, the key concept is penality, and the focus is on how the criminal law and punishment are used by the state to manage the urban poor though a nexus of class and race. He argues that the criminal law is used as a distinct strategy in response to ‘the social insecurity spawned by the precarization of wage labour and to the ethnic anxiety generated by the destabilization of established hierarchies of honour’.49 His account thus challenges the treatment of criminalization and punishment as distinct but related phenomena, the latter a direct response to the former. Instead he argues that penalization should be seen as a response to political and economic development, rather than to crime or insecurity, and consequently that criminalization must be understood in terms of state transformation and strategy for the control or management of sections of society rather than

in the Criminal Law: A Reply to Professor Dan-Cohen’, Journal of Criminal Law and Criminology, 77 (1986), 69. However one reads the decision in Purdy, it clearly rejected such acoustic separation. 46  N. Lacey, ‘Historicising Criminalisation: Conceptual and Empirical Issues’, Modern Law Review, 72 (2009), 936; see also her ‘What Constitutes Criminal Law’, in The Constitution of the Criminal Law (n. 1), 12. 47  Lacey, ‘Historicising Criminalisation’ (n. 46), 942–3. 48  ‘Marginality, Ethnicity, and Penality’, in this volume. See also Punishing the Poor (Durham, NC, and London: Duke University Press, 2009); Deadly Symbiosis: Race and the Rise of the Penal State (Cambridge: Polity, forthcoming 2015). 49  In this volume p. 278.

Introduction

15

as a primarily normative question. This is an important challenge to moral and political philosophical approaches to criminalization, which tend to neglect this institutional or social dimension altogether. One further dimension to the issue of complexity should be noted here. The simple focus on legislation assumes that national parliaments are responsible for criminal legislation. It is clear, however, both that national parliaments are frequently legislating to fulfil international obligations, and that international and transnational bodies are now a significant, independent source of new criminal norms. This means that questions about the relationship between the justification of political legitimacy and the justification of criminalization now play a crucial role not only at the domestic level (a topic to be discussed in Section III), but also in relation to issues of international and transnational criminal law. Indeed, it might be argued that the need for an account of the relationship between legitimacy and criminalization is particularly pressing at the international and at the transnational level, for two reasons. First, the relationship between legitimacy and criminalization has received even less attention at this level than it has at the domestic level. While, for example, there is sophisticated debate about how we should conceptualize,50 or justify,51 domestic political legitimacy, legal and political theorists have only recently begun to explore the idea of international legitimacy (particularly that of international institutions).52 Second, it is at the international level that the relationship between issues of political legitimacy and of criminalization becomes most clear. Providing an account of international and transnational criminal law requires that we provide an account of the distinction between crimes that are the exclusive business of the domestic political community, and crimes that are (also) the business of other states or of international institutions:53 when can domestic

50 See e.g. A. J. Simmons, Moral Principles and Political Obligations (Princeton: Princeton University Press, 1979); J. Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986); L. Green, The Authority of the State (Oxford: Oxford University Press, 1988). 51  e.g. C. H. Wellman and A. J. Simmons, Is There a Duty to Obey the Law? (New York: Cambridge University Press, 2005); G. Klosko, Political Obligations (Oxford: Oxford University Press, 2005); M. Renzo, ‘State Legitimacy and Self-Defence’, Law and Philosophy, 30 (2011), 575, and ‘Associative Responsibilities and Political Obligation’, Philosophical Quarterly, 62 (2012), 106. 52  See e.g. A. Buchanan, Human Rights, Legitimacy, and the Use of Force (Oxford: Oxford University Press, 2010), part 2; T. Christiano, ‘Democratic Legitimacy and International Institutions’, in S. Besson and J. Tasioulas (eds.), The Philosophy of International Law (Oxford: Oxford University Press, 2010), 119; P. N. Pettit, ‘Legitimate International Institutions: A Neo-Republican Perspective’, in Besson and Tasioulas (eds.), The Philosophy of International Law, 139; D. M. Weinstock, ‘Prospects for Transnational Citizenship and Democracy’, Ethics & International Affairs, 15 (2001), 53. 53  For some different kinds of approach to this task, see e.g. L. May, Crimes against Humanity (Cambridge: Cambridge University Press, 2005); A. Altman and C. Wellman, A Liberal Theory of International Justice (Oxford: Oxford University Press, 2009), esp. ch. 4; C. Wellman, ‘Piercing Sovereignty’, in Duff and Green (eds.), Philosophical Foundations of Criminal Law (n. 41), 461; L. May and Z. Hoskins (eds.), International Criminal Law and Philosophy (Cambridge: Cambridge University Press, 2010); D. Luban, ‘Fairness to Rightness: Jurisdiction, Legality, and the Legitimacy of International Criminal Law’, in Besson and Tasioulas (eds.), The Philosophy of International Law, 569; R. A. Duff, ‘Authority and Responsibility in International Criminal Law’, in Besson and Tasioulas (eds.), The Philosophy of International Law, 590; M. Renzo, ‘Responsibility and Answerability in the Criminal Law’, in The Constitution of the Criminal Law (n. 1), 209.

16

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

political institutions justifiably claim exclusive jurisdiction over criminal wrongdoing, and when may other states or international bodies justifiably insist on their right to intervene?54 Two points are worth noticing here. First, given that our understanding of transnational and international criminal law and our understanding of domestic criminal law are so closely interrelated, it is only to be expected that any answer to the question of how transnational and international crimes should be conceptualized will affect to some extent the way in which we understand domestic crimes, possibly leading us to rethink the way in which the structure and the boundaries of domestic criminal law are understood.55 Indeed, one of the most interesting problems raised by the creation of international crimes is that, to the extent that these crimes are increasingly being incorporated into domestic legislations, they end up having a ‘double-layered’ structure: they constitute at the same time municipal criminal offences (insofar as they are part of domestic criminal codes) and international offences (insofar as they are enshrined in international treaties and jus cogens rules).56 How to conceptualize this double-layered structure is a major challenge for scholars working on international and trans­ national criminal law. Secondly, there are obvious limits on how much progress can be made in addressing these issues through general philosophical discussion. Thinking about specific problems, such as the way in which criminal law should deal with war crimes and terrorism, has proved to be a particularly fruitful way to think about how to conceptualize some of the fundamental categories of international and transnational criminal law.57 We will return to complexity in Section IV, and in particular to the complexities introduced by the need to locate criminal law (institutionally and normatively) in the wide array of modes of legal regulation that characterize contemporary states. First, however, we should say more about some of the questions raised by the issue of legitimacy—an issue that bears as much on domestic criminal law as it does on international and transnational law: by what right do the various official actors involved in the making and enforcing of criminal law act as they do?

54  On the significance of issues of jurisdiction for criminalization, see also ‘Symposium on Criminal Jurisdiction: Comparison, History, Theory’, University of Toronto Law Journal, 63 (2) (2013). 55  See e.g. A. A. Haque, ‘International Crime: in Context and in Contrast’, in The Structures of the Criminal Law (n. 1), 106; Renzo, ‘Responsibility and Answerability in the Criminal Law’ (n. 53). 56  See A. Cassese, ‘Remarks on G. Scelle’s Theory of Role-Splitting in International Law’, European Journal of International Law 1 (1990), 210, ‘The Rationale for International Criminal Justice’, in A. Cassese et al. (eds.), The Oxford Companion to International Criminal Justice (Oxford: Oxford University Press, 2009), 123. 57  See e.g. J. McMahan, ‘War Crimes and Wrongdoing in War’, in The Constitution of the Criminal Law (n. 1), 151, on the relationship between in bello morality and in bello law; C. A. J. Coady, ‘Terrorism and the Criminal Law’, in The Constitution of the Criminal Law, 185, on the definition of terrorism and its implications.

Introduction

17

III.  Starting Points for a Theory of Criminalization: Moral or Political? Any normative account of the proper scope and operations of the criminal law must clearly depend on an account of the legitimacy of the state. The criminal law is part of the apparatus of the state—a part that involves very obvious exercises of the state’s coercive power. If we are to justify a system of criminal law, we must therefore be able to justify (or to appeal to a justification of ) that coercive power. Legislators, judges, police officers, prosecutors, prison officials, and all the other actors who play official roles in the making and enforcement of the criminal law, in legislation, in law enforcement, in the criminal process, in the administration of punishment, all exercise that power. But by what right do they do so? A normative theorist of criminal law might agree that her theorizing depends in this way on some account of the state’s legitimacy, but argue that she need not herself provide such an account, or commit herself to any particular such account. For, she might argue, theories of criminal law need not be shaped or structured by any particular account of state legitimacy; they need only presuppose that some such account is available. This reflects a more general assumption that can be discerned (if only by implication) in much theorizing about criminal law—that criminal law theory is largely independent of political theory. Although in recent years there has been a growing, more explicit engagement with issues in political theory,58 and although theorists often appeal to some usually vague and under-explained idea of a ‘liberal’ criminal law, they have too often treated criminal law theory more as a species of applied moral philosophy, without paying sufficient attention either to the implications for criminal law of different political theories, or to the institutional framework and structure of the criminal law itself.59 We will focus here on one central question about the way in which political theory, and conceptions of the state and political society, should figure in a theory of criminalization. Briefly stated, the question is: should a theory of criminalization start with an account of wrongs, understood independently of the criminal law and its institutional structure, and justify criminal law as an appropriate response to (some of) these wrongs? Or should we recognize that the wrongs with which the criminal law deals are always already embedded in a political-legal institutional framework, and that a theory of criminalization must therefore be grounded in a normative account of those institutions? This is one of the questions on which the editors have disagreed;60 but the disagreement has been productive. 58  This is reflected especially in the essays by Dimock and Pettit in this volume. 59  Thus, for just one instance, Fletcher began his seminal book by saying that ‘[c]‌riminal law is a species of political and moral philosophy’ (G. Fletcher, Rethinking Criminal Law (Boston: Little, Brown, 1978), xix); but as the book develops, the moral philosophy dominates. But contrast Fletcher, ‘Political Theory and Criminal Law’, Criminal Justice Ethics, 25 (2006), 18. 60  Compare, for instance, R. A. Duff and S. E. Marshall, ‘Public and Private Wrongs’, in J. Chalmers et al. (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010), 70; V. Tadros, ‘Wrongness and Criminalization’, in A. Marmor (ed.), Routledge Companion to Philosophy of Law (London: Routledge, 2010), 157; and Farmer’s chapter in this volume.

18

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

Any non-institutional approach must begin by identifying some set of pre-institutional concerns that need to be addressed, and then show how criminal law, as an institution of the state, is an (or the) appropriate way to address those concerns, or mischiefs. Those mischiefs might initially be identified in non-moral terms: on some versions of the Harm Principle, for instance, we should begin by identifying a range of harms, understood perhaps as setbacks to interests;61 we then have reason to criminalize a type of conduct if doing so is likely to be an efficient way of reducing the incidence of such harms.62 Or they might be identified in moral terms: we might begin not with some pre-moral notion of harm, or of offence, but with the idea of wrongful harm or offence, harms or offences that wrong those who suffer them;63 or we might begin with some more precise moral value, for instance that of sovereignty, or dignity, and take as the relevant mischief conduct that violates such values;64 or we might begin simply with the notion of (culpable) wrongdoing, and take the whole broad realm of moral wrongdoing as our starting point.65 For simplicity’s sake, to clarify the general point at issue here, we will focus on the legal moralist’s claim that the relevant category of (pre-legal, pre-institutional) mischiefs is constituted by moral wrongdoing. The claim that concerns us here is not the widely accepted ‘negative’ moralist principle that only wrongdoing should be criminalized (which does not take the wrongfulness of a kind of conduct to give us, by itself, any positive reason to criminalize it), but rather the ‘positive’ moralist principle that the wrongfulness of a kind of conduct gives us reason to criminalize it.66 That reason might, depending on the theorist, be preventive—that we have good reason to prevent wrongdoing, and criminal law can help to prevent it; or retributive—that culpable wrongdoing deserves punishment, which the criminal law can provide; or that wrongdoers should be called to account, which the criminal process can achieve. Wrongfulness is at most, of course, a good reason for criminalization, not a conclusive reason. No one supposes that, in the end and all things considered, we should criminalize 61  As Feinberg famously defined harm: J. Feinberg, Harm to Others (New York: Oxford University Press, 1984), ch. 1. For an alternative recent account, see Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), ch. 3. Feinberg’s account of harm as a basis for criminalization, however, also builds in wrongfulness: what gives us reason to criminalize a type of conduct is that doing so will efficiently prevent wrongful harms. 62  This is how the Harm Principle is explicated in its canonical formulations: see e.g. J. S. Mill, On Liberty (London: Parker, 1859), ch. 1, para. 9; Feinberg, Harm to Others (n. 61), 26. In actually applying the Harm Principle, however, theorists often instead take the key question to be whether the conduct to be criminalized itself causes or might cause harm: but see J. Gardner and S. Shute, ‘The Wrongness of Rape’, in J. Horder (ed.), Oxford Essays in Jurisprudence, 4th Series (Oxford: Oxford University Press, 2000), 193. 63 See e.g. Feinberg, Harm to Others (n. 61); Feinberg, Offense to Others (New York: Oxford University Press, 1985); Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), chs 3, 6. 64  e.g. A. Ripstein, ‘Beyond the Harm Principle’, Philosophy & Public Affairs, 34 (2006), 215 (sovereignty); M. Dan-Cohen, ‘Defending Dignity’, in M. Dan-Cohen, Harmful Thoughts: Essays on Law, Self and Morality (Princeton: Princeton University Press, 2002), 150 (dignity). See further Section VI. 65  e.g. Moore, Placing Blame (n. 15), and in this volume. 66  On ‘negative’ and ‘positive’ forms of legal moralism, see R. A. Duff, ‘Towards a Modest Legal Moralism’, Criminal Law and Philosophy, 8 (2014), 217.

Introduction

19

every kind of culpable wrongdoing. Some of the reasons that militate against criminalization may reflect other moral or political principles that bear on and constrain the criminal law: for instance the principle of liberty.67 Others are more pragmatic, to do with the feasibility, efficacy, and costs of criminalizing a type of conduct.68 Such pragmatic reasons are highly context sensitive, and no philosophical account of the criminal law can be expected by itself to yield determinate conclusions about what ought to be criminalized in a particular social, political, and historical context—though such accounts should have something to say about the kinds of consideration, both practical and principled, that should be relevant, and how they should be evaluated. However, we can focus here not on the question of what kinds of conduct should in the end be criminalized, but on the more modest question of what gives us good reason to criminalize a type of conduct, and on the legal moralist’s claim that such good reasons are (always or only) grounded in the wrongfulness of the conduct in question. Some legal moralists argue that there is a pro tanto reason in favour of criminalizing any and every kind of culpable moral wrongdoing: no kind of wrongdoing is in principle and ab initio outside the purview of the criminal law, although quite often this reason in favour of criminalizing a given type of wrongdoing is outweighed by other reasons against doing so.69 Others argue that only certain kinds of wrongdoing are even in principle apt candidates for criminalization: others are simply and from the start not the criminal law’s business. If we ask why some kinds of wrong should be thus excluded as candidates for criminalization, we will get different kinds of answer from different theorists. Some of those answers will appeal to moral rather than to distinctively political values and considerations. Mill’s advocacy of the Harm Principle, for instance, was grounded in the value of individual liberty, and our duty not to interfere with each other’s freedom except to prevent harm to others. That principle of non-interference of course applied to the state—with particular force, given the extent of the state’s coercive power. But it was not itself a political principle, since it applied as much to social pressures as it did to legal or political coercion: it should ‘govern . . . the dealings of society with the individual in the way of compulsion and control, whether the means used be physical force in the form of legal penalties, or the moral coercion of public opinion’.70 Similarly, it might be argued that we have a right to do wrong: a right to commit (certain kinds of ) wrong, and to

67  See Moore, in this volume. 68  Compare J. Schonsheck, On Criminalization: An Essay in the Philosophy of the Criminal Law (Dordrecht: Kluwer, 1994), on the ‘filters’ through which any proposal to criminalize a type of conduct must pass. 69  Moore (Placing Blame (n. 15) and in this volume) is again the clear contemporary defender of this view. Some would also take Devlin to be this kind of legal moralist: no kind of immorality is in principle not the business of the criminal law; see P. Devlin, The Enforcement of Morals (Oxford: Oxford University Press, 1965). Devlin’s concern, however, was not with immorality as such, but with what was believed or deeply felt to be immoral by the members of the particular society; and he was concerned with such perceived or felt immorality only insofar as a failure to criminalize it might lead to the harm of social collapse. 70 Mill, On Liberty (n. 62), ch. 1, para. 9.

20

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

bear the consequences, without interference from others.71 That right might be grounded in the value of autonomous agency, and a claim that respect for autonomous agency requires us to leave each other free to commit such kinds of wrong. That right, that duty of non-interference, has implications for the state: plausibly, if citizens have a right to do wrong, the state must not interfere with that right’s exercise; which implies, inter alia, that it should not criminalize wrongs that we have a right to commit. But this right to do wrong is not itself a political right, and is not grounded in any particular conception of the state and its proper aims: it is a moral right, which constrains the conduct of all agents—including state agents. By contrast, other accounts of why certain kinds of wrong are not, even in principle, the business of the criminal law appeal to a conception of the state and its proper aims and limits. The most familiar kind of account along these lines appeals to a roughly liberal conception of political community, and of the state’s role as the institutional manifestation or mechanism of such a political community. If we understand a political community as consisting, in part, in a set of relationships among citizens structured by a set of values that constitute their collective civic morality, the liberal argument is that that civic morality should not include all the values by which citizens, as moral agents, should guide their own lives: that some values, important though they might be for individual lives, should not be understood as civic values in which the polity has a collective interest; and that those matters of ‘private’ morality are therefore not the business of the criminal law, as an institution of the state. Many liberals would say something of this kind about adultery, for instance. Fidelity in interpersonal relationships is, they might agree, an important moral ideal, which they try to exemplify in their own lives: but it is not an ideal that should figure in the civic or political morality of a liberal polity, as a ‘public’ value by which all citizens must qua citizens be bound (an adulterer can be a good citizen); violations of that value are therefore in principle not the kinds of wrong that should concern the criminal law—they are in principle not apt candidates for criminalization.72 If we ask why we should set such limits on the values that are to count, and be enforceable, as the ‘public’ values of the political community, a familiar liberal answer is Rawlsian. If we are to sustain a stable political community in which citizens with very different conceptions of the good can respect each other, whose basic institutions all citizens can be expected to endorse, and in which the crucial set of basic liberties are protected, that community’s public, self-defining values must be limited in this way; they can include only those

71  See J. Waldron, ‘A Right to Do Wrong’, Ethics, 92 (1981), 21; D. Enoch, ‘A Right to Violate One’s Duty’, Law and Philosophy, 21 (2002), 367; O. J. Herstein, ‘Defending the Right to Do Wrong’, Law and Philosophy, 31 (2012), 343. 72  Echoes of the Wolfenden Committee’s comments on kinds of wrong that are ‘in brief and crude terms, not the law’s business’ (Report of the Committee on Homosexual Offences and Prostitution (London: HMSO, 1957), para. 61) should be evident here. The terminology of ‘public’ and ‘private’ wrongs, which has a long history in criminal law theory (see, e.g., Blackstone’s Commentaries on the Laws of England (available at ), Bk IV, ch. 1, p. 5) is of course problematic: see Moore, in this volume, pp. 198–200.

Introduction

21

values that all citizens, whatever their conceptions of the good, can be reasonably expected to endorse.73 The two approaches sketched above differ on whether every kind of wrong is in principle a candidate for criminalization, but they agree in a crucial starting point: for both begin with a general category of moral wrongs, understood and identified independently of the criminal law and its institutional structures (indeed independently of the state), and then ask which of those wrongs we have good reason to criminalize.74 Both therefore face the criticism that we should not begin our theorizing about the criminal law in this pre-legal, pre-institutional, even pre-political, way, with a general category of moral wrongs: that we must instead begin with the political and the institutional, if we are to understand the criminal law and its proper or legitimate scope.75 This criticism can be developed in different ways, but its central claim is that criminalizable wrongs take their character as wrongs that merit criminalization not—or not primarily or solely—from any pre-legal and pre-institutional wrongness, but essentially from their institutional setting and meaning: to identify and understand them as wrongs that we (a ‘we’ that is now already the political ‘we’ of a political community) have good reason to criminalize, we must understand them as wrongs committed within such an institutional setting—a setting that partly determines their meaning and their implications. One way to develop this line of thought is to argue that any understanding of specific wrongs determinate enough to guide decisions about criminalization, determinate enough to show that and why we have reason to criminalize them, must be the product of a political process of norm formation—a process that itself involves the institutions of the law.76 One obvious example of this point is that of wrongs connected to property: different normative understandings of property and of our interest in it, understandings which themselves reflect different political structures, will generate different views about whether, why and how such wrongs should be criminalized.77 But, it might be argued, the point applies well beyond 73 See J. Rawls, Political Liberalism (New York: Columbia University Press, 1993). See also M. Matravers, ‘Political Neutrality and Punishment’, Criminal Law and Philosophy, 7 (2013), 217. 74  It is worth noting that this kind of approach makes ‘mala in se’ central to criminal law—as they are central to so much criminal law theory; they are precisely wrongs that can be identified as wrongs independently of the (criminal) law. So-called ‘mala prohibita’ will then form a secondary category, of offences whose wrongfulness might seem by comparison more doubtful or tenuous. Critics will argue that this is one of the distortions to which an over-moralized approach to criminal law theory is prone. 75  See Duff, ‘Towards a Modest Legal Moralism’ (n. 66). For a useful exploration of this contrast between moral and political starting points in the philosophy of punishment, see F. Tanguay-Renauld, ‘Victor’s Justice: The Next Best Moral Theory of Punishment?’, Law and Philosophy, 32 (2013), 129. 76  See M. D. Dubber, ‘Criminal Law between Public and Private Law’, in The Boundaries of the Criminal Law (n. 1), 191; L. Farmer, ‘Criminal Wrongs in Historical Perspective’, in The Boundaries of the Criminal Law (n. 1), 214; and Bottoms and Farmer in this volume. 77  See e.g. the debate between Ripstein and Bird over the example of the ‘harmless nap’, where the different starting points (in a Kantian idea of the sovereign individual or in a Millian idea of harm) generate different understandings of the nature of the wrong: see Ripstein, ‘Beyond the Harm Principle’ (n. 64); C. Bird, ‘Harm Versus Sovereignty: A Reply to Ripstein’, Philosophy & Public Affairs, 35 (2007), 179; A. Ripstein, ‘Legal Moralism and the Harm Principle: A Rejoinder’, Philosophy & Public Affairs, 35 (2007), 195.

22

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

such clearly institutionally structured wrongs. There is, for instance, wide scope for disagreement, not about whether sexual assault and domination are wrongs, but about what kinds of conduct count as sexual, as assault, as domination; about the character, and the scope, of such wrongs; about whether they should be understood as ‘public’ or as ‘private’ and if as ‘public’, how the law should define and deal with them. Such disagreements implicate our understandings of the nature and scope of political community: they can be worked through, in a way that can generate a determinate-enough conception of such wrongs to feed into deliberations about whether and how to criminalize them, only as part of a political process of norm formation. This approach is developed in the two chapters in this volume which take MacCormick’s account of law as an ‘institutional normative order’ as their starting point.78 For both authors, one of the most distinctive features of MacCormick’s account is the claim that criminal law contributes to ‘securing the conditions of social peace and civility’.79 The first chapter, by Anthony Bottoms, focuses on the particular role played by the institution of criminal law in the building and sustaining of social trust, as a key component of social peace or order. While recognizing that trust is a generalized social phenomenon, Bottoms argues that the role played by the criminal law in the articulation and enforcement of public wrongs is crucial to the understanding and maintenance of social peace. However, he makes an important qualification to MacCormick’s argument by showing that the breach of social peace is insufficient to justify the creation of a criminal law, and that criminal law must draw on a conceptual vocabulary of harms and wrongs as a means of identifying and expressing social interests. This theme is developed in a slightly different way by Farmer, who argues that MacCormick’s claim should be read in terms of the purposiveness of the institution of criminal law. Criminal law protects certain objects or goods not because of their pre-legal moral value, but because they contribute towards certain social ends or goods. An institutional account of criminalization accordingly requires that we focus not only on the goods to be protected, but also on the ends which the law thereby hopes to bring about. Thus both accounts focus on the way that wrongs and harms are articulated within an existing institutional structure. Another way to develop the claim that we must begin with the political, not (merely) with the moral, is to argue that an account of the criminal law, as a particular kind of essentially coercive institution, must begin with an account of the proper functions and powers of the state; and that what makes any conduct criminalizable, what constitutes it as a criminalizable wrong, is the way in which it seeks or threatens to undermine the state’s functions. Thorburn, for instance, has been developing this kind of argument in a number of papers. He offers a (Kantian) liberal account of the essential function of the state as being ‘to secure for all of us the conditions of freedom as independence’: to secure the conditions 78  D. N. MacCormick, Institutions of Law (Oxford: Oxford University Press, 2007). See also now N. Lacey, ‘Institutionalising Responsibility: Implications for Jurisprudence’, Jurisprudence, 4 (2013), 1. 79 MacCormick, Institutions of Law (n. 78), 221.

Introduction

23

under which we can live together as free and equal agents, not vulnerable to arbitrary interference from others.80 The criminal law, as part of the apparatus of the state, is then properly concerned with actions that ‘demonstrate a willingness . . . to displace the legal rules themselves’, and that thus constitute ‘an injury to . . . the very idea of living together under law’.81 This kind of account draws a sharp distinction between the pre-legal, pre-institutional character of such moral wrongs as rape, murder, and other kinds of attack on the person, and their character as criminalizable wrongs in the context of a state. What makes them criminalizable is not their pre-institutional moral character as wrongs against or attacks on individual victims, but their character as denials or violations of ‘the very idea of living together under law’. Our question is not whether we should accept something like Thorburn’s Kantian account of the state and of the criminal law, but whether this is the right kind of approach to take to questions of criminalization. We can put the question in terms of the traditional idea of crimes as public wrongs.82 On one kind of approach, which we labelled ‘pre-institutional’ above, we begin with a general category of wrongs, and ask which of them (understood still as the kinds of wrong that we initially identified) should count as ‘public’, i.e. as the proper business of the state and of the criminal law: the features in virtue of which we see them as criminalizable (for instance that they cause or threaten harm to others, or that they violate another’s moral rights or sovereignty) are features that they already have, and that ground the determination that they should be counted as public wrongs. On the other kind of approach, which is exemplified by Thorburn’s argument, the wrongfulness of the wrongs that we have reason to criminalize is already a public wrongfulness: what constitutes the relevant kind of conduct as wrongful, in a way that could concern the criminal law, is precisely its impact on, its implications for, or its meaning in the context of, the public realm of the polity. A simple version of this approach is found in the idea that a public wrong is just a wrong that has some harmful impact on ‘the public’—on the citizen body as a whole: as Blackstone put it, crimes are breach and violation of the public rights and duties, due to the whole community, considered as community, in its social aggregate capacity. . . . [They] strike at the very being of society, which cannot possibly subsist, where actions of [that] sort are suffered to escape with impunity. In all cases the crime includes an injury: every public offence is also a private wrong, and somewhat more; it affects the individual, and it likewise affects the community.83 80  Compare Braithwaite and Pettit, Not Just Deserts (n. 14), 61–9, and Pettit’s chapter in this volume, on ‘non-domination’ as the key good for a republican political theory. 81  See M. Thorburn, ‘Constitutionalism and the Limits of the Criminal Law’, in The Structures of the Criminal Law (n. 1), 85 (the quotes are from 98 and 100); also his ‘Justifications, Powers and Authority’, Yale Law Journal, 117 (2008), 1070; ‘Criminal Law as Public Law’, in Philosophical Foundations of Criminal Law (n. 41), 21. For a Kantian conception of the state, see also A. Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy (Cambridge, Mass.: Harvard University Press, 2009). Compare too A. Brudner, Punishment and Freedom (New York: Oxford University Press, 2012). 82  See n. 72. 83 Blackstone, Commentaries on the Laws of England (n. 72), Bk IV, ch. 1, p. 5.

24

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

But there are other ways of characterizing the essentially public nature of the kinds of wrong that could concern the criminal law. In their contributions to this volume, Philip Pettit and Susan Dimock offer alternatives to the Kantian approach by appealing to two other prominent conceptions of ‘the political’. Dimock draws on the contractarian tradition in political philosophy, which understands political societies as cooperative schemes whose purpose is to realize the mutual advantage of their members.84 Within this framework, criminalization is justified instrumentally as a way of ensuring mutually beneficial cooperation among rational persons living together. Thus, for Dimock, legitimate forms of criminalization are only those that could be reasonably agreed to by members of society. Pettit, on the other hand, appeals to the republican tradition to whose revival he has powerfully contributed.85 In this tradition, the fundamental value on which the justification of political legitimacy rests is nondomination – individuals’ capacity to freely exercise their choices without being vulnerable to a power of interference by others. This capacity, Pettit argues, can be guaranteed only within a system of public law in which everyone is protected against interference with their basic liberties. Criminalization is a crucial element of this system, because through the imposition of costs accompanied by public condemnation on those who interfere with basic liberties, it provides individuals with the distinctively public security required by the republican conception of freedom. Obviously, the plausibility of these theories of criminalization largely depends on the plausibility of the wider political theories within which they are embedded. We will not find Dimock’s or Pettit’s views convincing unless we think that political societies should indeed be conceived in contractarian or in republican terms. In this sense, for these authors the debate about criminalization can start only after we have addressed the more fundamental question of how we should understand political societies (although presumably they would agree that which conception of political society we should adopt is partly determined by how plausible the theory of criminalization that it generates is). However, not every attempt to characterize the public nature of criminal wrongs is necessarily married to a specific political theory like Kantianism, Contractarianism, or Republicanism. Some have argued in more general terms that the public wrongfulness of the kinds of conduct that we have good reason to criminalize consists, for instance, in their tendency to cause ‘social volatility’, or to undermine the trust on which the possibility of social life depends.86 84  T. Hobbes, Leviathan, ed. R. Tuck (Cambridge: Cambridge University Press, 1996); D. Gauthier, Morals by Agreement (Oxford: Oxford University Press, 1986); P. Vallentyne (ed.), Contractarianism and Rational Choice (Cambridge: Cambridge University Press, 1991). 85  Braithwaite and Pettit, Not Just Deserts (n. 14); P. Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 2007), and On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge: Cambridge University Press, 2012). 86  L. C. Becker, ‘Criminal Attempts and the Theory of the Law of Crimes’, Philosophy & Public Affairs, 3 (1974), 262; R. Nozick, Anarchy, State and Utopia (Oxford: Basil Blackwell, 1974), 65–71; S. Dimock, ‘Retributivism and Trust’, Law and Philosophy, 16 (1997), 37.

Introduction

25

A familiar objection to such approaches is that they distort the character of the wrongs with which the criminal law deals (at least when those wrongs can be classed as mala in se): that our primary reasons for criminalizing murder, rape, and other attacks on the person lie in the (pre-legal, pre-political, pre-institutional) character of such wrongs as serious attacks on their individual victims, rather than in their impact on, or implications for, ‘the public’ or the polity as a whole. A major challenge for those who would begin in this way with the political is to show how they can do justice to what they take to be the essentially political character of the wrongs that can be the business of the criminal law, whilst at the same time explaining how the criminal law—in its definitions of such crimes, in the way that those who commit them are called to account and censured—is properly focused on the substantive wrongs that are done to the individual victims of crime. The claim that criminal law theory must begin with the political, and with the character of criminal law as a particular kind of state institution, raises further, methodological issues that we noted in Section I. If we are to theorize criminal law not in the a priori abstract, but as a concrete political institution, we must see it as always being situated in some particular legal system, some particular political and social order, with a particular, contingent history. Normative theorizing about criminal law must also itself begin from within some particular tradition of thought, rooted within a particular political, social, and legal context. The question then is whether, or how far, it can hope to transcend such historical contingencies: either, at its most ambitious, towards a universal and a priori grounding for such institutions, and thus for the criminal laws that they structure; or, more modestly, towards some larger normative conception of criminal law and its institutional setting that, whilst still falling well short of universality, can make possible a critical appraisal of any particular set of legal and political institutions.87 That question also brings us back to the question of the relationship between criminal law theory and political theory. Whether criminal law theory should ‘begin with the political’ or not, it is clear that political theory must figure prominently in any theory of criminalization: even if we begin by thinking about moral wrongdoing and the kinds of response that it might invite or require, we must at some point ask what kind of interest the state can properly take in what kinds of wrongdoing. If we are to begin with the political, however, political theory must figure much earlier in the discussion, since we must begin with an account of the state, and of the structure and institutions of a political community. One question then concerns the proper ambitions of normative political theory: how far can political theorists sensibly aspire to a universalist account of the form that political community must take; how far should they instead aspire, more modestly, only to articulate what are admittedly historically contingent accounts of the forms that it 87  Compare Lacey’s comments (‘Historicising Criminalisation’ (n. 46), at 941) on ‘the conditions of existence of social practices’. A related question concerns the extent to which a theory of criminalization should be substantive, specifying the proper content and scope of the criminal law; or procedural, concerned with the political processes through which issues bearing on criminalization are deliberated and decisions reached. For a move towards a more proceduralist account, see Duff, ‘Towards a Modest Legal Moralism’ (n. 66).

26

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

might take? A more relevant question for our present purposes, however, concerns the closeness of the connection between political theory and criminal law theory, in particular theories of criminalization. How far, in what ways, will different kinds of political theory have different implications for the aims, structure, and scope of the criminal law? Can liberal-minded criminal law theorists simply appeal, as many are prone to appeal, to an under-specified notion of ‘liberal democracy’ to work out what kind of criminal law could be appropriate for the citizens of such a polity: or do they need a more substantive picture of what a liberal democracy might be, and of the relationships between its citizens and between citizens and state, if they are to reach tolerably determinate conclusions about the role and the scope of a liberal criminal law?88 Thus, in this volume, Marshall, for example, discusses the roles which citizens, in particular the victims of crime, inhabit in the criminal law process, and the civic duties of participation which those roles involve. In this context she also raises questions about how we should understand the relationship between these citizen roles and the roles and responsibilities of officials. We must turn now, however, to a further kind of complexity with which any theory of criminalization must deal.

IV.  Criminal Law and Other Modes of Regulation Discussions of criminalization are sometimes conducted as if the choice facing legislators, or other officials whose decisions determine what is treated as criminal, is simply to criminalize or to do nothing: but that is, of course, far from the truth. Even when we have got to the point of identifying a social problem, or a socially problematic kind of conduct, that it would be possible to bring within the reach of the criminal law, the question is not simply whether or not we should do so, since criminalization is just one among many possible kinds of response to social problems; just one among other kinds of legal regulation and control. Legislators deciding what kinds of conduct they should formally criminalize, and officials deciding whether to treat as criminal conduct that has been formally criminalized, and citizens deciding whether to mobilize the criminal law in response to formally criminal conduct that they have suffered or witnessed, all face choices not between criminalizing and doing nothing, but between criminalizing and a range of other possible responses; a theory of criminalization must have something to say about the kinds of principle or consideration that should guide such choices. We noted this point above as one of the complicating factors that face any attempt to develop a theory of criminalization;89 in this section we will say a little more about some of these other kinds of response to problematic conduct, and about what is involved in the choices that have to be made between them. 88 Compare M. Philips, ‘The Justification of Punishment and the Justification of Political Authority’, Law and Philosophy, 5 (1986), 393, against M. Davis, ‘The Relative Independence of Punishment Theory’, Law and Philosophy, 7 (1989), 321, on the relevance of political theory to theories of punishment. 89  See text at nn. 10–12.

Introduction

27

On one view, those choices are theoretically fairly simple—though in practice of course highly complex. The question to be answered is: which kind of response will be most cost-effective or efficient as a means to the relevant goals, the most obvious goal being to reduce the incidence of social harm or mischief, or to repair such harm or mischief as has been done. Criminal law is, on this view, one among other kinds of regulatory mechanism, to be used as one among other instruments to achieve our social goals; when and how it is to be used thus depends on decisions about its efficiency as compared to other available instruments.90 This view does seem to characterize the way in which governments often respond to perceived social problems, but it leaves a number of questions unanswered. In particular, it leaves no room for the thought that the criminal law might be intrinsically, rather than instrumentally, appropriate (or inappropriate) as a response to some perceived social problem. That thought might initially be suggested by the common slogan, which has achieved the status of a legal principle in some jurisdictions, that criminal law should be a matter of ‘last resort’ (ultima ratio).91 There are different ways of understanding this slogan: read literally, it requires that criminal law not be used until every other kind of available measure has been tried and found wanting; read less literally, it could be taken to mean no more than that the criminal law should be used only if and when it amounts to a proportionate response, or that the use of the criminal law carries a heavier burden of justification than other kinds of legal regulation.92 The strict literal reading is not obviously plausible: we should at least consider whether criminalization cannot sometimes be an appropriate first resort, for instance because of the symbolic importance of such a response to some egregious kinds of wrong. Other, less literal readings might be taken as no more than useful reminders for legislators or prosecutors of the seriousness of the step that they are taking by criminalizing conduct. But they might also, by emphasizing the need for proportionality or the distinctive burden of justification, be pointing to something distinctive about criminal law as compared to other modes of legal regulation: although the criminal law does, of course, seek to regulate conduct, it is misleading to describe it simply as one among other systems of regulatory prohibitions that aim to reduce the incidence of the conduct that it prohibits.93 We should rather, on this view, recognize that the criminal law operates in a quite distinctive way: for instance that it performs a particular kind of communicative function, speaking in a moral voice to define a range of public wrongs and to censure those 90  See e.g. Law Commission of Canada, What is a Crime? Challenges and Alternatives (Discussion Paper) (Ottawa: Law Commission of Canada, 2003); N. des Rosiers and S. Bittle (eds.), What is Crime? Defining Criminal Conduct in Contemporary Society (Vancouver: UBC Press, 2004); also Horder’s chapter in this volume. 91  See K. Nuotio, ‘Theories of Criminalization and the Limits of Criminal Law: A Legal Cultural Approach’, in The Boundaries of the Criminal Law (n. 1), 238, at 255–7. 92  See generally N. Jareborg, ‘Criminalization as Last Resort (Ultima Ratio)’, Ohio State Journal of Criminal Law 2 (2005), 521; D. Husak, ‘The Criminal Law as Last Resort’, Oxford Journal of Legal Studies, 24 (2004), 207. 93  See the discussion in N. Lacey, ‘Criminalization as Regulation: The Role of Criminal Law’, in C. Parker et al. (eds.), Regulating Law (Oxford: Oxford University Press, 2004), 144.

28

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

who commit them.94 This suggests, however, that when we ask whether we have reason to mobilize the criminal law as a way of dealing with some social problem or mischief, we should attend not simply to its likely efficacy or inefficacy as a means of control or regulation, but rather to its intrinsic or non-instrumental appropriateness as a response to that mischief: given its distinctive character, given the way in which it characterizes the kinds of conduct that it defines as criminal (as public wrongs) and the nature of the response that it provides to such conduct (formal prosecution and punishment), is this an apt way to portray and respond to that mischief? We can see the importance of this sort of question, and of this dimension of debate about criminalization, by looking briefly at some of the other ways in which we might collectively respond to problematic kinds of conduct or situation that might seem to be apt candidates for criminalization: this should help us to identify more clearly the kinds of question that must be answered, the kinds of choice that must be made, on the path towards criminalization. Some would argue that we should, in principle, not even embark on that path. If, for instance, part of what is distinctive about criminal law is its connection to criminal punishment (that to criminalize conduct is to make it liable to punishment), and if criminal punishment cannot be justified, then criminalization cannot be justified.95 Or, if to criminalize conduct is to condemn it authoritatively as wrong, and we should refrain from such would-be authoritative condemnation, criminalization cannot be justified.96 We will not discuss such radically abolitionist ideas here,97 but will instead look briefly at some of the alternatives that face us even if we are in principle willing to criminalize.

A.  ‘Restorative’ responses One kind of alternative is exemplified by ‘restorative justice’ programmes and processes, though it also includes other kinds of informal response that might not be included under that label.98 Restorative justice procedures can of course figure within the criminal justice system: they can operate after conviction, as alternatives to or alongside punishment; they can operate as modes of diversion from

94  See e.g. Duff et al., The Trial on Trial (3) (n. 5), part II; Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), ch. 1. 95  For recent abolitionist arguments about criminal punishment, see D. Golash, The Case Against Punishment: Retribution, Crime Prevention, and the Law (New York: New York University Press, 2005); D. Boonin, The Problem of Punishment (Cambridge: Cambridge University Press, 2008); and M. Zimmerman, The Immorality of Punishment (Peterborough: Broadview Press, 2011). 96  This is a dimension of one central strand of abolitionist thought: see e.g. N. Christie, ‘Conflicts as Property’, British Journal of Criminology, 17 (1977), 1; L. Hulsman, ‘Critical Criminology and the Concept of Crime’, Contemporary Crises, 10 (1986), 63; H. Bianchi, Justice as Sanctuary: Toward a New System of Crime Control (Bloomington: Indiana University Press, 1994). 97  But see R. A. Duff, Punishment, Communication and Community (New York: Oxford University Press, 2011), 30–4, 56–74. 98  See generally G. Johnstone, Restorative Justice: Ideas, Values, Debates (2nd edn.; London: Routledge, 2011); G. Johnstone (ed.), A Restorative Justice Reader (2nd edn.; Cullompton: Willan, 2012).

Introduction

29

prosecution, after a person has been arrested for, and perhaps charged with, an alleged criminal offence. To the extent that they are available, criminal justice officials need to decide whether and when to divert cases from the criminal process into a restorative justice process. But such processes, whether formally organized or purely informal, can also constitute more radical alternatives to criminalization: problems that could in principle be treated within the framework of the criminal law, as involving criminalized or criminalizable conduct, can instead be treated outside that framework altogether. A decision that this is the best way to respond to a problem can be made quite informally by those most directly involved: rather than calling the police, they decide to deal with the matter informally among themselves. Or it could be made by officials: those with the authority to mobilize the resources of the criminal law might decide not to do so, either in particular cases on the basis of ad hoc decisions, or in classes of case as a matter of general policy; they might instead offer a state-supported restorative justice programme, or encourage those involved to seek their own informal resolution. Or the decision could be made by a legislature, which may decide not to formally criminalize a certain kind of conduct, or to make formal provision for a non-criminal, restorative response to certain crimes. In some cases, the process will still be focused on what is formally defined and perhaps seen as a crime: the decision will be to seek a non-criminal response to that crime. In other cases, the participants might adopt a more radically non-criminal perspective: rather than seeing the situation as one in which what is salient is a criminal wrong that merits condemnation, they see it as one in which people have a ‘conflict’ that must be resolved, or have ‘trouble’ in their social relationships that needs to be addressed in a way that might repair those relationships.99 To decide between a criminal law response and a restorative justice response (whether as a basis for legislation, or as a matter of official policy or in relation to a specific situation), we need to ask not just which kind of response is likely to be a more efficient means to some agreed end, but which is more appropriate to the salient features of the conduct or situation in question. In fact, it is not even clear that these different processes can be seen as means to the same end. To answer our question, then, we need to gain a clearer idea of the central or defining features of each response. We might think, for instance, that a criminal response focuses attention on an alleged wrong that was committed, and on the alleged perpetrator of that wrong as someone who should be called to account and censured for it; whereas, by contrast, a restorative approach focuses on some harm that has been caused or suffered, and seeks a less confrontational or accusatory process of mediation and negotiation to find ways to repair that harm and to reconcile those who were in conflict about it. What matters here is not whether such characterizations of either a criminal law response or of restorative justice are apposite (and at least in the case of restorative justice there are by now so many different kinds of process claiming that title that 99  See Christie, ‘Conflicts as Property’ (n. 96); Hulsman, ‘Critical Criminology and the Concept of Crime’ (n. 96).

30

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

it would be hard to find any non-vacuous general characterization that could apply even to most of them). The point rather is, first, that to decide whether or not to criminalize a particular type of conduct can involve deciding between criminalization and a range of other kinds of possible response; and second, that in order to make such a decision we must be able to grasp the characteristic features, aims and operations of each kind of response. If it is right to say that a criminal response is focused on calling the alleged perpetrators of public wrongs to punitive account, whereas a restorative justice response is focused on mediating between those who are in conflict, and on repairing harms that have been caused, we must ask which kind of response is appropriate to the situation at hand (a question which does not rule out the response ‘both’ or ‘neither’).100

B.  Professional discipline Many kinds of wrong that are, or could be, criminalized are actually dealt with, as wrongs, outside the criminal law. This often happens quite informally (when parents, for instance, discipline children within the family); but it also happens more formally when professional bodies or institutions operate their own disciplinary procedures. Doctors, lawyers, and many other professions have their own professional organizations, part of whose role is to lay down codes of conduct for members of the profession, and to discipline those who violate the code (indeed, it might be seen as one identifying feature of a ‘profession’ to have such a code and to enforce it).101 Sometimes, of course, misbehaving professionals will find themselves facing both a criminal prosecution and a professional tribunal for the same wrong: a doctor who attacks a patient might be criminally prosecuted for assault and disciplined or struck off by his professional body (and it is worth asking why this does not amount to an improper kind of double jeopardy). In other cases, however, conduct that could be treated as criminal is instead dealt with as an ‘internal’ disciplinary matter for the professional body: either it is not defined as criminal at all; or, although it fits the criminal law’s definition of an offence, it is left to be dealt with by the profession, rather than being put into the hands of the criminal law. Consider, for example, plagiarism committed by students or by members of academic staff in a university. Arguably, at least some types of plagiarism fit the criminal law’s definition of ‘fraud’: the plagiarist ‘dishonestly makes a false representation’ (that this essay or article is his own work), intending thereby ‘to make a gain for himself ’;102 and although in English law that ‘gain’ must be ‘in money or other property’,103 that would be true when, for instance, the plagiarism is 100  For some discussion of this question, see Marshall in this volume. 101  We leave aside the armed forces, whose systems of military law administered by courts martial are even closer in character to the criminal law, though still separate from it: similar questions arise about which kinds of wrong should be left in the hands of the military, to be dealt with as a military matter, and which should instead (or as well) be put in the hands of the criminal justice system. Many employers will also have their own disciplinary procedures for dealing with ‘internal’ misconduct: these raise the same kinds of question. 102  Fraud Act 2006, s. 2(1). 103  Fraud Act 2006, s. 5(2).

Introduction

31

intended to help the plagiarist win a prize or secure a promotion. Typically, however, even when the plagiarist fulfils the conditions of criminal liability, plagiarism is dealt with by the university concerned as an ‘internal’ disciplinary matter. The process might be analogous to the criminal process. Plagiarism is formally defined as an academic offence; a procedure is specified for dealing with allegations of plagiarism, including a hearing at which the alleged plagiarist is called to answer the accusation; if his plagiarism is established, there is then a formal finding that must be heard not just as a factual determination but as a condemnatory verdict; and this is typically followed by the imposition of a penalty (which, in the case of academic staff, could be dismissal). But why should such wrongdoing be treated in this way (with at least the tacit approval of the criminal law) as an ‘internal’ matter for the university, rather than as a criminal offence? One answer to this question would appeal simply to considerations of cost-effectiveness in harm prevention: such wrongdoing is not so serious, or so threatening to people outside the academic institution, that it is worth expending the resources of the criminal law to deal with it; it is dealt with more efficiently and more economically as an internal matter. There might be more to it than that, however. Plagiarism, as seen through the academy’s eyes, is not just a kind of fraud: it has a distinctive character as an attack on one of the institution’s defining values. If it is that character, rather than its nature as one among other kinds of fraud, that is properly salient in our understanding of the wrong, we can see reason to leave ownership of the wrong with the academic institution itself. Once again, our aim here is not to settle such questions about either the character or the proper allocation of the kinds of wrong with which professions might claim the authority to deal as internal, disciplinary matters. It is rather to point out, first, that decisions about what to define or to treat as criminal can also involve deciding whether certain kinds of wrong are best dealt with under the criminal law, or as internal matters for an institution, a profession, or an employer; and, second, that such decisions about the proper allocation of wrongs might depend not merely on considerations of cost-effectiveness, but on a view of the character of the wrong and thus of whose business, whose wrong, it properly and primarily is.104

C.  Government inspectorates A common mode of regulation for complex activities in contemporary societies is for the government to create a specialized body—an inspectorate—one of whose central tasks it is to administer and enforce a body of regulations governing the conduct of those involved in that activity: obvious examples in England include tax inspectors who administer the regulations governing how individual 104  It would in this context also be worth considering the reasons given by the Director of Public Prosecutions for his decision not to prosecute two doctors who had allegedly agreed to arrange ‘sex-selection’ abortions: the case was complicated, but among the reasons was the fact that the doctors had also been referred to the General Medical Council. .

32

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

and corporate tax is assessed and paid; and inspectors working for the Health and Safety Executive who administer the rules and requirements concerning health and safety that apply to a wide range of businesses and activities.105 The regulations that such officials have to enforce might be quite separate from the criminal law—we discuss this kind of case in Section IV.D; or they might be backed by the criminal law in that it is made a criminal offence to violate the regulations or to impede or seek to undermine their implementation.106 Even when the regulations are part of or backed by the criminal law, however, inspectors typically exercise a wide discretion about whether and when to bring a criminal prosecution: they may instead negotiate with the (alleged) violator to reach an agreement that avoids prosecution (and that involves no admission of liability). Someone who is found to have evaded his taxes could face a criminal prosecution; but he might instead be offered an arrangement under which he pays the tax owed, plus a further substantial penalty payment, and avoids any criminal prosecution or liability. This is indeed the most common way of dealing with discovered tax evasion: in 2010–11 there were fewer than 500 criminal convictions for tax evasion, while 2,000 people were put in the ‘Managing Deliberate Defaulters’ programme (without being prosecuted),107 and many others were dealt with through ‘the cost effective Civil Investigation of Fraud (CIF) procedures’.108 Similarly, health and safety inspectors will often not prosecute firms which they believe have been violating health and safety regulations: they will instead seek to negotiate with the firm to work out effective agreements and procedures to improve safety and to guard against future violations—agreements that are likely to involve no admission of liability, even if they do include substantial payments to repair past harm or to put in place future precautions. An interesting procedure in this context is that of the ‘Deferred Prosecution Agreement’, under which prosecutors agree to suspend prosecution of a corporation so long as it avoids future commissions of the relevant kind of offence and makes such changes in its operations as the agreement requires.109 In both these kinds of case, officials have to decide whether to treat conduct that (as they have good reason to believe) constitutes the commission of a criminal 105  See generally I. Ayres and J. Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford: Oxford University Press, 1992); C. Wells, Corporations and Criminal Responsibility (2nd edn.; Oxford: Oxford University Press, 2001); J. Gobert and M. Punch, Rethinking Corporate Crime (Cambridge: Cambridge University Press, 2003), ch. 9; K. Hawkins, Law as Last Resort: Prosecution Decision-Making in a Regulatory Agency (Oxford: Oxford University Press, 2003). 106  See, for just one instance, the provisions of the Health and Safety at Work etc. Act 1974. 107  See HM Revenue and Customs, Closing in on Tax Evasion (London: HMRC, 2012), 8 (also available at ). 108  See e.g. HMRC Code of Practice 9, 2011 (available at ) on the CIF process; also HMRC Criminal Investigation Policy () on the considerations that guide HMRC decisions about whether to use the CIF process or a criminal prosecution: prosecutions are ‘reserved for cases where HMRC needs to send a strong deterrent message or where the conduct involved is such that only a criminal sanction is appropriate’. 109  See D. N. Husak, ‘Social Engineering as an Infringement of the Presumption of Innocence: The Case of Corporate Criminality’, Criminal Law and Philosophy, 8 (2014), 353.

Introduction

33

offence as criminal:110 whether, in effect, to criminalize that conduct or to deal with it through some non-criminal negotiation and agreement (although of course the background threat of criminal proceedings can provide the supposed perpetrator with a powerful incentive to reach an agreement). So we must again ask not only what kinds of consideration should lead a legislature formally to define such conduct as a criminal offence, but also what kinds of consideration should guide officials in deciding whether to treat the conduct as criminal—i.e. whether to prosecute it in a criminal court. Practical considerations must clearly loom large in such decisions. Some will be the kinds of consideration that also apply in the context of criminal plea bargaining: what are the chances of a successful prosecution; what would it cost to pursue the case? Others are rather to do with securing future benefits or preventing future harms: is a CIF procedure more likely to result in the recovery of larger amounts of unpaid tax; what is likely to be the most effective way to persuade a company to improve its safety procedures? But, once again, we should not assume that such practical considerations are the only relevant ones— that criminal law should be seen as an instrument or technique alongside such other mechanisms, and that our choice of technique should be dictated purely by considerations of efficiency in pursuing the relevant social aims. We must at least ask whether there are other relevant considerations, for instance to do with the importance of calling public wrongdoers to public account, which bear on such decisions. To answer that question, however, we must attend more carefully to what is distinctive about the criminal law as a particular mode of regulation.

D.  Ordnungswidrigkeiten and ‘regulatory’ offences We have already noted the existence in some legal systems of a category of non-criminal, ‘regulatory’, or ‘administrative’ offences, such as the German Ordnungswidrigkeiten.111 Such ‘regulatory infractions’ supposedly differ from ‘real’ criminal offences in at least three ways: they do not attract the formal censure that attaches to criminal convictions; while they may be sanctioned by fines, or disqualifications from the activities in which the infraction occurred, imprisonment is not a possible penalty; and the procedures through which they are dealt with are simpler than those of a criminal trial (and provide fewer protections for the alleged violator). One question is whether, and on what basis, such a system of regulatory infractions is to be seen as genuinely distinct from criminal law. The European Court 110  Which in the case of deaths at work, for instance, could involve prosecution either for a breach of health and safety regulations, or for corporate manslaughter under the Corporate Manslaughter and Corporate Homicide Act 2007: see S. Antrobus, ‘The Criminal Liability of Directors for Health and Safety Breaches and Manslaughter’, Criminal Law Review (2013), 309. There are of course further issues, which we cannot address here, about whether it should be corporations, or individual managers or directors, or both, who are prosecuted. 111  See at nn. 10–11. See also Law Commission of Canada, What is a Crime? (n. 90), and Our Criminal Law (Ottawa: Law Commission of Canada, 1976); for more critical discussion, see Duff et al., The Trial on Trial (3) (n. 5), 189–98.

34

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

of Human Rights had to discuss this question in relation to the application of article 6 of the European Convention on Human Rights. Is someone accused of a regulatory infraction facing ‘a criminal charge’, in which case he is entitled to the various protections laid down in article 6 as entailed by the ‘right to a fair trial’? Or can it be argued that these protections are unnecessary, since he is not facing a criminal charge? The court held that a state could not render a law or an offence non-criminal simply by relabelling it: to determine whether it should count, for the purposes of the ECHR, as a criminal offence attention must also be paid to the nature of the offence, and to the nature and severity of the available penalty; and the court had the authority to hold that what a member state had classified as a non-criminal infraction should be treated as a criminal offence.112 This introduces a further complication to debates about criminalization, and to the question of what it is to criminalize a type of conduct: if we attend not (merely) to the formal classification of a regulation as ‘criminal’ or as ‘non-criminal’, but also to the substantial role that it plays in the legal system, we might see reason not only to argue that some offences formally classed as criminal are not ‘real’ crimes,113 but also to argue that some offences formally classed as non-criminal are or should be treated as ‘real’ crimes, in virtue of their character, or the reasons for defining them as offences, or the kinds of penalty that they attract.114 However, suppose that a suitably designed system of regulatory infractions can be clearly distinguished from criminal law. The next question is whether we should create or maintain such a system alongside the criminal law; and, if we should, how we should determine which kinds of violation should be defined as criminal, and which as (merely) regulatory. But we cannot answer these questions without getting a clearer idea of the proper aims of either kind of system. If we could see both in purely instrumental terms, as means of regulating conduct and preventing future harmful or wrongful misconduct, our task would be theoretically quite simple. We might argue, for instance, that while the criminal law, with its stigmatizing or censuring significance and the seriousness of the sanctions that it imposes, could be efficient as a means of dealing with seriously wrongful or harmful kinds of conduct, it is too coercive and costly an instrument for dealing with less serious kinds of conduct that we nonetheless need to regulate or deter; and that, while due respect for the rights of those facing criminal charges requires us to construct a criminal process that offers them the kinds of protection captured in article 6 of the ECHR, we would not need to provide such costly protections for those facing only a regulatory or administrative charge. This line of thought could favour the creation of a distinct system of regulatory infractions, as a cheaper and more efficient way of regulating less seriously harmful or wrongful kinds of conduct. As we have suggested before, however, such a purely instrumentalist perspective seems inadequate. We need to ask more carefully about the proper aims of each kind of system (are they both best understood as institutional techniques serving the same ends?), and about the meanings and value that they can be seen to 112 See Öztürk v Germany (1984) 6 EHRR 409; Lauko v Slovakia (2001) 33 EHRR 40. 113  See at n. 29.    114  Compare also Ashworth and Zedner, ‘Preventive Orders’ (n. 12).

Introduction

35

embody. For instance, should we see a system of regulatory infractions as serving a primarily deterrent purpose—as imposing penalties whose primary aim is to make compliance with the regulations prudentially attractive? Or should we see it instead, or also, as a way of making sure that the costs of various kinds of dangerous or harmful conduct are exacted from those engaging in such conduct?115 In either case, is this an aim that we should pursue in this way? Only when we have a clear view of the aims that such a system can properly serve, and of the proper aims of a system of criminal law, can we tackle the questions of whether we should maintain both kinds of system and, if we should, how we should decide the proper scope of each.

E.  Private or public law? The different kinds of regulatory regime discussed so far belong to the realm of public law: regulations are laid down by a public body, and are enforced by public officials; cases are brought by those officials, acting in the name and on behalf of ‘the public’, the polity as a whole. However, we must also attend to the realm of private law, most obviously tort law and contract law: for here we face yet further questions about the relationship between criminal law and other types of law. We can focus here on tort law, although similar questions arise about aspects of contract law (including the question of why breach of contract should be criminal only when it involves fraud or coercion). Some torts are of course also crimes: if, by recklessly lighting a bonfire, you burn down my garden shed, the police could charge you with criminal damage, and I could sue you to recover the cost of replacing my shed. Others are not: in many countries defamation and libel are not criminal offences, but those who are defamed or libelled can bring a tort suit;116 and if you cause me harm by conduct that falls below the appropriate legal standard of care, but that does not involve the kind of fault that is required for criminal liability, you may be tortiously but not criminally liable. So we need then to ask about the proper scope of each kind of law, and about the proper relationships between them. Which kinds of case are apt (only) for tort law, and which should fall (instead or also) within the criminal law? If such distinctions are important, we need an account of the proper aims of each kind of law.117 Suppose, for instance, that we see tort law as being primarily concerned with allocating the cost of harm that has been caused: it enables those who have suffered 115  See generally Tadros, ‘Criminalization and Regulation’ (n. 30); also Horder, in this volume. 116  Article 19, a body which campaigns for ‘freedom of expression and information’, has data about the ways in which different countries treat defamation as a criminal or as a civil matter (). 117  See generally M. Dyson (ed.), Unravelling Tort and Crime (Cambridge: Cambridge University Press, 2014). There are further questions about the ways in which tort law and criminal law can interact or interweave: for instance about the ways in which victims-plaintiffs can become involved as prosecutors of, or as parties to, criminal cases, or can join their claims for civil compensation to the criminal prosecution of the person who wrongfully harmed them.

36

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

harm to transfer the costs of that harm (the cost of repairing it, or of compensation for it) onto the person whose faulty conduct caused it. There would then of course be further questions about the basis on which these allocations should be made: should they be based on considerations of economic efficiency, for instance, or of justice? But if we could say that that is the proper concern of tort law, whereas criminal law is rather concerned with calling to public and punitive account those who commit what it defines as public wrongs, we could see how the two kinds of law differ quite radically in their aims, and thus see how we could begin to determine which kinds of case should belong to which of them. But suppose we instead see tort law as theorists of ‘civil recourse’ portray it:118 as enabling those who are wronged to call those who wronged them to account, and to obtain redress? Now the task of distinguishing criminal law from tort law becomes much harder: both are dealing with wrongs, and with wrongs that are ‘public’ in the sense that they are formally recognized as wrongs that merit some legal response;119 both make it possible to call a wrongdoer to public account for his wrong. So how would they now differ in their proper aims, and how should we decide which kinds of wrong belong to which? We do not suggest that either the simple cost-allocation model or the civil recourse model provide adequate accounts (either descriptive or normative) of tort law—indeed, we doubt that they do. But the point here is, again, that a normative theory of criminalization must include an account of the criminal law’s relationship to other modes of legal regulation, other kinds of legal process through which legally cognizable harms or wrongs might be addressed; and that any such account must involve an explanation of the proper functions of these other kinds of legal provision. If we are to work towards a decision about which kinds of wrongful, harmful, or otherwise undesirable conduct should be criminal, we must ask not only which kinds should attract the law’s formal attention, but which should fall within the purview of the criminal law, rather than being assigned, for instance, to the realm of private law, or to a regime of non-criminal regulation. But to answer that kind of question we must understand what these other modes of law are or should be for, as well as what criminal law is or should be for. This is not to suggest that a normative theory of criminalization must be a normative theory of all of law (as well as of all the kinds of extra-legal process and provision noted in this section). It is, rather, to emphasize that the criminal law cannot be theorized in isolation; and to suggest that it would be a mistake to view all these different modes of regulation in simple instrumentalist terms, as so many techniques through which states can seek to control the conduct of their citizens. The different ‘techniques’ have their own distinctive characters and purposes; we must therefore ask not just which is more likely to be cost-effective as a means, but also which is appropriate to the character of the mischief that is to be addressed.

118  See B. Zipursky, ‘Rights, Wrongs, and Recourse in the Law of Torts’, Vanderbilt Law Review, 51 (1998), 1; J. Goldberg and B. Zipursky, ‘Torts as Wrongs’, Texas Law Review, 88 (2010), 917. 119  See A. Y. K. Lee, ‘Public Wrongs and the Criminal Law’, Criminal Law and Philosophy, 8 (2014).

Introduction

37

V.  Blurring the Boundaries: Preventive Justice A discussion of the relationship between criminal law and other modes of legal regulation or control must include or presuppose an account of the distinctions between the different kinds of law or process. We have already noted that those distinctions—for instance, that between criminal law and non-criminal ‘administrative’ regulation—might not be sharp or clear: they might not be formally distinguished; and even if they are formally distinguished in the law, there might be room for argument about whether they are substantively distinct. What is also worth noting is the way in which governments can blur such distinctions in practice, by introducing measures and policies that in effect interweave or even merge criminal law with others modes of regulation and control. One example of this phenomenon which has become more significant in recent years is the way in which criminal law and immigration law are used together: criminal law is used increasingly to enforce immigration restrictions and controls; immigration law (notably the power to deport) is used increasingly to back up the criminal law; immigration and criminal officials work closely together as if engaged in the same enterprise.120 But we will focus here on another example: the range of practices captured by the label ‘preventive justice’, which threaten to disrupt our traditional understandings of criminal law itself and of the boundaries between criminal law and other forms of legal regulation and control.121 Some idea of prevention is inextricably bound up with any system of criminal law: for to define a type of conduct as criminal is to mark it as conduct in which those bound by the law should not engage, and therefore as conduct that they collectively have reason both to eschew and to prevent. Such prevention might not itself be a matter of criminal law: we can seek to prevent crime in a variety of ways, including education, ‘early interventions’ in contexts in which we can identify criminogenic factors, improvements in welfare provision and in social justice, ‘situational crime prevention’,122 and so on. But the criminal law itself can plausibly be seen as having two kinds of preventive aim. First, some idea of prevention is intrinsic to the operations of a legal system. Once a type of conduct is defined as criminal, the law permits or prescribes measures that aim to prevent it: part of the function of a police force is crime 120  See e.g. D. A. Sklansky, ‘Crime, Immigration, and ad hoc Instrumentalism’, New Criminal Law Review, 15 (2012), 157; A. Aliverti, ‘Making People Criminal: The Role of the Criminal Law in Immigration Enforcement’, Theoretical Criminology, 16 (2012), 417; A. M. McLeod, ‘The U.S. Criminal-Immigration Convergence and Its Possible Undoing’, American Criminal Law Review, 49 (2012), 105. 121  See especially A. J. Ashworth and L. Zedner, ‘Just Prevention: Preventive Rationales and the Limits of the Criminal Law’, in Philosophical Foundations of Criminal Law (n. 41), 279, ‘Prevention and Criminalization: Justifications and Limits’, New Criminal Law Review, 15 (2012), 542; ‘Preventive Orders’ (n. 12); A. J. Ashworth, L. Zedner, and P. Tomlin (eds.), Prevention and the Limits of the Criminal Law (Oxford: Oxford University Press, 2013). 122  On which see A. von Hirsch, D. Garland, and A. Wakefield (eds.), Ethical and Social Perspectives on Situational Crime Prevention (Oxford: Hart Publishing, 2000).

38

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

prevention; and the police (and others) may intervene, if necessary coercively, to prevent the commission of crime. The criminal process of trial and punishment to which those accused of committing crimes may be subjected might not be theorized in primarily preventive terms, since some will explain its justifying purposes as responsive rather than preventive: but at least part of the aim of this aspect of a criminal justice system must be to reduce the incidence of future crimes. Many of the provisions made for those undergoing punishment are quite properly aimed at preventing reoffending (even if such prevention is not the primary justifying purpose of their punishment); if an offender who has been punished reforms himself because, as he sincerely says, his punishment brought him to recognize the need for such self-reform, we would count that as a success. These remarks are not intended to deny the importance of the long-running debates between ‘retributivist’ and preventive accounts of criminal punishment; they are simply intended as reminders of some of the ways in which a system of criminal law cannot but aim at prevention.123 Second, prevention can also figure as the, or an, aim of criminalization: if we ask why we should criminalize a certain type of conduct, one answer will be that we should do so in order to prevent . . .; but to prevent what? Skating over the important substantive differences between various answers to this question, we can say that this conception of criminalization as serving a preventive purpose must begin with an account of some ‘non-trivial harm or evil’ that is to be prevented:124 we then have reason to criminalize a certain type of conduct if doing so will effectively prevent (or at least reduce the incidence of ) that harm or evil. Already, however, a complication should be noted—a distinction between two ways in which the notion of a non-trivial harm or evil could function as a ground for criminalization. First, it might be that what gives us reason to criminalize Φ is that by criminalizing it we will prevent a non-trivial harm or evil (or that if we do not criminalize it some non-trivial harm or evil will ensue): the focus is on whether the criminalization itself is efficiently harm preventive. Or, second, it might be that what gives us reason to criminalize Φ is that Φ itself constitutes or causes a non-trivial harm or evil: the focus is on the harmfulness of the conduct to be criminalized. Orthodox renditions of the harm principle typically express it in the first way;125 but when the principle is actually applied, it is often implicitly interpreted in the second way. These two readings will of course often generate similar conclusions: an efficient way of reducing the incidence of a harm or evil is often to criminalize, and thereby reduce the incidence of, conduct that constitutes or causes that harm or evil. But they can diverge, in particular when (which will be important in what follows) we

123  Although there is much more to be said about the very idea of prevention, and about the significance of the differences between various modes of prevention: for instance, between normative persuasion that prevents by showing the person good reasons why she should not offend; deterrence that prevents by making crime imprudent; and incapacitation that prevents by making crime impossible. 124  See Husak, Overcriminalization (n. 22), 65–72, for the argument that we can find within the criminal law a principle that ‘[c]‌riminal liability may not be imposed unless statutes are designed to prohibit a nontrivial harm or evil’. 125  See n. 62 and references given there.

Introduction

39

can more efficiently reduce the incidence of a non-trivial harm or evil by criminalizing conduct that might not, or not directly, constitute or cause it. Speed limits and other kinds of driving regulation, backed by the criminal law, illustrate this possibility: what justifies imposing a strict speed limit is not that everyone who drives at a speed in excess of the limit either causes harm or creates an unreasonable risk of harm, but that this is a more efficient way of preventing the harms that dangerously fast driving causes than by criminalizing only driving that can be shown to have been dangerously fast. Our concern in this section, however, is not with these familiar ways in which prevention might figure in normative theories of criminal law, but with some ways in which a concern to prevent what are perceived as non-trivial harms of evils has fuelled significant changes in the scope and operations of the criminal law—changes on which the study of ‘preventive justice’ focuses. As we will see, the phenomena captured by ‘preventive justice’ are not wholly new; but they have expanded significantly in recent years, in ways that raise difficult normative questions about the proper scope and aims of the criminal law. We can identify two kinds of ‘preventive’ strategy that have been adopted by governments anxious to address (and to be seen to address) some kind of harm, evil, or threat that is perceived as serious and for which (it is thought) orthodox criminal law responses are inadequate.126 The perceived threats that have, in Britain, driven a number of such strategies are terrorism and ‘anti-social behaviour’: we can illustrate the problems raised by ‘preventive justice’ by provisions addressing each of these two kinds of perceived threats. One kind of provision involves expanding the scope of the substantive criminal law—the creation of new offences, or the expansion of existing offences. On a traditional view of the criminal law, the paradigm offence consisted in conduct that itself constitutes, or that directly causes, the non-trivial harm or evil (the mischief ) at which the law is directed: this was true of the most salient mala in se, such as murder, rape, and other injuries to the person.127 Now the criminal law’s reach has commonly extended far beyond that paradigm to capture conduct that is in some appropriate way related to the relevant mischief, but that does not or might not constitute or directly cause it: we criminalize attempts, conspiracy, and incitement—the three standard inchoate crimes;128 we criminalize many kinds of 126 On these two kinds of strategy, in relation to terrorism, see also V. Tadros, ‘Justice and Terrorism’, New Criminal Law Review, 10 (2007), 658. 127  Although even here there are complications. Theft, for instance, might require an intention to deprive the owner permanently of the property that one appropriates, but not an actual permanent deprivation (see e.g. Theft Act 1968, s. 1): unless the mischief at which the law is aimed is simply such dishonest appropriation with such an intention, the conduct criminalized will not always bring about the relevant mischief (that of actually permanent loss of property). Similarly, the common law offence of assault in English law consists in conduct that causes another ‘to apprehend immediate and unlawful personal violence’ (see D. Ormerod, Smith and Hogan’s Criminal Law (13th edn.; Oxford: Oxford University Press, 2011), 619): we must then either argue that the mischief at which the law is aimed is simply such apprehension of immediate violence, or recognize that this offence also criminalizes conduct that might not bring about the relevant mischief, but need only arouse apprehension of it. 128  See Ormerod, Smith and Hogan’s Criminal Law (n. 127), ch. 13.

40

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

dangerous conduct, even when it does not in fact cause harm;129 and we criminalize conduct that makes it easier for others to commit crimes, or that is preparatory to the agent’s own intended crime.130 But recent years have seen a further expansion in the range and reach of such offences, especially in relation to terrorism. There are offences consisting in possessing ‘an article in circumstances which give rise to a reasonable suspicion’ that the possession is for terrorist purposes, or collecting, recording, or possessing ‘information of a kind likely to be useful to a person committing or preparing an act of terrorism’;131 a broadly defined offence of ‘encouraging’ terrorism, which requires no intention to do so;132 offences of facilitating or handling what is or is suspected of being terrorist property;133 offences of belonging to, professing to belong to, or supporting, or wearing the uniform of, a proscribed organization;134 and offences of failing to report information bearing on terrorist activity.135 The creation of these kinds of offence is purportedly justified as making the early detection and thus prevention of (potential) terrorist activities easier—as assisting the state’s discharge of its duty to ensure the security of its citizens: but they raise some obvious questions that bear more generally on the conditions under which it can be legitimate to use the criminal law to prevent such prospective harms or wrongs.136 Such measures cannot be justified merely by claiming that they are efficient means to the end of ensuring security or preventing terrorist activity (even if that claim is true): those who are convicted of such offences are censured as wrongdoers who deserve, or are liable to, the punishments they receive; we must therefore ask what punishable wrongs they could be thought to commit. In particular, can it be argued that the conduct thus criminalized is already pre-legally wrongful; or that once it is legally prohibited, it becomes wrongful as a breach of the duty that 129 See D. Husak, ‘The Nature and Justifiability of Nonconsummate Offenses’, Arizona Law Review, 37 (1995), 151; Duff, Answering for Crime (n. 32), ch. 7; Simester and von Hirsch, Crimes, Harms, and Wrongs (n. 7), 53–88. 130 Including the criminalization of many offences of possession, either with or without the intent that what is possessed will be used to commit a substantive crime: see M. D. Dubber, ‘Policing Possession: The War on Crime and the End of Criminal Law’, Journal of Criminal Law and Criminology, 91 (2001), 829; A. J. Ashworth, ‘The Unfairness of Risk-Based Possession Offences’, Criminal Law and Philosophy, 5 (2011), 237. 131  Terrorism Act 2000, ss. 57–8. In both cases, whilst the offence is defined strictly, the law allows a defence of innocent purpose or reasonable excuse: such burden-shifting provisions, which spare the prosecution the burden of proving what would normally (on a classical view of the structure of criminal offences) count as an element of the offence, and instead lay on the defence the burden of ‘disproving’ it (a burden that might, however, involve no more than adducing evidence ‘sufficient to raise an issue with respect to’ the matter in question; see Terrorism Act 2000, s. 118(2)), are another common feature of recent preventively oriented legislation; see further Tadros, ‘Justice and Terrorism’ (n. 125), 670–5. 132  Terrorism Act 2006, s. 1. 133  Terrorism Act 2000, ss. 14–18.    134  Terrorism Act 2000, ss. 11–13. 135  Terrorism Act 2000, ss. 19, 38B (as amended by Anti-terrorism, Crime and Security Act 2001, § 117(2)). 136  They also, of course, raise questions about the very idea of ‘security’, which looms so large in the rhetoric of anti-terrorist measures: see L. Zedner, ‘Securing Liberty in the Face of Terror: Reflections from Criminal Justice’, Journal of Law & Society, 32 (2005), 507; Security (London: Routledge, 2009); P. Ramsay, ‘Preparation Offences, Security Interests, Political Freedom’, in The Structures of the Criminal Law (n. 1), 203.

Introduction

41

the law, in prohibiting it, legitimately lays on citizens to contribute in this way to the prevention of terrorism? If the conduct cannot be shown to be in one of these ways wrong, should we then conclude that its criminalization is illegitimate? Or should we recognize that the criminal law must be adapted in the face of this kind of threat, and that what some portray as principled constraints of justice on its scope must now be compromised? The second kind of preventive justice measure to be noted here involves a kind of hybrid of criminal and non-criminal provisions, and is best exemplified by the notorious ‘anti-social behaviour orders’, which were billed as a more effective way of dealing with various kinds of anti-social conduct that, though at least very often criminal, was hard to prosecute.137 Other provisions, most notably the ‘control orders’ that could be imposed on people suspected to be involved in terrorist activities,138 display the same logical structure. The structure of this kind of provision is that there is an initial, formally non-criminal process, in which a court is given reason to believe that a person has been engaged in, and/or is likely in the future to engage in, some kind of undesirable, usually criminal, activity (anti-social behaviour; terrorism), and that it is necessary to subject him to restrictions in order to prevent (or to reduce the risk of ) future behaviour of that kind. The court can then impose a range of restrictions: on where the person may go or when he may travel (including imposing a curfew), on whom he may meet, and on a range of activities in which he might otherwise engage. Once the restrictive order is made, it is a criminal offence to breach it. Such individualized orders, aimed at preventing future offending, imposed by a court and backed by the criminal law, are not unusual;139 indeed, one could see the ancient procedure of binding someone over to keep the peace as such an order.140 We should note, however, first, that the conduct from which the person is ‘bound over’ to refrain is itself criminal conduct—whereas what contemporary orders like ASBOs prohibit might well be non-criminal conduct; and second, that breach of the requirements of a binding over cannot by itself render a person liable 137  See Crime and Disorder Act 1998, ss. 1–4 (ASBOs are to be replaced by ‘criminal behaviour orders’: see Anti-Social Behaviour, Crime and Policing Bill 2013–14, Part II). The literature on ASBOs is voluminous, but see especially A. J. Ashworth, ‘Social Control and “Anti-Social Behaviour”: The Subversion of Human Rights?’, Law Quarterly Review, 120 (2004), 263; Ashworth and Zedner, ‘Preventive Orders’ (n. 12); Simester and von Hirsch, Crimes, Harms, and Wrongs (n. 7), 212–32; P. Ramsay, The Insecurity State: Vulnerable Autonomy and the Right to Security in the Criminal Law (Oxford: Oxford University Press, 2012); R. A. Duff, ‘Perversions and Subversions of Criminal Law’, in The Boundaries of the Criminal Law (n. 1), 88; A. Cornford, ‘Criminalising Anti-Social Behaviour’, Criminal Law and Philosophy, 6 (2012), 1. 138  Prevention of Terrorism Act 2005; see Tadros, ‘Justice and Terrorism’ (n. 125), 666–70. Such orders have now been replaced by ‘Terrorism Prevention and Investigation Measures’ (TPIMs); see Terrorism Prevention and Investigation Measures Act 2011. 139  For just two other examples, see ‘non-molestation orders’ and ‘restraining orders’ (Domestic Violence, Crime and Victims Act 2004, ss. 1, 12); see A. J. Ashworth and L. Zedner, ‘Defending the Criminal Law’, Criminal Law and Philosophy, 2 (2008), 21, at 29–31, 35–7. 140  The power to bind over goes back to the Justices of the Peace Act 1361, but is now governed primarily by Justices of the Peace Act 1968, s. 1(7), Magistrates’ Courts Act 1980, s. 115 (when it is not a sentencing matter: see Powers of Criminal Courts (Sentencing) Act 2000, ss. 1, 12): see the CPS guidance on Binding Over Orders ().

42

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

to imprisonment (the breacher simply forfeits the amount of the recognizance on which the binding over was conditioned), whereas the breach of an ASBO can bring a sentence of up to five years in prison. However, these kinds of order have become more draconian, both in the severity of the restrictions that can be imposed and in the punishments that may then be imposed on those who breach such an order; and they pose another serious challenge to the classical conception of the criminal law as an enterprise of defining various kinds of wrongful conduct for which perpetrators would be liable to public censure and punishment. For, first, the criminal law is now being used to respond to breaches not of general and public laws, but of individualized orders prohibiting conduct that is often itself entirely harmless. Second, instead of subjecting a person to coercive restrictions only if and when he is proved to have committed an offence, in line with the classical conception of the criminal law, the law (a law beyond the criminal law) is now used to restrict individuals because of what it is thought they might otherwise go on to do.141 We must therefore ask both whether this is an appropriate use of the criminal law, and whether it is legitimate in these ways to bypass the limits that a classically conceived criminal law sets on the state’s use of its coercive powers.

VI.  The Shape of a Theory of Criminalization One point that should have emerged from the previous sections is that a normative theory of criminalization must begin by locating the criminal law—the criminal law understood not just as a legislated code or set of criminal statutes (the law in the books), but as it is deliberated, created and developed, interpreted, applied, and enforced by all the officials and citizens upon whose agency it depends for its real existence. A theory of criminalization must locate the criminal law within the larger framework of a political theory of political community, of the state, of citizenship, and of the relations between community, state, and citizens. It must also locate the criminal law within the larger structure of the law, as one among other modes of legal regulation, and offer an account (which will necessarily be a partly normative account) of its distinctive character or role. And it must address the fact that the criminal law is always the law of a particular polity, located within its own particular, contingent history. Only once we have in these ways located the criminal law can we ask the kinds of question with which many normative discussions of criminalization begin (and with which we also began): what are ‘the principles and values that should guide decisions about what to criminalize?’;142 ‘What can give us good reason to criminalize a type of conduct?’ (bearing in mind now that such values, principles, and reasons will need to be grounded in some appropriate 141  The use of these and other kinds of pre-emptive restriction, including the indefinite detention of those who are judged to be dangerous, might also (and dangerously) come to seem more palatable if we are impressed by the potential of developing accurate predictive techniques, especially those that draw on neuroscience: on this see M. Hildebrandt, ‘Proactive Forensic Profiling: Proactive Criminalization?’, in The Boundaries of the Criminal Law (n. 1), 113. 142  See at n. 4.

Introduction

43

political theory, and will need to point us towards criminal law in particular rather than merely towards some kind of legal regulation). A normative theory of criminalization must, however, tackle such questions: it must be a theory of criminalization—an account of how we should determine what kinds of conduct we should criminalize (which must include some account of who this ‘we’ are). But that ‘how’ is ambiguous: it can be answered either procedurally or substantively. A proceduralist account focuses on the processes through which criminalization decisions should be made, rather than on the substantive content of those decisions. One kind of proceduralist account is political: it aims to specify the political processes (processes, for instance, of public deliberation, or of expert scrutiny) through which such decisions should be made. Another kind is rational: it aims to specify the logical structure of the deliberations that should lead to such decisions. A substantive account, by contrast, focuses on the substantive principles, considerations, criteria that should determine the content of decisions about criminalization, whoever makes those decisions and through whatever processes they are made. We should not expect to find purely proceduralist accounts, accounts that set no limits on the possible content of the decisions made, in this context, any more than we should expect to find purely proceduralist accounts of political deliberation and decision making in general:143 but we can identify accounts that lie towards the proceduralist end of the spectrum. One such recent account is Douglas Husak’s.144 He identifies seven ‘constraints’ that proposed criminal statutes must satisfy if they are to be adequately justified. One constraint is purely procedural: ‘the burden of proof [that criminalization is justified or required] should be placed on those who favour criminal legislation’.145 Others point towards substantive conditions: for instance that criminal statutes must be ‘designed to proscribe a nontrivial harm or evil’, that conduct may be criminalized only if it is wrongful, and that criminalization must serve ‘a substantial state interest’.146 But ‘nontrivial harm or evil’, ‘wrongfulness’, and ‘substantial state interest’ are not yet substantive constraints: they are relatively formal ideas which must be given their substantive content by, presumably, a process of political deliberation. Husak’s account is in that way proceduralist: it tells us the rational process through which a legislature should go in deciding whether to criminalize a type of conduct; it is thus both politically proceduralist, in that it focuses on the legislative process through which criminal law is made,147 143 For a useful discussion, see D. Archard, ‘Political Reasonableness’, Canadian Journal of Philosophy, 35 (2005), 1. 144 Husak, Overcriminalization (n. 22): see also his chapter in this volume, applying that account to the question of whether polygamy should be criminal; also ‘Convergent Ends, Divergent Means: A Response to my Critics’, Criminal Justice Ethics, 28 (2009), 123; ‘Repaying the Scholar’s Compliment’, Jerusalem Review of Legal Studies, 1 (2010), 48; ‘Reservations about Overcriminalization’, New Criminal Law Review, 14 (2011), 96. For another proceduralist account see Schonsheck, On Criminalization (n. 68), ch. 3 on the three ‘filters’ (principles, presumptions, pragmatics) through which criminal laws must pass. 145 See Overcriminalization, 100–2. 146 See Overcriminalization, 66–77, 132–45. 147  Although Husak emphasizes the important role played by prosecutors, and prosecutorial discretion, in determining the effective scope and impact of the criminal law (see e.g. Overcriminalization, 19–32), his theory of criminalization is focused on the legislative process of passing criminal statutes.

44

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

and rationally proceduralist in that it specifies the logical structure of the proper legislative deliberations. A normative theory of criminalization should be at least partly proceduralist: it should, that is, have at least something to say about the processes, both political and rational, through which the scope, content, and structure of the criminal law should be determined. But it must at some point turn from procedure to substance: what kinds of substantive principle, value, or consideration should guide the decisions of those who make and enforce the criminal law? More precisely, there are at least two kinds of substantive question that we must ask at this stage. First, what can give us good reason to criminalize a type of conduct (or, for that matter, what cannot give us good reason to criminalize)? Second, what other kinds of principle, value, or consideration must come into play in deciding whether a good reason to criminalize is, all things considered, a conclusive or good enough reason? The first question is the ‘in principle’ question: it focuses on the kinds of consideration that could (or could not) give us any proper reason to contemplate criminalizing a type of conduct. If we determine, via our answer to that first question, that we have no good reason to criminalize a given type of conduct, we will not need to turn to the second question, since criminalization is off the table. If, however, we see that we do have good reason to criminalize a given type of conduct, we must then turn to the second question: we must consider, in particular, whether there are other values or principles that give us countervailing, and perhaps better, reason not to criminalize it;148 and whether it is practicable to criminalize it (a question which itself encompasses a range of more precise questions about the practicalities and costs of legislation and enforcement). Normative theories of criminalization typically focus on the first kind of question, asking what can give us good reason, in principle, to criminalize: thus both Mill and Feinberg couch their principles in the logic of reasons.149 It is here that we will find the kinds of consideration that bear directly and specifically on criminalization, since the considerations that bear on the second kind of question are likely to be relevant not so much to criminal law in particular as to the state’s exercise of its powers, especially its coercive powers, more generally. It is also here that we find the attempts that dominate so much normative theorizing about the scope or boundaries of the criminal law—the attempts to find the key to criminalization, that principle or set of principles which will tell us precisely when we have good reason to criminalize. The best known of these would-be master principles remains the Harm Principle, to which some would add an Offence Principle allowing the criminalization of (grossly) offensive conduct even if it is not harmful.150 Alongside the Harm Principle, we can set the German doctrine of Rechtsgüter, according to which criminalization must be aimed at protecting those legally 148  The value of liberty, for instance: see Moore’s chapter in this volume and, far more extensively, his Placing Blame (n. 15), chs 16, 18; see at nn. 66–9 in this chapter. 149  Mill: ‘The only purpose for which power can be rightfully exercised’ (On Liberty, ch. 1, para. 9). Feinberg’s ‘liberty-limiting principles’ are set in terms of what constitute ‘good’ or ‘morally relevant’ reasons for criminalization: see Harmless Wrongdoing (New York: Oxford University Press, 1988), xix–xx. 150  See especially Feinberg, Offense to Others (n. 63).

Introduction

45

recognized interests that constitute the set of Rechtsgüter.151 In the same camp we can place Alexander and Ferzan’s argument that the criminal law serves the aim of harm prevention by proscribing and punishing conduct that culpably (recklessly) creates an unjustified risk to a legally protected interest.152 Another would-be master principle, with the same generality and abstraction, is Moore’s version of Legal Moralism: ‘criminal legislation must exclusively aim at preventing or punishing moral wrongs, and this it can do by prohibiting all and only those behaviors that are in fact morally wrong’.153 Other candidate principles include the Dignity Principle, that ‘the main goal of the criminal law ought to be to defend the unique moral worth of every human being’;154 the Sovereignty Principle, that ‘the legitimate basis for criminalization’ lies in ‘violations of equal freedom’ or sovereignty;155 and the Dominion Principle, that the ‘target. . . for the criminal justice system should be ‘the maximization of the dominion of individual people’.156 We will not discuss these principles in any detail here, but will just indicate, briefly, why we do not think that the search for such an overarching principle, or set of principles, offers a promising way of developing a substantive theory of criminalization. The values to which these principles appeal (values such as freedom from harm, sovereignty, dignity, autonomy) might indeed have a role to play in deliberations about criminalization, but not as master principles which can determine the scope, even the ‘in principle’ scope, of the criminal law. Their role would rather be as providing non-exclusive signposts or frameworks for those deliberations: by trying to give such concepts more determinate meanings, by thinking about their possible implications (which are matters for political deliberation), we might come to see reasons for (or against) criminalizing certain kinds of conduct. One question, of course, concerns the grounding of such principles: why should we posit the protection of dignity, or sovereignty or dominion, or the prevention of harm, as the proper aim of the criminal law? Given our earlier discussion, it should be clear that an answer to this question must be political: these principles or values are relevant to criminalization because they are, or should be, integral to the political structure of the polity whose criminal law is in question. They might even figure explicitly in the con­stitution of the polity: thus dignity, for instance, figures as a central, indeed ‘inviolable’, value in the German, Swedish, and Polish constitutions.157 Constitutions are certainly 151  See R. Hefendehl et al. (eds.), Die Rechtsgutstheorie: Legitimationsbasis des Strafrechts oder dogmatisches Glasperlenspiel? (Frankfurt: Nomos Verlag, 2003); N. Persak, Criminalising Harmful Conduct (Dordrecht: Springer, 2007), ch. 5; C. C. Lauterwein, The Limits of Criminal Law (Farnham: Ashgate, 2010), ch. 1. 152  Alexander and Ferzan, Crime and Culpability (n. 16). 153  Moore, in this volume, p. 192; see at nn. 65–9 above. 154  Dan-Cohen, ‘Defending Dignity’ (n. 64), 150; for critical discussion, see D. Baker, The Right not to be Criminalized (Aldershot: Ashgate, 2011), ch. 4. 155  Ripstein, ‘Beyond the Harm Principle’ (n. 64), 216; for critical discussion, see Bird, ‘Harm Versus Sovereignty’ (n. 77); Baker, The Right not to be Criminalized (n. 154), 164–71. Compare Thorburn on the goal of ‘secur[ing] for all of us the conditions of freedom as independence’ (text at nn. 80–1 above), and Dubber’s espousal of an Autonomy Principle in M. D. Dubber, Victims in the War on Crime (New York: New York University Press, 2002). 156  Braithwaite and Pettit, Not Just Deserts (n. 14), 54; see also Pettit’s chapter in this volume. 157 See German Constitution, art. 1 (‘Human dignity is inviolable. To respect and protect it is the duty of all state authority’); also Swedish Constitution, art. 2, Polish Constitution, art. 30.

46

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

one way in which polities can formally declare, and seek to entrench, their constitutive values—values that will presumably have implications for their criminal laws; they might also be a source of constraints on the criminal law, both substantive and procedural.158 We still need to ask why, or how, a constitution should come to include these particular values (a question in political theory); but, more appositely for our present purposes, we also need to ask why it should be these values that govern the criminal law. To ask this is to ask two questions. First, given that constitutions typically declare the polity’s commitment to a fairly large set of principles, values, and goals,159 why should we pick out these particular value(s), such as dignity, to guide the criminal law? Second, do the specified value(s) point us towards the criminal law, as a distinctive method of securing or protecting them? These questions can be illustrated by reference to the Harm Principle, which (although it has been much criticized) remains the most prominent and popular master principle for those who seek a theory of criminalization. The classic modern discussion remains Feinberg’s four-volume The Moral Limits of the Criminal Law,160 although he did not believe that the Harm Principle is the only principle that gives us reason to criminalize: he thought that we also need an Offence Principle (we have reason to criminalize if this would be a necessary and effective way of preventing serious offence to others);161 and he remained uncertain about whether the prevention of non-harmful, ‘free-floating’ evils could give us good reason to criminalize.162 More recently, others have revived more ambitious claims that the Harm Principle specifies a necessary condition of criminalization.163 Our first question is why we should initially specify the prevention of harm as one of the proper aims of a polity. The problem is not that such a declared aim lacks the grandiloquence to which constitutional declarations typically aspire, but that it is radically indeterminate until we are given some clearer idea of what is to count as ‘harm’, and of what kinds of harm it is to be the polity’s business to prevent—since it is surely not plausible to suppose that it should aim to prevent every kind of harm. If, with others, we follow Feinberg’s lead and define harm in terms of setbacks to interests, we make some progress, though we still face questions about whose interests are to count, and whether we are to attend not only

158  See e.g. Baker, The Right not to be Criminalized, drawing on the US constitutional tradition, and V. Tadros ‘A Human Right to a Fair Criminal Law?’, in J. Chalmers et al. (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010), 103. Husak also draws on the US constitution for his ‘external’ constraints on criminalization: Overcriminalization, ch. 3. 159  As a glance at e.g. the German, Swedish, and Polish constitutions (see n. 157) will show. 160  New York: Oxford University Press, 1984–8 (Harm to Others, 1984; Offense to Others, 1985; Harm to Self, 1986; Harmless Wrongdoing, 1988). 161 See Offense to Others. 162 See Harmless Wrongdoing, especially 318–38. 163  See e.g. Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7; although they argue for an Offence Principle along with the Harm Principle, they also think that only harmful offence should be criminal; see ch. 7); Persak, Criminalising Harmful Conduct (n. 151); Baker, The Right not to be Criminalized (n. 154). Husak now also seems to take this view: what was the ‘nontrivial harm or evil’ constraint in Overcriminalization (n. 22, at 66–72) now appears simply as ‘the harm constraint’ in his chapter in this volume (p. 220).

Introduction

47

to the interests of humans, but also to those of other animals;164 but it still seems implausible to say that we should make it our collective business as a polity to protect all the interests of all the beings in question against setbacks. So we may need to specify the interests that are to be protected rather more precisely: for instance by distinguishing ‘welfare’ from ‘ulterior’ interests,165 or by talking of ‘resources’,166 or providing a list of the interests that the polity (or its laws) should protect.167 But why should the prevention of harm, the protection of interests against setbacks, point us towards the criminal law? It is worth recalling that Mill’s classical Harm Principle was not about the criminal law: the prevention of harm was ‘the only purpose for which power can be rightfully exercised over any member of a civilized community against his will’; and ‘power’ meant any kind of ‘compulsion and control, whether the means used be physical force in the form of legal penalties, or the moral coercion of public opinion’.168 Even if we attend only to legal power or coercion, there are many kinds of legal provision through which the polity can seek to prevent many kinds of harm: so why look to the criminal law? Part of the answer is, of course, that since the criminal law operates by guiding human conduct, it can help towards the prevention of those kinds of harm that are caused, or can be prevented, by human action; but that is only the start of an answer, since we still need to know what makes this distinctive mode of legal regulation appropriate, and when. Similarly, if we hold that a polity should aim to protect and promote dignity, or sovereignty, or autonomy, or dominion, we have as yet no reason to think that it should use the criminal law for this purpose. Another more promising answer to our question brings in wrongfulness. Advocates of the Harm Principle often build a wrongness requirement into their specification of the Principle itself: thus Feinberg defines ‘harm’, in the sense relevant to criminalization, as ‘setbacks of interests that are wrongs’;169 Simester and von Hirsch point out that ‘[t]‌he Harm Principle provides for protection against only those setbacks that D was not entitled to inflict on V’;170 Husak argues that ‘[i]n the clearest cases, the bad consequence that a person suffers through human agency should not be regarded as a harm unless his rights are implicated’.171 Now this does bring us closer to criminal law, if criminal law can be identified in part by its distinctive concern with wrongs: although there are other ways than criminalization of responding to, or trying to prevent, wrongful setbacks to legally protected interests (including tort law), we can say that we have some reason now to consider 164  Not to mention, as some ecological theorists would add, the interests of other aspects of the non-human natural world. 165  See Feinberg, Harm to Others (n. 61), 37–8. 166  See Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), ch. 3. 167  See Alexander and Ferzan, Crime and Culpability (n. 16), 269–77: if crime consists in recklessly creating unjustified risks to legally protected interests, ‘the criminal law must identify those interests that it will protect’—though they do not themselves then identify those interests completely or in detail. Compare too the German doctrine of Rechtsgüter (see n. 151), which must be grounded in a specification of the goods that are to count as Rechtsgüter. 168 Mill, On Liberty (n. 62), ch. 1, para. 9.    169  Harm to Others (n. 61), 36. 170  Crimes, Harms and Wrongs (n. 7), 38–9. 171  In this volume, p. 220. But see Tadros, ‘Wrongness and Criminalization’ (n. 60).

48

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

criminalization as an appropriate response. In talking of wrongful harms rather than simply of wrongs, however, we slide from one version of the Harm Principle to a different version.172 It is plausible to insist that, if we are going to criminalize harmful conduct, conduct that causes or might cause setbacks to legally protected interests, we should criminalize it only if it is wrongfully harmful; if we can provide a list of those interests that should be legally protected, it is also plausible to say that we have reason to criminalize conduct that wrongfully harms or threatens to harm them. However, the Harm Principle, in its canonical formulations, talks not about criminalizing harmful conduct, but of criminalizing conduct in order to prevent harm; and whilst we can indeed hope to prevent harm by criminalizing and thus reducing the incidence of harmful conduct, we can also do so by criminalizing conduct that might not itself be harmful. For instance, if our aim is to reduce the number of harmful road accidents, it is plausible that we can do this more efficiently by setting strict speed limits, and criminalizing breaches of them, than by simply criminalizing dangerous driving; but in doing so we will criminalize some instances of speeding that do not in fact create any wrongful risk of harm. If our aim is to reduce the number of harmful accidents in factories, it is plausible that we can do this more efficiently by creating regimes of safety regulation that impose detailed duties on employers (including duties of record keeping and inspection), and criminalizing breaches of them, than by simply criminalizing the creation of dangerous working conditions; but in doing this we will criminalize some breaches that do not in fact create any risk of harm. These kinds of law (many so-called mala prohibita are of this kind) would be sanctioned by the Harm Principle in its canonical formulations, but are much harder to justify if the Harm Principle is read as sanctioning the criminalization only of wrongfully harmful conduct: partly because the harms to be prevented need not be wrongful harms, and also because the conduct criminalized need not itself be harmful or dangerous. One response to such examples would be to insist that we may criminalize conduct only if the conduct itself can be shown to be both a source of harm and wrongful in virtue of its connection to harm: we might in this way be able to justify some offences involving conduct that is only ‘indirectly’ or ‘remotely’ connected to harm, but will no doubt also have to reject many existing offences, at least as they are currently defined.173 The fact that a theory of criminalization would condemn as unjustified large swathes of our existing criminal law is not itself, of course, an argument against the theory; indeed, if the frequent claim that we are suffering a crisis of overcriminalization is right, that is just what we should expect of a normative theory. However, such a swift dismissal of so 172  See nn. 62, 125, and accompanying text. For more detailed discussion of this topic, and the suggestion that we should talk not about ‘the Harm Principle’, but about the two Harm Principles, see R. A. Duff and S. E. Marshall, ‘ “Remote Harms” and the Two Harm Principles’, in A. P. Simester, Ulfrid Neumann, and Antje du Bois-Pedain (eds.), Liberal Criminal Theory: Essays for Andreas von Hirsch (Oxford: Hart Publishing, 2014), 205. 173  See Husak’s discussion of mala prohibita in Overcriminalization (n. 22), 103–19; and Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), chs 4–5, on ‘remote harms’ and the need to be able to ‘impute’ responsibility for such harms to the agents whose conduct is to be criminalized.

Introduction

49

much criminal law seems hasty, especially since we can offer a different, but still harm-based, justification for at least some such offences by appealing to the other, canonical version of the Harm Principle. For if we can justify the creation of regulations of the kind noted above (such as speed limits and factory regulations), on the grounds that they will efficiently reduce the incidence of serious kinds of harm without imposing unreasonable burdens on those subject to them, we can then say that individuals ought to obey such regulations, and do wrong if they fail to do so; why then should we not also say, by appeal to the Harm Principle and the requirement that only wrongful conduct be criminalized, that we have reason to criminalize such wrongs? We need not say that the conduct itself is wrongfully harmful; its wrongfulness rather lies in the failure to obey regulations that are themselves aimed at promoting safety and preventing harm: but it is consistent with the spirit of the Harm Principle to recognize it as an (in principle) candidate for criminalization. A similar issue arises for other would-be master principles noted above. If sovereignty, or dignity, or dominion, is the master value, we could argue that we have reason to criminalize conduct that wrongfully violates or threatens sovereignty, or dignity, or dominion: that would seem to give us a very limited criminal law; but surely if our concern is with the protection of such values, we also have reason to criminalize kinds of conduct that do not directly violate or threaten them. For we should, surely, aim not merely to protect such values, as instantiated in human lives, against direct violation: we should also be concerned to promote them, and to seek to secure the conditions under which they can be fully actualized and recognized;174 and we might sometimes best do that by creating regulatory systems that citizens ought to obey, and breaches of which will be in principle criminalizable wrongs. Our comments so far are not intended as critiques of, or as reasons for rejecting, any such master principles: the point is simply that even we begin with an apparently unitary principle, we will find more than one way in which it could give us reason to criminalize conduct, more than one route from the specified value to criminalization. This does then lead on, however, to the three kinds of criticism that any such monolithic theory of criminalization must face. First, there is the charge that even when developed in the way suggested above, any such theory will be under-inclusive: it will not be able to show that we have reason to criminalize some kinds of conduct that we surely do have good reason to criminalize. In the case of the Harm Principle, purported counter-examples of this kind include conduct that is profoundly offensive but could not plausibly be described as harmful;175 coercively paternalistic conduct that infringes the autonomy of the person coerced precisely in order to prevent him harming himself;176 conduct that violates other rights without causing harm;177 desecration of 174  See e.g. Braithwaite and Pettit, Not Just Deserts (n. 14), 54; see at n. 156 above. 175  Hence the addition of the Offence Principle to the Harm Principle: see Feinberg, Offense to Others (n. 63); but also Simester and von Hirsch, Crimes, Harms and Wrongs (n. 7), chs 6–8. 176  See Feinberg, Harm to Others (n. 61), 78; Harm to Self (n. 159), chs 18–19. 177  e.g. Ripstein, ‘Beyond the Harm Principle’ (n. 64); Gardner and Shute, ‘The Wrongness of Rape’ (n. 62—though they also argue that we must criminalize because not to do so would be harmful); H. Stewart, ‘The Limits of the Harm Principle’, Criminal Law and Philosophy, 4 (2010), 17.

50

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

corpses; and what Feinberg calls (misleadingly) the kind of ‘free-floating evil’ that is exemplified by commercial gladiatorial combat to the death.178 Other kinds of counter-example could also be offered to the other candidate principles, to the same effect. Committed advocates of a master principle might well be able to argue that, on carefully closer inspection, we can find in the principle grounds to criminalize the kinds of conduct that figure in the counter-examples. If we look carefully at the longer term or less direct effects of such conduct, for instance; or provide a richer, more expansive account of the relevant value (of what should count as ‘harm’, or what could be said to violate or undermine sovereignty or dignity); or think about the wide range of regulations that might serve to protect or promote that value, and that might then be backed by the criminal law: we will find that we can have reasons to criminalize at least many of the kinds of conduct that the critic offers as counter-examples. No doubt this will often be possible; but such a strategy will also face two further, familiar kinds of objection. The second kind of objection is that under-inclusiveness is replaced by indeterminacy or over-inclusiveness. As we expand the conception of the core value, or the range of kinds of conduct that can be portrayed as somehow threatening it, in order to meet the charge of under-inclusiveness, we risk depriving the principle of any determinate meaning, and thus of any real efficacy as guide to, or a constraint on, criminalization. This is a familiar complaint about the way the Harm Principle has been developed and applied in recent years:179 the idea of harm has been stretched to cover wider types of mischief, conduct lying ever more remote from the harm that is to be prevented is argued to be criminalizable, and it begins to seem that the Principle could be (ab)used to justify the criminalization of any kind of conduct that we want to criminalize. Efforts to extend the scope of other candidate master principles, in order to meet the charge of under-inclusiveness, are likely to face similar objections. Third, such expansions of the master principle, so that it can capture cases that seemed to constitute counter-examples, will also be vulnerable to charges of distortion. By insisting that what makes any kind of conduct criminalizable must be, for instance, its connection to harm, or that it violates or threatens dignity or sovereignty; by insisting that it is always that feature that characterizes a type of conduct as a criminalizable wrong: theorists face the charge that they are distorting the character of many of the wrongs that we have reason to criminalize, by trying to make them all fit this single specification. That charge might be forcefully brought, for instance, against Alexander and Ferzan’s account, according to which ‘we have only one crime—manifesting insufficient concern for others’ legally protected interests’; they would ‘do away with the special part of the criminal code’, which defines the different substantive crimes, since every criminal action consists, 178  See Feinberg, Harmless Wrongdoing (n. 160), 318–38; and, more generally, Duff, Answering for Crime (n. 32), 128–35. 179 See e.g. B. Harcourt, ‘The Collapse of the Harm Principle’, Journal of Criminal Law and Criminology, 90 (1999), 109.

Introduction

51

qua crime, in recklessly creating an unjustified risk of harm to a legally protected interest.180 One question is whether all and only the types of conduct that create such a risk are types of conduct that we have good reason to criminalize: that question underpins the first two kinds of objection noted above. But even when a type of conduct that we see good reason to criminalize can be said to create such an unjustified risk to another’s interests; even if there is to this extent an extensional equivalence between the types of conduct that their principle gives us reason to criminalize and the types of conduct that we do have reason to criminalize: we might still object that such an abstract specification does not reveal the varied substantive reasons that we have for criminalizing different kinds of wrong. It might be true, for instance, that a violent attacker and someone who recklessly creates a risk of serious physical injury to another person both ‘manifest insufficient concern for others’ legally protected interests’ (just as it might be true that both poisons and deep-fried Mars bars are ‘insufficiently nutritious’); but it does not follow, and seems implausible to claim, that we have just that same reason for criminalizing both types of conduct.181 Similar kinds of objection are likely to be brought against other efforts to ground criminalization in single master principles: that they distort or conceal, rather than revealing more clearly, the reasons we have to criminalize different types of conduct. We have not tried in this section to show that the search for a master principle, or a set of master principles, of criminalization is doomed to fail, or to show that the principles we have noted all fail: we have not provided a detailed enough account of the kinds of objection that such principles will face, or examined the responses that their proponents might offer to such objections. Our aim has rather been to raise more general doubts about this kind of approach to criminalization: why should we expect to be able to produce a plausible normative theory of criminalization that grounds all the in principle reasons we could have to criminalize types of conduct in a single principle, or in a small, ordered set of principles? Given the wide range of values that we can expect to figure in the public realm of a contemporary, pluralist polity, as values that require legal recognition and protection; given the diversity and complexity of the social and institutional formations of such a polity; given the variety of ways in which the criminal law can figure as a possible method of addressing different kinds of legally defined wrong: should we not more plausibly expect to find a diversity of grounds for, and of routes towards, criminalization—a diversity that cannot be captured in any neat theoretical structure of coordinated principles? Such diversity would of course make for messy theory: we would end up not with a pleasingly ordered structure of principles from 180  Crime and Culpability (n. 16); the quotes are from 246, 263. 181  The issue here is connected to some familiar issues about the definitions of offences: should the criminal law’s special part aim to define offences in terms that reflect the ‘thick’ ethical concepts which structure our extra-legal understandings of the wrongs involved; or should it seek a more austerely descriptive, and more general, mode of definition? Contrast e.g. J. Gardner, ‘Rationality and the Rule of Law in Offences Against the Person’, Cambridge Law Journal, 53 (1994), 502, with P. H. Robinson, Structure and Function in Criminal Law (Oxford: Oxford University Press, 1997); see also R. A. Duff, ‘Theorizing Criminal Law’, Oxford Journal of Legal Studies, 25 (2005), 353.

52

R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros

which we could then derive conclusions about what we have in principle reason to criminalize, but with a more disorderly set of considerations that might bear on the question of criminalization (including, no doubt, some of the values that figure in would-be master principles), of starting points from which and routes by which we might be led to see reasons to criminalize certain types of conduct.182 But perhaps such a messy theory is what the messy worlds in which we live require.

VII.  Where Do We Go from Here? As the previous sections of this Introduction should have made clear, our own work on this project has so far been largely of a preparatory, ground-clearing kind. We have not yet tried, either collectively or individually, to develop a complete theory of criminalization; rather, we have tried to gain a clearer view of the phenomena to be theorized, and of what any theory will need to include (the need, for instance, to ground it in political theory; the need to show what distinctive role the criminal law should play in relation to other kinds of law and other modes of legal regulation). The contributors to the four volumes of papers that the project has produced, including this volume,183 have done substantial constructive work towards a more adequate theorizing of criminal law; our own constructive and systematic contributions will be made in the three monographs that will complete the project’s published outputs. In this final section we therefore briefly describe these monographs, as they are currently planned, to give an idea of the directions in which we expect to progress from here. Lindsay Farmer’s The Institution of Criminal Law has two main aims. First, it advances a theory of criminal law as an institution. Understanding criminal law as an institution requires that we recognize that it can in part be differentiated from other social and legal practices by its distinct aims and social functions. Criminalization is accordingly to be understood not only in terms of the interests or goods protected, but also in terms of the broader aims that the criminal law promotes and the way that the criminalization of particular practices or conduct contributes to those broader aims. Understanding modern law as an institution, however, also requires us to see it as a practice that is committed to certain kinds of values, such as the rule of law, respect for individual liberty. In criminal law these values have been institutionalized in distinctive ways—the commitment to legality or to requiring mens rea, for example—and these commitments have in turn come to shape the scope and aims of the law. The second aim of the book is to argue that understanding the practices and scope of criminalization in modern criminal law requires that we view the law in historical perspective. This is necessary in part because our understandings of interests or goods, how they must be protected, and what they must be protected against, have changed over time, and in part

182  See Duff, ‘Towards a Modest Legal Moralism’ (n. 66).

183  See n. 1.

Introduction

53

because the aims and functions of the criminal law have themselves changed over the course of the modern period. Antony Duff’s The Realm of Criminal Law will ground an account of criminalization in a liberal-communitarian conception of the role that the criminal law should play in the political structures of a contemporary democratic polity. This will involve developing an account of the ‘public’ realm, the res publica, of such a polity—an account that will consist largely in an account of the process of public deliberation through which the polity’s members will work out an understanding of the civic enterprise in which they are involved as citizens. Within that realm, the criminal law’s distinctive role—the role that marks it out from other modes of legal regulation—is, first, to define a set of public wrongs that merit a formal collective, and public response which makes their wrongfulness salient; and, second, to structure that public response through the criminal process that culminates in the criminal trial and verdict. If we then ask, as a theory of criminalization must ask, how we are to determine what is to count as a public wrong, and which such wrongs are to be in principle criminalizable, we will need to recognize that no simple answer is available. The most that a normative theorist will be able to do is to sketch the different starting points from which, the different routes by which, and the diverse values in the light of which, the citizens of such a polity could come to see that they have good reason to criminalize a type of conduct. Victor Tadros’s Wrongs and Crimes will outline a non-consequentialist approach to criminalization. The book will consider a number of different relationships between moral wrongdoing and the criminal law. It will investigate the nature of moral wrongdoing, and the appropriate response to it. In the light of this, some familiar principles of criminalization will be found wanting. These include principles concerning harm, sovereignty, liberty, and wrongdoing. It will then explore the implications of some central non-consequentialist ideas for the decision whether to criminalize conduct. It will show that it is sometimes permissible to criminalize conduct even if doing so does more harm than good. The role of consent to criminalization will be investigated, as will the best way to understand and justify inchoate offences and mala prohibita.

2 Quantifying Criminalization James Chalmers and Fiona Leverick*

Criminalization is a subject that has attracted a growing degree of academic interest in recent times. But while a considerable amount of attention has been devoted to what the criminal law should look like in terms of the principles that might guide criminalization decisions,1 the question of what it actually looks like has been comparatively neglected. Our aim in this contribution is to go at least some way to rectifying this imbalance by providing quantitative data on the number of criminal offences created in the United Kingdom within two sample time periods. At the outset, it is necessary to say something about what we mean by criminalization, as the term is one that can be understood in a number of different ways. We can start from the positivist position that a crime is simply something which is, by virtue of a legal rule, capable of being prosecuted by the state in the criminal courts.2 On that basis, we might say that everything which is capable of prosecution is criminalized, but this is too blunt an analytical tool to be of much value. At this point, we could adopt the helpful distinction offered by Nicola Lacey between ‘formal’ criminalization (the law in the books) and ‘substantive’ criminalization (the law in action).3 *  We are grateful to Peter Duff, Lindsay Farmer, Pamela Ferguson, and Findlay Stark for their helpful comments on an earlier draft of this chapter. 1  Aside from the chapters in this volume and in the previous collection of papers stemming from the same project (R. A. Duff et al. (eds.), The Boundaries of the Criminal Law (Oxford: Oxford University Press, 2010)), see D. Husak, Overcriminalization: The Limits of the Criminal Law (New York: Oxford University Press, 2008). 2  G. H. Gordon, The Criminal Law of Scotland (3rd edn. by M. G. A. Christie; Edinburgh: W. Green, 2000), 7. See also G. Williams, ‘The Definition of Crime’, Current Legal Problems, 8 (1955), 107, 130; L. Farmer, ‘The Obsession with Definition: The Nature of Crime and Critical Legal Theory’, Social and Legal Studies, 5 (1996), 57. The matter is complicated slightly by the fact that domestic classification is not conclusive for the purposes of the European Convention on Human Rights, so that if a state classifies a matter as civil in terms of its own domestic procedure but attaches consequences to that procedure which are penal in nature, the matter would be regarded as criminal for the purposes of the Convention. See R. A. Duff et al., ‘Introduction: The Boundaries of the Criminal Law’, in Duff et al., Boundaries of the Criminal Law, 3–4. 3  N. Lacey, ‘Historicising Criminalisation: Conceptual and Empirical Issues’, Modern Law Review, 72 (2009), 936, 943. See also N. Lacey and L. Zedner, ‘Legal Constructions of Crime’, in M. Maguire, R. Morgan, and R. Reiner (eds.), The Oxford Handbook of Criminology (5th edn.; Oxford: Oxford University Press, 2012), 159, 162.

Quantifying Criminalization

55

These categories, while analytically valuable, are not as straightforward as they might first appear. In its simplest form, formal criminalization takes place through the creation of criminal offences by legislative instrument, and the recognition of common law offences by the courts in jurisdictions which admit of such a possibility. However, the scope of existing offences may be modified by judicial decision. The courts or the legislature may modify general doctrines of the criminal law, consequentially altering the scope of a broad range of criminal offences.4 Similarly, evidential rules of general application may be modified, thus making it easier or more difficult to prove certain offences.5 Are developments of this sort also examples of formal criminalization (or decriminalization)? Even if we choose not to regard them as such, there is no doubt that they effect real changes to the scope of the criminal law. Substantive criminalization most obviously takes the form of prosecutions for the offence in question, but practices short of this might also be considered a form of criminalization. Most obviously, these might include alternatives to prosecution (such as prosecutor fines, police cautions, formal or informal warnings).6 Official agencies of various kinds might use the threat of prosecution as a tool to modify behaviour, either warning individuals away from behaviour which might lead to prosecution or threatening proceedings in respect of behaviour which has already occurred in order to achieve other goals.7 These may include official agencies which have powers of prosecution, others which have an established role of reporting potential wrongdoing to police or prosecutors, and others which have no such established role but might report a criminal offence in the same manner as a private individual. Substantive criminalization—or at least something akin to it—might even occur without any kind of official action at all, where actors take steps to comply with the criminal law of their own volition, or as a response to pressures from other private actors. It is, perhaps, intuitively odd to describe the last of these as ‘criminalization’, and a different term could legitimately be employed. However, it undoubtedly represents a substantive consequence of formal criminalization,8 and so is worthy of note in this context. It follows from all this that there is no necessary link between formal and substantive criminalization. In some cases, formal criminalization may result in little or none of the substantive variety. Official actors may lack the inclination or resources

4  See e.g. R v G [2003] UKHL 50 (on recklessness); Criminal Justice Act 1967, s. 8 (on criminal intent). 5  See Lacey, ‘Historicising Criminalisation’, 952. 6  The use of which has increased substantially in the UK in recent times. See J. McEwan, ‘From Adversarialism to Managerialism: Criminal Justice in Transition’, Legal Studies, 31 (2011), 519, 537–8 (England and Wales); R. M. White, ‘Out of Court and Out of Sight: How Often are “Alternatives to Prosecution” Used?’, Edinburgh Law Review, 12 (2008), 481 (Scotland). 7  On the regulatory use of criminal law in this way, see K. Hawkins, Law as Last Resort: Prosecution Decision-Making in a Regulatory Agency (Oxford: Oxford University Press, 2002). 8  This could, however, occur regardless of formal criminalization, for example where individuals faced with complex legislative regimes act in an overcautious manner in order to ensure that they do not fall foul of the criminal law.

56

James Chalmers and Fiona Leverick

to enforce the law on the books, or juries may refuse to convict. Elsewhere, the practical realities of securing evidence of the commission of certain crimes may make prosecution only a remote possibility. The occurrence of substantive criminalization may be remote in time from formal criminalization, as when an archaic statutory provision is dusted down and used in response to a new social problem,9 or where official actors decide to clamp down on behaviour formerly tolerated. Conversely, a criminal statute might be rigorously enforced for many years but fall into disuse as social values change, although the formal position remains identical throughout. For the purposes of this chapter our concern lies primarily with formal criminalization: the law as it exists on the books. Given the preceding discussion, one might wonder whether this focus is misplaced. We would argue that it is not. The formal content of the criminal might not map perfectly onto the law as it is applied, but this does not mean that it is unworthy of study. Even if an offence is rarely prosecuted or otherwise enforced by official authority, it may nevertheless have real-world consequences in restraining private behaviour. (Indeed, if one takes the view that the function of the criminal law is primarily a deterrent,10 it might be said that enacting a provision that is never prosecuted is the perfect outcome for legislators if it means that actors have changed their wrongful behaviour so as not to fall foul of the provision.) Yet while empirical data on at least some aspects of substantive criminalization—prosecutions, convictions, punishment—is readily available,11 we know very little even at the most basic level about the formal content of the criminal law, a fact that has often been lamented.12 It is sometimes claimed that we face a crisis of over-criminalization,13 but empirical evidence of this in terms of the number of new offences enacted is surprisingly elusive, at least in the UK context (and, it seems, elsewhere14). It is this issue that concerns us here, and in the remainder of the chapter we assess existing attempts to ‘quantify’ the criminal law in the UK before going on to present our own work in this area.

9  Such as section 127 of the Communications Act 2003, used to prosecute Paul Chambers, who posted a joke on Twitter about planting a bomb at an airport (although his conviction was ultimately overturned on appeal). Section 127 was originally enacted in 1935 to protect female telephonists working at the Post Office from nuisance callers: see J. Rowbottom, ‘Terror Tweet Case Threatens Free Speech’, The Guardian, 4 October 2010. 10  See e.g. V. Tadros, ‘Criminalization and Regulation’, in Duff et al., The Boundaries of the Criminal Law, 163. This view is, of course, a contested one. 11  In England and Wales, see Ministry of Justice, Criminal Justice Statistics (available at on an ongoing basis). In Scotland, see Scottish Government Statistical Bulletin, Criminal Proceedings in Scotland 2012–13 (Edinburgh: Scottish Government, 2013). 12 See e.g. K. Reid, ‘Strict Liability: Some Principles for Parliament’, Statute Law Review, 29 (2008), 173, 178; Lacey, ‘Historicising Criminalisation’, 951. 13  Husak, in Overcriminalization, has been the most notable proponent of this argument. See also A. Ashworth, ‘Is the Criminal Law a Lost Cause?’, Law Quarterly Review, 116 (2000), 225. 14  See e.g. in the US context Husak, Overcriminalization, 9–10.

Quantifying Criminalization

57

I.  Quantifying the Criminal Law in the UK: The Work to Date In 1980, JUSTICE published a report analysing the content of English criminal law. Recognizing the difficulty in identifying a complete set of offences, it adopted the pragmatic solution of analysing all those offences dealt with in the 1975 edition of Stone’s Justices’ Manual.15 As JUSTICE noted, there were three limitations to this exercise: (a) it did not include offences which existed only at common law; (b)  Stone’s Justices’ Manual is not a complete compendium, ‘and only describes in detail those offences which are sufficiently common for Magistrates’ Courts to need a ready guide for them’; (c) it did not include offences created by Public Local Acts or local by-laws.16 Even with these limitations, the research identified 7,208 discrete offences, ‘created by 466 statutes and 37 statutory instruments’,17 with the caveat that it was ‘probable that a complete list might contain far more than that’.18 Attempts have been made since then to analyse systematically the content of the criminal law, but have either tended to count only those offences created by primary legislation which, as our analysis will go on to show, accounts for a minority of criminal offences, or have relied on a filtered source. So, Ashworth and Blake, in a study of strict liability and reverse burdens, analysed ‘all offences recorded in the 1995 edition of Archbold as triable in the Crown Court’,19 a total of 540.20 Likewise, Ashworth identified 188 offences created by Westminster in 1995 and 165 in 2005,21 while Ferguson offered a ‘conservative estimate’ of 400 ‘offence provisions’ in Acts of the Scottish Parliament in a ten-year period.22 These studies were not, of course, designed to draw conclusions about the total number of offences in existence or created in a specific period: such figures as they produced were a by-product of other work. The Law Commission has suggested a rather different sort of measure, noting that the expansion of the criminal law can be evidenced by the number of pages in Halsbury’s Statutes of England and Wales: ‘more than 2 and a half times as many pages were needed . . . to cover offences created in the 19 years between 1989 and 15 JUSTICE, Breaking the Rules (London: JUSTICE, 1980); C. T. Latham and J. Richman (eds.), Stone’s Justices’ Manual (107th edn.; London: Butterworth, 1975). 16 JUSTICE, Breaking the Rules, para. 2.17. 17 JUSTICE, Breaking the Rules, 36. 18 JUSTICE, Breaking the Rules, para. 2.18. 19  A. Ashworth and M. Blake, ‘The Presumption of Innocence in English Criminal Law’, Criminal Law Review (1996), 306, 307, citing P. J. Richardson (ed.), Archbold: Criminal Pleading, Evidence and Practice (1995 edn.; London: Sweet and Maxwell, 1995). The authors noted (at 307) that Archbold did not purport to be a comprehensive account of such offences, but that ‘it may be taken to include the offences prosecuted most frequently’. 20  Ashworth and Blake, ‘The Presumption of Innocence in English Criminal Law’, 309. 21  A. Ashworth, ‘Ignorance of the Criminal Law, and Duties to Avoid It’, Modern Law Review, 74 (2011), 1, 7. 22  P. R. Ferguson, ‘Criminal Law and Criminal Justice: An Exercise in Ad Hocery’, in E. E. Sutherland et al. (eds.), Law Making and the Scottish Parliament: The Early Years (Edinburgh: Edinburgh University Press, 2011), 208, 216.

58

James Chalmers and Fiona Leverick

2008 than were needed to cover the offences created in the 637 years prior to that’.23 This is also a measure that would account only for offences created by primary legislation. Halsbury does produce a companion volume of statutory instruments, but this was not the Law Commission’s focus. Not all attempts to quantify the criminal law of the UK have ignored the role of secondary legislation. In August 2006, it was reported in the press that the UK government had created 3,023 offences since May 1997, comprising ‘1,169 introduced by primary legislation—debated in Parliament—and 1,854 by secondary legislation’,24 figures described as having been ‘uncovered’ by Nick Clegg, then home affairs spokesman for the Liberal Democrats.25 Exactly how these figures were arrived at is obscure. In the months preceding this report, Clegg had tabled a series of written questions in Parliament asking individual departments to list the criminal offences created by their departments, but the responses were generally limited to primary legislation and listed relatively few offences.26 The following month, Channel 4’s FactCheck examined the claim and concluded that it was on ‘pretty solid ground’, while suggesting that some offences would be amendments of existing ones, or would now be ‘effectively unused’, having been enacted, for example, in response to the foot and mouth crisis of 2001.27 The figure of 3,023 new offences attracted significant academic attention,28 but as we will shortly demonstrate it was almost certainly an underestimate. A more comprehensive set of figures was published in December 2011 by the Ministry of Justice, covering the creation of criminal offences by the Coalition government in England and Wales. According to these statistics, 174 criminal offences were created in the 12 months ending May 2011, compared to 712 in

23  Law Commission, Consultation Paper on Criminal Law in Regulatory Contexts (Law Com. CP No. 195, 2010), para. 1.17. 24  N. Morris, ‘Blair’s “Frenzied Law Making”: A New Offence for Every Day Spent in Office’, The Independent, 16 August 2006. 25  This seems to have continued work undertaken by Simon Hughes, Nick Clegg’s predecessor in this role: see H. Kennedy, Just Law: The Changing Face of Justice and Why it Matters to Us All (London: Chatto and Windus, 2006), 23. 26  The response offered by the Department for Environment, Food and Rural Affairs acknowledged that offences might have been created by subordinate legislation, but said that information on this ‘could be provided only at disproportionate cost’: HC Deb. 26 June 2006 col. 14W. Chris Huhne was later to attribute a figure of ‘no fewer than 3,600 new offences’ since May 1997 to ‘parliamentary answers to questions tabled by Liberal Democrats’, but at the same time as complaining of departments refusing to give full answers: HC Deb. 4 December 2008, col. 172. See also HC Deb. 8 February 2010, col. 606. 27  Channel 4 News, ‘Has Labour Really Created 3,000 New Crimes?’, 19 September 2006, available at . The foot and mouth crisis was an outbreak in the UK of a serious disease affecting cattle and sheep, which led to mass slaughter of affected animals. 28  See e.g. A. Ashworth and L. Zedner, ‘Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure and Sanctions’, Criminal Law and Philosophy, 2 (2008), 21, 22, 32; C. Pantazis, ‘The Problem with Criminalisation’, Criminal Justice Matters, 74 (2008), 10; A. Crawford, ‘Governing through Anti-Social Behaviour: Regulatory Challenges to Criminal Justice’, British Journal of Criminology, 49 (2009), 810, 826.

Quantifying Criminalization

59

the 12 months ending May 2010.29 The Ministry’s figures include primary legislation and secondary legislation and, helpfully, are accompanied by the raw data on which they are based.30 But even these figures underestimate the extent of criminalization, for several reasons. First, they (deliberately) omit offences created by statutory instrument but applying only at a ‘limited local level’: that is, by-laws, transport and Work Act orders, and harbour orders.31 Our own figures indicate that 52 offences were created by such legislation in 2010–11.32 Secondly, the Ministry’s figures appear simply to omit 14 pieces of secondary legislation which should properly have been included.33 These 13 instruments created 164 offences. So far as we can tell, there is no pattern to the non-inclusion of these instruments: it is not connected either to the date of their creation or their subject matter. Thirdly, in relation to both primary and secondary legislation, the number of offences has been undercounted, as the Ministry treated any section of the legislative instrument which creates an offence as a single offence regardless of how many offences it actually creates. As we will go on to argue,34 this is not a defensible method of quantifying criminal offences, given that a single offence-creating section of a statute may contain numerous clauses prohibiting diverse types of conduct, targeted at distinct audiences and resulting in different consequences (such as maximum penalties). The fact that we know so little about how many criminal offences exist might seem embarrassing. It suggests that those of us who teach criminal law do so with a rather sketchy knowledge of what the criminal law actually is—our understanding of the subject being based on certain general principles and working assumptions which may or may not be reflected in the body of law as a whole. But does the number of offences matter?

II.  The Limitations of Numbers In order to answer this question, we must first note some caveats. While the fact that we cannot readily quantify the number of offences in existence might be cause for alarm about the state of the criminal law, knowing it might not take us very far. First, even the most careful assessments of the number could vary, because it can be surprisingly difficult to identify accurately the number of offences created by a piece of legislation, a matter to which we return later in this chapter. But leaving aside that difficulty, there is no identifiably correct number of offences which should ideally exist. Even if we accept that there are at present too many offences, 29  Ministry of Justice, New Criminal Offences: England and Wales 1st June 2009–31st May 2011 (London: Ministry of Justice, 2011). 30  At . 31  Ministry of Justice, New Criminal Offences, 4. 32  The creation of criminal offences at a ‘limited local level’ is a curious phenomenon which we cannot explore further here. Our methodology records when offences are limited in this way, so as to allow for future analysis. 33  These are as follows: SI 2010/1554, SI 2010/1882, SI 2010/2079, SI 2010/2225, SI 2010/2228, SI 2010/2281, SI 2010/2035, SI 2010/2503, SI 2011/84, SI 2011/657, SI 2011/659, SI 2011/1301, SI 2011/1296, SI 2011/1297. 34  See text accompanying nn. 46–58.

60

James Chalmers and Fiona Leverick

it would not follow from this that a reduction in their number would be necessarily a positive development. Suppose that the government insisted, in a possibility canvassed by the Law Commission, ‘that every relevant Government department reduce its stockpile of criminal offences by 25% over (say) five years’.35 The cure might be worse than the disease. It is conceivable that a government department might achieve this target not by decriminalization, but instead by amalgamating different offences together into broader provisions which in fact criminalized a greater range of conduct than the provisions which they replaced. As that suggests, the total number of criminal offences is not in itself an accurate measure of the extent of criminalization. Suppose that jurisdiction A has a single criminal offence of homicide, while jurisdiction B has offences of first-degree murder, second-degree murder, and manslaughter.36 The extent to which jurisdictions A and B criminalize homicide may be identical, save that jurisdiction B does so with three times as many criminal offences. There may even be good reasons, such as considerations of fair labelling37 and certainty,38 for preferring jurisdiction B’s approach. That said, there clearly is a relationship between the number of criminal offences and the extent of criminalization. Assuming a reasonable stability in drafting techniques, this relationship may allow for meaningful if somewhat limited comparisons over time within a single jurisdiction, although it is less likely that it will allow for meaningful cross-jurisdictional comparisons. But perhaps more importantly, the number of criminal offences is a mere aspect (and perhaps not a very interesting aspect) of any analysis. The problem is that we have little systematic knowledge about why or when lawmakers resort to the criminal law, and the fact that no one can say how many criminal offences exist is little more than a symptom of this much larger problem. As Douglas Husak comments:39 Penal statutes rarely are read, even by those who purport to theorize about them. In fact, it is safe to say that virtually no one systematically studies criminal codes. . . . Professors seldom assign them in class, so students are unlikely to examine them. Criminal theorists remain obsessed with the so-called general part of the criminal law, and they are far more fascinated by the philosophical foundations of the criminal law than by the criminal law itself. As a result, much of their theorizing loses touch with that body of law they purport to be theorizing about.

Or, as William Stuntz observes, criminal law ‘is not one field, but two’—‘core crimes’ and ‘everything else’—with the first dominating criminal law teaching and scholarship, and the second dominating criminal codes.40 35  Law Commission, Criminal Law in Regulatory Contexts, para. 3.16. The Commission noted that this proposal, ‘whatever might be said about its merits’, was not one which could be put forward by the Commission, because it would be a ‘policy decision’. 36  An example canvassed by the Law Commission, Criminal Law in Regulatory Contexts, para. 3.18. 37  On which, see J. Chalmers and F. Leverick, ‘Fair Labelling in Criminal Law’, Modern Law Review, 71 (2008), 217. 38  On which, see A. Ashworth and J. Horder, Principles of Criminal Law (7th edn.; Oxford: Oxford University Press, 2013), 62–5. 39 Husak, Overcriminalization, 33. 40  W. Stuntz, ‘The Pathological Politics of Criminal Law’, Michigan Law Review, 100 (2001), 505, 512. For a discussion of the implications of this distinction, and what the ‘core’ of criminal law might be thought to contain, see Husak, Overcriminalization, 33–4.

Quantifying Criminalization

61

That is not to say that criminal law scholarship is blind to Stuntz’s ‘everything else’. Criminal lawyers will frequently engage in critical analysis of overly broad provisions,41 or cite relatively obscure statutory provisions to demonstrate some absurdity.42 But such examples tell us little about the system of the criminal law. Are they isolated difficulties, or symptoms of systematic problems? What does ­‘everything else’, taken as a whole, look like? Without systematic analysis, we risk the danger of doing little more than trading anecdotes about whatever oddities have happened to catch our eye. But systematic analysis is complex and time consuming, and so it is unsurprising that relatively little has been done.

III.  Systematic Analysis: Making a Start In the research reported here,43 we have attempted to establish a methodology for a systematic analysis of the extent of the criminal law. Importantly, we have not relied on a ‘filtered’ source such as a published compendium of legislation or other text, nor have we restricted ourselves to primary legislation. Instead, we have reviewed all Acts of Parliament, Acts of the Scottish Parliament, statutory instruments, and Scottish statutory instruments over two specific time periods of one year.44 The time periods selected were the first year of the current Coalition government (that is, all offences created from 6 May 2010 to 5 May 2011) and, by way of comparison, the first year of the New Labour government (from 2 May 1997 to 1 May 1998).45 Our research does not, therefore, say anything about the total number of criminal offences in existence in the UK: that would require a much more comprehensive and longer-term project, although our research demonstrates that this 41  See e.g. J. Hodgson and V. Tadros, ‘How to Make a Terrorist out of Nothing’, Modern Law Review, 72 (2009), 984, discussing s. 58(1) of the Terrorism Act 2000. 42  See e.g. P. Alldridge, ‘Making Criminal Law Known’, in S. Shute and A. P. Simester (eds.), Criminal Law Theory: Doctrines of the General Part (Oxford: Oxford University Press, 2002), 103, 103 n. 5, noting that s. 362 of the Insolvency Act 1986 meant that anyone who becomes bankrupt was guilty of a criminal offence if they had bought a single lottery ticket in the previous two years. Commendably, this offence was repealed by the Enterprise Act 2002. The conduct previously cri­ minalized by s. 362 is no longer an offence in itself, but may be taken into account by the court in deciding whether to grant a bankruptcy restrictions order. See Enterprise Act 2002, s. 257 and Sch. 4A. 43  We are grateful to the University of Edinburgh School of Law (where James Chalmers worked during the initial stages of this research) for providing funding through its Strategic Investment Fund which allowed us to employ Peter Lewin as a research assistant, and we are in turn grateful to Peter for his diligent efforts at a stage where our methodology was still under development. 44  While this approach is significantly more comprehensive than other attempts, it remains incomplete, because criminal offences may be created by other means. The Law Commission estimated that there are ‘now over 60 national regulators’ with the power to make criminal law, alongside trading standards authorities and 486 local authorities: Law Commission, Criminal Law in Regulatory Contexts, para. 1.21. We excluded legislation which was applicable only to overseas or dependent territories, but included legislation applying to the Channel Islands and Isle of Man. 45  For these purposes, we treated the government’s first year in office as having begun the day after the relevant General Election, and treated an offence as having been created on the day on which the relevant statute received Royal Assent or the statutory instrument was made. Where appropriate, we recorded separately the date on which the relevant provision came into force and on which it ceased to be in force.

62

James Chalmers and Fiona Leverick

would be a feasible one. The aims of our current project have been twofold: first, to establish a methodology and secondly, to show how the criminal law is currently used as a legislative device. In this chapter, we report on three aspects of this project. First, we explain some of the basic methodology involved in our work. Secondly, we present an overview of what our research has demonstrated about the two time periods which we assessed. Thirdly, we offer some brief conclusions.

IV.  Methodology (1): What is an Offence? In order to collect data about each criminal offence created by legislation, it is necessary to determine just what ‘an’ offence is.46 In most cases, this will be reasonably clear. Where different sections of a legislative instrument contain distinct provisions defining distinct acts or omissions as offences, it is clear that multiple offences have been created. This is so even though some offences may overlap with others or even be redundant: for example, many regulatory statutes create offences of making false statements which seem simply to restate the general law of fraud in a specific context.47 Such offences have no effect on the total amount of conduct which is caught by the criminal law48 (which raises the question of why they should ever be enacted),49 but they are clearly distinct offences and we have analysed them as such. There can, however, be rather more difficulty in assessing whether a single section of a legislative instrument should be treated as creating a single offence or multiple offences. Take, for example, the innocuous looking regulation 91(a) of the Feeding Stuffs (Establishments and Intermediaries) Regulations 1998.50 This states simply that it shall be an offence for a person ‘without reasonable excuse, to contravene any of the regulations 33 to 81’. Regulations 33 to 81 are a series of distinct restrictions, normally commencing in the form ‘No person shall . . .’ or similar, before describing a prohibited activity. While at first glance, this might be thought to be a single offence (regulation 91(a)), this can only rationally be understood as creating 48 separate offences. There are 48 different prohibited actions which can lead to criminal punishment, set out in a ‘laundry list’ form before regulation 91(a) criminalizes every one of them. These prohibitions are directed to 46  The difficulties involved in this seemingly simple exercise are noted by Stuntz, ‘Pathological Politics’, 514 and Lacey, ‘Historicising Criminalisation’, at 951–2. 47  See Law Commission, Criminal Law in Regulatory Contexts, paras 3.123–3.128. 48  This is not true for offences which are ‘fraud-like’ but are easier to prove than the offence of fraud would be (perhaps because the mens rea requirement is a lesser one). In turn, that raises the question of whether it is appropriate to impose such a lower threshold of liability: from the extent to which this technique is seen in practice, it seems unlikely that this question receives the attention which it should as part of the drafting process. 49  We have discussed elsewhere possible justifications for creating specific criminal offences which overlap with more general ones in this way: see Chalmers and Leverick, ‘Fair Labelling’, both generally and in particular at 229 and 231. See also Husak, Overcriminalization, 36–8. 50  SI 1998/1049.

Quantifying Criminalization

63

varying audiences: sometimes ‘any person’ but sometimes ‘intermediaries’ or intermediaries who have been ‘approved’ in specific ways. The applicable penalties vary depending on which regulation has been contravened;51 some contraventions are prosecutable on indictment but others only on summary complaint;52 some contraventions are subject to a six-month time bar on prosecution but some are not.53 Elsewhere, such distinctions are harder to make. Regulation 21 of the Merchant Shipping and Fishing Vessels (Health and Safety at Work) Regulations 199754 sets out a number of ‘general duties’ of workers aboard a ship to which the Regulations apply, ranging from the general (e.g. taking reasonable care for the health and safety of themselves and other workers)55 to the more specific (e.g. not disconnecting, removing, or interfering with a safety device; reporting deficiencies in health and safety arrangements to the relevant person).56 Failure to comply with regulation 21 is punishable by a fine,57 but is this one offence or several? In our analysis, we have treated it as a single offence. There is not a clear distinction between the various duties set out in regulation 21, and a person could well be in breach of a number of them simultaneously. Nor are there any consequences, either in terms of sentencing or procedure, attached to precisely which aspect of regulation 21 the accused may be said to have breached. As these examples illustrate, our general approach has been to ask whether the relevant legislative provision sets out a clear and distinct act or omission, separable from other specified acts or omissions, which would amount to a breach of the criminal law. We refer to this as the test of separability. If so, that is a separate offence.58 This conclusion should not depend on drafting technique. For example, the Feeding Stuffs Regulations might have been drafted differently, so that each of regulations 33–81 were in the format ‘Any person who [engages in the specified conduct contained in this regulation] commits an offence’. It would be wrong to say that two drafting techniques with identical consequences should be regarded in one case as creating a single offence and the other 48. Because, by contrast, the various aspects of regulation 21 of the Merchant Shipping Regulations are not clearly separable in this way, we have treated them as amounting to a single offence. In part, our approach is motivated by a desire to err on the side of caution: we do not want to overstate the number of criminal offences created by legislation. We refer to this as the principle of parsimony.

V.  Methodology (2): What Data can be Captured in Relation to Each Offence? Once individual offences have been identified, the next step in our analysis was to characterize various aspects of them so as to allow for analysis. In respect of each offence, we recorded inter alia the following data: the applicable mode of trial 51  r. 92(1)–(2). 52  r. 92(1)–(2). 53  r. 93(1). 54  SI 1997/2962. 55  r. 21(1)(a). 56  rr. 21(2)(b) and 21(3). 57  r. 24(3). 58  A conclusion which is fortified where there are different procedural consequences depending on which particular act or omission is in issue.

64

James Chalmers and Fiona Leverick

(indictment, summary, or either way); the maximum penalty associated with the offence; the geographical extent of the offence; any special capacity associated with the offence; whether the offence was one of commission or omission; the extent to which mens rea was required; whether the offence applied extraterritorially; the ‘subject matter’ of the offence; and what the ‘presumed public policy’ associated with the offence appeared to be. In assessing the last of these, we adopted the categories of public policy identified by JUSTICE in their 1980 report, with a few additions to accommodate purposes which seemed to us to be apparent in modern practice.59 Some of these assessments involved the exercise of a degree of judgment. This is depressingly necessary in identifying even the basic elements of an offence: it is not always clear, for example, whether an offence creates a reverse burden of proof, to the extent that we were ultimately unable to record any useful data on this issue. There is no good reason why this should be left unclear in legislative drafting, but it is. It is also often unclear whether or not mens rea is required in relation to a particular offence. We recorded only whether a mens rea requirement was specified and did not speculate on whether one might be read into the offence by the courts.60 It should be remembered that we did not have to worry whether we might be at risk of committing these offences. Citizens who are at risk of being punished by the state should not be left in the dark as to their potential liabilities in this way. We were also faced with the problem that some criminal offences are created by remarkably inaccessible and tortuous forms of legislative drafting.61 Consider, for example, the Parliamentary Voting System and Constituencies Act 2011, which made provision for the Alternative Vote referendum.62 Schedule 4 of that Act provided that various provisions of the Representation of the People Act 1983, including many criminal offences, should apply in respect of the referendum. However, Schedule 4 also listed a whole series of ‘modifications’—amendments or substitutions of entire provisions—which should apply to the 1983 Act for this purpose. The consequence is that trying to work out exactly what electoral offences existed as a result of the 2011 Act is an extremely cumbersome process. There is no amended text of the 1983 Act available, because the 2011 Act does not actually amend the 1983 Act—it merely created a modified zombie version of the 1983 Act for a limited time. Anyone seeking to understand the offences created must read the two statutes side by side, substituting words and provisions where appropriate.63 59  See JUSTICE, Breaking the Rules, para. 2.13. 60 See the discussion of the presumption of fault by the Law Commission, Criminal Law in Regulatory Contexts, paras 6.9–6.18. 61 On the drafting of criminal provisions, see J. R. Spencer, ‘The Drafting of Criminal Legislation: Need it be so Impenetrable?’, Cambridge Law Journal, 67 (2008), 585; P. R. Ferguson, ‘The Drafting of Offence Provisions by the Scottish Parliament’, Statute Law Review, 32 (2011), 161. 62  A referendum on a proposal to change the voting system for elections to the United Kingdom Parliament: see Parliamentary Voting System and Constituencies Act 2011, s. 1. The proposal was rejected. 63  That in turn raises the question of whether the 2011 Act can be said to create new offences, or simply to amend existing ones. On balance, given the extent to which the 2011 Act amended the text of the 1983 Act for its own purposes, we felt it fair to treat it as creating new offences specific to the referendum.

Quantifying Criminalization

65

For the purposes of our research, such difficulties can be overcome with time, but this particular drafting technique seems to contribute little to the requirements of accessibility or fair notice. Similar difficulties arise with complex provisions which cross-refer to European legislation, and which give no indication on their own face of exactly what conduct is prohibited.64

VI. An Overview of the Research: How Many New Offences? Table 2.1 presents the number of criminal offences created by statute or statutory instrument in the two time periods under review. As Table 2.1 demonstrates, a total of 1,395 criminal offences were created in the New Labour sample and 1,760 in the coalition sample. These figures are astonishing: remember, as we noted earlier, the claim that the government had created 3,023 offences in the ten years following May 1997, a rate of just under one offence per day. Our research suggests that the Coalition government in fact created offences at a rate which was over four times faster than that. However, our research also suggests that the Labour government created criminal offences at a far faster rate than previously claimed. We identified 1,395 offences created solely in the year following 1997, which suggests that the figure of 3,023 over ten years is a significant underestimate. Because the data underlying the 3,023 figure appears nowhere to have been published, it is difficult to tell why this is so. Perhaps the most likely explanation is that it failed properly to identify when a single section of a statute actually creates multiple offences, as we discussed earlier. The figures presented in Table 2.1, however, overstate the extent of the criminal law created during that time period, because many of the legislative instruments concerned will extend only to part of the UK. In some cases, the implementation of European legislation will have resulted in our research identifying three distinct Table 2.1  Offences created in 1997–1998 and 2010–2011

Statute Statutory Instrumenta Total

1997–1998

2010–2011

18 1,377 1,395

247 1,513 1,760

a  Including Scottish Statutory Instruments.

64  And, in at least one case, to seemingly non-existent European legislation: see The Marketing of Fresh Horticultural Produce Regulations 2009, SI 2009/1361. Regulation 4(3) makes it a criminal offence to (inter alia) fail to comply with art. 4(1) of Commission Regulation 1580/2007 when read with art. 4(3) or art. 4(4) of the same regulation. The Commission Regulation has no art. 4(3) or 4(4). The Regulations also appear to make it a criminal offence for traders to fail to comply with art. 10(4) of the Commission Regulation, but art. 10(4) is not directed at traders.

66

James Chalmers and Fiona Leverick

Table 2.2  Offences applicable to England created in both the study years Geographical extent

England England and Wales Britain UK Total

1997–1998

None None 213 1,022 1,235

2010–2011

212 9 4 409 634

Table 2.3  Offences applicable to Scotland created in both the study years Geographical extent

Britain Scotland UK Total

1997–1998

2010–2011

213 3 1,022 1,238

4 810 409 1,223

offences covering the same act or omission: one applicable to Wales, one to England, and one to Scotland.65 This is a result of the devolution process which led to the establishment of the Scottish Parliament66 and the Welsh Assembly67 in 1999, both of which have law-making powers in relation to certain matters, known as devolved matters.68 Thus in an area such as agriculture, which is a devolved matter in both Scotland and Wales, the relevant measures may be implemented separately for England, Scotland, and Wales, so that there is a danger of triple-counting offences.69 Table 2.2 takes account of this difficulty by identifying only criminal offences which were applicable to England (regardless of whether they also extended to another part of the UK). Table 2.3 performs a similar exercise for Scotland. As Tables 2.2 and 2.3 demonstrate, even when the issue of geographical duplication is accounted for, the number of criminal offences created remains strikingly high and is notably higher than that contained in previous estimates. The Ministry of Justice, as we noted earlier, concluded that 174 new criminal offences were created in the 12 months ending May 2011 by primary or secondary legislation taking effect 65  Occasionally European legislation is also implemented separately for Northern Ireland, resulting in four such offences, although there were no examples of this in our two sample time periods (for an example that fell outside our time scale see n. 70). 66  Under the Scotland Act 1998. 67  Under the Government of Wales Act 1998. 68  The Scottish Parliament was given law-making powers upon its establishment and can legislate on any issue unless it is a ‘reserved matter’ (see Sch. 5 of the Scotland Act 1998). The Welsh Assembly did not initially have law-making powers but gained them under the Government of Wales Act 2006 (see Sch. 7 for a list of the devolved areas). 69  See e.g. the Beef and Veal Labelling Regulations 2010, SI 2010/983; the Beef and Veal Labelling (Scotland) Regulations 2010, SSI 2010/402; and the Beef and Veal Labelling (Wales) Regulations 2010, SI 2011/991 (W 145). The Beef and Veal Labelling (Northern Ireland) Regulations 2010 (SI 2010/155) implemented the same European legislation in Northern Ireland but did not receive Royal Assent during the relevant period so were not included in our database.

Quantifying Criminalization

67

in England and Wales.70 Our research, by contrast, identified 634 criminal offences applying to England which were created in a near-identical time period. Some of the reasons why our figures differ so dramatically from those produced by the Ministry have already been noted,71 and we do not intend to explore them further here.72 Two further points are, however, obvious from these figures: first, there were significantly fewer criminal offences created applying to England and Wales in 2010–11 than in 1997–8; and secondly, there were almost twice as many criminal offences created in 2010–11 applying to Scotland than was the case in England and Wales.73

VII.  The Growth of the Criminal Law: Whose Responsibility? One possible explanation for the growth of the criminal law is that governments are constrained by external obligations to criminalize. For example, the offence of ‘causing a nuclear explosion’ under the Nuclear Explosions (Prohibitions and Inspections) Act 199874 was publicly criticized, with the then Liberal Democrats’ home affairs spokesman, Chris Huhne, arguing that this could be covered by a ‘whole raft of offences’ already in existence.75 In response, the Home Secretary (Jack Straw) pointed out that the offence had been created in order to meet the UK’s obligations under an international treaty.76 For each of the criminal offences which we identified, we recorded whether its creation appeared to be required in order to implement a European Directive or an international obligation. We determined this by looking at the text of the legislation itself77 and any Explanatory Notes attached to it. This exercise may be incomplete: the text of legislation or any notes accompanying it do not necessarily disclose, even by implication, the factors which motivated it. Our figures may, therefore, be an underestimate, although if this is the case it is likely to be a

70  Ministry of Justice, New Criminal Offences. 71  See text accompanying nn. 31–4. 72 For a more detailed discussion, see J. Chalmers and F. Leverick, ‘Tracking the Creation of Criminal Offences’, Criminal Law Review, 543 (2013), 543. 73  Space precludes an exploration of this difference here. See further J. Chalmers and F. Leverick, ‘Scotland: Twice as Much Criminal Law as England?’, Edinburgh Law Review, 17 (2013), 376. 74  Section 1. 75  ‘Jack Straw rejects call to repeal “trivial laws” ’, BBC News Online, 22 January 2010. 76 ‘Jack Straw rejects call to repeal “trivial laws” ’, BBC News Online, 22 January 2010: the Comprehensive Nuclear-Test-Ban Treaty (adopted 10 September 1996) (1997) 35 ILM 1439, art. I(1) of which commits each state party ‘not to carry out any nuclear weapon test explosion or any other nuclear explosion, and to prohibit and prevent any such nuclear explosion at any place under its jurisdiction or control’. It might be claimed that existing laws already had this effect, but there are obvious reasons why the UK might want to be able to point to a specific piece of legislation implementing this obligation. 77  The long title of the legislation is often informative in this respect. For example, the long title of the Nuclear Explosions (Prohibitions and Inspections) Act 1998 is ‘An Act to enable effect to be given to certain provisions of the Comprehensive Nuclear-Test-Ban Treaty adopted in New York on 10th September 1996 and the Protocol to that Treaty; and for connected purposes.’

68

James Chalmers and Fiona Leverick

Table 2.4  Did the criminal offence implement an international obligation?

No Yes—European Directive Yes—international obligation Yes—international obligation implemented at EU level Total

1997–1998

2010–2011

128 (9%) 947 (68%) 320 (23%) None 1,395

529 (30%) 1,043 (59%) 84 (5%) 104 (6%) 1,760

relatively minor one. Nevertheless, this exercise gives us a useful sense of the source of new criminal laws, as indicated by Table 2.4. As Table 2.4 shows, the majority of criminal offences created in both the New Labour sample and the coalition sample arose from an international or European obligation placed on the UK. Most commonly this arose where the UK was implementing a European Directive, a process that accounted for 68 per cent of offences created in the New Labour sample and 59 per cent of offences in the coalition sample (although if international obligations implemented at EU level are included in this latter figure it rises to 65 per cent). Previous research into the impact of European regulation on domestic legislation in the UK has generally concluded that its impact to date has been relatively limited and certainly far lower than some sections of the UK media perceive it to be. Page, for example, concluded that only 15 per cent of (non-local) statutory instruments issued between 1987 and 1997 made reference to European legislation.78 Page’s study, however, counts only the number of legislative instruments that have stemmed from EU regulation and does not, as we have done, attempt to calculate the number of criminal offences contained within those instruments, which may explain why our figure is so much higher. The Law Commission has suggested that the UK is too quick to create criminal offences in order to implement European Directives.79 We do not intend to enter that debate here, but would note that European law, while responsible for the clear majority of new criminal offences, is not itself a full explanation. Indeed, the number of new criminal offences not required by European law increased by a far greater extent between the two study years: from 128 (9 per cent of the New Labour sample) to 529 (30 per cent of the coalition sample). This might suggest that the ‘influence’ of Europe on the UK’s criminal law is waning, but to reach any firm conclusion on this issue would require a greater range of longitudinal data than we presently have at our disposal. What is also clear is that the influence of the EU is confined to, or at least is most apparent in, a relatively limited number of areas. The vast majority of criminal offences ‘imposed’ upon the UK 78 E. Page, ‘The Impact of European Legislation in British Policy Making: A Research Note’, Public Administration, 76 (1998), 803, 805. See also V. Miller, ‘How Much Legislation Comes from Europe?’ (House of Commons Research Paper 10/62, 2010) who reaches a similar conclusion (at 21). Page’s figures are consistent with those from other European countries: see W. C. Müller et al., ‘Legal Europeanization: Comparative Perspectives’, Public Administration, 88 (2010), 75, 77. 79  Law Commission, Criminal Law in Regulatory Contexts, paras 3.103–3.108.

Quantifying Criminalization

69

by the EU were (unsurprisingly) in the areas of agriculture, food production, the environment,80 health and safety at work, and animal welfare.81

VIII.  What are All These New Criminal Offences About? In what areas is the criminal law used? To explore this question, we categorized every new offence according to the categories used in Halsbury’s Statutes,82 with a few additions to avoid gaps or inappropriate categorization.83 Pamela Ferguson, in her study of criminal offences created by primary legislation by the Scottish Parliament in its first ten years, noted that some criminal law statutes ‘paint a picture of Scotland as a largely rural nation concerned with land management and animal welfare, which is what one might expect in a country in which agriculture plays a prominent role’.84 As these figures show, the picture is broadly similar for the UK as a whole. The largest category by far in both samples is that of agriculture, which accounts for 30 per cent of the offences in the New Labour sample and 32 per cent in the coalition sample. If animal welfare and food production are added to agriculture, these figures rise to 45 per cent (New Labour) and 43 per cent (coalition). The vast majority of these offences, however, emanate from European obligations. Of the 420 criminal offences created in the area of agriculture in the New Labour sample, 400 of these were created to implement a European Directive. The proportion is similar for the coalition sample (560 of 569). One of the other substantial categories is that of terrorism, which accounts for 11 per cent of the offences in both samples. Once again, however, these offences have not tended to arise from purely domestic policymaking: every single one of the offences created in the New Labour and coalition samples stemmed from an obligation upon the UK to implement international sanctions aimed at preventing terrorism.85 While offence creation in the areas of agriculture and terrorism is consistently high across the two samples, this is not the case for some of the other subject areas. Health and safety at work, for example, accounted for 13 per cent of offences in the New Labour sample but only 3 per cent in the coalition sample. This is unlikely to represent any great societal shift in priorities: of the 348 health and safety offences in the New Labour sample, the vast majority were contained in four statutory instruments implementing the UK’s international obligations in 80  For a discussion of the impact of European law in this area, see A. Ross, H. Nash, and C. T. Reid, ‘The Implementation of EU Environmental Law in Scotland’, Edinburgh Law Review, 13 (2009), 224. 81  On this, see also Page, ‘Impact of European Legislation’, 805; Müller et al., ‘Legal Europeanization’, 82. 82  This is a commercially produced compendium of legislation in England and Wales, first published in 1929. 83  The additions are fishing, food production (excluding agriculture), parliament/elections, and terrorism/international sanctions. 84  Ferguson, ‘Criminal Law and Criminal Justice’, 216–17. 85  Which is not to say that the UK government has not created terrorist offences unilaterally in recent times. It clearly has (see, for example, the offences contained in the Terrorism Act 2000) but these did not fall into either of our two sample time frames.

70

James Chalmers and Fiona Leverick

various aspects of merchant shipping.86 The over-representation of offences relating to elections in the coalition sample (170 offences compared to none in the New Labour sample) can be attributed to the fact that two referendums took place during this time period: the Alternative Vote referendum87 and the referendum on increased law-making powers for the Welsh National Assembly.88 The former resulted in the creation of 75 offences under the Parliamentary Voting System and Constituencies Act 2011 (the tortuous method of drafting used in this legislation was discussed earlier) and the latter in the creation of 38 offences under the National Assembly for Wales Referendum (Assembly Act Provisions) (Referendum Question, Date of Referendum Etc.) Order 2010. Likewise, the disparity between the two samples in the area of health and care regulation (no offences in the New Labour sample compared to 131 in the coalition sample) can be attributed predominantly to the 91 criminal offences contained in a single piece of legislation, the Independent Health Care (Wales) Regulations 2011.89 At this point, we offer these figures simply as an indication of the spheres in which governments are most likely to resort to criminalization. An extension of our study to cover a broader time span would allow us to explore in more detail how patterns of criminalization have developed over time.

IX.  How Punitive are These Offences? Although the figures in Table 2.5 demonstrate that criminalization—at least in terms of the number of offences created—is directed in large part towards a number of specific fields such as agriculture and food production, we should not be too quick to dismiss these offences as purely ‘regulatory’ in nature. Any conviction for a criminal offence is a serious matter, no matter how ‘regulatory’ it might seem90 and many so-called ‘regulatory’ offences actually criminalize significantly harmful (or potentially harmful) conduct.91 86 The Merchant Shipping (Cargo Ship Construction) Regulations 1997 (147 offences); the Merchant Shipping (Fire Protection: Large Ships) Regulations 1998 (115 offences); the Merchant Shipping and Fishing Vessels (Health and Safety at Work) Regulations 1997 (31 offences); and the Merchant Shipping (Safe Manning, Hours of Work and Watchkeeping) Regulations 1997 (30 offences). 87  See n. 62. 88 The National Assembly for Wales Referendum (Assembly Act Provisions) (Referendum Question, Date of Referendum Etc.) Order 2010, SI 2010/2837. The proposals were approved. 89  SI 2011/734 (governing the regulation and inspection of healthcare providers in Wales). 90  Particularly so if it results in someone with no previous convictions acquiring a criminal record. For a recent example, see J. Hall, ‘Olympic Flag Seller Given Criminal Record’, The Telegraph, 3 October 2012, reporting on the case of an architecture graduate convicted of unlawful street trading after assisting a friend’s business by selling ‘assorted flags’ during the Olympic games. The £45 fine (albeit accompanied by a £15 victim surcharge and £336 towards prosecution costs) may well be a trivial matter compared to the consequences of a criminal record for someone newly entering the job market. The newspaper report indicates that the defendant, after being sentenced, ‘raised her hand and asked if she now had a criminal record. The judge replied that she did and she left the courtroom in tears.’ 91  S. Green, ‘Why it’s a Crime to Tear a Tag off a Mattress: Over-Criminalisation and the Moral Content of Regulatory Offences’, Emory Law Journal, 46 (1997), 1533, 1565; Ashworth, ‘Is the Criminal Law a Lost Cause?’, 243.

Quantifying Criminalization

71

Table 2.5  Criminal offences by subject matter 1997–1998 2010–2011

Agriculture (including farming and horticulture) Animals (general animal welfare, veterinary medicine) Armed forces (including weapons) Children Courts and legal services Criminal law (general) Environment (including energy conservation and pollution control) Fishing Food production (excluding agriculture) Health and care regulation Health and safety at work (including on ships) Land, tenants, and housing Nature conservation (including forestry but excluding animals) Parliament / elections Police (including prisons, private security) Roads, railways, and transport Sale of goods Shipping and navigation (including port management) Terrorism / international sanctions Water (supply of, excluding nature conservation issues) Othera

420 134 3 None None 8 127 52 88 None 348 1 3 None 9 13 3 19 158 None 9

569 32 33 8 9 23 54 41 148 131 45 22 20 170 7 21 34 8 188 171 26

a

  1997–8: aviation (4); financial services (4); matrimonial law (1). 2010–11: companies, commerce, and competition (4); culture, entertainment, and sport (4); education (4); local government (2); matrimonial law (1); nationality and immigration (1); planning and compulsory acquisition (5); registration concerning the individual.

An indication of the relative seriousness of the offences in our samples92 is contained in Table 2.6, which sets out the maximum penalty available on conviction for the offences created in the two study years. As Table 2.6 shows, the penalties available on conviction for the many offences created annually are significant: most of them are imprisonable (65 per cent of the New Labour sample and 56 per cent of the coalition sample). Measured by this yardstick, it is encouraging to see that the criminal offences created in 2010–11 appeared to be slightly less punitive than their counterparts in 1997–8. This dif­ ference is also reflected in the maximum term of imprisonment available on conviction: 24 per cent of imprisonable offences created in 1997–8 were punishable by a maximum of six months or less, compared to 38 per cent in 2010–11. One feature of legislative practice is that a significant number of offences with very heavy maximum penalties are now created by means other than an Act of Parliament. This is surprising. As the Law Commission has suggested, ‘the question of when a criminal offence should be created is one with some 92  As Ashworth has noted, maximum penalties are not necessarily a perfect indicator of ‘seriousness’: see ‘Is the Criminal Law a Lost Cause?’, 243.

72

James Chalmers and Fiona Leverick

Table 2.6  Maximum penalty available on conviction

Imprisonment Fine (unlimited) Fine (standard scale)a Fine (nominate value)

1997–1998

2010–2011

906 (64.9%) 202 (14.5%) 270 (19.4%) 17 (1.2%)

993 (56.4%) 123 (7%) 641 (36.4%) 3 (0.2%)

a

  The ‘standard scale’ is fixed by primary legislation (see Interpretation Act 1978, Sch. 1 for the relevant references) and presently has five levels ranging from £200 to £5,000. A statute may set the maximum penalty by reference to one specific point on the scale, or alternatively to the ‘statutory maximum’ or ‘prescribed sum’. In England and Wales and Northern Ireland, the ‘statutory maximum’ or ‘prescribed sum’ is £5,000, equal to the highest point on the standard scale. In Scotland, however, this is now £10,000: Criminal Proceedings (Reform etc.) (Scotland) Act 2007, s. 48. We have nevertheless treated all references to the statutory maximum or prescribed sum as ‘standard scale’ fines.

Table 2.7  Maximum penalty for imprisonable offences created by statutory instrument

3 months 6 months 51 weeks 1 year 2 years 5 years 7 years 10 years

1997–1998

2010–2011

6 (0.7%) 202 (22.5%) None 2 (0.2%) 664 (74.1%) 1 (0.1%) 21 (2.3%) None

334 (38.5%) 4 (0.5%) 33 (3.8%) 8 (0.9%) 355 (40.9%) 130 (15%) 1 (0.1%) 2 (0.2%)

constitutional significance’.93 Given this, it might reasonably be expected that offences which attract potentially lengthy periods of imprisonment should only be created by a statute. This, however, is not the case, as demonstrated by the Table 2.7, which, for all those offences created by statutory instrument which can result in imprisonment, sets out the maximum length of the period of imprisonment permitted. As Table 2.7 demonstrates, a substantial number of criminal offences created under secondary legislation are relatively serious ones: 686 of the offences in the New Labour sample (43 per cent of all offences created by statutory instrument during that time period) and 488 in the coalition sample (32 per cent of all offences created by statutory instrument during that time period) carried maximum penalties of two years’ imprisonment or more. It is also of some concern that the number of offences created by statutory instrument carrying maximum sentences of five years or more was considerably higher in 2010–11 than in 1997–8. Such offences fall into two categories. First, they include Orders in Council made

93  Law Commission, Criminal Law in Regulatory Contexts, para. 3.157.

Quantifying Criminalization

73

under section 1 of the United Nations Act 1946.94 These are orders made to give effect to decisions of the Security Council that measures not involving the use of armed force should be employed in response to a threat to the peace, breach of the peace, or act of aggression.95 Given the nature of such orders, and the fact that they follow from a decision of the Security Council (which the UK, of course, would have been in a position to veto) the fact that these offences are created by statutory instrument may be relatively unproblematic. All of the 22 offences created by statutory instrument in 1997–8 which carried penalties of five or more years’ imprisonment fell into this category. Of the 133 offences with such high maximum penalties created by statutory instrument in 2010–11, however, only three fell into this category.96 The remaining 130 were contained in a single Scottish Statutory Instrument, the Water Environment (Controlled Activities) (Scotland) Regulations 2010.97 This instrument was made under the authority of primary legislation which specifically permits the creation of criminal offences carrying up to five years’ imprisonment,98 but it is not clear why it was thought appropriate to permit such punitive powers to be exercised by subordinate legislation,99 particularly when most (but not all) of the offences thereby created have no express requirement of mens rea.

X.  To Whom is the Criminal Law Directed? The Question of Special Capacity A final aspect of our analysis which we wish to note relates to what might be termed ‘special capacity’. Frequently, criminal offences will not be addressed to the general public, but instead only to a particular set of people: an important point given that media representations (and indeed many academic discussions) of criminalization suggest otherwise. There are some offences which are incapable of commission other than by people holding certain licences; others are not formally limited in this way but are restricted to persons undertaking certain roles or engaging in certain activities. Exceptionally, there are even some offences which can

94 Such Orders in Council are statutory instruments and not primary legislation: Statutory Instruments Act 1946, s. 1(1)(a). 95  Charter of the United Nations, 24 October 1945, 1 UNTS XVI, art. 42 (‘These [measures] may include complete or partial interruption of economic relations and of rail, sea, air, postal, telegraphic, radio, and other means of communication, and the severance of diplomatic relations’). 96  This seems not to be because the category has become less common: in fact, a number of such offences were created towards the end of May 2011, a matter of days after the end of our sample period. 97  SI 2011/209. It should be acknowledged that this instrument is in fact a consolidation of the Water Environment (Controlled Activities) (Scotland) Regulations 2005, SSI 2005/348 and subsequent amendments. The UK government has created offences by statutory instrument with similarly high penalties, but not within our sample period. See e.g. the Groundwater (England and Wales) Regulations 2009, rr. 13(1) and 18(2). 98  Water Environment and Water Services (Scotland) Act 2003, Sch. 2 para. 20(2). 99  Neither the Explanatory Notes nor the Policy Memorandum for the Bill as introduced in the Scottish Parliament explain this.

74

James Chalmers and Fiona Leverick

only be committed by specified persons.100 Table 2.8 sets out the extent to which ‘special capacity’ was required for the offences created in the two study years. The categorizations in Table 2.8 require some brief explanation. Some criminal offences explicitly provide for special capacity: an example is section 31(1) of the Crofting Reform (Scotland) Act 2010, which makes it a criminal offence for the owner of land on which common grazing is situated to fail to register a transfer of ownership of that land. Others do not explicitly provide for special capacity, but nonetheless carry this implication, and we have included these in the table in two categories: those where special capacity is implied because ‘ordinary people’ would never undertake the activity in question and those where ‘ordinary people’ would be highly unlikely to undertake this activity. An example of the former is section 3(1) of the Contaminants in Food (England) Regulations 2010, which makes it a criminal offence for a person to place on the market foodstuffs containing contaminants in excess of certain prescribed limits. An example of the latter is section 17(a) of the Equine Identification (Wales) Regulations 2009, which makes it a criminal offence for a person to destroy or deface an identification document on a horse. The attribution of an offence to one or the other of these two ‘implied special capacity’ categories was a difficult judgment call at times and the distinction is a very narrow one. The main point is that both are clearly not targeted at the public at large. As Table 2.8 shows, the vast majority of the criminal offences created in both samples are not of ‘general’ application but are targeted instead at those acting in some form of special capacity (almost 98 per cent of the New Labour sample Table 2.8  Special capacity required for each criminal offence created

None Role (by virtue of engaging in an activity) Role (by virtue of being awarded a licence or by registration) Role (status, e.g. ‘a debtor’) Implied special capacity (ordinary people never undertake this activity) Implied special capacity (ordinary people highly unlikely to undertake this activity) Prior circumstances (e.g. receiving information or a donation) Imposed (where a prior requirement or direction has been imposed on the accused) Corporate offence Specific body (e.g. ‘the harbour trust’) Familial

1997–1998

2010–2011

33 (2.4%) 728 (52.2%) 87 (6.2%) None 256 (18.4%)

200 (11.4%) 652 (37%) 158 (9%) 18 (1%) 345 (19.6%)

47 (3.4%)

117 (6.6%)

20 (1.4%)

39 (2.2%)

203 (14.6%)

187 (10.6%)

9 (0.6%) 12 (0.9%) None

35 (2%) 6 (0.3%) 3 (0.2%)

100  So, for example, r. 17(2) of the Porthmadog Harbour Revision Order 1998 creates an offence of failure to notify destruction or decay of tidal works to the Corporation of Trinity House of Deptford Strond, which can only be committed by Gwynned Council.

Quantifying Criminalization

75

and almost 89 per cent of the coalition sample). This is a significant finding. ‘Regulatory’ criminal law is all but ignored by most criminal law texts and journals101 but, in our two samples at least, it dominates the criminal law.102 Having said that, the number of offences of general application created in the 2010–11 sample is far higher than in the 1997–8 sample (200 compared to 33). The limited time frame for which we have data at present means we cannot draw any reliable longitudinal conclusions but this is something which we intend to explore in a subsequent paper. A final point worth making about Table 2.8 is the substantial number of special capacity offences stemming from a prior requirement or direction imposed upon the accused: 203 in the New Labour sample and 187 in the coalition sample (representing almost 15 per cent and almost 11 per cent respectively of all offences created within that time period). The offences included in this category fall into two main subcategories: those where the accused has been specifically requested to provide certain information and those where he has been served with some sort of prior notice prohibiting certain conduct. An example of the former is section 16(1)(b) of the Marketing of Fresh Horticultural Produce Regulations 2009, which makes it a criminal offence for any person to fail to provide any information or record that an authorized officer acting for the purpose of the enforcement of those regulations may reasonably require. An example of the latter is section 103(3) of the Marine and Coastal Access Act 2009, which makes it a criminal offence for a person to fail to comply with a ‘stop notice’ issued by an enforcement authority. Stop notices can be issued under the Act where a person is undertaking a marine activity that is causing or is likely to cause any of the various harmful effects set out in section 102(5). This two-stage construction of criminal liability is a development in the criminal law that has received comparatively little attention,103 although parallels might be drawn with the use of Anti-Social Behaviour Orders and other similar preventive orders.104 101 Some notable exceptions are R. Glover, ‘Regulatory Offences and Reverse Burdens: The “Licensing” Approach’, Journal of Criminal Law, 71 (2007), 259; G. Richardson, ‘Strict Liability for Regulatory Crime: The Empirical Research’, Criminal Law Review, 295 (1987), 295; J. Rowan-Robinson, ‘Crime and Regulation’, Criminal Law Review (1988), 211. For a collection of essays on the interface between regulation and criminal justice, see H. Quirk, T. Seddon, and G. Smith (eds.), Regulation and Criminal Justice: Innovations in Policy and Research (Cambridge: Cambridge University Press, 2010). 102  That is not to say that regulatory offences do not receive attention: there is a vast literature on regulation generally, which includes discussion of the role of criminal measures. See e.g. I. Ayres and J. Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford: Oxford University Press, 1992); R. Baldwin and J. Black, ‘Really Responsive Regulation’, Modern Law Review, 71 (2008), 59; T. Prosser, Law and the Regulators (Oxford: Clarendon Press, 1997); J. Black, ‘Talking about Regulation’, Public Law (1998), 77. The regulatory criminal law is also discussed at length in specialist journals: see e.g. A. Ogus and C. Abbot, ‘Sanctions for Pollution: Do we have the Right Regime?’, Journal of Environmental Law, 14 (2002), 283. 103 The use of such ‘stop notices’ by regulators is provided for under s. 46 of the Regulatory Enforcement and Sanctions Act 2008 and stemmed from the recommendations of the Macrory Review, Regulatory Justice: Making Sanctions Effective (2006). 104 On which there exists a voluminous literature: see e.g. A. Ashworth, ‘Social Control and “Anti-Social Behaviour”: The Subversion of Human Rights?’, Law Quarterly Review, 120 (2004),

76

James Chalmers and Fiona Leverick

Another variant of the two-step process involves criminal liability being imposed on a person who breaches the specific terms of a licence or permit:105 87 criminal offences in the New Labour sample (just over 6 per cent) and 158 in the coalition sample (9 per cent) fell into this category. Constructing criminal liability in this way lies some distance from the ‘normal’ process of setting out defined offences of general application in publicly available statutes and while it clearly cannot be criticized on fair warning grounds, questions can certainly be raised in terms of a lack of transparency and the potential for inconsistency.

XI.  Some Preliminary Conclusions The data we present above offers a snapshot of current practice and provides the starting point for what we hope will be a far more comprehensive programme of research. At this stage, we offer preliminary conclusions in three areas, all of which we intend to develop in future work. First, we hope to have established in this chapter that it is possible sensibly to quantify governmental use of the criminal law, and to have shown that this project sheds some light upon criminalization as practice: how do governments use the criminal law as a legislative tool? As we noted at the outset, the number of criminal offences created is only one ‘measure’ of the extent of criminalization and many of the offences covered by our sample will rarely or never be prosecuted. But this is hardly the point. Even if an offence is never prosecuted, it imposes regulatory costs on those who seek to comply with it and it may create a fear of prosecution, which may (justifiably or otherwise) be used itself as a regulatory tool.106 Furthermore, if only arbitrarily and inconsistently prosecuted, it may lead to substantial injustice in the cases where it is applied.107 Our project may, then, be concerned with criminalization in only one particular sense, that of formal criminalization (the ‘law on the books’), but these preliminary findings have convinced us that it is a valuable one and that it would in principle be possible to ‘fill in the gaps’ in our database and to monitor criminal offence creation in the UK on an ongoing basis. Secondly, while our data is only part of a far bigger picture, the sheer number of criminal offences in the two samples raises questions over whether the criminal law was in all of these instances the most appropriate mechanism for regulating the conduct in question. This is clearly not a concern in every case. As we have seen, many of the offences are far from minor: the majority in both samples are imprisonable offences. There is little room for debate over whether the criminal law is 263; P. Ramsay, ‘What is Anti-Social Behaviour?’, Criminal Law Review (2004), 908; A. P. Simester and A. von Hirsch, ‘Regulating Offensive Conduct through Two-Step Prohibitions’, in Simester and von Hirsch (eds.), Incivilities: Regulating Offensive Behaviour (Oxford: Hart Publishing, 2006), 173. 105  For discussion of this method of regulation, see e.g. C. T. Reid, ‘Regulation in a Changing World: Review and Revision of Environmental Permits’, Cambridge Law Journal, 67 (2008), 126. 106  See Richardson, ‘Strict Liability for Regulatory Crime’, 300. 107  Black, ‘Talking about Regulation’, 93.

Quantifying Criminalization

77

the right place for imprisonable offences.108 The most pressing concern here is that so many have been created by subordinate legislation: the creation of such serious offences should be a matter for Parliament rather than for administrative regulation.109 But not all of the offences in our samples are ‘serious’ ones (however that might be defined) and here the question is whether they would more appropriately have been the subject of alternative methods of regulation, such as civil penalties, or indeed whether they should have been regulated at all. The rise in quasi-criminal measures has been the subject of much critical debate and it is not our intention to contribute to this literature here.110 One possibility that does, however, strike us as worthy of attention is the creation of a new tier in the criminal law of the various UK jurisdictions comprising administrative offences, perhaps along the lines of the German Ordnungswidrigkeiten111 or the Spanish infracciones administrativas.112 Duff describes such administrative offences as ‘subversions’ of the criminal law, at least insofar as they are used to deal with ‘public wrongs’,113 but Ashworth and Zedner urge that such alternatives be considered as an alternative to the expansion of the criminal law ‘proper’.114 We intend to explore this possibility in future work. Thirdly, it is difficult to escape the conclusion that the results of criminalization are frequently so inaccessible as to breach basic principles of fair notice. Fair notice can be seen as one component of the overarching concept of the principle of legality,115 which requires ‘fair warning of clear and certain criminal laws that do not operate retrospectively’.116 This is of particular importance when, as is the case in the UK context, the defence of ignorance of the law is so restrictive as to be almost non-existent.117 On the basis even of our preliminary findings, it has to 108  See e.g. Ashworth, ‘Is the Criminal Law a Lost Cause?, 255. 109  See Law Commission, Criminal Law in Regulatory Contexts, para. 3.157, quoted at n. 93. There is a secondary question of whether all criminal offences should be created by primary legislation, but that is not one that we intend to address here. 110  In the UK context, see e.g. R. M. White, ‘“Civil Penalties”: Oxymoron, Chimera or Stealth Sanction?’, Law Quarterly Review, 126 (2010), 593; Ashworth and Zedner, ‘Defending the Criminal Law’, 21; R. A. Duff, ‘Perversions and Subversions of Criminal Law’, in Duff et al. (eds.), The Boundaries of the Criminal Law, 88. This debate mirrors one that has been ongoing in the US context for far longer: see K. Mann, ‘Punitive Civil Sanctions: the Middleground between Criminal and Civil Law’, Yale Law Journal, 101 (1991–2), 1795 and the responses to Mann’s paper published in the same issue of that journal. 111  See M. Bohlander, Principles of German Criminal Law (2008), 14; T. Weigend, ‘The Legal and Practical Problems Posed by the Difference between Criminal Law and Administrative Penal Law’, Revue internationale de droit pénal, 59 (1988), 67. 112  See M. F. Aebi and M. Balcells, ‘The Prosecution Service Function within the Spanish Criminal Justice System’, European Journal on Criminal Policy and Research, 14 (2008), 311 at 313. For discussion of the general expansion in the creation of ‘administrative offences’ in European criminal law, see J.-M. Jehle, M. Wade, and B. Elsner, ‘Prosecution and Diversion within Criminal Justice Systems in Europe: Aims and Design of a Comparative Study’, European Journal on Criminal Policy and Research, 14 (2008), 93, 95. 113  Duff, ‘Perversions and Subversions’, 106. 114 Ashworth and Zedner, ‘Defending the Criminal Law’, 33, 49. See also Ashworth, ‘Is the Criminal Law a Lost Cause?’, 255. 115  For a discussion of this principle and what its component parts might be, see P. Westen, ‘Two Rules of Legality in Criminal Law’, Law and Philosophy, 26 (2007), 229. 116  Ashworth, ‘Ignorance of the Criminal Law’, 4. 117  Ashworth and Horder, Principles of Criminal Law, 218. In the Scottish context, see J. Chalmers and F. Leverick, Criminal Defences and Pleas in Bar of Trial (Edinburgh: W. Green, 2006), ch. 13.

78

James Chalmers and Fiona Leverick

be questioned whether the principle of fair notice is satisfied by the criminal law in the UK. Our research has thrown up multiple examples of offences defined by reference to other offences,118 or by reference to multiple lengthy European directives.119 One might respond that this is of little concern because the vast majority of criminal offences are regulatory ones,120 and, as Ashworth has put it, there is ‘every reason to expect people to acquaint themselves with laws relevant to their particular activities, whether it be motoring, commerce, building, or whatever’.121 It is true that, in both of our samples, the majority of criminal offences created were targeted at those operating in some form of ‘special capacity’, although many were not. Even for special capacity offences, however, there is surely a limit to what the state can reasonably require in terms of discovering the law? Ashworth is quite right: it is not unreasonable to ask those engaging in a particular activity that has the potential to cause harm to others to take reasonable steps to acquaint themselves with the relevant criminal provisions. But he underestimates the scale and difficulty of this task. It was a task that took us, as legally qualified academics, considerable time and intellectual effort. It should be remembered that many ‘regulatory’ offences are targeted not at ‘big business’ (or at least not exclusively so) but at small, family run or sole trader businesses122 or at individuals engaging in particular activities in a personal and/or non-profit-making context. Even if it is accepted that the criminal law of the UK raises fair notice concerns, how this might be addressed is another matter. All primary and secondary legislation enacted since 1991 is publicly available to those who have internet access via the UK Statute Law Database, but as we have shown it is an extremely lengthy and complex task to extract from this source a comprehensive list of criminal offences. It has sometimes been suggested that there should be a consolidated list of all stat­utory offences publicly available.123 There would be considerable practical diff­i­culties in achieving this, but this in itself is not a reason not to do so: if any government wishes to inflict sanctions on its citizens, it must accept the task of providing those citizens with fair notice of their obligations.124 The real objection to such a proposal is that there are so many criminal offences that the consolidated list would

118  See e.g. Coroners and Justice Act 2009, Sch. 6 Part 2 para. 8(1). 119  See e.g. the Animals and Animal Products (Import and Export) Regulations, r. 33 read with r. 5(1), under which anyone exporting bovine animals or swine commits an offence unless they comply with 45 different directives. 120 The debate over precisely what constitutes a ‘regulatory’ offence is not one we intend to enter here. For discussion see G. Smith, T. Seddon, and H. Quirk, ‘Regulation and Criminal Justice: Exploring the Connections and Disconnections’, in Quirk, Seddon, and Smith, Regulation and Criminal Justice, 2–4. 121  Ashworth, ‘Ignorance of the Criminal Law’, 21. 122  A point made in a slightly different context by Horder in his contribution to this collection, pp. 123–4. 123  See e.g. Ferguson, ‘Criminal Law and Criminal Justice’, 223–4 (suggesting a consolidating statute); JUSTICE, Breaking the Rules, 53–4 (suggesting an official publication available in public libraries and government bookshops). 124  Ashworth, ‘Ignorance of the Criminal Law’, 10; D. Husak, ‘Ignorance of Law and Duties of Citizenship’, Legal Studies, 14 (1994), 105, 115; M. Travers, ‘Mistake of Law in Mala Prohibita Crimes’, University of Chicago Law Review, 62 (1995), 1301, 1320.

Quantifying Criminalization

79

be of little practical use to anyone, particularly as many regulatory offences might appear almost meaningless were they to appear stripped of context.125 What cannot be doubted is that the relative inaccessibility and incoherence of much criminal law can lead easily to injustice, as evidenced by the case of R v Chambers.126 That case concerned a prosecution for fraudulent evasion of duty127 in relation to tobacco products which relied in part on the Excise Goods Regulations 1992.128 These regulations had in fact been superseded in relation to tobacco products by the materially different Tobacco Products Regulations 2001,129 the latter having come into force over five years before the events giving rise to Chambers’s prosecution. The error—of which Chambers’s prosecution was not an isolated example—was, by chance, discovered at the last minute before his appeal against a confiscation order was to be disposed of. The Court of Appeal allowed that appeal,130 and Toulson LJ commented that:131 This case also provides an example of a wider problem. It is a maxim that ignorance of the law is no excuse, but it is profoundly unsatisfactory if the law itself is not practically accessible. To a worryingly large extent, statutory law is not practically accessible today, even to the courts whose constitutional duty it is to interpret and enforce it. There are four principal reasons. First, the majority of legislation is secondary legislation. Secondly, the volume of legislation has increased very greatly over the last 40 years. . . . Thirdly, on many subjects the legislation cannot be found in a single place, but in a patchwork of primary and secondary legislation. Fourthly, there is no comprehensive statute law database with hyperlinks which would enable an intelligent person, by using a search engine, to find out all the legislation on a particular topic. This means that the courts are in many cases unable to discover what the law is, or was at the date with which the court is concerned, and are entirely dependent on the parties for being able to inform them what were the relevant statutory provisions which the court has to apply. This lamentable state of affairs has been raised by responsible bodies on many occasions . . . .

No one concerned with the criminal law should be sanguine about this lamentable state of affairs, and we hope that the project outlined in this paper can contribute to remedying it.

125  See Law Commission, A Criminal Code for England and Wales (Law Com. No. 177, 1989), para. 3.3. 126  [2008] EWCA Crim. 2467. The case is discussed by G. L. Gretton, ‘Of Law Commissioning’, Edinburgh Law Review, 17 (2013), 119. We are grateful to George Gretton for bringing this case and Toulson LJ’s remarks to our attention. It is, sadly, not an isolated problem. See further Spencer, ‘The Drafting of Criminal Legislation’. 127  Under s. 170 of the Customs and Excise Management Act 1979. 128  Excise Goods (Holding, Movement, Warehousing and REDS) Regulations 1992, SI 1992/3135. 129  SI 2001/1712. 130  Perhaps unsurprisingly, given that the deficiency in the prosecution had not been noticed until this point, no appeal appears to have been brought against either Chambers’s conviction or the community sentence which had been imposed on him. 131  At paras 64–8.

3 Criminal Law as an Institution Rethinking Theoretical Approaches to Criminalization Lindsay Farmer*

I.  Introduction: The Criminal Law Vanishes In section 1.02-(1) of the US Model Penal Code (MPC) the drafters address the question of the purposes of criminal law.1 They begin by stating that the first objective of the law is to prevent the commission of offences, and that the principle which should guide the definition of offences is ‘to forbid and prevent conduct that unjustifiably and inexcusably inflicts or threatens substantial harm to individual or public interests’. The content of the law thus appears to be based on a combination of individual and public interests, while the purpose of the law is the prevention of harm to the identified interests. These issues—the type of interests to be protected, the type of harms that they are to be protected against, the relation to the dominant aim of the criminal law understood as deterrence or retribution, and the relation between wrongdoing and defences—are those that have been taken to be the necessary elements of any theory of criminalization. However, if we examine this statement more closely we can see that, in spite of apparent clarity, there is a certain vagueness in this formulation with respect to the

*  Earlier versions of this chapter have been presented at seminars in Uppsala, Edinburgh, Queen’s University, Kingston and at the Criminal Law Sciences Club at the University of Toronto. I am very grateful to the participants in those seminars for their comments. I would particularly like to thank Sarah Armstrong, Markus Dubber, Niki Lacey, Claes Lernestedt, and Scott Veitch. As ever this would not have been possible without my collaborators on the Criminalization project—Antony Duff, Sandra Marshall, Massimo Renzo, and Victor Tadros. 1  The full section reads: (1)  The general purposes of the provisions governing the definition of offenses are: (a) to forbid and prevent conduct that unjustifiably and inexcusably inflicts or threatens substantial harm to individual or public interests; (b) to subject to public control persons whose conduct indicates that they are disposed to commit crimes; (c) to safeguard conduct that is without fault from condemnation as criminal; (d) to give fair warning of the nature of the conduct declared to constitute an offense; (e) to differentiate on reasonable grounds between serious and minor offenses.

Criminal Law as an Institution

81

specification of the nature and scope of the criminal law. The resort to criminal law must be understood in terms either of prevention, or as a response to the quality of the harms to the protected interests—that is to say those harms which are serious or substantial. However, this does not explain any specific qualities of the criminal law: the aim of preventing crimes (or, more precisely, preventing the harms by reference to which crimes are defined) does not point towards criminal law; and the section does not specify why or when certain types of harm should be regarded as criminal. The point here is not so much that other areas of the law may aim at the prevention of, even serious, harms, but that there is a failure even to attempt to specify the nature of the criminal law (what is criminal law and why it should be used in these instances?). Thus, in the section which purports to define the purposes of the criminal law there is only vagueness and evasion. There is a sense in which the criminal law vanishes just at the point at which we would expect to see it come into focus. I will argue in this chapter that this failure to reflect on the nature or purpose of the criminal law is a characteristic of theories of criminalization more generally. Theories of criminalization need to give more attention to the question of what makes the law criminal, or the distinctive character or aims of the criminal law.2 If we are to have a normative theory of criminalization—what conduct is properly subject to regulation by the criminal law—it must surely be essential to have a richer account of what the criminal law is, what it is for, and what is to be achieved by resort to the criminal law. This, I will argue, should not be approached as a prima facie moral question, as something that can be defined a priori and independently of the legal institution, but must be understood in terms of the development of the modern criminal law. The chapter is in three main parts. In Section II there is a discussion of different theories of criminalization which will develop the claim made here that there has been a failure on the part of such theories to attend to the nature or purpose of the criminal law. In Section III I go on to set out an understanding of criminal law as an institution as a framework that can combine a normative perspective on issues of criminalization with a richer understanding of the nature and purpose of the criminal law.3 In the concluding section I then return to the question of how this account, which gives more weight to the nature and function of the modern criminal law, can contribute to normative theories of criminalization.

2  I am aware of the extensive literature on the definition of crime but, as I shall argue, this reproduces the structure of the above argument. This is because it starts by trying to identify the nature of a crime rather than the nature of criminal law. And if, as is taught in introductory criminology lectures, crime is first and foremost a legal or social construct, it would make more sense to start with the law. See also M. D. Dubber, ‘Criminal Law between Public and Private Law’, in Duff et al., The Boundaries of the Criminal Law (Oxford: Oxford University Press, 2010), 191: ‘This means that an account of criminalization needs an account not of crime simpliciter, but of law in general, and of criminal law within it.’ 3  See e.g. N. MacCormick, Institutions of Law: An Essay in Legal Theory (Oxford: Oxford University Press, 2007). See also N. Lacey, ‘Institutionalising Responsibility’, Jurisprudence 4 (2013; 1–19).

82

Lindsay Farmer

II.  Criminalization and Criminal Law In very general terms there are two main strands in liberal thought about criminalization: those which begin from questions of harm and harm reduction or prevention, and those theories, often called legal moralism, which are centred around questions of wrongdoing and retribution. The first derives from, and is animated by, John Stuart Mill’s famous ‘harm’ principle: that ‘the only purpose for which power can be rightfully exercised against any member of a civilised community, against his will, is to prevent harm to others’.4 In this tradition efforts have focused on attempting to define the concept of harm: what is it? does it include forms of potential as well as actual harm? Are forms of paternalism to prevent self-harm justified? This has achieved perhaps its fullest contemporary expression in the four-volume work of Joel Feinberg, in which harm was defined as a ‘setback to interests’, and which took particular aim at what he saw as the illegitimate criminalization of various forms of non-harmful conduct simply on the basis that it was regarded as immoral or offensive.5 The use of the criminal law was thus seen as justified primarily, though not exclusively, as a means of preventing serious setbacks to the interests of others. The alternative approach—legal moralism—argues that we should begin from the idea of wrongdoing, since it would be unjustifiable for the state to punish a person for conduct which is not morally wrong.6 This approach begins with a more explicit link to the ends of punishment, since punishment is understood as being justified only when imposed for wrongful conduct. Thus, for example, Moore asserts that the proper aim of the criminal law is to inflict deserved punishment on culpable wrongdoers—and thus that the wrongfulness of conduct provides prima facie positive reasons for criminalizing it.7 From this perspective, then, it would be unjustifiable either to punish non-culpable actors or to criminalize conduct which was not wrongful. While there are significant differences between these approaches, which have been discussed at length in criminal law theory, it is less often remarked that there are a number of striking similarities between them. This can be seen both in the concessions each make to the other’s position, and in the acknowledgements of the limits of ‘principle’ in this area. Defenders of the harm principle, such as Feinberg, 4  John Stuart Mill on Liberty and Other Essays (1859), ed. John Gray (Oxford: Oxford World Classics, 1991), 14. 5  The Moral Limits of the Criminal Law, 4 vols. (Oxford: Oxford University Press, 1985). He did allow that offensiveness could sometimes justify criminalization (vol. ii), and that some non-harmful wrongs might be criminalized (vol. iv). For discussion see R. A. Duff, Answering for Crime (Oxford: Hart Publishing, 2007), ch. 6; H. Stewart, ‘The Limits of the Harm Principle’, Criminal Law and Philosophy, 4 (2010), 17; B. E. Harcourt, ‘The Collapse of the Harm Principle’, Journal of Criminal Law and Criminology, 90 (1999), 109. 6  See the illuminating discussion in R. A. Duff, ‘Towards a Modest Legal Moralism’, Criminal Law and Philosophy 7 (2013), 217 distinguishing between different versions of legal moralism. See also e.g. D. Husak, Overcriminalization: The Limits of the Criminal Law (Oxford: Oxford University Press, 2008). 7  See e.g. M. Moore, Placing Blame (Oxford: Oxford University Press, 1997), ch. 1.

Criminal Law as an Institution

83

concede that a criminal law which punished all harmful conduct would be too broad, and therefore introduce a limiting principle, which is that only ‘wrongful’ harms should be subject to criminal punishment—with the measure of ‘wrongfulness’ depending on either the rights that are to be protected or the mode of infringement.8 It is also acknowledged that the specification of ‘harms’, and thus the content of the law, will further depend on community values or standards as these are determined through the political and legislative process. Conversely, legal moralists concede that not all prima facie wrongful conduct (lying, cheating at games) should be the subject of the criminal law because, once again, the law would be too broad. They accordingly argue the need for a limiting principle, such as that the criminal law should be concerned only with harmful wrongs— though this throws them back onto acknowledging the unspecific or conventional nature of the concept of harm.9 It is also worth noting that many legal moralists also concede that some conduct will be wrongful by virtue of having been made so by law, and so must either allow that the category of wrongs is broader than merely ‘pre-legal’ moral wrongs or that the criminalization of non-wrongful conduct may in certain instances be justified.10 In either case—wrongful harms or harmful wrongs—the language and the structure of the arguments have a great deal in common. The initial category (harm or wrong) is defined independently of law and is thought to shape the proper scope of the law; the aim of criminal law is understood as an aim of punishment (deterrence or retribution); and in each case it is acknowledged that the initial organizing principle is too broad, and that the eventual content of the criminal law will depend on factors which have little to do with categories of moral thought. I shall say more about each of these points as a way of developing my central claim about the failure of these theories to address the nature of the criminal law. Most contemporary theories understand the scope of the criminal law as being properly determined by the interests which are to be protected—wrongful harms or harmful wrongs. Their starting point is that these interests or moral wrongs can, and should, be identified a priori and independently of law, and they are identified primarily in terms of individual rights or interests, the protection of which is seen as the foundation or core of the criminal law.11 It is sometimes recognized (as in the Model Penal Code), that there are also public interests which must be protected, such as public order or the state, and that these are not simply reducible to private interests, but these are generally presented as secondary to the individual interests.12 And while there are some interests, such as property, which might appear to fall somewhere between public and private—they are directed either at the protection of individual interest in personal private property, or at the 8  See e.g. J. Feinberg, Harm to Others (Oxford: Oxford University Press, 1984), 31–6. 9  e.g. Moore, Placing Blame (n. 6), chs 1 and 16. 10  See Duff, ‘Towards a Modest Legal Moralism’ (n. 6), 3–5; Duff, Answering for Crime (n. 5), 89–93. cf. Husak, Overcriminalization (n. 6), 103–19 for a more sceptical account. 11  For criticism of this see L. Farmer, ‘Criminal Wrongs in Historical Perspective’, in Duff et al., The Boundaries of Criminal Law (Oxford: Oxford University Press, 2012). 12  See discussion in Dubber, ‘Criminal Law between Public and Private Law’ (n. 2), at 206–11.

84

Lindsay Farmer

protection of the system of property rights or interests—it is once again the protection of the underlying private interest that is given priority.13 What these views share is an understanding of the scope of criminal law as being determined by private right, and of the function of criminal law as the protection of these rights. On this view, criminal law appears as something adjunct to private moral or legal rights: the role of the law is to protect the pre-existing rights.14 Criminal law then appears as in some sense ‘adjectival’, protecting goods, interests, or rights that have been defined elsewhere against certain kinds of serious and culpable interferences.15 This is also linked to an account of the development of the law. Criminal law is understood as being founded in a core which based on the protection of individual rights, and then collective or welfare interests have been added on—and the problem of over-criminalization is then understood as that of limiting or pinning back those crimes which are not directly linked to individual rights.16 Given the dubious historical accuracy of this account, it is best understood as a theoretical claim about the primacy of certain individual rights or interests. However, this approach is limited in a number of ways. It takes wrongs against individuals as fundamental and accordingly struggles to explain collective or welfare interests; rights or interests which are created by law may be seen as problematic;17 and, perhaps most importantly, if the function of criminal law is only that of protecting private rights, there is little or no space for an account of public function of criminal law. Where the aim of criminal law is discussed, it is generally seen in terms of the ends of punishment, either retribution or deterrence, or some combination of the two. In deterrence-based accounts, this means that there is only a contingent relation between the aim and the content of the law: the aim of harm prevention or reduction does not depend on the content of any particular harms. This might seem surprising because it means that it is hard to specify anything distinctive about criminal law—it is just another instrument which might be deployed by the 13  See MPC ss. 220.1–230.5. See also e.g. S. Shute, ‘Appropriation and the Law of Theft’, Criminal Law Review (2002), 445–58. See also S. Gardner, ‘Property and Theft’, Criminal Law Review (1998), 35. See generally Dubber, ‘Criminal Law’, 206–7. Although it is worth noting that this kind of formulation might apply more generally—offences against the person could be understood as protecting the system of personal rights and so on. 14  cf. the role of Rechtsgut in German criminal law. For discussion see K. Nuotio, ‘Theories of Criminalization and the Limits of Criminal Law: A Legal Cultural Approach’, in Duff et al., The Boundaries of the Criminal Law (Oxford: Oxford University Press, 2010), at 244–52; M. D. Dubber, ‘Theories of Crime and Punishment in German Criminal Law’, American Journal of Comparative Law, 53 (2006), 679. 15  The term ‘adjectival’ is taken from Jeremy Bentham, On the Principles and Morals of Legislation (1789; Oxford: Clarendon Press, 1996). For discussion see Farmer, ‘Reconstructing the English Codification Debate: The Criminal Law Commissioners 1833–45’, Law & History Review, 18 (2000), 397 at 419–22. 16  This also implicitly underpins the account of over-criminalization in Husak, Overcriminalization (n. 6). See also A. P. Simester and A. von Hirsch, Crimes, Harms and Wrongs (Oxford: Hart Publishing, 2011), ch. 1 where they make the claim that these kind of individual wrongs are ‘archetypal’—joining the sense of historical priority to that of being theoretically fundamental. 17  Here I am referring to offences such as the failure to pay taxes, where the crime is the result only of the creation of a tax regime and cannot readily be linked to a private moral or legal right. See refs. at n. 10 where these kind of mala prohibita are discussed.

Criminal Law as an Institution

85

state for the prevention of harm. The picture is more complex with wrong-based accounts. These accounts can generate strong reasons why we should only punish morally responsible actors, and in certain cases (e.g. homicide or rape) can identify wrongs that might uncontroversially be criminalized. However, beyond this narrow core it can struggle to generate reasons why punishment is a justifiable response to a broader range of ‘wrongs’ or why some wrongs are appropriately addressed by the criminal law while others are not. The identification of the wrong thus needs to be supplemented by reasons, such as the degree of harm or the effectiveness of a particular mode of enforcement or definition, that have little to do with the aim of retribution as such. It is this step—the identification of the legal wrong and its definition in criminal law—rather than the identification of the pre-moral wrong, that is the central step in criminalization and whose theoretical significance requires proper acknowledgement. The key point here then is that while consideration of the ends of punishment might contribute to an explanation of the moral or political justification for the exercise of the criminal law, it radically under-determines the scope of the criminal law.18 This should lead us to ask whether the aims of criminal law can be reduced to those of punishment. Writers such as Marshall and Duff have suggested that the relation between punishment and criminal law is at best historically and culturally contingent; that is to say that the aim of the criminal law is primarily to censure or condemn conduct of which the community disapproves and that this does not necessarily require punishment.19 If we accept this argument, then it would imply that there is a public interest in resort to criminal law—either an inherent quality or an aim of the criminal law—which cannot be explained solely in terms of the punishment of an individual for a wrong. This is an argument that has been addressed most fully by Duff, who seeks to identify the public interest in terms of the requirement that crimes be ‘public wrongs’.20 On his account crimes should be understood as wrongs which concern us all as members of the community, and can thus be distinguished from private wrongs, which concern only those individuals directly involved (and so would be a matter only of private redress). For this kind of ‘modest’ legal moralism the scope of the criminal law should not depend primarily on the ‘wrongness’ of conduct, but on the identification of those wrongs which properly count as public in the sense in which ‘we are criminally responsible as citizens to our fellow citizens’.21 There is therefore a need to begin with the public realm: A justification of criminalisation will need to begin by specifying some value(s) that can be claimed to be public, as part of the polity’s self-definition; show how the conduct in 18  See comments in Duff et al., ‘Introduction’, in R. A. Duff et al., The Boundaries of the Criminal Law (Oxford: Oxford University Press, 2010), 6–7. 19  S. E. Marshall and R. A. Duff, ‘Criminalization and Sharing Wrongs’, Canadian Journal of Law and Jurisprudence, 11 (1998), 7–22. 20  See Duff, Answering for Crime (n. 5), esp. chs 4 and 6; R. A. Duff and S. E. Marshall, ‘Public and Private Wrongs’, in J. Chalmers, F. Leverick, and L. Farmer (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010); See also Duff, ‘Towards a Modest Legal Moralism’ (n. 6), 6–7. 21 Duff, Answering for Crime (n. 5), 142.

86

Lindsay Farmer

question violates that value or threatens the goods that it protects; and argue that that violation or threat is such as to require or demand a public condemnation.22

The importance of this account is that it seeks to go beyond narrow accounts of the criminal law as the protection of private interests, to identify a ‘public’ feature of the law based in the political values of a given community. But it is not clear that this goes far enough. First, it seems that many public values, or values which are part of the public realm, might equally well be protected by other forms of law. If, for example, tort law is understood as a form of redress for wrongs, it is hard not to see why there is not also a form of public interest in ensuring that this redress takes place such that they are also public wrongs. The answer to this might be that criminal law is a particular form of public censure or condemnation. But while this might explain part of the function of the criminal law—identifying public wrongs and condemning individuals for their wrongdoing—it does not help us to identify which wrongs might be criminal or how they might be defined in the criminal law.23 The broader problem with this approach might be summed up in this way: it starts by trying to identify a specific quality of the wrong but in doing so it neglects, or does not fully develop, an account of the aims or function of criminal law. If we are to begin to develop such an account, one place that we can look is criminological approaches to criminalization. A distinctive feature of work here is that in identifying specific qualities of the criminal law in practice it focuses more on the social function of the criminal law. Here there are, broadly speaking, two kinds of approach. The first looks at the question of who is criminalized and how this process is exercised, seeing both over- and under-criminalization, to see the social function of law in practice.24 Over-criminalization focuses on the criminalization of particular groups, communities, and locations. This has looked at the police practices of using stop and search powers against particular ethnic communities or groups defined by gender or age. It examines the use of powers, such as those under the anti-terrorism legislation, to police public order and so on. Or it looks at the intersections between, for example, housing or immigration policy and the criminal law to trace the increasing use of the criminal law in these areas.25 Under-criminalization, by contrast, has generally argued that certain areas of social life, typically those connected with certain forms of economic activity conducted by relatively powerful social groups, have

22 Duff, Answering for Crime (n. 3), 143. 23  See also Dubber here criticizing the move to the publicness of wrongs as ignoring the ‘lawness of criminal law’, ‘Criminal Law between Public and Private Law’ (n. 2), 206. 24 See e.g. the discussion in C. Pantazis, ‘The Problem with Criminalisation’, Criminal Justice Matters, 74 (2008), 10–11. See also Muncie in the same issue (13–14) tracing the roots of the analysis of criminalization within criminology. See also P. Scraton, Power, Conflict and Criminalisation (London: Routledge, 2007). 25 See e.g. A. Aliverti, ‘Making People Criminal: The Role of Criminal Law in Immigration Enforcement’, Theoretical Criminology, 164 (2012), 17–34; L. Zedner, ‘Security, the State, and the Citizen: The Changing Architecture of Crime Control’, New Criminal Law Review, 13 (2010), 379–403.

Criminal Law as an Institution

87

been under-criminalized. It has been argued that the ‘social harm’ caused by certain activities—typically in financial markets or business practices—matches or exceeds the harm caused by more ‘traditional’ criminal activities, and that this is evidence of double standards in the recognition of the harms caused by, or censure of, certain activities.26 ‘Harm’ is understood as a means of illustrating the symbolic equivalence of certain types of activities and as a potentially powerful argument in support of criminalization. There are a number of important features of this kind of approach. First, and importantly, the focus is not usually on legislation or the law-making process, but on how the law is used and enforced, or the use of police powers—even when this does not lead to prosecution.27 This then produces a powerful empirical picture of state practices of criminalization (or of the social consequences of the failure to criminalize). Second, this kind of approach is highly sensitive to questions of social exclusion and the exercise of social power. It looks at the groups or communities who are either criminalized or, conversely, neglected by the law, or the ways that criminal law supports these broader processes of social exclusion.28 This allows the linking of questions of criminalization to broader trends and developments not only in the use of the criminal law but in social and economic development that are often hard to detect or absent from the perspective of more traditional normative approaches to criminalization. This points to the fact that the criminal law is used instrumentally by the state or other socially or economically powerful groups. A theory of criminalization must thus address the fact not only that conduct to be criminalized cannot always be identified independently of other legal, social, and institutional structures but also that the use of criminal law in these circumstances is about more than just censure. The defining of conduct as criminal through the law is also an appeal to the symbolic value and power of the criminal law. However, notwithstanding these strengths, there are problems with this approach in terms of thinking about criminalization more generally. The first is that this kind of approach can sometimes seem narrow in focus, looking at particular instances or areas of the application of the criminal law in such a way that it can be unclear what this might tell us about the application or use of the criminal law in general. If certain communities are criminalized through the use of stop and search powers this does not necessarily tell us anything about the acceptance of norms of criminal law more generally, or about the attitudes of the criminalized community to those broader norms. As a result, the normative implications of this 26  See e.g. P. Hillyard et al., Beyond Criminology: Taking Harm Seriously (London: Pluto 2004). See also essays by Pemberton, White, and Snider in Criminal Justice Matters, 74 (2008). On the weaknesses of the social harm perspective see Harcourt, ‘The Collapse of the Harm Principle’ (n. 5). 27  cf. e.g. Simester and von Hirsch, Crimes, Harms and Wrongs (n. 16), ch. 1, who define criminalization solely in terms of the legislative decision, suggesting that such an approach is ‘uncontroversial’. This is discussed at greater length in the Introduction to this volume at pp. 6–16. 28  See e.g. J. Young, The Exclusive Society: Social Exclusion, Crime and Difference in Late Modernity (London: Sage, 1999); L. Wacquant, Punishing the Poor: The Neo-Liberal Government of Social Insecurity (Durham, NC: Duke University Press, 2009). See also P. Squires and J. Lea, Criminalisation and Advanced Marginality: Critically Exploring the Work of Loïc Wacquant (Bristol: The Policy Press, 2012).

88

Lindsay Farmer

kind of analysis remain unclear.29 There is, at best, a failure to attempt to develop a conceptual language that can bring together the normative and empirical dimensions of criminalization.30 Overall, then, while the focus on social power and the uses of the criminal law is important, it can also be argued that the criminal law as a whole does not come clearly into focus. The second kind of approach is focused on the social function or meaning of the criminal law more broadly. Here criminal law is seen as distinctive in terms of its social functions. Thus, for example, the French sociologist Émile Durkheim argued that punishment played the particular social function of articulating and expressing forms of social disapproval for the breach of particular social norms, but that the form and degree of repressiveness of punishment would change according to the form of social solidarity. Thus the criminal law allowed for the continual process of social reintegration of communities through the identification and punishment of those who transgressed its norms.31 Important as this is, though, it is worth distinguishing between the aims of the criminal law and those of pun­ishment for, as penologists such as Garland have pointed out, criminal law and punishment are distinct social and cultural practices.32 While for Garland this points to the need for sociologists to study punishment as a cultural practice in its own right, and not simply as an instrumental response to crime, the point might also be reversed: the aims or social function of punishment are not necessarily the same as those of the criminal law. Thus rather than seeing criminal law as simply the obverse of punishment (or that the aims of criminal law flow from the aims of punishment), it is necessary to understand it as a distinct social practice. This is a point recognized implicitly by some Marxist analyses of criminal law, which point to the significance of the form of criminal law in protecting the interests of particular social classes or repressing others in certain historical periods.33 Such approaches have provided important accounts of how the legal form can be used to promote the interests of a certain class or why, say, property has been accorded particular significance by the criminal law in certain periods. Their importance goes beyond this to pointing out that the origins and function of the criminal law at particular points in time might depend less on moral theory than the ability of social groups 29  See e.g. A. Millie, ‘Value Judgments and Criminalization’, British Journal of Criminology, 51 (2011), 278–95. Interestingly, though, he wants to expand beyond moral judgments (or moral philosophy) to include prudential, economic, and aesthetic judgments. 30  See N. Lacey, ‘Historicising Criminalisation: Conceptual and Empirical Issues’, Modern Law Review, 72 (2009), 936–60 at 944. 31  E. Durkheim, The Division of Labour in Society (London: Macmillan, 1984); ‘Two Laws of Penal Evolution’, Economy and Society, 2 (1973), 285–308. See also discussion in D. Garland, Punishment and Modern Society (Oxford: Clarendon, 1990), chs 2 and 3. 32 Garland, Punishment and Modern Society (n. 31), esp. at 16–22: ‘Like architecture or diet or clothing or table manners, punishment has an instrumental purpose, but also a cultural style and a historical tradition . . .’ (19). See also D. Garland and P. Young, The Power to Punish (London: Heineman, 1983), ch. 1. 33 See E. B. Pashukanis, Law and Marxism (London: Pluto Press, 1977), ch. 7. See also D. Hay, ‘Property, Authority and the Criminal Law’, in D. Hay et al., Albion’s Fatal Tree (Harmondsworth: Penguin, 1975); A. Norrie, Crime, Reason and History. A Critical Introduction to Criminal Law (2nd edn.; Cambridge: Cambridge University Press, 2001).

Criminal Law as an Institution

89

or classes to harness the power of the law. In short, an account of criminalization, and an understanding of the nature of particular wrongs and harms, must be open to this particular combination of understanding how the law is used and enforced, the interests that it might serve, and the social functions that it might perform. This brings us then to the question of how we might begin to integrate these insights in such a way as to retain sight of the distinctive social and legal character of criminal law. I shall argue in the Section III that one way in which we might do this is through an understanding of law as an ‘institutional normative order’.

III.  The Institution of Criminal Law The institutional theory of law argues that that law must be understood as a form of institutional normative order.34 Law is not conceived of as a system of norms endowed with institutional existence; rather it is institutions that constitute legal norms. Thus, an understanding of law should not be based on an understanding of moral norms such as right and wrong, but on the norms that have become institutionalized in particular forms, agencies, and rules of law.35 Furthermore, an explanation of the institution requires an account of the relevant rules set out in the light of the point or end of that institution.36 Institutionalization occurs, on MacCormick’s account, whenever we encounter a ‘two-tier’ normative practice—that is, when there is not only the practice, but also more or less formal rules for ordering the practice.37 In the case of law, he thus seeks first to identify and explain how basic forms of obligation and duty arise between persons, and then how the process of institutionalization requires recognition of particular powers and competences on the part of those enforcing the norms and managing the institutional order.38 The account thus proceeds from an investigation into the quality of norms and how these are ordered and make interpersonal conduct or social life to some extent orderly, towards an explanation of how these norms become institutionalized. While the formulation of law as an 34  I shall develop my argument in this section primarily through an engagement with MacCormick, Institutions of Law (n. 3). 35 MacCormick, Institutions of Law (n. 3), 34–7; ‘Law as Institutional Fact’, Law Quarterly Review (1974), 102–29 etc. See also the essays in M. del Mar and Z. Bankowski (eds.), Law as Institutional Normative Order (Aldershot: Ashgate, 2009). For other version of institutional theory see O. Weinberger, Law, Institution and Legal Politics (Dordrecht: Kluwer, 1991), esp. at 163–72 for a brief account of the history of institutional legal theory; M. la Torre, Law as Institution (Dordrecht: Springer, 2010). cf. N. Luhmann, Law as a Social System (Oxford: Oxford University Press, 2004). 36 MacCormick, Institutions of Law (n. 3), 294–8; Weinberger, Law, Institution and Legal Politics (n. 35) at 114, 158. 37  Institutions of Law (n. 3), ch. 2. While the existence of institutions is grounded in ‘pre-legal’ norms, this operates as an anthropological assumption about the conditions for social existence rather than providing the basis for the substance of particular moral or legal norms. Weinberger, Law, Institution and Legal Politics (n. 35), for example, denies the ‘subsiduary [sic] validity of moral concepts as determinants for the law’ (116). 38  cf. the distinction between primary and secondary rules in H. L. A. Hart, The Concept of Law (Oxford: Clarendon, 1961).

90

Lindsay Farmer

institution is deliberately broad, to include rules, institutional arrangements and competences, practices and values, this account is also linked to a more schematic history of legal persons, rights, and things, and the development of modern forms of law, the modern state and civil society. The point of this, MacCormick argues, is to recognize that while law might be (and indeed has been) institutionalized in many different ways, the dominant form of institutionalization is through state law and that it is accordingly necessary to acknowledge its central importance.39 While it is not possible to give a full account of this approach here, we can identify certain core features. At risk of oversimplification this approach seeks to combine a positivist recognition that particular legal systems develop in distinctive ways, such that the content and mode of expression of legal norms may vary between legal systems, with the view that there are, notwithstanding this, certain inherent features and values of modern law.40 This is captured in the idea of law as ‘institutional fact’. Here MacCormick argues that the ascription of legal personality or the recognition of the existence of a legal relationships requires the interpretation and application of the relevant legal norms. This requires not only knowledge of the law in a given system, but also of the values or ends to which the institutions were oriented. It is not possible to understand rules of contract or property, without also understanding them as part of a system of rules, including all areas of law, that strives towards a basic coherence, and which is oriented towards the intrinsic ends of the enterprise of governance under law.41 There are two dimensions to this claim. First, the resort to law or legal rules is purposive, both in the sense that law is understood as an instrument that can bring about socially desirable ends, and that, as one amongst a variety of means for addressing social problems in the modern state, the resort to law should be understood as a choice (even if this often appears to be unreflexive).42 Second, that a commitment to modern law as a means of bringing out certain social ends also entails a commitment to the forms and values of the institution of law. These both shape and constrain the way that certain ends can be brought about and, arguably, also entail a commitment to a certain form of human agency.43 Thus MacCormick argues that ‘a coherent account of the character of any modern legal system [will] have to take seriously the very general values that are inherent in the character of the legal 39  Institutions of Law (n. 3), 285–9. He puts a particular stress on Weber’s definition of the state as holding the monopoly of legitimate violence. 40  See 289–93 where he acknowledges the similarities to Luhmann’s systems theory, in which each functionally differentiated social system is organized according to its own code. See Luhmann, Law as a Social System (n. 35). 41  There is a striking affinity with the ideas of Fuller here in terms of the relation between legal form and the social purpose of law. See Lon L. Fuller, The Morality of Law (rev. edn.; New Haven: Yale University Press, 1969). See also K. Rundle, Forms Liberate: Reclaiming the Jurisprudence of Lon L Fuller (Oxford: Hart Publishing, 2012). 42  I would go further here and argue that even if, as in the classic liberal formulation, law is understood only as a framework for allowing individuals to pursue their own ends, this is still purposive in the sense that the sense of individual liberty is the purpose which shapes and orients the understanding of law as a structure of rules. On the conception of the modern state as a combination structure of rules and as purposive entity see M. Oakeshott, On Human Conduct (Oxford: Oxford University Press, 1975), 185–236 at 199–206. 43  See Rundle, Forms Liberate (n. 41), ch. 1.

Criminal Law as an Institution

91

enterprise’.44 The values that are associated with modern law are the commitment to the rule of law, understood as the articulation and adjudication of express and binding rules for the governance of social conduct, and somewhat more vaguely, a commitment to justice and the common good ‘according to some reasonable conception of these’.45 Two further points should be noted. First, ‘institutional facts’ require to be interpreted in a basic hermeneutic that seeks always to reconstruct—whether from the position of a judge or an academic—the best possible understanding of the rule and its place within the institution. It is key to this claim that such reconstructions acknowledge the systematicity of law and the values and ends to which the institution is (or should be) oriented. This is a complex undertaking in which there is always a critical mediation between the rules identified, the consistency with other rules in the system, the principles of the particular system, and the values and ends of the institution of law more broadly conceived.46 Second, while MacCormick contends that this process of interpretation and reconstruction cannot be value free—that is, that it must recognize both the values to which law is committed and the commitment to value in the process of reconstruction—he is insistent on the recognition of the separation of law from morality and that we recognize the ‘posited’ character of modern law. This requires that when we are dealing with legal institutional facts we must recognize that, even where these are shaped by, or recognize certain moral values, these must be understood primarily from the perspective of the institution of law. There are certain features of this approach that I want to stress. First, it begins from the social fact of law not morals.47 Second, the hermeneutic account of law then gives an important place to values in the self-understanding of the legal system, and looks at the way mechanisms for institutionalizing a commitment to these values have been built into modern legal systems. Third, while recognizing certain irreducible features of law as a mode of governance, this does not seek to produce a stipulative definition of the nature or scope of law or particular areas of law, but rather acknowledges that this is continually in the process of being worked out within the institution. Finally, this points to a necessary element of temporality and change. MacCormick focuses on certain features of modern law as positive law, while recognizing that the paradigm of modern sovereign states is itself in the process of evolution and change. It is then, in an important sense, a ‘post-positivist’ theory.48 What does this mean for the criminal law? MacCormick gives only a brief and rather sketchy account of the criminal law, but this contains significant pointers as to how an argument about criminalization might be developed.49 His suggestion 44  Institutions of Law (n. 3), 294. 45  Institutions of Law (n. 3), 304. The argument about justice and the common good is articulated in chs 14 and 15 (see esp. 274–7). 46  Institutions of Law (n. 3), 290–8. 47  This is not to say that moral values should not inform the law or even that the law should speak in a moral voice. See Institutions of Law (n. 3), chs 14 and 15. See also N. MacCormick, ‘Against Moral Disestablishment’, in Legal Right and Social Democracy (Oxford: Clarendon, 1982). 48  Institutions of Law (n. 3), 5. 49  Institutions of Law (n. 3), ch. 12.

92

Lindsay Farmer

is that in thinking about the conditions for a justifiable system of criminal law we should not start from the requirement of a wrong (that punishment can only be justified when an individual has committed a public wrong), such that securing civil peace is only seen as a side effect of the imposition of justified punishment on individuals. Instead, he argues that this relation must be reversed: that peace and civility should be seen as preconditions for justice, and that the just imposition of punishment can only take place once institutions of law have been established.50 The aim of the criminal law is, thus, understood for MacCormick in terms of securing the conditions of civil society.51 This, it is argued, is a matter of facilitating relative peace and mutual trust between strangers: The collective sense of security and solidarity in a relatively peaceful society is likely to depend on a fairly high degree of confidence among law-abiding persons that those who do not abide by the law, engaging in violent or dishonest behaviour, will be effectively restrained.52

The aims of the criminal law are accordingly linked to the operation of an effective system of punishment, but they have conceptual priority and are broader in scope than simply the justification of punishment. Thus while MacCormick wishes to recognize the existence of the right to punish, this is understood as an institutional right which serves other broader aims or purposes of the criminal law, rather than as an aim of the criminal law in its own right. These broader aims are described by MacCormick as those of ending forms of private vengeance and bloodfeud and securing a state monopoly over the legitimate use of violence, further institutionalized through the development of a specialized body of criminal law and agencies to enforce it.53 Security and civility are thus key themes running through his analysis. Indeed he goes on to argue that this sense of securing the civility of civil society is key to understanding the nature of criminal law as an institution for dealing with serious and wilful wrongdoing, as the civility of civil society is at least partly dependent on the existence of a fair and effective system of criminal law.54 However, the criminal law operates within the broader framework of institutional normative order that is state law, and this provides limits to the criminal law both in terms of functional competence (establishing boundaries between areas of private and administrative law) and through the increasingly widespread application of values such as the restraint of public actors which are developed in public law. Criminal law, on this view, is always expressive, or even constitutive, of the prevailing social morality adopted by the state—that is to say that it makes sense to conceive of criminal law as speaking in a moral voice, expressing community disapproval of conduct—but 50  Institutions of Law (n. 3), 218. 51  p. 207 and ch. 12 more generally. 52  p. 208; going on to point out that it also requires confidence that wrongdoers will be tried and prosecuted fairly. 53  Institutions of Law (n. 3), 207–9. This is discussed further in Section III.A. 54  There are, of course, a range of other institutions, both legal and non-legal, which also contribute to this general end. I shall discuss the sense in which securing civil order might be an aim specific to the criminal law in Section III.A.

Criminal Law as an Institution

93

this moral voice is always mediated through the institutions of law.55 This accordingly grounds the idea of public wrong in the aim of the institution and formulates it in a way which potentially has implications for the content of the criminal law.56 Crucially, however, it starts from the claim that criminal law has some specific public institutional purpose which is more than a right to punish or respond in some way to individual wrongs. While MacCormick presents the role of the legal theorist as that of rational reconstruction, developing the most coherent possible account of the rules, principles, and aims of institution, this might also be read as requiring a more radically historicized account.57 I take it as central to this account that, rather than attempting to identify features of criminal law in general, it is based on the identification of specific features of the modern criminal law, as this has evolved and is evolving in relation to certain political forms and forms of social organization. An account of the criminal law can only reflect its institutional form at a particular point in time, and criminal law theories are rational (to a greater or lesser degree) reconstructions of the institution. Thus, if we are looking at the institution of criminal law within the modern constitutional state this should be done by exploring how the modern institution of criminal law developed—the habits, customs, theories, and practices of criminal law that have shaped the modern law.58 This is more than just establishing the contingency of particular understandings of crime, or even wrongdoing, or even of particular institutional arrangements or practices— important as it is to be aware of these factors.59 It is about the development of the criminal law as a distinct body of rules and of the reflexive understandings of the nature and purpose of the criminal law. That is, the development of criminal law theory is not a matter of an inexorable progress towards an underlying truth or understanding, but is at best an attempt to rationally reconstruct an understanding of the nature and form of criminal law at particular points in time. This requires that we recognize the limitations of theory—precisely because criminal laws are institutional facts and as the institution evolves, so too does our understanding. Notwithstanding these arguments, it is important to recognize that MacCormick’s account of criminal law as an institution is undeveloped in certain key respects,

55  ‘Criminal law is always and inevitably expressive, or perhaps it is better to say constitutive, of a prevailing social morality adopted and enforced by the state’; MacCormick, Institutions of Law (n. 3), 211. 56  Institutions of Law (n. 3), 216. He also seeks to formulate this in terms which pull it away from a connection to something like the harm principle because the protection of peace or civility might include instances of criminalizing conduct where there is no clear harm (though recognizing that wrongful intention will be required). 57  Institutions of Law (n. 3), ch. 16. See also D. N. MacCormick, ‘Reconstruction after Deconstruction: A Response to CLS’, Oxford Journal of Legal Studies, 10 (1990), 539–58. 58  See also Lacey, ‘Historicising Criminalisation’ (n. 30). 59 See Farmer, ‘Criminal Wrongs in Historical Perspective’ (n. 11). For an attempt to trace patterns in the development of criminal law theory see N. Lacey, ‘Contingency, Coherence and Conceptualism: Reflections on the Encounter between “Critique” and the “Philosophy of Criminal Law”’, in R. A. Duff (ed.), Philosophy and the Criminal Law (Cambridge: Cambridge University Press, 1998).

94

Lindsay Farmer

and can at best be a starting point for theoretical development.60 I want to take up three points here as a way of trying to build on these basic methodological foundations. These are, first, the appropriateness of the identification of securing civil order or the conditions of civil society as the aim of criminal law; second, the question of whether this offers an adequate foundation for identifying the distinctive character of criminal law; and, lastly, I shall comment on how I see this kind of account feeding into normative theories of criminalization. I should also stress that my aim here is less that of clarifying MacCormick’s theory than to show how it might offer a basis on which I can develop a new kind of approach to thinking about criminalization.

A.  Securing civil society There is some confusion about the nature of the claim about the role of the criminal law in securing civil society. It appears in part as a threshold claim, drawing on Locke, about the need to establish peace in order to prevent private vengeance as a precondition for the development of legal institutions.61 However, in addition to this it is then treated as though it is also a specific aim of the institution of the modern criminal law and can shape its content in some way. The first version of the claim would deny any specific organizing role to the idea of securing civil order or society, seeing it only as a basic condition for establishing institutions of law: once civil peace was established by the entering into of a social contract, then the criminal law would require an independent justification. This has led Ulväng to express the concern that in fact the justification of the right to punish is more central to this account than MacCormick wants to concede—for this becomes the operative aim of criminal law once civil order has been established.62 I shall explain below why I think that there are good reasons for holding onto the second type of claim—that securing civil order is a continuing aim of the criminal law—but before doing so I want briefly to address this point. My argument is not denying the need for the justification of punishment, but it is suggesting that by treating securing civil order rather than punishment as the aim of the criminal law, questions about the justification of punishment are primarily institutional in character. This has two dimensions. First, as I have already argued, on this account civil order does not appear as a side effect of justified punishment, but as an aim of the institution. It may then turn out that the best way of sustaining civil society is through just punishment, justified through law—so that the ends and the means may combine in a way that acquires a force or weight of its own—but this is not a necessary relation. This does not mean that the demands of justice should be neglected, but that these demands should, at least in the modern criminal law, be seen as having 60  For critical discussion of MacCormick on criminal law see V. Tadros, ‘Institutions and Aims’ and M. Ulväng, ‘Criminal Law and Public Peace’, both in M. del Mar and Z. Bankowski, Law as Institutional Normative Order (Dartmouth: Ashgate, 2009). See essays in the book generally for discussion of the institutional analysis of law. 61  J. Locke, The Second Treatise of Civil Government (1690), ch. 2. 62  Ulväng, ‘Criminal Law and Public Peace’ (n. 63), 133–6.

Criminal Law as an Institution

95

emerged and been given shape with the development of the institution.63 As has been recognized from Hobbes on, there is no necessary connection between the right and the power to punish. A second set of concerns then cluster around the generality of the aims of the criminal law in relation to criminalization.64 On the one hand, it can be argued that the general aim of securing civil order is simply too thin or vague to contribute to any meaningful justification of why we might criminalize particular acts or omissions, or to explain why we might resort to criminal law rather than other forms of regulation.65 On the other, it is so broad as to be incapable of generating any limits to criminalization: almost any measure might be justified on the grounds that it contributes to securing civil order, and thus this can do no work in setting limits to the scope of the criminal law.66 Thus, for example, if we talk about securing civil peace this might mean anything from the establishment of the basic conditions of social life to the micro-management of civility in contemporary society.67 However, rather than seeing this as a weakness, I am inclined to view this as a potential strength as it opens up a range of questions about how ideas of peace and civil order have shaped the criminal law, how criminal law as an institution has sought to secure these ends, and how the criminal law has interacted with other areas of law or forms of regulation. At one level, the generality of the claim is important as it is broad enough to allow us to group together the wide range of social practices that have been called criminal law, but to see how the aim of securing civil order has been thematized in different historical periods and the level of specificity with which it has been pursued.68 Once again, questions of values and justice are an important part of this story, but the approach must be to look at the particular ways that respect for values has been institutionalized—and how certain values have come to be seen as setting limits on the scope or form of the criminal law. In addition, it directs us to look at 63  In the modern criminal law, the requirement for just punishment has been institutionalized in some codes and human rights instruments, regulated by appeal courts and review commissions. The values, and the demand for justice are important, but this should not obscure the fact that both the modern understanding of just punishment (proportional, certain, humane) and the particular form of its institutionalization are relatively recent, and followed on from the establishment of law. 64  A further question is whether we should see the aim of the criminal law as being fixed, or whether this also changes over time—and what this might mean for this theory. I do not address this here as I would argue that ‘securing civil order’ is a sufficiently broad way of describing the aim of the criminal law in the modern period. 65  Or put in another way, securing civil order might be seen as an aim of law in general rather than being specific to the criminal law. 66  See Ulväng, ‘Criminal Law and Public Peace’ (n. 63), 136–40. See also Tadros’s claim that this is inherently conservative—justifying existing institutions: ‘Institutions and Aims’ (n. 63), 84. See also F. Pollock and F. W. Maitland, The History of English Law (2nd edn. reissued; Cambridge: Cambridge University Press, 1968), ii. 453. cf. Dubber who regards this lack of limits as a characteristic of police power rather than the criminal law: M. D. Dubber, The Police Power (New York: Columbia University Press, 2007). 67  See e.g. Bottoms in this volume. 68  Central to this kind of account must also be an awareness of how ideas about civility have themselves been transformed over the modern period and how law has been used in advancing the civilizing project. See e.g. N. Elias, The Civilising Process, 2 vols. (Oxford: Basil Blackwell, 1982). The argument is discussed in K. Watson (ed.), Assaulting the Past: Violence and Civilization in Historical Context (Cambridge: Cambridge Scholars Publishing, 2007).

96

Lindsay Farmer

how the law has sought to secure these ends. For MacCormick, this is presented as a narrative of public institutions replacing private vengeance and the monopolization of force.69 This is clearly an important dimension of the emergence of modern criminal law, but it is not the only dimension. However, research suggests that the movement from private to public is not unidirectional: recent years have seen an increase in both public and private police; and moves towards increased public power might lead to increased regulation of the private domain.70 But once again this potentially opens up questions about the relationship between private and public, as forms of law and as more general questions of the distribution of social and political space that need to be explored. This also brings us to the question of limits. Once again the question here is the priority given to the institution: it is not that values are not part of the law, but it is important to understand how they have developed. Thus we must address the question of how criminalization becomes formulated as an issue for criminal law theory, or how concerns about limits become institutionalized with the development of the modern criminal law. Criminalization, in the modern sense of searching for a justification for the application of state criminal law to certain conduct, requires the prior emergence of an understanding of the criminal law as a unified body of rules with a distinct aim and character.71 In England the idea of criminal law first emerged only in the late eighteenth century as a way of organizing groups of laws relating to the King’s Peace and the police power. It brought together treasons, felonies, and misdemeanours that had previously been dealt with in different jurisdictions and under different procedures, and the label ‘criminal law’ emerged from a range of alternative formulations, such as pleas of the crown and penal law.72 The best-known way of thematizing criminal law from this period remains Blackstone’s idea of criminal law as ‘public wrong’, and the importance of this is that it captures the sense of the need to explain what makes this field of law distinctive by identifying a unifying concept capable of capturing the diverse practices of the emerging field.73 Later writers have taken up this challenge in different ways, and this modern understanding of the criminal law has developed together with an understanding that it was an area that was itself structured by values or principles.74 Explicit debates over 69  Thus he is not seeking to answer the general question of ‘why punish’ but merely that of ‘why states punish’. 70  See e.g. R. Matthews (ed.), Privatizing Criminal Justice (London: Sage, 1989); D. Sklansky, ‘Private Police and Democracy’, American Criminal Law Review, 43 (2006), 89; J. Suk, ‘Criminal Law Comes Home’, Yale Law Journal, 116 (2006), 2. 71  This is also reflected in dictionary definition of the term, with the word criminalization not being used in its modern sense until the mid-nineteenth century. See for further discussion. 72  See discussion in D. Lieberman, ‘Blackstone and the Categories of English Jurisprudence’, in N. Landau, Law, Crime and English Society, 1660–1830 (Cambridge: Cambridge University Press, 2002); L. Farmer, ‘Of Treatises and Textbooks: The Literature of Criminal Law in Nineteenth Century Britain’, in M. D. Dubber and A. Fernandez, Law Books in Action (Oxford: Hart Publishing, 2012). 73  Commentaries on the Laws of England (1769) (Chicago: University of Chicago Press, 1979), iv. 5. 74  Though the idea of ‘principle’ was itself still evolving. See L. Farmer, ‘The Idea of Principle’, in J. Chalmers et al. (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010).

Criminal Law as an Institution

97

a theory of ‘criminalization’ did not themselves emerge until the second half of the nineteenth century—notably in the clash between J. S. Mill and Stephen over liberty—perhaps as the idea of a criminal law became more settled, and certainly in response to changes in state function. But the central point here is that criminalization does not make sense in isolation from an understanding of modern character of the institution of criminal law. Before leaving this point we should note one final way in which the question of aims is relevant to an understanding of criminalization: that the aims of the criminal law in general, or of particular areas of law will shape understanding of the nature of particular wrongs and the way these are protected in law. The point can be made with a well-known example. It is, as I noted above, a common claim in criminal law theory that there are certain core wrongs, such as homicide or rape, which are recognized as such by all legal systems.75 Even if we were to concede the general point—that homicide and rape are crimes in all legal systems at all times—it quickly becomes apparent that the way that these are criminalized is linked to the aims of the criminal law. Thus, although it appears that rape has always been treated as a serious crime, the justification for the criminalization of rape has been very different in different periods and this shapes the scope of the crime—who is protected and what they are protected against. It has been argued that the crime of rape in early medieval English law was an offence against sexual property—or the property of the father in the chastity (and hence marriageable value) of his daughter—and consequently the law distinguished between the rape of different types of women, treating the rape of a virgin as the most serious.76 By the eighteenth century rape was seen as a crime of violence, with the consequence that non-consensual intercourse fell outside the scope of the crime.77 And in many jurisdictions the crime of rape is now conceived of as an offence against sexual autonomy—the choice of the individual to decide when and with whom to have sexual intercourse.78 This has the consequence of making consent (or its absence) the central element in the definition of the crime and broadening the definition of the actus reus (to include oral and anal penetration) as it is the wrong against sexual autonomy which is central, rather than the gender or status of the victim. The example is schematic but illustrates the central point that the wrong cannot be

75  See text at n. 14. 76 See Bracton on the Laws and Customs of England, translated with revisions and notes by Samuel E. Thorne (Cambridge, Mass.: Harvard University Press, 1968) at 415: ‘Punishment of this kind [castration] does not follow in the case of every woman, though she has been forcibly ravished, but some other punishment does follow, according as she is married or a widow living a respectable life, a nun or a matron, a recognised concubine or a prostitute plying her trade without discrimination of person.’ 77  See Blackstone, Commentaries on the Laws of England (n. 72), 201: Rape is the ‘carnal knowledge of a woman forcibly and against her will’. And notoriously a man could not rape his wife because consent was deemed to have been given on marriage: ‘for by their mutual matrimonial consent and contract the wife hath given up herself in this kind unto her husband, which she cannot retract.’ Sir M. Hale, History of the Pleas of the Crown (London, 1736), 629. 78 See e.g. Scottish Law Commission, Report on Rape and Other Sexual Offences (Rept. No. 209) (London: HMSO, 2007) and Sexual Offences (Scotland) Act 2009.

98

Lindsay Farmer

understood as an intrinsic quality of the conduct, but is linked to the aims of the criminal law which in turn shapes the scope of the legal protection.

B.  The distinctiveness of criminal law As noted in Section III.A, a criticism of taking the aim of securing civil society as an aim of criminal law is that it is too general. Can it not be said that all forms of law are in some sense directed towards the end of securing peace or civil order? If this is the case then this account is vulnerable to the same criticisms that I made of other theories in Section II: that it cannot identify anything that is distinctive about criminal law; or that in order to provide an account of the distinctiveness of criminal law it must fall back on something like the ends of punishment, as the only distinctive feature of criminal law is the routine application of punishment. This is a serious criticism, and in addressing it we need to turn to the literature on the nature of crime and criminal law. As an initial point it is important to distinguish between two different possible senses of the claim about being distinctive. The first is general, and the second is the more specific one of seeking to identify the positive characteristics that might be unique to the criminal law. Distinctiveness in the first sense takes the form of the claim that criminal law is not just adjectival, protecting rights that are defined elsewhere. It has a public function of its own, but it is also public law in a relevant sense. By this I mean that it is a means of protecting legally defined interests (rather than either private rights or moral wrongs).79 Distinctiveness is this sense is intended to capture the difference between law and moral theory.80 In relation to the second sense, we should begin by noting that attempts to produce a definition of crime and criminal law have largely struggled to identify satisfactorily characteristics which are unique to the criminal law or a substantive unity to the criminal law. We might as a consequence be wary of this attempt to identify unifying features or organizing principles, as it is not clear that this is a question capable of being answered in a straightforward way. Such attempts have routinely fallen into one of two traps. On the one hand, some definitions are too broad and positivistic. The best known of these is perhaps Glanville Williams’s claim that a crime is an act capable of being followed by criminal proceedings.81 This has been criticized on the basis that while it might be descriptively accurate, it 79  cf. the idea of Rechtsgut, or legally protected interests in German criminal law. Criminal law is not about the protection against moral wrongs, but interests which have been defined in law. On the struggle to reconcile the descriptive and normative in this concept see M. D. Dubber, ‘Theories of Crime and Punishment in German Criminal Law’, American Journal of Comparative Law, 53 (2006), 679. 80  cf. M. Thorburn, ‘Constitutionalism and the Limits of the Criminal Law’, in R. A. Duff et al. (eds.), The Structures of the Criminal Law (Oxford: Oxford University Press, 2011) at 97–102 for a recent account which seeks to argue that the constitutional state establishes the framework for civil life and that criminal wrongs are those which attack the very idea of living together under law. 81  ‘The Definition of Crime’, Current Legal Problems, 8 (1955), 107. For discussion see L. Farmer, ‘The Obsession with Definition: The Nature of Crime and Critical Legal Theory’, Social & Legal Studies, 5 (1996), 57–73.

Criminal Law as an Institution

99

is circular and says nothing about what might make proceedings criminal, and so is of little normative value. On the other hand, some theories have sought to identify a characteristic of the concept of crime that might provide unity to the idea of criminal law.82 A significant problem with these approaches is that by treating crime as an object which can exist independently of criminal law, unity might be achieved at the cost of sociological understanding.83 One way out of this dilemma, rather than focusing on the search for the single factor that is able to define the distinctive character of criminal law, is rather to note that there are instead a range of factors each of which has been accorded a certain weight and significance, and which might be capable of doing a certain amount of normative work in thinking about criminalization. The focus of the institutional theory is not necessarily to identify a single defining characteristic. Instead it is necessary to recognize that there are different kinds of norms in the criminal law and that the range of regulation, the scope of wrongs that have been criminalized and so on, have changed over time. In doing this, however, we should recognize that there is not necessarily a single modern criminal law. There are different stages or periods of development and the current criminal law is a combination of different institutions, practices, and so on from different periods.84 Rather than producing an a priori definition of criminal law, the first task of the theorist should be to trace the range of possible factors and justifications and their changing relation to the justification of punishment. However, while this might produce a descriptively rich account of criminalization, it leaves open the question of how this would feed into a normative theory of criminalization.

IV.  Rethinking Theoretical Approaches to Criminalization In concluding I want to address this question of how the approach set out here might contribute to normative theorizing about criminalization. First of all, the approach set out in this chapter has been to suggest that in thinking about crim­ inalization we need to pay more attention to the institution of law, and specifically to the distinctive character or aims of the criminal law. In doing so I am not suggesting here that we should disregard the central moral questions that have dominated writing about criminalization—wrongs, harm, punishment. However, I am arguing that neglect of the institutional features of criminal law has led to certain odd features in the literature on criminalization, and further that it may offer a way bridging the gap between this literature and important criminological 82  See e.g. G. Lamond, ‘What is a Crime?’, Oxford Journal of Legal Studies (2007), 609–32. 83  It can, of course, be argued that these are different kinds of question. The former is descriptive and analytical, and the latter is engaged in a philosophical attempt to identify principles or criteria which might assist in determining what might properly be criminal. However, they are closely related, and for reasons expressed in Section II I have reservations about approaches which seek to begin by identifying the character of wrongs without first acknowledging the role of law in shaping our understandings of those wrongs. 84  See D. Garland, Culture of Control (Oxford: Oxford University Press, 2001), ch. 8.

100

Lindsay Farmer

writings on criminalization. A recognition of the distinctive institutional features of the criminal law has been absent from existing normative theories of criminalization, and the methodology outlined here is intended in the first instance to be a corrective to this neglect. The purpose of a theory of criminalization is to come to terms with legal questions about the proper scope and limits of the criminal law. Moral theories have become an important resource in thinking critically about the limits and scope of criminal law, but should not necessarily be accorded priority. Second, while it is clear that my approach is one that aims to present a descriptively richer account of the modern criminal law, we might ask whether this is likely to be theoretically useful. When building a theory, we might decide to picture criminal law in a certain way for strategic reasons, and provided that this is not too remote from ‘reality’, there are clear advantages to such selectivity in terms of reducing complexity—and moreover a description, however rich, is not a theory. But the question here is that of how we can be sure that in doing so we are taking the most relevant or salient features of the criminal law. It is clear that a certain kind of orthodoxy has developed within criminalization theory—about there being a core of moral wrongs, about the relationship between general part and particular crimes and about the rational character of the criminal law. But does it make sense to see criminal law as a form of ‘rational coercion”, appealing to a subject’s responsible agency?85 Can or should all forms of criminal law fit the model of moral condemnation? Are there core or paradigmatic crimes? Many criminological and sociological accounts of criminal law would question such assumptions and so the aim of beginning with this kind of descriptively rich account of the criminal law to try and build up a more robust account of the relevant starting points for normative theory. Finally, a concern that is addressed to positivist (or post-positivist) approaches such as this is that in defining the scope and aims of the law in terms of a description of field of the criminal law, we must accept criminal law’s own account of its scope—a concern which can be seen to be linked to Tadros’s claim that MacCormick’s institutional theory generally neglects the demands of justice.86 The argument here is that this kind of approach will be incapable of generating the necessary critical bite when discussing criminalization, as the tendency will be to produce an account which will describe and legitimate existing institutions. But the risk here is the same as that run by any theoretical account of law: that of falling between irrelevance and apology. The answer to this lies in the critical space which is always opened between the institution and its values, and the process of reflecting on both the ideals and the practice in order to produce a critical understanding of the modern institution of the criminal law.

85  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 16), 6. 86  Tadros, ‘Institutions and Aims’ (n. 60), 87–8.

4 Bureaucratic ‘Criminal’ Law Too Much of a Bad Thing? Jeremy Horder

I.  Re-moralizing the Idea of Criminal Law: The ‘Counter-Reformation’ Recent years have seen the emergence of what might be called a ‘counter-reformation’ in criminal law scholarship.1 Counter-reformation thinking advocates a return to the idea that the criminal law ought to be employed to try people and punish them only for serious kinds of wrongdoing.2 In that regard, such thinking is strongly associated with the case for the confinement of criminal wrongdoing to wrongdoing accompanied by fault,3 and exemplified mainly by wrong actions rather than by culpable omissions (other than in exceptional cases4). To these articles of faith should be added the claim that it should always be the state’s burden, in criminal proceedings, to prove beyond reasonable doubt that such wrongdoing was committed by the defendant. I want to raise some questions about the strength of the case for a counter-reformation, insofar as it relates to at least some of these building blocks in the attempt to narrow the legitimate scope of the criminal law.5 To understand these questions, though, we must first sketch (and a sketch is all 1  Andrew Ashworth, ‘Is the Criminal Law a Lost Cause?’, Law Quarterly Review, 116 (2000), 225; Douglas Husak, Over-Criminalisation (Oxford: Oxford University Press, 2008); R. A. Duff, Answering for Crime (Oxford: Hart Publishing, 2007); A. P. Simester and Andreas Von Hirsch, Crimes, Harms and Wrongs: On the Principles of Criminalisation (Oxford: Hart Publishing, 2011). 2  In developing this thesis, I should not be taken to be denying that there have always been writers who have argued in favour of the ‘counter-reformation’ view; there have. My use of the term ‘counter-reformation’ is merely a convenient way to describe a certain cast of thought, and how it reacts to and compares with a different, ‘reformation’ cast of thought. 3  I take the point that, for some, there can be no ‘wrongdoing’ without fault, but I will not take up that issue here: see A. P. Simester, ‘A Disintegrated Theory of Culpability’, in Dennis Baker and Jeremy Horder (eds.), The Sanctity of Life and the Criminal Law: Essays in Honour of Glanville Williams (Cambridge: Cambridge University Press, forthcoming). 4  On which see Andrew Ashworth, ‘The Scope of Criminal Liability for Omissions’, Law Quarterly Review, 105 (1989), 424. 5 See further, Jeremy Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’, in G. R. Sullivan and Ian Dennis (eds.), Seeking Security: Pre-empting the Commission of Criminal Harms (Oxford: Hart Publishing, 2012), 79–102.

102

Jeremy Horder

that is possible here) the developments that led to the original reformation of the understanding of criminal law, radically departing from the traditional conception of that phenomenon.

II.  The Reformation in the Understanding of Criminal Law Year by year the subordinate government of England is becoming more and more important. . . . We are becoming a much governed nation, governed by all manner of councils and boards and officers, central and local, high and low, exercising the powers which have been committed to them by modern statutes.6

These are not the words of a Confederation of British Industry Chairman launching yet another broadside against the supposedly over-mighty state in the era of European law, but the words of no less a figure than F. W. Maitland, lecturing in the late 1880s. In that regard, the background to the ‘reformation’ view of criminal law was the significant expansion of regulatory criminal offences during the mid-nineteenth century.7 Amongst other catalysts for regulatory activity that proved potent in Victorian England, governments caught what Carolyn Steadman has aptly called ‘inspection fever’.8 Important examples of the fever’s symptoms being displayed were the powers given to Factory inspectorates (Factories Acts, from 1833), and Poor Law Commissioners (Poor Law Amendment Act 1834), but also illustrated through the establishment of numerous other bodies such as ‘undertakers’ empowered to manage and control markets and fairs (Markets and Fairs Act 1847). In many such instances, the executive bodies were themselves granted delegated powers, under what Bentham termed more generally ‘a sort of imperfect mandate [left] . . . to the subordinate power-holder to fill up’,9 to make rules breach of which could be a criminal offence.10 Nonetheless, according to Nicola Lacey,11 up to and including the time of Sir James Stephen,12 many criminal law writers of the seventeenth and eighteenth centuries, such as Blackstone or Hale, regarded the criminal law as having a narrower but distinctive morally legitimate field of 6  F. W. Maitland, The Constitutional History of England (Cambridge: Cambridge University Press, 1906), at 501. 7  W. H. Greenleaf, A Much Governed Nation, vol. ii (London: Routledge, 1987), at 534. 8 Carolyn Steadman, Policing the Victorian Community: The Formation of English Provincial Police Forces 1856–1880 (London: Routledge, 1984), at 54. See also Oliver Macdonagh, ‘The Nineteenth-Century Revolution in Government: A Re-appraisal’, Historical Journal, 2 (1958), 52, at 58. 9  C. W. Everett (ed.), Jeremy Bentham, The Limits of Jurisprudence Defined (New York: Chicago University Press, 1945), at 109, cited by Greenleaf, A Much Governed Nation (n. 7), ii. 532. 10  See e.g. Markets and Fairs Act 1847, section XLII (granting powers to make rules), and section XLIII (making breach of these rules a criminal offence). In 1819–20, ten statutes passed in that year conferred rule-making authority on ministers, commissioners, or other executive bodies, but in 1860 33 statutes passed had the same effect: Greenleaf, A Much Governed Nation (n. 7), vol. ii. 11  Nicola Lacey, ‘Legal Constructions of Crime’, in M. Maguire, R. Morgan, and R. Reiner (eds.), Oxford Handbook of Criminology (4th edn.; Oxford: Oxford University Press, 2007), at 185. 12  Sir James Stephen, A History of the Criminal Law of England, 3 vols. (London: McMillan, 1883).

Bureaucratic ‘Criminal’ Law

103

operation.13 For these writers (so the argument runs), what made the central case of a ‘truly criminal’ offence central was its connection to certain kinds of serious moral wrongdoing: broadly speaking, offences against religion, against the state, against the person, or against property.14 By the time Stephen was writing his history of the criminal law in the late nineteenth century, the sheer number of regulatory offences on the statute book that did not very obviously fit within this field of operation had become all-too evident. Stephen noted that so large in number and varied in character were these offences that, ‘it would be practically impossible in such a work as this to give anything like a full account of them within any moderate compass’,15 even though, in Stephen’s words, some of these offences, ‘relate to matters of the utmost importance and the deepest historical interest’.16 Like his predecessors, though, Stephen felt able to shrug off the need for even a cursory analysis of such offences because, in his view (repeatedly emphasized), they, ‘have so very faint and slight a connection with the criminal law [“properly so-called”, he might have added] that it would be out of place to enter upon that history at length in a work like the present’.17 That self-confidence in the proper scope of the criminal law remained reflected in the content of criminal law courses and of the textbooks that came to serve them during the twentieth century: including Kenny’s Outlines of Criminal Law,18 and, most famously, Glanville Williams’s Textbook of the Criminal Law.19 However, the further expansion of the regulatory state in the first half of the twentieth century undermined the confidence of many—including Williams himself—that, putting aside pedagogic considerations, criminal law could still be given a central-case analysis in terms of serious wrongdoing. In fact, it was Williams himself who came to influence a generation of post-war lawyers and theorists with his process-driven definition of crime in terms of the kinds of legal proceedings (civil or criminal) that may follow breaches of legal rules and standards.20 Williams was unsparing in his criticism of distinguished commentators who, in the early part of the twentieth century, had sought to defend the view that there was such a thing as criminal law properly so-called: ‘wrongdoing which directly and in serious degree threatens the security or well-being of society . . . [when] it is not safe to leave it redressable only by compensation of the party injured’.21 For Williams, there could be no ‘essence’ of crime in a world, as he saw it, dominated by mala prohibita, ‘with the close 13  For more on Bentham’s views, see n. 41. 14  Although these writers were, of course, perfectly well aware of the existence of regulatory criminal offences that fell outside this field of operation. 15 Stephen, A History of the Criminal Law of England (n. 12), iii. 263. 16 Stephen, A History of the Criminal Law of England (n. 12), iii. 264. 17 Stephen, A History of the Criminal Law of England (n. 12), iii. 264. 18  C. S. Kenny, Outlines of Criminal Law (London: Macmillan, 1902). 19  Glanville Williams, Textbook of the Criminal Law (London: Sweet & Maxwell, 1978). We should note, though, that the offences against religion and against the state gradually reduced in importance, as criminal law courses settled into a secular, individualist model still predominant today. 20  Glanville Williams, ‘The Definition of a Crime’, Current Legal Problems, 8 (1) (1955), 107. 21  Sir Carleton Allen, ‘The Nature of Crime’, in his Legal Duties (Oxford: Oxford University Press, 1931), 233–4. See also Jerome Hall, General Principles of Criminal Law (Indianapolis: Bobbs-Merrill Co. Inc, 1947), at 213.

104

Jeremy Horder

social and economic regimentation that seems to be an inseparable part of modern society’.22 By contrast, for Williams: A crime then becomes an act that is capable of being followed by criminal proceedings, having one of the types of outcome (punishment etc) known to follow these proceedings.23

Williams’ much-cited argument on this issue drew, in intellectual terms, on earlier process-driven definitions such as that of Kenny, who focused on the state’s power to halt prosecutions or pardon offenders in criminal (but not in civil) cases, or that of Winfield, for whom what was important was the fact that an offender could not bargain his or her way out of punishment in a way that is possible in civil proceedings.24 Such process-driven definitions of crime, with their Weberian emphasis on the roles and standards adopted and applied in virtue of office by officials,25 said nothing about what kinds of conduct should, or should not, be followed by ‘criminal’ proceedings. This—reformation—view nonetheless become very much an orthodoxy during the second half of the twentieth century (and rightly so),26 when so much important work was being done on the way in which criminal cases falling outside the traditional fields, such as health and safety legislation, came to court or were dealt with in other ways.27 As Nicola Lacey has put it: Criminal law . . . concerns itself with the formally established norms according to which individuals or groups are adjudged guilty or innocent. . . . In a system in which criminal law is regarded as a regulatory tool of government and in which (as in the UK) there are very weak constitutional constraints on what kinds of conduct can be criminally proscribed—a world in which everything from terrorism through dumping litter to licensing infractions and ‘raves’ can be criminalised—there is little that can be said by way of substantive rationalisation of the nature of criminal law.28

The reformation view focused on the criminal law as—in the spirit of Kelsen29—a normative field of meaning: a field in which we can say that the norms30 are empty vessels into which any content could be poured by government or, increasingly, by 22  Williams, ‘The Definition of a Crime’ (n. 20), at 115. 23  Williams, ‘The Definition of a Crime’ (n. 20), at 123. 24  See Kenny, Outlines of Criminal Law (n. 18), and Sir Percy Winfield, The Province of the Law of Tort (Cambridge: Cambridge University Press, 1931), ch. VIII, both discussed in David Ormerod, Smith and Hogan’s Criminal Law (13th edn.; Oxford: Oxford University Press, 2011), at 13. 25  Famously reflected in the central place given in the mid-twentieth-century analysis of a legal system to a rule of recognition accepted by officials, by H. L. A. Hart, The Concept of Law (2nd edn.; Oxford: Clarendon Press, 1961). There is, though, perhaps a hint of old-fashioned Austinian positivism in Williams’s unwitting shift from normative analysis (‘an act capable of being followed . . .), to empirical, predictive analysis (‘the types of outcome [punishment etc.] known to follow these proceedings’). 26  See e.g. the section on the definition of crime in Ormerod, Smith and Hogan’s Criminal Law (n. 24), at 3–15. 27  See e.g. W. G. Carson, ‘Symbolic and Instrumental Dimensions of Early Factory Legislation’, in R. G. Hood (ed.), Crime, Criminology and Public Policy (London: Heinemann, 1974). 28  Lacey, ‘Legal Constructions of Crime’ (n. 11), at 181 and 184–5. 29  Hans Kelsen, General Theory of Law and State, trans. A. Wedberg (New York: Russell & Russell, 1961), at 61. 30  Many of which have typically been power-conferring or declaratory in nature, imposing primary duties by implication rather than expressly.

Bureaucratic ‘Criminal’ Law

105

its bureaucratic agencies. The field was thought to be made meaningful as such (in a way that has now become problematic) by the presence of certain key procedural elements: the need for proof of the facts beyond reasonable doubt, the fact that proceedings were undertaken—or could be taken over—by agents of the state, by the availability of punishment following a finding of guilt, and so on.31 It could, of course, be argued that these key elements themselves represent a distinctive morality of the criminal law; but even if that were true, it would be—in the spirit of Fuller32—a morality of legal procedure, and not one of substantive law.

III. The Foundations of the Counter-Reformation A. Introduction Regulatory criminal law-making has long been an important dimension to state and legislative power. How else, one might ask, can one explain the presence in Magna Carta not only of the guarantee of key liberties, but also of regulatory criminal law-related rules dictating precise standard measurements for the sale of wine, ale, and cloth?33 One of the features of the (process-driven) reformation view of criminal law is that it has little difficulty accounting for the presence of bureaucratic or regulatory offences within the criminal law.34 So long as such offences are dealt with through a process involving the key elements of criminal procedure (just mentioned), then they would be criminal offences. What is more, reformation thinking is no mere ex post facto rationalization of early twentieth-century regulatory expansionism. With rare exceptions, courts throughout the nineteenth century, for example, consistently took the view that regulatory offences, even strict liability offences or those involving only a fine as punishment, were criminal offences.35 By contrast, modern writers have begun reasserting and elaborating on the views of Sir Carleton Allen36 that there is such a phenomenon as criminal law properly so-called, a phenomenon the existence of which calls into question the place of many regulatory offences in a criminal code.37 The emergence of the 31  See e.g. Benham v UK (1996) 22 EHRR 293. The meaningfulness of the field has become problematic, because it has become a matter of contention whether any particular one, or all, of the key elements must be present if proceedings are to be regarded as criminal. 32  Lon Fuller, The Morality of Law (New Haven: Yale University Press, 1965). 33  Clause 35 reads: ‘There shall be standard measures of wine, ale, and corn (the London quarter), throughout the kingdom. There shall also be a standard width of dyed cloth, russet, and haberject, namely two ells within the selvedges. Weights are to be standardised similarly.’ In itself, of course, this clause does not create a ‘criminal offence’ in the modern sense. However, it would have been taken as the standard to be applied under local weights and measures laws in, for example, the medieval market courts of piepowder, where sales in breach of such measures could be treated as an offence. 34  A definition of such offences can be found in the Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts. See also Ashworth, ‘Is the Criminal Law a Lost Cause?’, at 228. 35  See e.g. the cases discussed in Williams, ‘The Definition of a Crime’ (n. 20), at 111–12. 36  See n. 21. 37  See n. 1.

106

Jeremy Horder

counter-reformation is attributable, in part, to modern angst, shared by some politicians and sections of the media as well as by many scholars, about the number of criminal offences—especially regulatory offences—that have come on the statute book.38 This modern angst bears a striking similarity to the outpouring of criticism in the inter-war period of the growth of regulatory law in general (and of delegated power to make such law, in particular), criticism predictably accompanied by a ‘back-to-basics’ campaign.39 Ashworth himself starts his article seeking to shock us out of our complacency, by suggesting that, ‘there are probably around 8,000 criminal offences now, mostly created over the last 150 years, under the varying influences of . . . the expansion of regulatory mechanisms, and so forth’.40 With this background in mind, the counter-reformation view of criminal law tries to reignite the back-to-basics campaign as follows: Legions of strict liability offences, for example, penalise relatively minor omissions or wrongful acts. But a core element of criminal law, from a normative point of view, is that the criminal sanction should be reserved for substantial wrongdoing [involving harm and culpability].41

Ashworth’s account of criminal law, like that of its far less sophisticated early twentieth-century counterparts,42 departs from the traditional pre-reformation writers’ accounts in employing a more generalized notion of ‘substantial wrongdoing’ as the central case [‘core element’] of a criminal offence. Unlike the traditional accounts, it does not pick out categories of such cases—offences against religion, offences against the person, offences against property, and so on (although some of these are naturally included within the notion of substantial wrongdoing).43

38  For some political angst, see e.g. Daniel Hannan and Douglas Carswell, The Plan: Twelve Months to Renew Britain (London: Daniel Hannan and Douglas Carswell, 2008), ch. 8. In that regard, the authors rail against the mushrooming of so-called ‘Quangos’. The increasing importance of these bodies as regulators in the 1980s and 1990s owed a good deal to the burgeoning of sectoral regulation following the privatization of a number of sectors of the economy formerly subject to state control: see e.g. Maher M. Dabbah, ‘The Relationship between Competition Authorities and Sector Regulators’, Cambridge Law Journal, 70 (2011), 113. The authors’ complaints are much the same as those which met the expansion of various boards, ‘undertakers’, inspectors, and other officials—including the police—involved in regulation, from the middle of the nineteenth century. 39  See e.g. Lord Hewart, The New Despotism (London: Ernest Benn, 1929); Sir Carleton Allen, Bureaucracy Triumphant (Oxford: Oxford University Press, 1931). See also the view expressed by Glanville Williams, ‘The Definition of a Crime’ (n. 20). The back-to-basics campaign, so far as the ‘evil’ of delegated powers is concerned, is reviewed briefly in Greenleaf, A Much Governed Nation (n. 7), vol. ii, ch. 6. 40  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 226. 41  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 240 (my emphasis). It would be possible to date the origins of some elements of this view to enlightenment thinking, and in particular to Bentham’s view that a criminal law must address only ‘mischiefs’ so grave that the punishment of them produces overall less evil than would the continued toleration of the mischief. Part of this assessment for Bentham, as for Ashworth, involved consideration of whether leaving the matter to the civil would be as efficacious in deterring the mischief: Jeremy Bentham, Principles of Penal Law (ii), 1.4, cited by Kenny, Outlines of Criminal Law (n. 18), at 24. 42  See Allen, ‘The Nature of Crime’ (n. 20). 43  I have sought to criticize the focus on ‘substantial wrongdoing’ as an effective constraint on criminalization in Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’ (n. 5).

Bureaucratic ‘Criminal’ Law

107

In taking that course, Ashworth’s account is obviously better fitted for—indeed, perhaps inspired by—the ideals of the non-perfectionist, secular liberal state, in which criminal justice is broadly seen in night watchman terms as founded on the equal protection of important liberties, a value regarded as lexically prior to concerns about security more generally and (even more clearly) prior to concerns about moral or spiritual disintegration.44

B.  Playing the numbers game: some historical notes At this point, it is worth subjecting to scrutiny one driver behind counter-reformation thinking. This is the more or less clearly articulated idea that there are now ‘too many’ criminal offences. It seems reasonable to infer that Ashworth believes that 8,000 criminal offences on the statute book (and at common law) is a great deal too many, and he is not alone amongst distinguished criminal law scholars in holding such a view.45 The belief that citizens are hemmed in on all sides by extensive and intrusive criminal legislation (and by regulatory requirements more generally), and hence dependent on the favourable discretion of prosecutors to maintain their freedom, is, of course, not a new belief. In 1762, Oliver Goldsmith spoke of the gap he perceived to exist between the number of enforceable criminal laws, and the number of instances in which they were actually enforced: There is scarcely an Englishman who does not almost every day of his life offend with impunity against some express law, and for which in a certain conjuncture of circumstances he would not receive punishment. Gaming-houses, preaching at prohibited places, assembled crowds, nocturnal amusements, public shows, and an hundred other instances are forbid and frequented. These prohibitions are useful; though it be prudent in their magistrates, and happy for their people, that they are not enforced, and none but the venal or mercenary attempt to enforce them.46

Goldsmith’s views were noted by A. V. Dicey over 120 years on, and similar observations about the perceived scale of criminalization were being made at that later

44  Ashworth is famous as a critic of the notion that concerns for liberty and justice must merely be ‘balanced’ against a concern for security and public order: see Andrew Ashworth, ‘Criminal Procedure, Human Rights and Balance’, in Thomas Elholm et al. (eds.), Ikke kun Straf: Festskrift til Vagn Greve (Jurist- og Okonomforbundets Forlag 2008). On the so-called ‘lexical’ priority of liberty, see John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1999), at 266. 45  See e.g. Douglas Husak’s magisterial treatment of the subject as it applies to Federal law in the USA in his Over-criminalisation (n. 1); John Spencer, ‘The Drafting of Criminal Justice Legislation— Need it be so Impenetrable?’, Cambridge Law Journal, 67 (2008), 585. 46  Oliver Goldsmith, Works, vol. cxi (London: John Murray, 1854), 194–5. For similar contemporary criticisms, see Archdeacon Paley’s Principles of Moral and Political Philosophy (1785 edn.) at 504, and Sir William Blackstone, Commentaries on the Laws of England (1765–9), iv. 2–4. In highlighting the lack of enforcement of existing laws, Goldsmith is alluding to the ‘common informer’ system under which a private individual bringing an offender to justice would share in any fine imposed (‘the venal or mercenary’). The system was ineffective in a number of ways. For various reasons I cannot go into here, from Tudor times onwards (and probably long before), only some 2–3 per cent of the courts’ time was occupied by such offences.

108

Jeremy Horder

date by F. W. Maitland, in his analysis of what he called active and negative duties, imposed by the state: [T]‌urn from active duties to negative duties, duties which consist in forbearance only and see how we are surrounded by prohibitions: the list of offences for which one may be punished summarily by justices of the peace is enormous.47

As Maitland was well aware, many of these offences were by the late Victorian period when he was writing already commonly created through the use of secondary legislation, whose use became increasingly controversial in the early twentieth century. This is what Cecil Carr said about the phenomenon (and he was not alone), in his famous lectures on delegated legislation published in 1921: In mere bulk the child now dwarfs the parent. Last year, while 82 Acts of Parliament were placed on the statute book, more than ten times as many ‘Statutory Rules and Orders’ of a public character were officially registered under the Rules Publication Act. The annual volume of public general statutes for 1920 occupied less than 600 pages; the two volumes of statutory rules and orders for the same period occupy about five times as many. The excess in mere point of bulk of delegated legislation over direct legislation has been visible for nearly thirty years.48

It is interesting to speculate on what Carr’s view would have been, had he focused specifically on the criminal laws passed in 1920. Adopting the interpretation of separate criminal offences set out in R v Courtie,49 between 150 and 190 offences— the vast bulk of which are regulatory in character—were created by primary legislation alone in 1920, with at least a further 80 created in secondary legislation passed in that year.50 Naturally, civil servants in the mid-twentieth century were well aware of the potential for controversy in the use of the criminal law to help achieve regulatory ends. Writing in 1954, Sir Frank Newsam, Permanent Under-Secretary of State for the Home Department, claimed that: It is part of the Home Secretary’s general duty to watch that the penalty provisions included in legislation promoted by other Departments are not out of keeping with those in existing 47 Maitland, The Constitutional History of England (n. 6), at 505. 48  See Cecil T. Carr, Delegated Legislation; Three Lectures (Cambridge: Cambridge University Press, 1921), 2. 49  [1984] AC 463 (CA). Ashworth (‘Is the Criminal Law a Lost Cause?’ (n. 1), n. 8) does not take that approach to counting offences. He takes a less expansive approach that involves counting as one offence individual provisions within statutes that create offences, rather than looking to the number of separate offences within the provision. As he rightly concedes, that entails an underestimation—in my view, a very large and significant one—of the numbers of offences created in any given year. The difference of approach illustrates some of the methodological difficulties facing scholars seeking to play the numbers game. Indeed, I will later criticize the adoption of a Courtie approach as liable to create distortions in analysis: see text at n. 74. 50  Not the least of the difficulties facing any scholar seeking to decide how many criminal offences were created in a single year is that, as the state has sought to extend its governance through law, statutes have become more systematized, becoming interdependent and closely related. So, for example, a statute in year X may make it henceforth a criminal offence for anyone to attempt to breach—or to become complicit in the breach of—regulations made criminal as substantive offences in year X minus 1; or, conversely and much more commonly, a year X statute may make it a criminal offence to breach regulations that may be passed in the future under powers granted by the year X statute; and so forth.

Bureaucratic ‘Criminal’ Law

109

statutes, and that they are reasonable in themselves. The creation of new criminal offences and penalties may sometimes have unexpected consequences, and the Home Office is constantly on its guard to foresee and prevent them.51

The same claim continues to be made over 50 years later by the Ministry of Justice, which issues guidance on principles of restraint in the creation of criminal offences by other departments of state.52 Even so, Newsam conceded that the power to create criminal offences had been considerably widened by the removal from the Home Office’s regulatory jurisdiction of a wide range of social and economic activity, which was now the responsibility of officials in mushrooming departments such as Health, Transport, Housing, Labour, Agriculture and Fisheries, and Education. Such divisions of labour between increasing numbers of separate departments were motivated in part by a perceived need for specialization, and each department that saw itself in those terms accordingly exercised rule-making powers, including powers to create criminal offences, on the basis of an ‘expert’ theory of authority. This was a development that had long been analysed or predicted by amongst others Karl Marx, who saw such divisions of labour—at least when justified by expert or technical considerations—as a necessary evil in modern society.53 However, in spite of this, in Newsam’s view, these developments had not diminished in any way more generally the volume of work with which Home Office officials were expected to cope, leaving the Home Office hard-pressed to control and direct the criminal law-making activities of more or less newly minted Departments of state.54 No doubt, few Home Office ministers since then would beg to differ on that point.

C. Playing the numbers game: the normative (in)significance of numbers What does this tell us about the over-proliferation thesis? An initial difficulty for anyone seeking to defend the over-proliferation thesis is (as the Law Commission has observed55) to know how many offences would be about right, or how many would be for that matter too few, knowledge of which is surely important if the thesis is, as such, to gain significant traction. Perhaps 3,500 offences, would be about right? Would, by contrast, 1,500 offences be too few? These questions only have to be asked for their unhelpfulness to be plain for all to see. Secondly, over-proliferation theses typically beg important questions. For example, putting 51  Sir Frank Newsam, The Home Office (London: George Allen & Unwin, 1954), at 125 (my emphasis). 52 : ‘A gateway established to prevent the proliferation of unnecessary new criminal offences’. 53  Karl Marx, Capital: A Critique of Political Economy, vol. i (New York: Vintage Books, 1977), at 781–94. See also Max Weber, The Theory of Social and Economic Organization (New York: Oxford University Press, 1947), at 339: ‘Bureaucratic administration means fundamentally the exercise of control on the basis of knowledge. This is the feature of it which makes it specifically rational’ (my emphasis). 54 Newsam, The Home Office (n. 51), at 125–6. 55  Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), at para. 3.18.

110

Jeremy Horder

it crudely, are fewer numbers of broadly defined offences better (and in what respect?) than large numbers of highly specific ones? Fewer specific offences generally entails a smaller number of more broadly defined ones; but that reduction in quantity may lead to a corresponding deterioration in the quality of the criminal law. It would, for example, perhaps be possible to abolish a swath of specific animal welfare offences, and to seek to rely instead on a single broad crime of causing, or risking,56 unnecessary suffering to protected animals.57 To take that (ideologically speaking) ‘common law’ course, though, would be to sacrifice a large measure of legal certainty respecting what conduct amounts to a criminal offence on the altar of crime number reduction, a certainty highly prized by many of the adherents of the over-proliferation thesis themselves.58 Such a policy would place a heavy burden on the courts adequately to develop the meaning of the broader offence in its new role. Yet, we can have little confidence that this burden will be effectively discharged, not least because the courts have been conspicuously poor at settling on a coherent policy to determine the meaning of such offences, as has been the case with public nuisance.59 Quite simply, as amateurs dabbling in the complex, highly varied and expert-dominated world of regulatory control, Appeal Court judges will never be up to the task, even if the lis inter partes were suitable—and it is manifestly wholly unsuitable—as a means of developing a regulatory strategy for criminal offences (or, for that matter, civil penalties). It is possible to imagine a world in which guidelines on meaning and on law enforcement policy are issued by departmental and industry experts that relate to every context in which the general offence of causing or risking unnecessary animal suffering may take place: to name but a few such contexts, the commercial transport of animals,60 the disturbance of basking sharks,61 the possession of a live badger,62 the maintenance and running of dog kennels,63 pet shops,64 and circuses,65 the possession of dangerous wild animals,66 the failure to provide a proper living envir­ onment for a pet cat,67 or the sale of goldfish at fairs.68 However, when underpinned with an offence as vague as causing or risking the unnecessary suffering of a protected animal, such an approach can only serve to set up a whole series of unhelpful 56  Risking harm would have to be added to the existing general offence of causing unnecessary suffering to a protected animal contrary the Animal Welfare Act 2006, s. 4, in order to provide adequate coverage, not least when the concern is with the handling and upkeep of very sensitive animals or of large numbers of animals. 57  A similar point could be made about the vast numbers of specific environmental offences, which could in theory be dealt with solely through the common law offence of public nuisance. 58 Andrew Ashworth, ‘Ignorance of the Criminal Law, and Duties to Avoid it’, Modern Law Review, 74 (2011), 1. 59  See John Spencer, ‘Public Nuisance—A Critical Examination’, Cambridge Law Journal (1989), 55. 60  See e.g. the Welfare of Animals During Transport Order 1992. 61  Wildlife and Countryside Act 1981, s. 5. 62  Protection of Badgers Act 1992, s. 4. 63  Animal Boarding Establishments Act 1963. 64  The Pet Animals Act 1951. 65  The Performing Animals (Regulation) Act 1925. 66  The Dangerous Wild Animals Act 1976. 67  Animal Welfare Act 1986, s. 9, and see, in relation to this, the Code of Practice for the Welfare of Cats: . 68  Animal Welfare Act 2006, s. 11.

Bureaucratic ‘Criminal’ Law

111

tensions between the principle of legality and the rule of law,69 leaving citizens— and those tasked with regulating their conduct—in acres of legal no man’s land. What Glanville Williams once claimed to be a defect of mala prohibita would in all probability turn out to be the defect of turning back to a common law approach: ‘it is utterly inevitable that the citizen can only find out the limits of the permissible by bringing down the law on his own head’.70 It is one of the unsung virtues of clear and specific secondary legislation (including criminalizing legislation) made under primary legislation directed at regulating particular industries or practices, backed by an enforcement strategy informed by experts following consultation with industry specialists, that it is capable of coordinating behaviour in the interests of the common good in a way that general common law-style offences cannot hope to do. Accordingly, the legislature has for decades (indeed, longer) sought to specify in some detail what failures in point of animal welfare will attract criminal sanctions in regulatory contexts. Consider section 5 of the Welfare of Animals During Transport Order 1992, made under the Animal Health Act 1981, that makes it a criminal offence to fail to abide by the following conditions: Persons having control of animal transport undertakings 5. Subject to article 6 below, every person having control of any animal transport undertaking which transports animals in the course of business or trade shall— (a) ensure that the animals are entrusted only to persons possessing the knowledge necessary to administer appropriate care to the animals in transport; (b) ensure, in the case of animals travelling unaccompanied, that the consignee is prepared to receive them; (c) ensure that the animals are transported without delay to their place of destination; (d) ensure that during the journey the consignment is accompanied by a certificate signed by him or on his behalf stating: (i) the origin and ownership of the animals, (ii) their place of departure and place of destination, and (iii)  the date and time of departure; (e) draw up for journeys exceeding 24 hours a journey plan showing— (i) the arrangements for the animals to be rested, fed and watered, and (if necessary) unloaded and given accommodation appropriate to their species; (ii) the arrangements for feeding and watering in the event that the planned journey is changed or disrupted, and ensure that during the journey the consignment is accompanied by the journey plan; and (f ) ensure that copies of the journey plan and the certificate required by this ­article are kept for a period of six months from the end of the journey and produce them at the request of an inspector. 69  See, in a different context, J. M. Finnis, ‘Invoking the Principle of Legality against the Rule of Law’, New Zealand Law Review (2010), 601. 70  Williams, ‘The Definition of a Crime’ (n. 20), at 115. Williams probably took this view of mala prohibita because, at the time he was writing, there were no systems of the kind currently widespread in the world of regulation devoted to giving publicity to and guidance on a person’s obligations and on how to comply with the law: see Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’ (n. 5).

112

Jeremy Horder

Was the passing of this Order an example of the unnecessary proliferation of regulations backed by criminal sanctions, or alternatively, was it good statecraft attentive to the need for clarity and certainty concerning the taking of important steps to avoid animal suffering? I do not believe that, to preserve the purity of the ‘true’ criminal law, such rules may be justified only if they end only in some kind of civil penalty. The RSPCA, for example, said in response to the Law Commission’s proposal to replace a greater number of criminal sanctions with administrative penalties: The RSPCA also consider that the power to impose a system of civil sanctions already enacted in the Regulatory Enforcement Sanctions Act 2008 and introduced in the Environmental Sanctions Order 2010 should not be extended beyond environmental offences (which was the intended ambit of these pieces of legislation) onto the [Animal Welfare Act 2006]. Prosecutions concerning sentient beings are in a different category to other types of prosecutions. Offences under the [Animal Welfare Act] should remain criminal offences without the introduction of a system of civil sanctions.71

So, is unjustifiably risking animal suffering in itself necessarily a ‘serious’ wrong that thereby qualifies as a candidate for ‘true’ criminalization? Or, is the real truth that such wrongs vary enormously in terms of seriousness, and that there is no clear discernible tipping point at which they become ‘truly’ criminal?72 For Ashworth and Zedner, it is late twentieth-century legislators who must take a large slice of the blame for supposed over-criminalization. They say, ‘the pace of change appears to have quickened in the final quarter of the last century [the twentieth century] and this acceleration is continuing’.73 Yet, methodological difficulties beset any attempt to prove or disprove the numerical element to this thesis, whichever approach one adopts to decide the ‘how many?’ and ‘too many?’ questions.74 For example, here is one problem with adopting the approach to counting set out in R v Courtie, in order to prove the thesis.75 It is the difficulty of knowing whether the difference between formally separate criminal offences is so significant that their separateness is really an important factor in any critique of (supposed) over-proliferation. Consider the aforementioned Welfare of Animals During Transport Order 1992. A question arises concerning how appropriate it is to see the 1992 Order as having (needlessly?) added so many offences to the list. This is because the mischiefs separately addressed are so closely related. This is an

71 , at para. 1.26. 72  See the discussion of ‘seriousness’ as a benchmark in Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’ (n. 5), and see also the concluding section to the current chapter. 73 Andrew Ashworth and Lucia Zedner, ‘Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure and Sanctions’, Criminal Law and Philosophy, 2 (2008), at 21. I have conceded that increasing specialization and multiplication of executive and quango functions has led to an increase in the volume of criminal legislation over the last 25 years: see n. 38. The Law Commission has made the same point: see Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts, at paras 1.21–1.24. 74  See text at n. 49. 75  [1984] AC 463 (CA). See text at n. 49. The Courtie approach was not, of course, created by the Court of Appeal for the purpose for which it is now being used.

Bureaucratic ‘Criminal’ Law

113

issue that must be addressed in considering quite literally thousands of regulations creating criminal offences in the UK. When offences are closely interrelated and mutually interdependent in the constitution of a coherent scheme of protection, as in clause 5 (set out above), is it right to criticize each separate offence for having on its own contributed in a some significant way to an undesirable ‘proliferation’ of criminal offences? Bear in mind that the separation of the wrongs in section 5 is meant primarily to enhance clarity and specificity for the purposes of fair and effective regulation, so far as both enforcers and transporters are concerned. Accordingly, it can be argued that it is misleading in this instance—as it would equally be in thousands more instances of similar secondary legislation—to point to this section of the 1992 Order as an example of the ‘lost cause’ that our supposedly runaway criminal law-making process is claimed to be. In this, as in many other respects, much modern legislation simply follows the examples set by legislation creating criminal offences in regulatory contexts going back many years. The legislature has long preferred highly specific individual criminal wrongs, moving away from continued reliance on very broad general offences of a common law type (other than as a back-stop).76 This does no more than reflect what is now a largely unquestioned feature of good statecraft: the modern duty (the origins of which lie in the Victorian era) to provide for clear, fine-tuned regulatory governance in the promotion of safety and public welfare in an ever-widening variety of contexts.77 The legislation of 1920 is replete with statutes providing a highly context-specific criminal law basis for, or dimension to, such governance, examples being: the Census Act 1920, the Dangerous Drugs Act 1920, the Ready Money Football Betting Act 1920, the Seeds Act 1920, the Roads Act 1920, the Firearms Act 1920, the Official Secrets Act 1920, the Employment of Women, Young Persons and Children Act 1920, and many more (and this is not to mention all the regulations and orders also passed in that year to the same end, such as the American Gooseberry Mildew Order 1920). Many of these Acts and Orders drew in to the process of regulatory governance already existing professional bodies, officials, and employers. The Acts and Orders sometimes included a role for such people in enforcement or in licensing, such as Chief Police Officers (Firearms Act 1920), the Registrar-General (Census Act 1920), medical practitioners (Dangerous Drugs Act 1920), and employers (Employment of Women, Young Persons and Children Act 1920). However, when it comes to the delivery of regulatory goals it is also worth noting that whilst in 76  An excellent analysis, where public nuisance is concerned, can be found in Spencer, ‘Public Nuisance—A Critical Examination’ (n. 59). Perhaps unmatched for the level of detail provided is the Locomotives Act 1861, s. 3, which saw the creation of a one-sentence, 158-word offence concerned in great detail with the size and composition of permitted Locomotives and their tyres. 77  Care must be taken in interpreting this claim. It is obviously no part of the claim that the nineteenth and twentieth centuries saw, from modest beginnings, a carefully stage-managed march towards wide-scale regulatory control. Historians still debate the question of the extent to which the nineteenth century in particular saw any state ‘planning’ of an overall strategic kind. For A. J. P. Taylor, for example, England simply, ‘stumbled into the modern administrative State without design’: A. Taylor, Laissez-faire and State Intervention in Nineteenth-Century Britain (London: MacMillan, 1972), at 236.

114

Jeremy Horder

1902 there were only 50,000 non-industrial civil servants, by 1939 the number had risen to 163,000, and by 1944 it was 505,000.78 Against that background, moreover, we should note that some of the 1920 legislation demonstrates that long before accession to the EU, the passing of regulatory criminal legislation had been the product of international agreements or of international standards, rather than of some peculiarly English taste for ‘the teasing vigilance of the perpetual superintendence of law’.79 Examples of such quintessentially white collar, civil service-led criminal law are the Dangerous Drugs Act 1920, which sought to give effect to the Hague International Opium Convention of 1912, and the Employment of Women, Young Persons and Children Act 1920, which sought to give effect to International Labour Organization standards. In terms, then, of governments’ ability to deliver regulatory strategies (including widespread use of criminal offences), through increases in the human resources of officialdom needed to carry out that activity, it is highly arguable that the really significant modern period was the first half of the twentieth century, and not the last third.

D. Over-proliferation theses, the EU ‘problem’, and the concept of ‘corporate’ crime According to Julia Black, regulation is, ‘the sustained and focused attempt to alter the behaviour of others according to standards or goals with the intention of producing a broadly defined outcome or outcomes, which may involve mechanisms of standard-setting, information-gathering and behaviour modification’.80 Precisely because they focus on aspects of strategic planning in the role of the employer, of the doctor, of the Vet, of the public official, or of whoever else is charged with playing a part in policy implementation, regulatory criminal laws are preventive offences, aimed primarily at risk reduction rather than with harm already done. The last point is picked up by Ashworth and Zedner who treat regulatory criminal law with some suspicion as a largely unwelcome encrustation on good old-fashioned, ‘truly criminal’ law: The historic orientation of the criminal justice system towards reactive policing and post-hoc punishment is [now] overlaid by a pro-active, preventative rationale that seeks to avert harms before they occur.81

In that regard, though, it is telling that, in his powerful argument decrying the expansionary tendencies of the criminal law, Ashworth does not take as one of his main case studies a piece secondary regulatory legislation. Instead, he concentrates 78  Strikingly, there are now only 435,000 non-industrial civil servants: . 79  An early nineteenth-century MP’s phrase to describe the evils of codification: Hansard, xix, col. 647 (29 March 1811). 80 Julia Black, ‘Critical Reflections on Regulation’, Australian Journal of Legal Philosophy, 27 (2002), 1. 81  Ashworth and Zedner, ‘Defending the Criminal Law’ (n. 73), at 40. In fairness, their critique of pre-emptive crimes is also concerned with their use in non-regulatory contexts.

Bureaucratic ‘Criminal’ Law

115

a great deal of his fire on a preventive measure that is, strictly speaking, epipheno­ menal in criminal law terms, namely the anti-social behaviour order, a two-step prohibition not regarded by the courts as in as of itself a criminal offence.82 Whether or not anti-social behaviour orders were ever justified, that focus is telling. It is an illustration of what might more broadly be considered to be the intense concern of many criminal lawyers interested in criminalization with individual or group ‘trouble-making’ (and violence) in towns and cities—especially when engaged in by disadvantaged young people—and with the state’s continuing commitment to what is viewed as a disproportionate and often unnecessary criminal law response to such ‘trouble-making’ or violence.83 Whatever its merits in its context (and I do not underestimate its moral importance), this liberal-individualist Dickensian focus, as common on the ‘right’ as on the ‘left’ of the spectrum of criminological thought,84 leaves too little examined what became during the twentieth century— and still remains—the biggest driver of criminalization in modern Britain. This is the attempt to regulate the activities of businesses, and of small businesses in particular, not least (in more recent times) in order to deal with the challenges of creating a single European market. For example, historically, a great deal of legislation underpinned by criminal offences has always been aimed at farming and allied trades, even though this sector constitutes only 2–3 per cent of UK enterprises, and the advent of the single European market has sustained that trend. As a recent illustration, let me take 2008 as an example. In that year, 440 offences were brought on to the statute book. Perhaps as few as 30 of these offences were aimed at private individuals. Of the rest, aimed at business activity, 179 offences were created by the DEFRA, 133 of which were aimed specifically at the agricultural sector. One of the main influences here may be (there is not enough evidence to prove it) the impact of European law in seeking to create a level playing field for trading and the harmonization of good trade practices throughout the EU. Notoriously, the EU is responsible for far less UK legislation than the general public supposes, even though 92 per cent of European rules and standards are incorporated into law in the UK by secondary legislation. For example, between 1987 and 1997, of the 27,999 statutory instruments issued only 7.9 per cent made reference to European legislation (although this rises to 15 per cent if local SIs are excluded).85 Having said that, the government department that was most concerned, in its own activities, with turning European rules and standards into secondary legislation (‘Euro SIs’) was what is now the Department for the Environment, Food and Rural Affairs (DEFRA), but was formerly the Ministry of Agriculture, Fisheries 82  See Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1) 83  See Husak, Over-Criminalisation (n. 1). 84  See e.g. George L. Kelling and James Q. Wilson, ‘Broken Windows: The Police and Neighbourhood Safety’, ; contrast: R. J. Sampson and S. W. Raudenbush, ‘Seeing Disorder: Neighborhood Stigma and the Social Construction of “Broken Windows”’, Social Psychology Quarterly, 67 (4) (2004), 319–42. 85  Edward Page, ‘The Impact of European Legislation in British Public Policy-Making: A Research Note’, Public Administration, 76 (1998), 803.

116

Jeremy Horder

and Food (MAFF). Of all the secondary legislation passed under MAFF’s aegis between 1987 and 1997,86 no less than 51.3 per cent of these laws can be classed as Euro SIs. By contrast, 28 per cent (still a substantial figure) of the secondary legislative activity of what was then the Department for Trade and Industry can be called Euro SIs, together with 21 per cent of the secondary legislative activity of the Department of Transport. These figures are significant, because the criminal law has historically had to be a mainstay of EU implementation strategy in many areas of business activity for almost all UK government departments, given that there has until relatively recently been no tradition of creating a category of administrative or ‘civil’ penalties to supplement or replace ‘criminal’ legislation.87 A cursory examination of Euro SIs created by secondary legislation in any given year since accession to the EU (further research is needed on this point) often reveals a high proportion that create criminal offences, the bulk of which are usually included by what was MAFF and is now DEFRA. For example, in 2008, it seems likely that over 190 of the 440 offences created in that year were meant to give effect to common European standards relating to the single European market.

E. Administrative penalties: the counter-reformation thinkers’ holy grail? At this point, I must directly confront the argument that it is not the number of offences that is the problem, so much as the fact that so many of the wrongs concerned are regarded as ‘criminal’ offences, when they should be treated instead as civil or administrative offences.88 According to Ashworth and Zedner, ‘[t]‌here is a strong argument for confining criminal liability to offences that require fault, and of creating a separate category of MAPs (Monetary Administrative Penalties) . . .’89 to punish and deter other kinds of wrong. Certainly, few people seriously doubt that there is plenty of legitimate work for civil or administrative penalties to do in some cases;90 but is the answer really as simple as they suggest? Regrettably, it is not. To begin with, such penalties—MAPs—have been condemned by some as neither fish (‘truly’ criminal), nor fowl (civil law wrong), or have been branded cheap and not-so-cheerful substitutes for ‘truly’ criminal law, yet carrying much more severe fixed punishments in some cases than would ever have been imposed

86  Such as the Welfare of Animals during Transport Order 1992, discussed in Section III.C. 87  Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34). I should not be taken here to be endorsing a clear distinction between ‘criminal’ law and ‘administrative penalties’. 88  See e.g. R. A. Duff, ‘Perversions and Subversions of Criminal Law’, in R. A. Duff, Lindsay Farmer, S. E. Marshall, Massimo Renzo, and Victor Tadros (eds.), The Boundaries of the Criminal Law (Oxford: Hart Publishing, 2010), at 102–5; Victor Tadros, ‘Criminalisation and Regulation’, in Duff et al. (eds.), The Boundaries of the Criminal Law, at 174. 89  Ashworth and Zedner, ‘Defending the Criminal Law’ (n. 73), at 33. 90  See Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34).

Bureaucratic ‘Criminal’ Law

117

through the exercise of a criminal court’s sentencing discretion.91 In that, ironically, the critics of such penalties share common ground with some of the penalties’ supporters, who have decried the criminal courts’ tendency to take regulatory crime far less seriously than they should, by imposing fines amounting to far less than the costs saved by the offender in committing the offence.92 Further, in wishing away regulatory offending as best dealt with by civil or monetary penalties rather than by the ‘true’ criminal law, counter-reformation thinkers fail to acknowledge that the spread of such penalties can be seen as challenging— rather than embodying or symbolizing—the idea that there is something morally distinctive about ‘criminal’ sanctions. The European Court reserves the right to decide the essentially contested question whether or not an offence is in substance criminal, or a civil wrong, according to a range of criteria it determines for itself. That creates a theoretical and practical vicious circle in relation to the answering of the question: ‘what is a criminal offence’? So, any cordon sanitaire that a counter-reformation thinker might seek to erect around ‘truly’ criminal offences, to preserve them from adulteration by mere civil penalties, must inevitably remain porous in practice and contentious in theory. The implications of this have been set out by Reid (if I can be forgiven for citing his work at length): The important fundamental distinction could be seen as marking out not the criminal law, but those laws through which the state (in whatever form) imposes penalties. The criminal law is only a subset of this bigger genus. Given the proportion of criminal offences nowadays dealt with by fixed penalties and a range of other statutory diversions from prosecution, how much of even the mainstream criminal law lives up to the traditional paradigm of criminal law which entails those accused of offences being brought to trial with punishment being imposed only by a court after due process designed to protect the rights of the accused? The question then becomes one of identifying the fundamental elements, substantive and procedural, which justify the state in imposing penalties of any sort and then finding appropriate ways of applying these in different contexts, only some of which may be ‘criminal’ in our minds, but all of which must provide appropriate safeguards against abuse of state power. Most discussion over ‘criminalisation’ misses out this vital preliminary stage, suggesting that the options are criminalisation or nothing, as opposed to seeing the criminal law as part of this wider penalty-imposing framework.93

One way of construing MAPs is simply as a formalized out-of-court disposal by punishment of offending behaviour. If that seems plausible, then the understanding of society as subject to over-‘criminalisation’ is better replaced by a different focus. This is the question whether, in regulatory contexts, people are now potentially subject to over-penalisation respecting all manner of wrongdoing and risk

91  Robin White, ‘“Civil Penalties”: Oxymoron, Chimera and Stealth Sanction?’, Law Quarterly Review, 126 (2010), 593; and see the response of Professor Colin Reid to the Law Commission’s Consultation paper (text at n. 92): . 92  See Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), appendix A. 93 , at para. 1.26.

118

Jeremy Horder

taking, rather than simply to over-criminalization (as counter-reformation thinkers understand that notion). It seems that counter-reformation thinkers do not have a ready answer to that question.94 Yet, if Reid is right about the terms in which the debate about the legitimacy of state coercion should be conducted, the latter question is as—or more—important than the former.95 There have been two recent attempts to theorize administrative penalties in a sophisticated way, with a view to establishing a basis for maintaining the purity of the ‘truly criminal’ prohibition. The first is focused on the idea that administrative penalties are deterrents unaccompanied by official condemnation or censure (parking fines writ large) whereas, normatively speaking, conviction for a genuine crime involves—or should involve—such condemnation or censure.96 One difficulty with the theory is that it runs together two slightly different ideas. First, there is the idea of ‘censure’, which whilst obviously a response to wrongdoing, is a narrow concept closely associated with an official rebuke to a public servant for inappropriate behaviour, historically including sexual misconduct and the use of bad language in the legislative chamber: what could be called ‘officials’ censure’.97 Whilst officials’ censure is unlikely to be imposed except in cases of serious wrongdoing, so that it remains distinct from lesser measures such as official reprimands, there is nothing in the idea of officials’ censure that ties it to serious wrongdoing, as such. Less serious wrongdoing by a very senior trusted official, such as the US President, might call for officials’ censure as much because of the wrongdoer’s status as because the wrongdoing was serious in itself. So, it is strongly arguable that it is misleading to use the concept of ‘censure’ to describe an intrinsic element in the criminal conviction of a private individual. Indeed, in English law, one of the few uses of ‘public censure’ as an intrinsic part of the punitive process—when the Financial Conduct Authority publicly censures instances of financial wrongdoing— is as a way of penalizing wrongdoing outside the traditional criminal courts. By contrast, some might consider the more general concept of ‘condemnation’ as rightly associated with cases where serious wrongdoing is in issue, whatever the status of the wrongdoer; but to use the term ‘condemnation’ to confine the scope of wrongdoing appropriate for criminal conviction seems to beg the question. To describe criminal conviction as concerned, or as ideally concerned, with wrongdoing appropriate for ‘condemnation’ is to foreclose discussion of the very issue to be decided: should criminal liability be confined to wrongdoing that ought not to be 94 Although I should mention, in fairness, that Ashworth and Zedner point to the possible net-widening effects of being able to impose fixed penalties, and to the fact that fixed penalties will, necessarily, bear down more harshly on offenders with low incomes: Ashworth and Zedner, ‘Defending the Criminal Law’ (n. 73), at 28. 95  A point noted by Tadros: ‘Criminalisation and Regulation’ (n. 88), at 175. 96 Duff, ‘Perversions and Subversions of Criminal Law’ (n. 88), at 102–5; J. Feinberg, ‘The Expressive Function of Punishment’, in J. Feinberg, Doing and Deserving (Princeton: Princeton University Press, 1970). There is, of course, nothing unique about the criminal law as a means of imposing official censure or condemnation. The removal of a knighthood, medals, or other honours, in response to wrongdoing, involves such censure or condemnation. 97  ‘Punishment in the House’, The New York Times, 18 November 2010, drawing on Congressional Research Services.

Bureaucratic ‘Criminal’ Law

119

merely criticized but condemned? Is condemnation not, in any event, merely very strong criticism, and thus a judgment of degree at one end of a spectrum rather than involving a difference of kind? Duff himself suggest that there can be such a thing as ‘mild’ condemnation, and hence concedes that: [T]‌he criminal law need not deal only with serious wrongs—it can also provide modest punishments for offences that are, while still genuine wrongs, relatively minor. . . . [R]egulations serve the common good; breaches of them are therefore breaches (often minor breaches) of our civil responsibilities, which merit (often mild) condemnation as wrongs.98

The conviction-as-condemnation theory has also been criticized as putting too much emphasis on what is in fact a contingent element of the punishment process. As Tadros puts it: We can impose suffering on people to deter other people without intending to communicate condemnation for that person’s wrongdoing. Also, punishment is not distinctive from civil penalties in imposing suffering on people . . . . Fines that are imposed only in order to compensate the victims may make the person who is liable to pay them suffer.99

For Duff, administrative penalties, ‘are not marked, addressed or condemned as wrongs’.100 However, if, following Tadros, we take the condemnatory element out of this claim as being normatively supererogatory or redundant, then it becomes much less clear how, on the one hand, ‘true’ crimes, and on the other hand, mere administrative transgressions, are to be qualitatively distinguished. Tadros himself has an alternative theory aimed at maintaining a realm of purely criminal law, distinct from the world of administrative penalties. On his account, penalties are not aimed at making people suffer in order to deter others. On the contrary, for him, penalty schemes have no pretensions to deter at all, for all the suffering that the imposition of the penalties may cause. For Tadros, penalties are aimed instead at ‘redistributing benefits and burdens’.101 He continues, ‘penalties are aimed at ensuring that the costs of breaching regulations are borne by those who breach it rather than those that suffer from the breach’.102 Ingenious though the theory is, I suggest that it turns out to be possibly even less convincing that the censure-or-condemnation theory it is meant to replace. To begin with, Tadros’s theory about official responses to regulatory wrongdoing— that ‘penalties are aimed at ensuring that the costs of breaching regulations are borne by those who breach it’103—does not reflect current practice, and it would be grossly 98 Duff, Answering for Crime (n. 1), at 173–4. Quite rightly, Duff adds that a system of administrative penalties can still coexist with a criminal law that extends to minor wrongs. 99  Tadros, ‘Criminalisation and Regulation’ (n. 88), at 174. 100  Duff, ‘Perversions and Subversions of Criminal Law’ (n. 88), at 103. Accordingly, Duff recommends the introduction of a specific category of ‘administrative offence’, a category for transgressions that do not involve ‘public wrongs’ worthy of censure, wrongs that, ‘properly concern us all as citizens’. Such a scheme is likely to be fraught with difficulty and controversy, whether or not it has been made to work in some jurisdictions: see the criticism of the idea by the Law Commission, Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), paras 3.28–36. 101  Tadros, ‘Criminalisation and Regulation’ (n. 88), at 174. 102  Tadros, ‘Criminalisation and Regulation’ (n. 88), at 174. 103  Tadros, ‘Criminalisation and Regulation’ (n. 88), at 174.

120

Jeremy Horder

unjust if it were to do so. Many of the significant costs of implementing regulatory penalties—the employment of officials, outlay in relation to any investigation and the making of findings, together with the costs of fine recovery and of conducting appeal hearings—must in fact commonly be borne by the state rather than by the offender. This is simply because it would be unduly harsh and out of proportion to make the wrongdoer bear all these costs, other than in exceptional cases.104 So, it is not strictly true to say that penalty schemes aim to ensure that, ‘costs of breaching regulations are borne by those who breach it,’ nor, in any legal system that aspires to do justice in a fair and proportionate way, should it ever be true. At a deeper level, though, the theory fails to identify what is in fact a distinctive role for penalties for breach of an obligation in a regulatory scheme, separate from the role of behaviour-influencing taxation. Suppose that it is true, as Tadros claims, that the point of penalties is solely to ‘redistribute[e]‌benefits and burdens’105 as between those governed by the regulatory scheme and those not governed (but possibly affected) by it. It then becomes unnecessary, and possibly distorting, to distinguish between a redistributive tax on certain kinds of regulated conduct and a penalty scheme for engaging in certain kinds of regulated conduct; but that seems wrong. Suppose we wish to improve the environmental conditions in a traffic-bound city centre, and we intend to make car users pay the costs of that. Amongst other things, we could choose either or both of charging an advance fee for bringing a car into the centre at certain times,106 or imposing a penalty on those found driving in the centre at certain times. Traditionally—and in my view rightly—the latter is a distinctive deterrent strategy based on the creation of a malum prohibitum. Morally speaking, amongst other things, when pursuing such a strategy it will be improper for an official to say, ‘Actually, we don’t care if people are left undeterred by and are willing to pay the fines, because that means more money for environmental improvements’. When the pursuit of regulatory penalties is not motivated in part by their potential to deter, it is wrong to seek to impose them.107 In particular, it will be improper to use them solely for redistributive purposes, in such circumstances. Naturally, when a penalty scheme is introduced as part of a redistributive revenue-raising exercise, it may be tempting to judge its overall success in that exercise by the financially enriching side effects that punishing law-breakers has. Taking that approach, a penalty scheme may turn out to be relatively inefficient as a revenue-raising mechanism if law-breakers are simply too few, as well as when they are too hard or too 104  See e.g. the Citizens’ Advice Bureau’s report, Uncivil Recovery, into the disproportionate nature of the costs imposed on employees guilty of minor infractions, when charged for civil recovery actions undertaken by private firms at the behest of the employer: . 105  Tadros, ‘Criminalisation and Regulation’ (n. 88), at 174. 106  This option assumes that no penalty scheme is needed, because the fee is effectively impossible to avoid: say, barriers or traffic officers automatically turn back all drivers who try to enter the centre without payment of the fee beforehand, or the like. 107  This claim intentionally overstates my case, to aid clarity in casting doubt on Tadros’s theory. In fact, the motivation for imposing a penalty can be just deserts based, but I will not go into that issue here.

Bureaucratic ‘Criminal’ Law

121

expensive to catch, or when extracting fines from them proves too difficult. As a penalty scheme, though, its function is to deter by the threat or imposition of a financial setback proportionate to the malum prohibitum in issue. So, the scheme should not be judged, as such, principally by the amount of money it raises through failures in point of deterrence; that would be perverse. By contrast, a strategy of charging an ex ante fee to drive in the city centre may have a simple redistributive function at its heart, revenue raising for environmental improvements being its intended effect. In such a case, a transport official may quite properly say, ‘if, in spite of the fee, people want to keep driving in the same numbers into the city centre, that is fine by us, because it means more revenue for environmental improvements’. In pursuing the fee-charge strategy the authority may also wish to deter entry into the city centre, something it may try to do by imposing a fee sufficiently high that more people than before will simply avoid coming into the centre; but all that shows is that it can be appropriate for officials to use taxes (or advertising, or other strategies) to influence behaviour, as well as to raise revenue.

IV.  Individualism and Corporate Criminal Liability In critically analysing 39 new offences of this kind created in a single year, 1997, Ashworth finds yet more to disagree with about supposedly modern trends in criminalization than the sheer number of offences and the proliferation of regulatory offences. He divides them for the purposes of analysis as follows: 1. Thirteen requiring proof of intention or recklessness, most concerned with the provision of false information. 2.  Nine strict liability offences, with defences placing the burden on the accused; 3.  Eight omission-based offences, with defences placing the burden on the accused; 4.  Six strict liability offences, with no applicable defences.108

He then goes on to say: [T]‌he bulk of new offences are characterised by three features—strict liability, omissions liability, and reverse onus provisions for exculpation. All these features lie a considerable distance from the conception of criminal laws held by many university teachers and criminal practitioners. Indeed, they are inconsistent with prominent elements of the rhetoric of English criminal law—that there is a presumption that mens rea is a prerequisite of criminal liability, that liability for omissions is exceptional, and that ‘one golden thread’ running through English criminal law is that the prosecution bears the burden of proving guilt.109

There are three points to be made about the claims just set out. First, there is absolutely nothing new (post-1945) about the presence of strict liability and omission-based liability, reverse burdens, or the criminalization, without more, 108  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 227–8. 109  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 228 (footnotes omitted).

122

Jeremy Horder

of the provision of false information. The 1920 legislation considered earlier has many such provisions, and Maitland was moved to remark on them in his work on constitutional law 30 years before that.110 They are part of what has for long been the ‘natural law’ of regulatory strategy and enforcement (broadly, rational choice theory) that has for hundreds of years cast a barely acknowledged shadow over the vaunted ‘golden thread’ thesis.111 Secondly, the foundations of Ashworth’s argument seem surprisingly contentious.112 Do we really know if it is true that reverse onus provisions, together with strict and omissions-based liability, detract equally and to the same extent from the conception of criminal law held by most university teachers and criminal practitioners? Even if they do, what do these groups—or majorities within them—have in common that makes their conception of the criminal law the right normative starting point, and not just a focal point for a liberal rallying cry against a perceived curse of modernization? Thirdly, Ashworth does not use as an organizing distinction whether or not the offences in question are regulatory in character, even though he acknowledges that, ‘the bulk of the new offences may be described as “regulatory”, in the sense that they form part of statutory schemes for the regulation of certain spheres of social or commercial activity, and are generally enforced by the regulatory authority rather than by the police’.113 This is a significant point in this context, because I suggest that some of the concerns that Ashworth raises about the breadth of criminal liability draw their moral strength from a supposition that it is an individual, not a business, facing conviction. Ashworth has a strong preference for the use of intention and recklessness as fault elements in any ‘criminal’ law; indeed, he goes so far as to say that these fault elements should be intrinsic to the ‘paradigm crime’.114 Further, he is not—with some important exceptions115—a supporter of the widespread use of criminal liability for omissions. Taking the issue of fault elements first, even when the criminal liability of individuals is in issue, one may question the claim that intention or recklessness should be intrinsic to the ‘paradigm crime’. In France—not especially noted for the brutality of its criminal code—criminal liability for negligently caused harm is possible for a number of stigmatic offences against the person.116 In English law, rape is a paradigm crime, but few would now be prepared to argue for the old view that the fault element should be an intention to have non-consensual sexual intercourse, or subjective recklessness as to whether such intercourse might take place, rather than some form of negligence as to the absence of consent.117 The fact that intention and recklessness may be highly morally significant in some contexts—in the civil as well as in the criminal law—does nothing to show that they 110  See text at n. 136. 111  See, for example, the early regulatory scheme set up under the Knackers Act 1786. 112  For an argument similar to Ashworth’s, see Husak, Over-Criminalisation (n. 1), at 34. 113  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 228. 114  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 241. 115  Ashworth, ‘The Scope of Criminal Liability for Omissions’ (n. 4). 116  French Penal Code, arts 221-6, 222-19. 117  For the old view, see DPP v Morgan [1976] AC 182.

Bureaucratic ‘Criminal’ Law

123

are, or should be, of intrinsic significance within the criminal law as a whole.118 This is well illustrated by the fact that when companies are accused of offences, there are in fact serious objections to the use of intention and recklessness as fault elements. If subjective fault elements are used in the definition of crimes aimed at companies, this almost inevitably entails the application of the ‘identification’ doctrine of corporate criminal liability to such elements.119 According to that doctrine, if the company is to be found guilty of an offence involving proof of fault, an individual company director (or equivalent person) must be found to have possessed the relevant fault element at the time in question.120 As is well known, the larger the company and the more diffuse, regionalized, or contracted-out its operations, the less likely it is that any director (or equivalent person) will have actually known or suspected that an employee might commit a criminal offence involving proof of subjective fault.121 It is not surprising, thus, that when corporate liability is in issue, governments of different political stripes have preferred to shift the burden and costs of proof of lack of fault to the company, by moulding a broader conception of fault to fit the goals of regulatory compliance. Governments have frequently done this by requiring the company to prove not that it was unaware of and intended no wrongdoing (a negative), but that it exercised all due diligence and took all reasonable steps to avoid the wrongdoing being committed (a positive).122 As many commentators have pointed out, this broader perspective on fault enables legal analysis to be applied to the way that corporate systems operate to reduce or increase risks, rather than to be applied simply to what particular individuals knew or did not know at a given moment. The broader perspective is also, though, sufficiently flexible to accommodate the fact that, in the case of sole proprietorships or their equivalent, it is not so much rule-governed ‘systems’ that guide business decisions but individual effort and care.123 Ashworth himself is amongst a number of commentators who have called for a more punitive approach to wrongdoing by companies,124 although the focus of such calls is almost always medium- to large-sized businesses that are far fewer 118  For criticism of Ashworth’s view that the importance of a principle can be broadly measured by the generality of its application, see John Gardner, ‘Ashworth on Principles’, in Lucia Zedner and Julian V. Roberts, Principles and Values in Criminal Law and Criminal Justice (Oxford: Oxford University Press, 2012), at 7–10. For more discussion of the role of intention and recklessness in regulatory contexts, see the Conclusion. 119  Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), part 5. 120  Tesco Supermarkets v Natrass [1972] AC 153 (HL). 121  For a full discussion, see Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), part 5. 122  See the Law Commission’s discussion of, in particular, the use of criminal legislation in the consumer protection field, Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34), paras 3.83–3.112. For an appraisal of the role of proving due diligence, see Jeremy Horder, Excusing Crime (Oxford: Oxford University Press, 2004). 123  This is acknowledged, for example, by the Law Commission in its recognition that less may be expected of small firms in terms of bureaucratic accountability, when it comes to putting in place adequate systems to prevent bribery: Law Commission, Reforming Bribery (Report No. 313), part 6. 124  Ashworth, ‘Is the Criminal Law a Lost Cause?’ (n. 1), at 251.

124

Jeremy Horder

in number than sole proprietorships or their equivalent (although they generate a greater share of GDP). Accordingly, in what follows, I shall also assume that the focus is medium- to large-sized businesses. At issue is the extent to which the understanding and shape of the criminal law should be driven by values central to individual human lives, such as treatment with concern (as capable of emotional suffering) and respect (as capable of giving an intelligent conception of the good intrinsic value in their lives), dignity, and personal autonomy. In Ashworth and Zedner’s account of criminal law, such values drive what they call their ‘liberal conception of the criminal law that treats individuals as autonomous moral agents’.125 However, the nature of companies is such that they do not need to be treated with concern in the sense just given, because the stigma of conviction cannot induce emotional suffering in these bodies as such, they do not have any dignity to be trampled on by reverse burdens of proof, and they do not have any personal autonomy to be threatened by omissions-based liability or by the imposition of risk-based penalization. Penal legislation directed at such companies must be both morally permissible and the best option in the circumstances, it must allocate burdens in a fair way, and it must be proportionate (all of which is also true in relation to individual criminal liability), but beyond that it need not reflect the much fuller range of concerns at issue when individuals face penalization. There is no space to deal with these issues adequately here, but let me by way of example look at two questions bearing on the issue of personal autonomy: the risk-based nature of much regulatory criminal law, and its reliance on omissions-based liability. As we have seen, regulatory criminal law frequently adopts an anticipatory perspective on penalization,126 being comprised largely of so-called prophylactic crimes, offences where, ‘the risk of . . . harm does not arise straightforwardly from the prohibited act. It arises only after further human interventions either by the original actor or by others.’127 Engaging in a form of regulated trading without a licence is perhaps a good example. In itself, such a practice may pose no risks of harm on a given occasion; but sooner or later, if people engage in the regulated trade without meeting the licensing conditions necessary to engage in that trade, the risk of harm from poor trading practices will rise to unacceptable levels.128 Even so, Ashworth and Zedner criticize the criminalization of much risk-based offending on the following basis: To hold a person responsible now for her possible future actions (i.e. without proof of an intent to do the actions), as may occur in respect of pre-inchoate liability and crimes of possession, is objectionable in principle, since such a prediction ‘denies my responsible agency 125 Andrew Ashworth and Lucia Zedner, ‘Prevention and Criminalisation: Justifications and Limits’, New Criminal Law Review, 15 (2012), 542. 126  See Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’ (n. 5). 127  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 1), at 79. 128  This claim assumes that the licensing conditions in question are indeed necessary and appropriate for reducing risks of poor trading practices to acceptable levels. In that respect, there are many studies that engage in economic analysis of licensing conditions. See e.g. John W. Borowski and Gerard C. S. Mildner, ‘An Economic Analysis of Taxicab Regulation in Portland, Oregon’, .

Bureaucratic ‘Criminal’ Law

125

by treating me as someone who cannot be trusted to guide his actions by the appropriate reasons’.129

One difficulty with this argument is its narrowness. It relies on there being a distinction of great moral significance between simple possession, and an intention to do something (wrong) in relation to the thing possessed; but in many instances this distinction is not morally significant.130 Very commonly, when the law imposes restrictions on possession, it does so because to possess the thing is thereby to become morally responsible for its treatment in a certain (harm-reducing or welfare-enhancing) way, irrespective of one’s intentions in that respect. An example might be the possession of animals, whether domestic pets (the keeping of which is governed by the Animal Welfare Act 2006), or dangerous wild animals (governed by the Dangerous Wild Animals Act 1976). To be in possession—and especially, of course, to be an owner—of a protected animal is in and of itself to be under an obligation to treat that animal in a certain way, whatever one’s intentions with regard to it; and the heavy regulatory structure governing the treatment of animals in all manner of contexts reflects this.131 So, to give a memorable example from the early Victorian era, simply by coming into possession of an eagle, a jackal, and a bear when he came up to Christ Church, Oxford, Frank Buckland (son of the famous naturalist, William Buckland) came under an obligation to treat those animals in a certain way. Breach of that obligation is now rightly treated as a criminal offence under the Dangerous Wild Animals Act 1976.132 A second, related difficulty with Ashworth and Zedner’s argument is its over-robust individualism. Buckland, no doubt, thought that his intentions were benign, and—given that he was acknowledged as an authority on animals—would certainly have regarded any regulatory state restrictions on his possession of the animals as, to use Ashworth and Zedner’s words, ‘objectionable in principle . . . by treating [him] as someone who cannot be trusted to guide his actions by the appropriate reasons’. The introduction of a standard-setting licence to engage in conduct only on certain conditions is an expression of the perfectly legitimate belief that, were no such standard enforced, too many individuals wishing to engage in that conduct may indeed turn out to be people who ‘cannot be trusted to guide [their] actions by the appropriate reasons’, in a context where a free-for-all approach may lead to unacceptable levels of risk of harm. In many areas of activity, individuals act in conditions of far more imperfect knowledge than the state and the experts who inform the state’s decisions, as well as differing greatly in the level of their 129  Ashworth and Zedner, ‘Defending the Criminal Law’ (n. 73), citing Duff, Answering for Crime (n. 1), at 165. 130  Elsewhere, Ashworth has in fact conceded that there may be a case for the criminalization of possession beyond cases in which the possession is motivated by an intention to do some future (harmful) action: see Andrew Ashworth, ‘The Unfairness of Risk-Based Possession Offences’, Criminal Law and Philosophy, 5 (2011), 237. 131  See the statutes mentioned in the text following n. 60. 132  For the full story, see Noel Annan, The Dons (London: Harper Collins, 1999), ch. 2. Buckland named the bear Tiglath-Pileser, and dressed ‘Tig’ in a cap and gown before taking him to wine parties where Tig could be hypnotized for the amusement of other guests.

126

Jeremy Horder

personal commitment to a concern for preventing harm, in the resources they have that can be devoted to promoting the safety of others, and so on. Trusting the individual to do the right thing, then, may quite often turn out to be a wrong and unacceptable option, particularly when the chances of an individual devising his or her own harm-free path inevitably depend on the way in which many others potentially affected seek to determine their own paths.133 This is not, of course, to advocate universal nanny-state paternalism. In introducing licensing schemes for a restricted range of activities, a government is not denying to an individual the right to determine for him or herself what ends (including that activity) to pursue in the exercise of his or her autonomy. For good harm-prevention reasons government is, instead, simply setting down a structure within which that end may more safely be pursued, in the light of the need to accommodate others’ possibly cheek-by-jowl pursuit of their own ends, and so on.134 More pertinently, the difficulty with the argument is that its application to corporate liability is far from secure. In relation to the liability of mentally competent and mature individuals, Ashworth and Zedner are only on strong ground when a pre-inchoate offence is concerned with conduct based on reasons the appropriateness of which it would be better to leave the individual him or herself to determine. An example might be the decision to buy potentially dangerous items such as a kitchen knife, weed killer, pain killers, or bleach. In such cases, the chances of misuse may be too small, the importance of expert guidance too negligible, and the cost of regulation too high bearing in mind the numbers of people engaging in the conduct perfectly safely, for state intervention to be worthwhile. These considerations may or may not be decisive in any given case, quite apart from any personal autonomy-based consideration militating against intervention. However, companies cannot enjoy personal autonomy. So, whatever other reasons there may be against the criminalization of activity companies engage in, the protection of personal autonomy will not be amongst them.135 Corporate activities may generate a great deal of autonomy-enhancing value for those who work in them and for the public at large,136 but that value is not value intrinsic to companies’ own ‘lives’. So, whilst governments may, perhaps wrongly, fail to trust companies to be guided by appropriate reasons, such a failure cannot manifest an improper interference with 133  In short, Ashworth and Zedner overlook the state’s need to impose rules that have binding force even when in particular circumstances individuals may be confident that will do better if guided by reasons they bona fide regard as appropriate: see J. M. Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980), chs IX, XI, XII.3. 134  It follows, I think, that licensing schemes may have to be ruled out completely, in that they inappropriately trespass on autonomy, only where the intrinsic value of the activity harmless-in-itself lies in the very fact that it is engaged in, for example, spontaneously, as an expression of creative freedom, through the dictates of religious faith, privately, and in other analogous examples; but that is not a point that can be argued here. 135  By that, I mean that companies, even though they engage in these activities, cannot generate for themselves any intrinsic value through, in Raz’s words, ‘controlling, to some degree, their own destiny, fashioning it through successive decisions throughout their lives’: Joseph Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986), at 369. 136  On this, see Jeremy Horder, ‘Strict Liability, Statutory Construction, and the Spirit of Liberty’, Law Quarterly Review, 118 (2002), 458.

Bureaucratic ‘Criminal’ Law

127

personal autonomy in the way that it can when a government treats a mentally competent individual person in such a way. To give a simple example, section 387 of the Companies Act 2006 requires a company to keep proper accounting records that show its financial position reasonably accurately. By contrast, an individual is under no such duty in relation to his or her financial affairs even though the imposition of such a duty might be socially beneficial; and the explanation for that is very much the one that Ashworth and Zedner offer, in terms of the importance of personal autonomy. An analogous point can be made regarding liability for omissions. Extensive liability for omissions in business contexts has been around for a very long time. In the late 1880s, in his discussion of active duties, F. W. Maitland was moved to observe that: [I]‌f one takes up any business or employment, if one begins to build a house or thinks to open a lodging-house, or keep a trading ship or be a baker or be a chimney sweep, straightway one comes into contact with a mass of statutory rules, and if one keeps all the rules expressly laid down by statute still one is not safe, one may come across the rules, orders and regulations which some Secretary of State or central board has been empowered to make . . . .137

Omissions-based liability may threaten personal autonomy. This is because, in requiring positive action, it severely restricts the range of choices or activities someone may engage in at a given moment. By contrast, a prohibition on actively doing something merely forbids someone from making a given choice or engaging in a particular activity, from an otherwise limitless number of permissible choices or activities.138 However, if, as entities, companies cannot enjoy personal autonomy, the special wrong involved when liability for omissions undermines individuals’ personal autonomy is absent when such liability extends to companies.139 There is, for example, a very great deal of difference between, on the one hand (to give the classic example), imposing on individuals enjoying a day at the beach a duty to help someone in peril at sea, and on the other hand, placing a duty on a ship’s master to provide assistance to another ship in distress at sea. The former may be objectionable as an unwarranted threat to the personal auto­ nomy of those enjoying their day at the beach, whereas the latter has long been a relatively uncontroversial legal requirement.140 Although the ship’s master’s obligation to rescue is an obligation placed on an individual, the obligation engages the master in his professional role as a representative of the company, as not as an autonomous private citizen.

137 Maitland, The Constitutional History of England (n. 6), at 505. 138  See, further, A. P. Simester, ‘Why Omissions are Special’, Legal Theory, 1 (1995), 311. 139  Moreover, risks associated with liability for omissions can be managed or outsourced by a company in a way that an individual, in the nature of things, cannot do. 140  ‘A master of a ship at sea which is in a position to be able to provide assistance, on receiving information from any source that persons are in distress at sea, is bound to proceed with all speed to their assistance, if possible informing them or the search and rescue service that the ship is doing so . . .’: 1974 International Convention for the Safety of life at Sea, ch. 5, regulation 33(1).

128

Jeremy Horder

V. Conclusion I believe that neutrality on the question of whether there are ‘too many’ criminal offences is the best approach for criminal lawyers, whose primary task is to analyse such claims in the different contexts in which they come to prominence, historically, politically, and so on. In that regard, though, it must be recognized that the criminal law is no longer the weapon of choice that it was, for those seeking to regulate conduct in a large variety of contexts. The Regulatory Enforcement and Sanctions Act 2008 was meant to put the criminal law firmly in its place as a means of pursuing only repeated or ‘serious’ wrongdoing in regulatory contexts, with other kinds of warning and sanction playing a much more prominent role at the front line.141 It is clear that many regulatory and enforcement bodies support that kind of approach in general, but in their responses to Law Commission’s proposals, trading standards bodies also unintentionally revealed the inevitable ambivalence and difficulty involved in deciding when and whether conduct constitutes ‘serious’ wrongdoing. This is hardly surprising, given the inherent vagueness of the term ‘serious’, and the sheer range of factors that may bear on an evaluation in that respect.142 The Association of Chief Trading Standards Officers said, for example: Sometimes [it may be] appropriate to prosecute less serious crime—[it] can be effective for tackling more serious issues eg under Cancellation of Contracts Concluded in a Consumer’s Home or Place to Work Regulations 2008. Initially [it] may appear waste of courts time and disproportionate to [the] scale of offence, but doorstep crime is usually overpriced, prevalent, can lead to serious fraud, and shown to be linked to distraction burglary. Prosecution for fraud or attempted [fraud] will be considered if possible but where [there is] no evidence, [and] circumstances suggest the offender would have progressed to more serious offending had enforcement agencies not stepped in, prosecution under Regulations may be only option for Trading Standards Services.143

Against that background, the Office of Fair Trading highlights the difficulty of trying to hive off criminal offences from other measures aimed at regulatory compliance, in essence because of the vagueness of ‘seriousness’ when it comes to an attempt to confine the criminal law to ‘serious’ offences: Our concern is that all or most breaches of the law that we and partner authorities enforce can sometimes demand prompt use of criminal sanctions. Most kinds of regulatory breach are capable of incorporating conduct that involves serious malpractice and/or a serious threat of harm to consumers, for example, rogue traders overcharging vulnerable consumers for unnecessary work . . . it would be undesirable (and challenging) to reframe the law such 141  See the discussion in Law Commission Consultation Paper No. 195 (2010), Criminal Law in Regulatory Contexts (n. 34). 142 On this, see Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalisation’ (n. 5). 143 , para. 1.90 (my emphasis).

Bureaucratic ‘Criminal’ Law

129

that parties are subject to criminal prosecution only where significant harm results and/or there is evidence of a high level of culpability. Under-enforcement would reduce the incentive for traders to comply not only with the law, but with the preliminary (or soft enforcement approaches) widely used by enforcement authorities as a mean of gaining compliance without resort to the courts.144

One important reason for thinking that the criminalization of conduct—such as, say, a negligent failure to return accounts on time—may be justifiable even though it is not ‘serious’ lies in these consultation responses. There is obviously a difference between, on the one hand, the seriousness of a piece of conduct as a wrong, and on the other hand, an evaluation of the seriousness of wrongdoing in context. The definition of a crime is better suited to capture the former, because the delineation of culpable conduct can normally be made a relatively clear-cut matter in regulatory contexts.145 The prosecutorial or sentencing decision-making stage is the best forum in which to address the latter, on which many incommensurable factors, and matters of degree, will have a bearing. In that regard, in regulatory contexts, the fact that wrongdoing was engaged in intentionally or recklessly may simply be one dimension respecting which the wrongdoing was serious in context; that fact alone cannot wholly convincingly be understood as involving the crossing of a unique threshold separating criminal from civil liability. Consider this example. Suppose (1) that Company D negligently fails to return accounts, (2) that such conduct fits the definition of an offence, and (3) that Company D has on three occasions negligently failed to return accounts in spite of warnings received. A regulator or prosecutor is more likely to say—and probably rightly so—that Company D is now guilty of ‘serious’ wrongdoing that warrants criminal proceedings. We can, of course, make no sense of the idea that repetition of the offence makes wrongdoing more serious unless we have a basic offence to be repeated; but a similar point could be made about, for example, an offender’s failure beforehand to seek the help of a regulator in an effort to reduce the risk of the offence occurring, or of an offender’s failure to put in place safeguards recommended by a regulator prior to the commission of the offence.146 In neither of these examples, though, is it true to say that it is the intentional or reckless character of the wrongdoing that plays a unique role in making it serious. Indeed the presence of these fault elements may play no such role at all. Suppose, in the example just given, that between each negligent failure to return accounts on time, the company was taken over and new directors appointed who were un­aware of previous failings. In such a case, the repetition of the offence by the company can still make the offending behaviour more serious, even if it is admitted

144 , para. 1.6. 145  Which increases its value, in terms of the need to comply with fair warning requirements of what is criminally prohibited, for the purposes of art. 7 of the European Convention. Naturally, I am aware that many offences draw on (un)reasonableness, or on similarly vague terms, to define wrongful conduct. 146  In the UK, some 70 per cent of small businesses seek official advice or guidance at some point on their business practices.

130

Jeremy Horder

that the offence was never committed intentionally or recklessly. The presence of intention or recklessness can sometimes add something to the seriousness of offending in regulatory contexts, but it is not as such the key to understanding the seriousness of such offending. So far as regulatory criminal offences are concerned what is needed is, first, an appreciation of the historic role they have played: forming part of the structure of a regulatory arm to the state that has made possible the promotion of public goods across many fields of human activity, at a level and with a breadth unimaginable 300 years ago. In saying that, I fully acknowledge that regulatory offences have rarely been enforced, and when they have been enforced it has, historically, often been in a corrupt, discriminatory or class-biased way.147 My point concerns the perceived importance to legislators of integrating criminal law into regulatory schemes of virtually all kinds. Secondly, counter-reformation thinkers need to confront the reality that the substantive and procedural lines between ‘truly’ criminal offences and ‘merely’ administrative penalties have begun to become, and will continue to become, more blurred. This will happen as elements of arbitrariness, anomaly, and inefficiency involved in keeping them wholly separate undermine the goal of proportionate enforcement practice, and create unfairness from the point of view of those targeted by one form of penalization rather than the other. In that regard, the introduction by schedule 17 of the Crime and Courts Act 2013 of Deferred Prosecution Agreements that (in spite of the agreement to defer) can incorporate a financial penalty, is a good example of how such blurring is being used to escape the arbitrariness or unfairness that may come about through an insistence on such separation. The counter-reformation has arrived, predictably enough, just when the changing landscape of sanctioning is threatening—for perfectly understandable reasons—to make a nonsense of the idea that (putting aside the special case of liability to a term of imprisonment), in seeking to identify what fairness in procedure requires, one must first decide whether or not the proceedings or the punishments are ‘criminal’. Thirdly, what should be acknowledged more clearly is the legitimacy of the ‘two cultures’ of criminal law that now, very broadly speaking, exist side by side as modes of governance (albeit with much blurring and cross-application). On the one hand, there is penal liability imposed on individuals, where both the limits and the nature of liability must be shaped by values such as dignity, personal autonomy, concern and respect. On the other hand, there is penal liability imposed on companies, where these values have little or no application, and where instrumental considerations of efficiency play a legitimate role in penalization decisions, even though values such as fairness and proportionality also play an important role. Counter-reformation thinkers have sought to generalize some of the values that are significant when determining individual liability, in order to make them definitive of criminalization as a whole, even though those values are not at stake when corporate criminal liability is in issue. Their omission, then, is to fail to acknowledge 147  See e.g. Steadman, Policing the Victorian Community (n. 8).

Bureaucratic ‘Criminal’ Law

131

the plurality of values that legitimately affect criminalization decisions (and penalization decisions more generally), something reformation thinkers did not fail to do, even if reformation thinkers themselves failed to draw sufficient attention to the general importance of values such as fairness and proportionality in the taking of such decisions.

5 Criminalization in Republican Theory Philip Pettit

There are many issues that call for discussion in the theory of criminal justice but some of the most important revolve around the idea of criminalization. Why should society resort to criminalization rather than regulation of some other kind? What acts, on the face of it, call to be criminalized? And how ought society to criminalize: what penalties should it impose, what procedures should it follow? I believe that addressing these questions properly requires reliance on an overall theory of the purpose of government and law. If we are to avoid trading ad hoc intuitions and judgments in discussion of these questions, then we must put up a general, broadly plausible theory and show how it generates answers that prove persuasive on reflective consideration, even if the responses initially challenge received views. The best test of a theory, indeed, will be its capacity to support such answers, achieving what John Rawls describes as a reflective equilibrium with our considered judgments.1 In this chapter I try to sketch out some responses to the questions raised by criminalization, drawing on the republican theory of government and law.2 I think 1 John Rawls, A Theory of Justice (Oxford: Oxford University Press, 1971). For an insightful account of the contingencies in the historical development of the criminal law, particularly in Great Britain, see N. Lacey, ‘What Constitutes Criminal Law?’, in R. A. Duff, L. Farmer, S. E. Marshall, M. Renzo, and V. Tadros (eds.), The Constitution of the Criminal Law (Oxford: Oxford University Press, 2013). A philosophical reconstruction of the system of criminalization, such as that which is sketched here, necessarily involves looking for a pattern—a pattern under which the system would be justifiable—that may at best have been only loosely adumbrated in the history of the system. 2  The recent movement, as I think of it, began from the historical work of Quentin Skinner, The Foundations of Modern Political Thought (Cambridge: Cambridge University Press, 1978), on the medieval foundations of modern political thought, and from his subsequent articles in the 1980s on figures like Machiavelli who wrote within the republican tradition identified by John Pocock, The Machiavellian Moment: Florentine Political Theory and the Atlantic Republican Tradition (Princeton: Princeton University Press, 1975). An up-to-date list of English works in contemporary republican thinking should include these books: P. Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 1997); Q. Skinner, Liberty before Liberalism (Cambridge: Cambridge University Press, 1998); W. Brugger, Republican Theory in Political Thought: Virtuous or Virtual (New York: Macmillan, 1999); M. Viroli, Republicanism (New York: Hill and Wang, 2002); J. Maynor, Republicanism in the Modern World (Cambridge: Polity Press, 2003); B. Pettit and B. Western, ‘Mass Imprisonment and the Life Course: Race Class Inequality in U.S. Incarceration’, American Sociological Review, 69 (2004), 151–69; I. Honohan and J. Jennings (eds.),

Criminalization in Republican Theory

133

of this theory as a research programme rather than an ideology and part of the point of the chapter is to see how well it fares in structuring our thought about certain issues of criminal justice. The lines I take derive in good part from work done elsewhere, in particular work with John Braithwaite, but inevitably the focus on criminalization introduces some rethinking and pushes me into positions I haven’t explicitly defended before.3 The chapter is in five sections. In the first section I give an account of criminalization that ought to be acceptable, I think, from the point of view of different theories. In the second section I outline the republican theory of government and law. In Sections III–V I use this theory to generate answers to the three crucial questions as to why we ought to criminalize any acts; what acts we ought to criminalize; and how we ought to pursue the project of criminalization. The chapter ends with a brief Conclusion in which I note some limitations on the scope of the discussion.

I. Criminalization The first thing to say about criminalization is that it involves the sanction-based regulation of acts.4 Criminalization is regulatory insofar as it is designed to reduce the incidence of the sorts of acts it targets; this is something on which we can all agree, even as we disagree on why such reduction is valuable. It employs sanctions for the purpose of regulating acts rather than relying on other devices: say, preventive measures that would screen out access to those acts.5 Preventive measures will have a place in any criminal justice system—think about the role of security checks and surveillance cameras—but they are not criminalizing. Finally, criminalization is primarily designed to regulate acts rather than anything else, its particular focus being the actus reus or guilty deed. If criminalization targets dispositions, for Republicanism in Theory and Practice (London: Routledge, 2006); C. Laborde, Critical Republicanism: The Hijab Controversy and Political Philosophy (Oxford: Oxford University Press, 2008); F. Lovett, Justice as Non-domination (Oxford: Oxford University Press, 2010); J. L. Marti and P. Pettit, A Political Philosophy in Public Life: Civic Republicanism in Zapatero’s Spain (Princeton: Princeton University Press, 2010); E. MacGilvray, The Invention of Market Freedom (Cambridge: Cambridge University Press, 2011); P. Pettit, On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge: Cambridge University Press, 2012); P. Pettit, Just Freedom: A Moral Compass for a Complex World (New York: W. W. Norton and Co., 2014); as well as the collections of papers that focus on republican theory or that put republican ideas to work, as in J. Braithwaite and P. Pettit, Not Just Deserts: A Republican Theory of Criminal Justice (Oxford: Oxford University Press, 1990), on criminal justice. For an overview see F. Lovett and P. Pettit, ‘Neo-Republicanism: A Normative and Institutional Research Program’, Annual Review of Political Science, 12 (2009), 18–29. 3  The works on which I mainly draw are Braithwaite and Pettit, Not Just Deserts; P. Pettit, ‘Republican Theory and Criminal Punishment’, Utilitas, 9 (1997), 59–79; P. Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 1997; P. Pettit, On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge: Cambridge University Press, 2012). 4  N. Lacey, ‘Criminalization as Regulation: The Role of Criminal Law’, in C. Parker, C. Scott, N. Lacey, and J. Braithwaite (eds.), Regulating Law (Oxford: Oxford University Press, 2004), 144–67. 5  P. Pettit, ‘Institutional Design and Rational Choice’, in R. E. Goodin, The Theory of Institutional Design (Cambridge: Cambridge University Press, 1996).

134

Philip Pettit

example, that will only be to the extent that the dispositions are manifested in acts. And if it targets relationships that will only be to the extent that they are conspiratorial relationships formed for planning or executing criminal acts. But there are many sanction-based ways in which the legal or political system might seek to regulate the acts of members—broadly, adult, able-minded members—which would not count by anyone’s lights as instances of criminalization. The two modes of sanction-based regulation that stand in clearest contrast are what I shall describe as opportunity-cost regulation and admission-cost regulation. Opportunity-cost regulation, as the name suggests, involves regulating against a certain type of action by raising the relative but not the absolute costs of taking it; that is, by offering rewards for taking an alternative, thereby imposing the loss of an enhanced opportunity on anyone who sticks with the action. Opportunity-cost regulation would involve rewarding people for not committing crimes rather than punishing them for offending. It might render offences less probable but it would not render them any the less accessible to potential offenders: it would not make them into impossible options, of course, but neither would it even make them so unattractive in themselves as to be ineligible. Clearly, no one would think of such an attempt to regulate crime as an instance of criminalization and I put aside this candidate without further argument. Admission-cost regulation, to go to the next candidate, involves imposing absolute rather than just relative costs on criminal acts: penalizing the acts rather than rewarding the choice of an alternative. The penalties imposed are absolute costs in the sense that they reduce the utility of resorting to such an act, regardless of the alternatives. Let them be high enough and they may make the offence into an option that is effectively ineligible for most people. I describe the penalties as admission costs on the grounds that strictly they are consistent with allowing people to offend provided that they are willing to pay the costs in the event of detection. However severe the penalties imposed may be, regulating acts on the basis of the deterrent effect of such associated costs, and on that basis only, fails to distinguish it from a regime that would allow the relevant acts to be performed, on the proviso that the offender is willing to cover the price of admission. Those associated with the law-and-economics tradition, insisting that deterrence is the main point of criminalization, often cast criminalization as an instance of admission-cost regulation. Thus adherents would treat it in the manner in which, according to anecdote, some people treat parking fines—as costs to be paid for parking illegally. The expected cost of parking on this approach would be the fine attached, discounted by the probability of detection. Those who treat the cost as a price, not as a penalty, presumably think that the price is low enough to make parking illegally into a commercially attractive deal. The idea that criminalization is nothing more than admission-cost regulation is hardly any more fetching than the idea that it is equivalent to opportunity-cost regulation.6 Imagine appearing as an offender in a criminal court and explaining 6 J. Feinberg, ‘The Expressive Function of Punishment’, The Monist, 49 (1965), 397–423; T. M. Scanlon, What We Owe to Each Other (Cambridge, Mass.: Harvard University Press, 1998), 266.

Criminalization in Republican Theory

135

to the judge that you are perfectly happy to accept the standard penalty, since you understand that you were only entitled to do what you did, provided you were willing to pay the associated costs. Imagine representing yourself, in other words, as entirely free of remorse or regret. You suggest that the assault or burglary or fraud, like the parking in our earlier example, was worth it: if not worth the actual cost to be imposed, at least worth the expected cost involved. Were you to deliberate and represent yourself in such a manner, it would be clear to all involved that for whatever reason, whether of stupidity or cynicism, you just did not understand what criminalization entails. Criminalization certainly involves imposing penalties on the acts it targets, and may also have the effect of rewarding conformity to the law, but it has an important aspect that is entirely missing in such forms of regulation. Where opportunity-cost regulation may channel people away from criminal acts and admission-cost regulation constrain their criminal acts, criminalization serves the further role of condemning the performance of such acts.7 The element of condemnation or reprobation is evident in the fact that no court would be impressed with a defendant’s readiness to bear the costs of a crime, short of this disposition counting as a sign of remorse and a token of the defendant’s determination not to offend again. It is not surprising that criminalization should have this condemnatory aspect. In indicting at least paradigm offences—murder, assault, and theft, for example— the criminal laws regulate against negative patterns of behaviour such that, absent excuse or justification, almost everyone is manifestly disposed to blame perpetrators.8 These laws promote positive patterns of behaviour such that almost everyone in the society, as a matter of common awareness, must be expected to approve of conformity on the part of others and to disapprove of deviance. The regularities that criminal laws support in paradigm cases, therefore, will attract general approval, and offences general disapproval, and the desire for approval and the fear of disapproval may help independently to support conformity.9 If criminal laws are supported in that way—supported, in effect, by approbatively grounded norms—then the act of criminalization will impose penalties on the indicted acts that serve not only to deter but also to signal the presence of general disapproval or reprobation. The considerations rehearsed in this discussion ought to be available from within any theoretical viewpoint and I hope that the account of criminalization that they support is one that can be endorsed on all sides. To criminalize an act is to impose absolute rather than just relative costs as sanctions and, at least in paradigm cases, it is to impose those costs on the manifest assumption that the acts penalized attract disapproval within the community. Thus it is at once to penalize and to reprove. Given this understanding of criminalization we are in a position 7  R. A. Duff, Punishment, Communication, and Community (Oxford: Oxford University Press, 2001). 8  J. Gardner, Offences and Defences: Selected Essays in the Philosophy of Criminal Law (Oxford: Oxford University Press, 2007). 9 G. Brennan and P. Pettit, The Economy of Esteem: An Essay on Civil and Political Society (Oxford: Oxford University Press, 2004); K. A. Appiah, The Honor Code: How Moral Revolutions Happen (New York: Notrons, 2010).

136

Philip Pettit

to address the questions as to why we should criminalize, what we should criminalize, and how we should pursue criminalization. But in order to deal with those issues, as I do in Sections III–V, we need to sketch out a theoretical standpoint from which to approach them and I attempt to do this in the Section II.

II.  Republican Theory Republican theory has a long history, going back at least to the period of the Roman republic. As a theory of the role of the state in relation to citizens, it was developed by Roman thinkers like Polybius, Cicero, and Livy. It was applied to their own societies by the burghers of the medieval and Renaissance cities of northern Italy—and further developed by writers like Niccolò Machiavelli in his Discourses on Livy. And it served a crucial role in the establishment of modern regimes like the Dutch Republic, the English republic, and, perhaps most saliently, the American Republic. The English authors who shaped the tradition included figures like James Harrington, John Milton, and Algernon Sidney in the seventeenth century and various successors who held onto the main republican ideas in the eighteenth century, arguing that they could be reconciled with the new constitutional form of monarchy. These eighteenth-century figures included radicals like the authors of Cato’s Letters in the early part of the century,10 and various supporters of the American—and often the French—revolution like the chemist Joseph Priestley and his mathematician friend Richard Price.11 The republican approach is characterized by three ideas, which were all present in the figures I mention, though often differently interpreted and weighted. The ideas are that the citizens of a society—these were often identified with propertied, mainstream males—should each enjoy the status of freemen, as they were called, not being subject to the dominating power of others in their personal choices; that the state which guards citizens against private domination should also guard against their public domination, accepting the constraints—the checks and ­balances—imposed in a ‘mixed constitution’; and that in order to ensure the reliable functioning of such a mixed constitution, the citizenry should be ever vigilant of public power and be ready to contest and challenge it at the slightest suspicion or sign of abuse. All of these ideas are relevant for thinking about the criminal justice system but the most crucial idea for a theory of criminalization is that of equal freedom as non-domination. If you are to enjoy freedom as non-domination in certain choices, so the idea went, you must not be subject to the will of others in how you make those choices; you must not suffer dominatio in the word established in

10  J. Trenchard and T. Gordon, Cato’s Letters (New York: Da Capo, 1971). 11  J. Priestley, Political Writings (Cambridge: Cambridge University Press, 1993); R. Price, Political Writings (Cambridge: Cambridge University Press, 1991).

Criminalization in Republican Theory

137

Roman republican usage.12 Thus it is not enough for freedom that you manage to enjoy non-interference. Freedom requires that you not be exposed to a power of interference on the part of any others, even others who are disinclined to use that power against you. The mere fact that I can interfere at little cost in your choices— the mere fact that I can track those choices and intervene when I like—means that you depend for your ability to choose as you wish on my will remaining a goodwill. You are not sui juris—not your own person—in the expression from Roman law. You are unfree, as Richard Price explained,13 because your access to the options will depend on an ‘indulgence’ or an ‘accidental mildness’ on my part. To quote from a seventeenth-century republican, Algernon Sidney, freedom in this tradition requires ‘independency upon the will of another’—an ‘exemption from dominion’ in relations with others.14 In an equivalent slogan from Cato’s Letters, ‘Liberty is, to live upon one’s own terms; slavery is, to live at the mere mercy of another.’15 In arguing that the state should be concerned in the first place with the equal freedom of its citizens, republicans held that citizens should each be assured of enjoying non-domination in a sphere of choice that came to be described as that of the fundamental or basic liberties.16 This might be identified, in contemporary terms, with the sphere of choice required for being able to function in the local society.17 The basic liberties are those personally significant choices that everyone can exercise at the same time as everyone else and do so with satisfaction: their exercise by others does not take away from the value of your exercising them in your own case. While each society will have to identify the exact liberties to be available for its citizens, setting down conventions to define rights of movement or association or ownership, for example, those liberties will certainly comprise choices in traditional categories like the following. The freedom to think what you like. The freedom to express what you think. The freedom to practise the religion of your choice. The freedom to associate with those willing to associate with you. The freedom to own certain goods and to trade in their exchange. The freedom to change occupation and employment. The freedom to travel within the society and settle where you will. The freedom to use your leisure time as you wish.

12  F. Lovett, Justice as Non-domination (Oxford: Oxford University Press, 2010), appendix. 13 Price, Political Writings, 26. 14  A. Sidney, Discourses Concerning Government (Indianapolis: Liberty Classics, 1990), 27, 340. 15  Trenchard and Gordon, Cato’s Letters, ii. 249–50. 16  J. Lilburne, The Legal Fundamental Liberties of the People of England, asserted, revived, and vindicated (London, 1646). 17 A. Sen, Commodities and Capabilities (Amsterdam: North-Holland, 1985); M. Nussbaum, Frontiers of Justice (Cambridge, Mass.: Harvard University Press, 2006).

138

Philip Pettit

But how is the state to provide protection against domination in those choices for each of its citizens: that is, in a contemporary version of the approach, for each adult, able-minded, more or less permanent resident? The natural lead to this question is given by the image of the free citizen—the liber or freeman, in received usage—that is at the core of the tradition. In this image, free persons can walk tall, and look others in the eye. They do not depend on anyone’s grace or favour for being able to choose their mode of life. And they relate to one another in a shared, mutually reinforcing consciousness of enjoying this independence. Thus, in the established terms of republican denigration, they do not have to bow or scrape, toady or kowtow, fawn or flatter; they do not have to placate any others with beguiling smiles or mincing steps. In short, they do not have to live on their wits, whether out of fear or deference. They are their own men and women and however deeply they bind themselves to one another, as in love or friendship or trust, they do so freely, reaching out to one another from positions of relatively equal strength. The status required for being a free person under the received, republican image imposes two sets of requirements, objective and subjective. On the one side it requires that you should be objectively secure against the intrusions of others, including the intrusions of the very government that protects you against others, in the enjoyment of the basic liberties. Being secure against intrusion in such activities means, in the ideal, that you are safeguarded against any agent or agency that might take against you, regardless of the improbability of such hostility. You are proof against the vagaries of alien wills, however unlikely such vagaries might be. But the status of the free person, as encoded in the traditional image, has an intersubjective as well as an objective side. Not only does it require you to be securely safeguarded against others; it also requires this safeguarding to be registered as a matter of common awareness. Everyone, yourself included, must be aware that you are safeguarded; everyone must be aware that this is a matter of general awareness; and so on. Your safeguarded status must be manifest and salient to all. Only if this is so, can you hope to achieve the relationship with others that the image encodes. Only then can you walk tall among your fellows, conscious of sharing in the general recognition that no one can push you around with an expectation of impunity. The idea that the objective status of republican citizenship might have an intersubjective register played an important role in the traditional model of the free citizen. The reason is that as Thomas Hobbes noted in another context, the ‘reputation of power is power’.18 The fact that potential offenders recognize that you have entrenched rights means that the protection you enjoy is thereby reinforced; their recognition of the protection provided is bound to inhibit and help to damp any temptation to offend against you. How are you going to achieve the objective and intersubjective status associated with being a free person in this sense? The republican tradition has always insisted that such a status is only going to be available under a public rule of law 18  T. Hobbes, Leviathan, ed. E. Curley (Indianapolis: Hackett, 1994), 10.5.

Criminalization in Republican Theory

139

in which each is treated as an equal, being offered the same resources of choice, and protection against intrusion, as others. But the general idea has also been that an effectively supportive law will inevitably be reinforced by corresponding norms or morals. Machiavelli remarks in Discourses 1.18 on the importance of such norm-based support for the law when he says that ‘just as good morals, if they are to be maintained, have need of the laws, so the laws, if they are to be observed, have need of good morals’.19 The laws and corresponding norms that are required will come in various forms. The law of property and contract will serve to define basic liberties, making clear to all citizens the limits on these fronts within which they are protected. The law of torts will serve a similar function, allowing the courts to determine the limits beyond which people are not entitled to impose risks on others. Family law and employment law will play a vital role in defining and protecting important liberties within the context of asymmetrical and potentially troublesome relationships. And criminal law will serve to establish the protections that each can expect to enjoy against particularly egregious forms of dominating interference that any others may seek to practise against them. How much protection should the law provide against domination? Building on the intersubjective aspect of the free-citizen ideal, republican theory offers a natural workable criterion. This criterion, which I describe as the eyeball test, requires that people should be so resourced and protected in the basic liberties that by local standards they can look others in the eye without reason for fear or deference. They do not depend for their security on the indulgence and condescension of others. They can walk tall and assume the status of an equal with the most powerful in the land. Or at least they can do so, provided that by local standards—even the most charitable, local standards—they do not count as excessively timid or paranoid. A system of law might serve people well in protecting them against private domination by others, of course, and yet fail to protect them in the public sphere against the very government that shapes and sustains the system. That is why republican theory emphasizes the need for a mixed constitution and a contestatory citizenry. The idea is that such institutional measures—such measures, ideally, of equally shared popular control—can ensure that while government interferes with its people in imposing laws, it does so in a way that is subject to their control and therefore not dominating; it does not expose them, as even a benevolent dictatorship would do, to the arbitrary or uncontrolled will of another.20 I shall assume in the remainder of this chapter that the point of government, as republican theory describes it, is to secure people in the undominated enjoyment of their basic liberties, where this requires protection against both private and public domination. It is not enough on this account that government organizes things so that overall there is not much interference by some citizens in the basic liberties of others. That might be so because the weak are good at ingratiating themselves with the strong, keeping them sweet and well disposed, or because the weak are 19  N. Machiavelli, The Complete Work and Others (Durham, NC: Duke University Press, 1965). 20 Pettit, On the People’s Terms.

140

Philip Pettit

lucky enough to live in the presence of a powerful but benevolent elite: at the limit, a benevolent despot. What matters in republican theory is that no one has to depend on the goodwill of others for being able to exercise their basic liberties and, in particular, that they do not have to depend on the goodwill of others for avoiding the intrusive incursions that get to count in most countries as crimes. With this theory in place I turn in the remaining sections to the three questions I raised earlier.

III.  Why Criminalize? Assume for the moment that there is no issue about what acts to criminalize: I turn to that question in Section IV. With this assumption in place, the question that naturally arises is, why criminalize those acts? Why not rely on one of the other modes of regulation rather than having recourse to criminalization? One alternative to criminalization would be a form of preventive regulation that screens out offensive acts, whether by means of the security device that makes it impossible for you to carry a gun on an airplane or by internment in the case where you are deemed to be a danger to others. Some preventive regulation clearly has a role in any criminal justice system—the security device is a good example— but any philosophy of government that puts a premium on freedom, especially freedom as non-domination, is bound to shrink from any widespread recourse to screening out offences as distinct from sanctioning them.21 Thus there is good reason to opt for criminalization rather than preventive internment in dealing with agents who are capable in the normal way of responding to sanctions and of benefiting from freedom; if there is room for internment, that can only be with those who are designated, under testing procedures, as dangerous to others and yet unresponsive to sanctions. Does the cause of freedom argue for criminalizing relevant acts as distinct from regulating them by means of other sanctions, as in opportunity-cost or admission-cost regulation? Not, it should be noticed, on the standard, non-republican theory of freedom. On this theory, freedom requires the absence of interference or likely interference, however obtained, and not the public security against interference in which non-domination consists. And on such a theory of freedom as non-interference, it does not necessarily follow that any threats to freedom argue for criminalization. It is true that criminalizing a form of interference may make people less likely to interfere with others and thereby increase freedom as non-interference. But it is equally true that opportunity-cost regulation and admission-cost regulation may also make people less likely to interfere with one another. And so it does not follow from the fact that a certain sort of act infringes 21  Some criminal justice measures taken against ‘inchoate crimes’—that is, acts involved in preparing to commit crime—are better cast as preventive but still acceptable measures, according to some recent theorists. See e.g. K. K. Ferzan, ‘Inchoate Crimes at the Prevention/Punishment Divide’, San Diego Law Review, 48 (2011), 1273–97.

Criminalization in Republican Theory

141

such freedom that it ought to be criminalized. Whether it should be regulated via criminalization or in some other way ought to turn on the contingent matter of which mode of regulation promises in the circumstances to be most effective. Things look very different, however, from within a perspective under which freedom requires, not just the lack or unlikelihood of interference, but the public security against interference that can enable people to walk tall and assume the status of equals with the best in their society. People will not enjoy any form of security if they have to depend on the rewards offered to potential offenders being enough to persuade them. And they will not enjoy a public form of security if there is nothing held against offenders who are willing to pay the admission cost for offending, however high that may be, just for the sake of enjoying the interference they practise against others. Opportunity-cost regulation does not take away the option of interference from would-be offenders, concentrating as it does on rewarding alternatives. Thus, however successful it may be in steering potential offenders away from interference, it does not deny them the capacity to interfere. And so it will allow those with superior strength or wealth, connections or cunning, to retain a power of interfering at will in the affairs of others. Those others will depend on the goodwill of such superiors for being able to act as they wish within the domain of the basic liberties and will be saliently denied the freedom as non-domination of republican citizens. Admission-cost regulation does replace the option of interference that would-be offenders may access by an option of interference-with-a-threat-of-penalty and it may thereby make interference into an option that is ineligible in the view of most. It may provide in that sense for a measure of objective security. But will this penalty be enough to enable potential victims to pass the eyeball test in relating to potential offenders? Will it be enough to provide them with the public, intersubjective form of security—and with the associated enhancement of objective security—that that test requires?22 No, for two reasons. Admission-cost regulation deprives potential victims of any public security against those who are manifestly, even openly willing to run the risk of the penalty, thereby paying the expected cost, for the sake of the gains of interference. And besides, it gives lucky offenders who manage to escape detection or conviction no publicly endorsed reason for remorse about what they did; for all that admission-cost regulation involves, they will be entitled to think about the offence as a gamble that turned out, from their point of view, for the best. People will be able to enjoy the public status of the free citizen only insofar as those who would offend against them in relevant ways face not only the absolute costs imposed on offences but also the reprobation of the community. Under such criminalization, potential victims will have a publicly affirmed security against those who are manifestly willing to run the expected cost of offending; they can 22  Apart from the considerations that I go on to mention, it is worth noting also that there is considerable evidence that what keeps most people from crime is not the fear of penal sanctions but rather the sense that criminal behavior is objectionable and unthinkable by received community standards. See T. R. Tyler, Why People Obey the Law (New Haven: Yale University Press, 1990).

142

Philip Pettit

invoke public condemnation against those who are so disposed to offend, claiming the protection of the community against them. And under such criminalization, potential offenders have to recognize that even if they succeed in avoiding de­tection or conviction for an offence, they are still subject to reprobation by community norms; let them accept those norms and they cannot just think of themselves as lucky gamblers, acknowledging no grounds for remorse. These observations mean that on the republican theory of freedom, criminalization is the right response to various categories of interference, and the right response for more than contingent reasons. Criminalization is not suitable just because it happens as a contingency of circumstance to promise better results in republican terms than any alternative form of regulation. It is suitable because, unlike opportunity-cost or admission-cost regulation, it is capable of providing for the enjoyment of the distinctively public security in which republican freedom consists. The public security that public condemnation can establish is not attractive just for the fact that it increases suitably the perceived safeguards that people enjoy against crime. It is attractive, more specifically, because of the way in which it increases those safeguards: viz., by making others aware of the grounds on which the law protects you or me. In condemning offences against us, the law makes clear to others that in the common perceptions that it articulates, we are sacred sites of agency: we are centres of rights that others invade, not just at their peril, but in defiance of community standards.

IV.  What to Criminalize? Given our understanding of what criminalization involves, and this republican account of why it makes sense to criminalize rather than regulate in other ways, we turn now to the question of what acts deserve to be criminalized. What acts are such that it makes sense for the state to impose absolute costs on their performance, communicating at the same time a high level of reprobation for resort to such acts? As the executor of criminal justice the state assumes its most dangerous role. It has discretion over what offences are criminal, how offenders should be identified, whether they should be prosecuted, and how they should be tried and, if convicted, punished. Such power is dangerous in the extreme since, if abused, it holds out the prospect of a deeply intimidating form of public domination. That intimidation may derive, for example, from the propensity of legislators to criminalize and penalize with a view to political advantage, of police or prosecutors to be selective or discriminatory in the presumptive offenders they target, and of prison guards to take out their personal frustrations or prejudices in their treatment of inmates. As we have seen, republican theory would argue for a mixed constitution and a contestatory citizenry—in effect, a multidimensional, popular democracy—to guard against abuses in any domain, including the criminal justice system. But the more restricted the criminal justice system is, the easier it will be to impose

Criminalization in Republican Theory

143

such democratic controls on its operation. Hence the inquiry into what acts the system should criminalize—and as we shall see, into how criminalization should be pursued—must start from a presumption in favour of parsimony. The onus of proof in any arguments about what to criminalize should fall on those who defend criminalization rather than on those who oppose. The presumption of parsimony is not just supported on the top-down side by the danger of excessive resort to criminalization. It is also supported from a bottom-up perspective by the fact that many offensive acts are likely to be containable, and perhaps even more effectively containable, without criminalization. It is true that all acts of interference in the basic liberties of others are offensive within a republican perspective, where we may take interference to involve removing one of the options, replacing the option with a penalized alternative, or misrepresenting the option in a deceptive or manipulative spirit. By acting in any of these ways, after all, I assume control of your choice or usurp your control in partial measure by imposing certain terms or limitations upon you. But it would make no sense to invoke the criminal justice system to oppose acts of interference that do little damage, such as the ordinary act of deception or promise breaking. The criminal justice cure might carry more dangers for freedom than the deception or promise breaking it was meant to restrict. And informal norms against deception and promise breaking, which rely on the pressure of communal condemnation on its own, might be more effective in restricting such offences. The presumption of parsimony argues for limiting criminalization to acts of interference in people’s basic liberties which, unlike these offences, are incapable of being effectively restricted in the absence of criminal penalty and condemnation. The most plausible examples of acts that ought to be criminalized are the offences that have traditionally counted among ordinary people as criminal and condemnable and that are often cast as belonging to the core of criminal law.23 They include fatal offences, offences against the person, offences against property, and offences against public order, strictly understood. In the traditional phrase, they mostly count as mala in se rather than mala prohibita. These are acts that are bad in themselves, by regular criteria, not just acts that are bad by virtue of breaking rules that might well have taken a different form. They would include reckless driving, for example, but not breaking a speed limit of 30 miles per hour when the limit might well have been set at 45 miles per hour.24

23  Lacey, ‘Criminalization as Regulation’; G. Fletcher, Rethinking Criminal Law (New York: Little Brown, 1978). 24  This is a tricky distinction to draw and I ignore an important qualification. It might be a malum prohibitum to drive at 35 mph but that very same act could count as a malum in se insofar as it involves breaching the local speed limits, where it is assumed that there have to be some such limits. In the same way an act of theft might count as a malum prohibitum insofar as it is a breach of a contingent, variable set of property conventions and as a malum in se insofar as it is a breach of local property conventions where again, and this time even more plausibly, it is assumed that there have to be some such conventions. I also ignore another issue to which Chris Bennett has drawn my attention: that a malum in se might be thought to be an act that harms another or an act that for whatever reason—perhaps because it hurts another—is morally wrong.

144

Philip Pettit

Apart from such acts of interference with the basic liberties, the candidates for criminalization are generally taken to include secondary acts that facilitate or probabilify criminalized acts of interference, or that jeopardize the system for protecting against such acts. Acts of preparing for crime, say by collecting required resources or organizing for the perpetration of the crime might count in this category. So might acts of inciting others to crime, whether in one-to-one exchange or in publicly offensive hate-speech. And so of course might acts of hindering arrest or of undermining court proceedings by perjury. But in the case of these acts, as with crimes in the primary category, the presumption of parsimony argues strongly for not allowing secondary crimes of these kinds to proliferate.25 There is a well-documented tendency in many legal systems to extend crimin­ alization to a wider and wider range of activities, allowing both the category of primary crimes and the category of secondary crimes to expand.26 This is unfortunate for a number of reasons. First, using the criminal justice system across a wider and wider bandwidth of acts can weaken its impact in more important ranges of the spectrum, undermining the condemnatory role of criminalization. Second, it runs the risk of doing more harm than good, putting in the hands of criminal justice authorities a power that may impact more deeply on prospects of non-domination than any of the offences against which it would protect. And finally, it can make conviction difficult in areas where it ought not to be difficult, extending the special protections appropriate only for those charged with serious criminal offences—see Section V—to those areas. I do not have the expertise to speculate on how the range of criminalization might be effectively reduced. One proposal worth considering is that other systems should follow the example of those countries that distinguish a range of regulatory statutes, in particular statutes directed against offensive behaviour, from criminal laws. In Germany, for example, a range of regulatory offences—Ordnungswidrigkeiten, as they are known—are distinguished from criminal offences and treated in a distinct manner. These offences include violations of restrictions on public gatherings, for example; creating a public nuisance by being drunk and disorderly; and breaking the speed limit and other traffic regulations.

V.  How to Criminalize? The third question to be addressed bears on the form that criminalization should assume and breaks down into two main issues. First, assuming that the criminal justice authorities are not to be given total discretion in the matter, what sorts of penalties should be attached in law to different crimes? And second, assuming that the criminal justice authorities are to enjoy some discretion, what restrictions should be placed on them in seeking to identify and convict offenders and in 25 Parsimony might raise similar worries about some of the preventive measures envisaged in Ferzan, ‘Inchoate Crimes at the Prevention/Punishment Divide’. 26  D. Husak, Overcriminalization (Oxford: Oxford University Press, 2008).

Criminalization in Republican Theory

145

applying penalties to those convicted? I address each of those issues briefly in this final, somewhat longer section. The answer to the first question is that acts should be subject to reprobative penalties that do best by a mix of goals: 1. the penalties reduce the aggregate incidence of such acts, amounting to more than token rebukes; 2. they are not so great or gruesome that the authorities that can impose them— and that might impose them in error or malice—are liable to dominate the public at large; 3. they do not make it likely that charged or convicted offenders will be exposed to humiliation and domination by agents of the system or members of the public; 4. they treat offenders as capable of regretting the ill they did—and not just as susceptible to intimidation—and they allow for reintegration of offenders into the community; 5. they vary in harshness with acts of varying seriousness so that potential offenders are likely to see the difference in the acts and register the difference in the penalties. These constraints argue for having deterrent penalties but penalties that avoid terrorizing citizens in general or offenders in particular.27 They argue in the abstract for establishing a hierarchy of penalties, from least to most harsh, and then matching those penalties with different types of offences, beginning with the least serious and advancing to the most serious. More realistically and plausibly, they argue for a hierarchy that allows for some judicial discretion—more on this in a moment— matching penalty ranges to ranges in the seriousness of an act. An act’s seriousness will be determined by the grievousness of the harm imposed, the culpability of the agent performing the act, and, I would say, the extent to which the agent has been convicted of similar offences in the past. Such a pairing of penalties with offences makes sense in virtue of the fifth consideration given, arguing against the danger of an offender thinking, in the old proverb: ‘might as well be hanged for a sheep as a lamb’.28 But it also makes sense on the separate ground, important under a republican perspective, that the sentences imposed on different offenders should not vary to the point where offenders have good reason to judge that they are not being treated as equals.29 What sorts of penalties are likely to meet these constraints? Capital and corporal punishment would be outlawed by a number of the constraints, as would branding or any form of long-term stigmatization.30 No surprise there. The constraints would 27  E. Kelly, ‘Criminal Justice without Retribution’, Journal of Philosophy, 106 (2009), 440–62. 28  C. de S. Montesquieu, The Spirit of the Laws (Cambridge: Cambridge University Press, 1989). 29  A. Ashworth and A. Von Hirsch, ‘Not Not Just Deserts: A Response to Braithwaite and Pettit’, Oxford Journal of Legal Studies, 12 (1992), 83–98. 30  J. Braithwaite, Crime, Shame and Reintegration (Cambridge: Cambridge University Press, 1989).

146

Philip Pettit

allow escalating penalties for repeat, and presumably dangerous, offenders, under the account of seriousness just given. But in view of the case for parsimony, they would favour penalties at the lighter end of the spectrum such as community service and fines, if only because prison offers so many opportunities for the domination of inmates by guards or by other prisoners. In particular, they would favour penalties that communicate the community’s recognition of the status of the victim and its disapproval of the offender’s action. This consideration will come up again when we address the mode in which penalties ought to be imposed and its relevance there argues for associating different offences with penalty ranges rather than precise penalties; such a loose linking with penalties allows a range of discretion to the sentencing judge. The issue of what penalties work well is a matter for continued monitoring in any society and common sense suggests that the legislature should set up a semi-autonomous sentencing commission to conduct such monitoring and to make recommendations for change in an incremental manner that allows for confirmation or retraction on the basis of careful study. This is particularly important in view of the fact that politicians are bound to be tempted to play politics with penalties, using the outrage caused by particular crimes to build an emotive, eye-catching campaign for ever stricter sanctions.31 I said that the principles enumerated argue in the abstract for a proportional pairing of penalties with offences. But in legislating for penalties certain variations from abstract proportionality are inevitable—ideally, variations that are intelligible to all and need not count as discriminatory. Consider for example the sort of criminal act that is manifestly very hard to detect. Should the penalty for such an act be increased with a view to keeping the expected cost of the penalty—this is determined by the harshness of the penalty times the probability of detection and conviction—at a level that can still plausibly inhibit would-be offenders? I would say that other things being equal, it ought to be increased towards that level. I prefer to regard the example as introducing an exception to the abstract rule of proportionality but it might be thought that the evidence for premeditation and culpability is higher—the probability of the offender having blundered into the crime is lower—the higher the presumptive confidence of the offender that the offence would escape detection. But there is another consideration that can argue, in the opposite direction, for reducing the penalty associated with an offence to a lower level than might otherwise seem appropriate. This is that reducing the penalty in that way increases the chances of detection and conviction—and ultimately the chances of reducing the aggregate number of offences. It might be, for example, that corporations would be more inclined to expose corruption in their ranks if the penalties incurred by offenders were relatively light, perhaps even being suspended in the event that internal disciplinary action is taken against the culprit.32 31  P. Pettit, ‘Is Criminal Justice Politically Feasible?’, Buffalo Criminal Law Review, Special Issue ed. Pablo de Greiff, 5 (2) (2002), 427–50. 32 B. Fisse and J. Braithwaite, Corporations, Crime and Accountability (Melbourne: Cambridge University Press, 1993).

Criminalization in Republican Theory

147

These considerations can sometimes come into conflict, raising pressing issues of institutional design and effective regulation in determining the penalties that ought to be legally attached to different offences. Consider insider trading, for example. It is an offence that can be very difficult to detect, thereby attracting an especially heavy penalty. But it is also an offence that might be much more readily reported and detected, if the penalty were on the lighter rather than the heavier end. I mention this case to underline the fact that philosophy and theory cannot rule on such concrete issues. It can only provide general, flexible guidelines, indicating the cases where exceptions may have to be made. Such flexibility will be perfectly acceptable on the broadly goal-centred way of thinking about criminalization and criminal justice that I am advocating here. The idea is that the system as a whole, including the pattern of criminalization, should be designed so that if the participants stick to their roles—if they only exercise discretion, for example, in the measure and the manner allowed—then the result will be the optimal promotion of the republican ideal: that people should be enabled equally to enjoy security in the exercise of their basic liberties.33 It should be no surprise that the system that does this will often have to be adjusted away from abstract, inflexible templates that permit no variation in the proportional pairing of penalties with offences.34 We have been discussing the issue of what sorts of penalties ought to be associated with different offences. The second issue that I mentioned bears on what restrictions should be placed on the exercise of discretion by criminal justice authorities on two fronts: in seeking to identify and convict offenders and in imposing penalties on those convicted. In regard to restrictions on the first front, republican theory broadly supports official, traditional practice, albeit a practice that often fails to be honoured under contemporary pressures of court business; more on this in the Conclusion. In line with such practice, it would draw on the presumption of parsimony to argue for a range of constraints on state procedure. Thus the police should be strictly limited in the surveillance they may practise in searching out offenders; the prosecution should carry the burden of proof in seeking to establish the guilt of offenders; the prosecution should be required, at least in general, to establish intent or recklessness or some such guilty attitude—mens rea—in the offender; and the criterion of proof should guard against the conviction of the innocent by requiring guilt beyond reasonable doubt. But what of the constraints that ought to guide criminal justice authorities in imposing penalties on the convicted: that is, their sentencing and punishment? The goal-centred approach adopted in republican theory argues against any mechanical application of abstractly assigned penalties to convicted offenders, recognizing that there are salient and relevant differences between individual offences and that these 33  J. Rawls, ‘Two Concepts of Rules’, Philosophical Review, 64 (1955), 3–32. 34  There a contrast in this respect with a retributivist theory, otherwise quite close to the republican, in which something like the pairing mentioned is given a canonical status that allows of few if any exceptions.

148

Philip Pettit

cannot be ignored in criminal justice practice. This consideration offers support for the prescription of penalty ranges rather than exact penalties for different offences and, as we shall see, for other forms of judicial discretion. There is a danger associated with giving too much discretion to a court, of course, since the capricious judge may wantonly abuse any discretion. But that danger can be countered, at least in principle, by the possibility of review and by the bad publicity that a capricious judge is likely to attract. Whatever discretion judges enjoy, however, they should not have a power of exceeding the upper limit on penalties, or ranges of penalties, which is generally allowed. This is because the possibility of exemplary punishment would give a judge the power of interfering at will in the fortunes of the offender in the dock. It would permit the judge to act on a personal antipathy towards that individual, or that individual’s religion or ethnicity, in choosing to breach the upper bound on normal penalties. There are possible protections against the abuse of such a power, since its exercise might have to be justified and would be subject to appeal by the defence. But these might well be thought insufficient to counter the domination that a ‘hanging judge’ can wield or seem to wield over a defendant. Whereas this consideration argues that judges ought not to be able to breach the upper limit on recognized penalties, no equally weighty consideration argues against allowing judges to exercise discretion and impose penalties that fall below the corresponding bottom limit. The individual offender in the dock is not exposed to the will of the judge in the same way when that judge has a power of mercy as distinct from a power of imposing an exemplary punishment. Or at least that will be so when the exercise of the power requires justification, is subject to appeal by the prosecution, and, in the nature of the case, is only exercised occasionally. Such measures can guard against a scenario in which mercy becomes the norm, and the opportunity not to display mercy becomes an opportunity for exemplary harshness. To say that there is no powerful argument against mercy, as there is against exemplary punishment, is not yet to make a case for allowing mercy. But there is a case that can be made in republican theory on behalf of such a possibility. The best way of making the case may be to think more generally about the function that the sentencing and punishment of convicted offenders ought to serve. It turns out that such general reflection offers a ground on which, among other measures, to defend the possibility of mercy. Taking freedom as non-domination to be a central value in political life, there are three distinctive ills that a crime in the primary category, as we conceive of that category, imposes. First, it represents an assumption of personal power on the part of the offender and ignores or rejects the victim’s status as a free, undominated citizen. Second, it deprives the victim of resources, whether of life or latitude or property, that are needed for the exercise of the basic liberties and the enjoyment of freedom as non-domination. And third, it reduces the confidence that others in the community are entitled to place in legal protections, thereby impacting on their enjoyment of such freedom.

Criminalization in Republican Theory

149

The attempt to rectify such ills ought to guide the sentencing and punishment of offenders, as indeed it ought to guide the determination of legally permitted ­penalties. Rectification of the denial of the victim’s status argues for a mode of sentencing in which the offender is made aware of the grounds on which the community disapproves of the offence and is given the opportunity to feel and register remorse. Rectification of the second ill—the deprivation of life, latitude, or property—argues for demanding restitution, where possible, and where not, a degree of compensation or reparation; this might support penalties that allow for such rectification, as in fines that benefit the victim or the victim’s family or community service that plays a similar role. And rectification of the third ill argues for imposing a sufficiently deterrent penalty to provide reassurance for others in the community that they are not worse off than before the offence occurred. With these points in place, the case for allowing mercy is straightforward. It may often be that while the offence of which the defendant has been convicted generated one of these ills—say, the denial of the victim’s status—it did not cause any others. Perhaps a burglar broke into your house but caused himself such an injury—perhaps a debilitating head injury—that he did not take any property and made himself incapable of reoffending. In such a case there is every reason why a judge might exercise mercy, and do so on grounds that allow other more harshly treated burglars to realize that they are not being discriminated against. This sort of argument can generalize to offer support for mercy in any of a variety of cases. It can also serve to explain why the unsuccessful attempt to commit a crime should not necessarily attract the same penalty as the successful crime. In many cases the difference between the successful and the failed attempt will be registered in legally assigned penalties but even if it were not, then there is reason why it should be registered in the exercise of a certain mercy by the court. The failed attempt will not inflict the second of the three ills associated with a primary crime and will not call on at least that account for redress. Still, short of the lack of success indicating rank incompetence on the part of the offender, the failed attempt to perpetrate an offence may attract a penalty quite close to what a successful counterpart would have earned.

VI. Conclusion This brief account should indicate the grounds on which I think that a republican political theory can offer a satisfactory account of criminalization, supporting judgments of criminal justice that are sometimes novel but always worthy of ­consideration—and, I would suggest, endorsement. Given a natural understan­ ding of criminalization, the theory offers plausible answers to the three questions as to why we ought to criminalize, what we ought to criminalize, and how we ought to criminalize. It does well by the reflective-equilibrium criterion for a satis­ factory theory. In conclusion, however, I note two important limitations on what I have attempted here. The first is that I have only considered crimes in which the offenders

150

Philip Pettit

are adult, able-minded, individual human beings. I have avoided the immense qualifications required when we consider the offences of children and of those who suffer from various cognitive or affective disorders. And I have ignored, for reasons of its complexity, the issue of how far corporate organizations can be treated as criminal offenders and what sort of treatment is called for in their case. For the record I do think that corporate entities are fit to be held criminally responsible but there are too many questions raised by this perspective for me to consider them here.35 The second limitation on what I have attempted appears in a degree of idealization implicit in the approach. It is common in the theory of distributive justice, following John Rawls,36 to argue that it presupposes more or less idealized ‘circumstances of justice’: in particular, conditions of moderate but not extreme scarcity and moderate but not extreme egoism. What I have sketched in this chapter is a theory of criminal justice—in particular, a theory of criminalization—that presupposes correspondingly idealized circumstances: specifically, conditions in which sufficiently few actions are criminalized, sufficiently few crimes are committed, and sufficiently few offenders are charged and convicted, to allow for a proper hearing to be given to each particular case. Unfortunately the circumstances of criminal justice, so described, are not often satisfied in contemporary societies and may rarely have been satisfied in the past. This argues for modifications to the arrangements considered here, if anything even half-satisfactory is to be approximated. In real-world situations, it may often be better, not to try to replicate imperfectly the arrangements described, but rather to introduce variations that allow for the difficulty or impossibility of replication.37 Thus it might make very good sense, as many have argued, to try to supplement the system with programmes of restorative justice in which certain offenders can plead guilty to their crimes and avoid the normal judicial process. Under such a programme the admitted offender enters a conference with the victim, in the presence of a number of associates chosen by each, to determine what the offender should do to make up for the offence.38 While I have not been able to explore the possibility here, restorative justice programmes may well promise in real-world circumstances—and indeed in more ideal conditions too—to serve criminal justice better than anything I have been able to consider.39

35 P. Pettit, ‘Responsibility Incorporated’, Ethics, 117 (2007), 171–201; P. Pettit, ‘Corporate Responsibility Revisited’, Rechtsfilosofie & Rechtstheorie, 38 (Special Issue: Philip Pettit and the Incorporation of Responsibility) (2009), 159–76; C. List and P. Pettit, Group Agency: The Possibility, Design and Status of Corporate Agents (Oxford: Oxford University Press, 2011). 36 Rawls, A Theory of Justice. 37  A. Vermeule, The System of the Constitution (New York: Oxford University Press, 2011). 38  H. Strang and J. Braithwaite (eds.), Restorative Justive: Philosophy to Practice (Aldershot: Dartmouth-Ashgate, 2000); J. Braithwaite, ‘Setting Standards for Restorative Justice’, British Journal of Criminology, 42 (2002), 563–77; G. Johnstone (ed.), A Restorative Justice Reader: Texts, Sources, and Content (Abingdon: Willan Publishing, 2003). 39  I am grateful to a number of people who discussed the ideas in this presentation at the University of Stirling Conference in Sept 2011 and at a workshop in Queen’s University, Belfast in June 2012. I would like in particular to mention Andrew Ashworth, Chris Bennett, Anthony Duff, Kim Ferzan, Nicola Lacey, Sandra Marshall, Victoria McGeer, Michael Moore, Konstantinos Papageorgiou, Massimo Renzo, Victor Tadros, and Lucia Zedner. I am especially grateful for written comments received from Ben Ewing, John Braithwaite and Massimo Renzo.

6 Contractarian Criminal Law Theory and Mala Prohibita Offences Susan Dimock

I. Introduction One seemingly perennial debate within criminal law theory concerns what, if anything, justifies the use of mala prohibita offences. While some of that debate may be attributed to disputes over the meaning of mala prohibita and how they should be contrasted with mala in se, and indeed whether there is a meaningful distinction to be drawn at all, most criminal law theorists continue to employ the distinction and share at least a core understanding of what they mean by the respective labels.1 I employ the most common and minimal description of the distinction in what follows: an offence is an instance of malum prohibitum when the conduct it proscribes is not wrongful prior to or independent of law, while a malum in se offence is an act that is wrongful prior to and independent of its legal prohibition. This way of drawing the distinction is compatible with contractarianism, whereas some other ways of drawing it are not. Most notably, some commentators think of mala in se offences as involving ‘acts that are wrong in themselves’ or as consisting of conduct that is intrinsically wrong. Such usage lends itself naturally to ‘a kind of moral realism’, as Alice Ristroph notes, and such realism is incompatible with the constructivist rational choice contractarianism I utilize in this paper.2 Malum in se conduct cannot be understood as wrong prior to the social contract, prior to or 1  For scepticism about the distinction, see Glanville Williams, Textbook of Criminal Law (2nd edn.; London: Stevens & Sons, 1983), 936; Wayne R. LaFave, Criminal Law (3rd edn.; St Paul: West Publ. Co., 2000), 35–6. No modern critic of the distinction goes quite so far as Jeremy Bentham once did when he remarked: ‘the acute distinction, between mala in se, and mala prohibita: which being so shrewd, and sounding so pretty, and being in Latin, has no sort of occasion to have any meaning to it: accordingly it has none.’ A Comment on the Commentaries (1776), in J. H. Burns and H. L. A. Hart (eds.), Collected Works of Jeremy Bentham, vol. iii (London: Athlone Press, 1977), 63. 2  Alice Ristroph, ‘Criminal Law in the Shadow of Violence’, Alabama Law Review, 62 (2011), 571 at 582, though this definition continues ‘acts that are wrong in themselves, independently of any statute or positive law’ (emphasis added). Thus it is compatible with contractarianism, because an act can be wrong independently of positive law if it violates a term of the social contract; an act can be wrong because it violates a term of mutually beneficial cooperation prior to or independently of any legal statute prohibiting it.

152

Susan Dimock

independently of the agreement to enter into mutually beneficial cooperation, but it can be understood as wrong prior to its legal prohibition, and it is that common understanding that forms the basis of the discussion to follow. When non-experts think about criminal law, the specific examples of criminal offences they most readily identify are undoubtedly mala in se: murder, rape, robbery, kidnapping, and the like. Nonetheless, the need to find a justification for using mala prohibita offences seems pressing, given that modern criminal codes contain many such offences, and it has been contended that their recent proliferation is at least in large measure responsible for our current crisis of over-criminalization and the related phenomenon of over-punishment.3 Here I argue that rational choice contractarianism of the type inspired by Thomas Hobbes and developed by David Gauthier provides a novel yet convincing account of how the inclusion of mala prohibita offences in criminal law can be justified.4 It can do so, moreover, while respecting some core requirements of penal justice, especially that criminal law should prohibit only wrongful conduct and should expose persons only to punishments that are deserved. And while all wrongs are, on this approach, dependent on a prior agreement to cooperate, I also suggest that contractarianism provides an illuminating way of further fleshing out the distinction between malum in se and malum prohibita conduct. The argument takes the following form. Some norms are so important to living cooperatively together and to realizing mutual fulfilment that a society must adopt them; some kinds of conduct, or life plans, are so malevolent that they cannot be permitted in any society united by mutually fulfilling cooperation. There are some requirements that any acceptable social structure must include, and some types of conduct it must prohibit and declare ineligible for adoption. Forms of interaction that violate the generalized Lockean proviso—conduct that betters one’s situation by worsening that of others—ought to be prohibited using criminal law.5 Forms of interaction that impose costs on some that they could not accept reasoning ex ante must be forbidden. The core criminal law offences include such conduct that must be prohibited in a society committed to mutual advantage. Conduct that must be condemned and prohibited, because a society that permitted it could not be construed as united for mutual advantage, conduct which if allowed would undermine

3  Douglas Husak, Overcriminalization (Oxford: Oxford University Press, 2008), 119. 4  Hobbes himself had a relatively underdeveloped and problematic theory of law, and Gauthier has not advanced a contractarian theory of (criminal) law or applied his contractarian theory to debates within criminal law theory. Thus both projects I undertake in this chapter—articulating a contractarian theory of criminal law and applying that theory to a specific substantive debate within criminal law theory concerning the justification of adopting mala prohibita offences—perforce extend beyond Gauthier’s own work. Though Claire Finkelstein has developed a contractarian approach to punishment in a series of papers, from which I have benefited enormously, theories of punishment are considerably narrower than theories of criminal law, and she has not applied her understanding of contractarianism to the specific problem of mala prohibita crimes. Thus the ground I cover here is largely terra nova. I have begun developing the foundations of a contractarian theory of criminal law in ‘Contractarianism and Criminalization: Some Foundational Issues’, forthcoming. 5  David Gauthier, Morals by Agreement (Oxford: Oxford University Press, 1986) and ‘Twenty-Five On’, Ethics 123 (2013), 601.

Contractarian Criminal Law Theory and Mala Prohibita Offences

153

the rationality of committing to cooperation, is malum in se, wrong independently of its legal status. Mala prohibita conduct, by contrast, consists of actions that are wrongful only within certain contingent and legally structured cooperative practices and such actions are thus not wrongful prior to or independently of the law which in part constructs those practices. To anticipate the main argument to come, contractarianism provides a rationale for criminalizing mala prohibita conduct, when doing so serves mutually beneficial cooperation among rational persons living together in moral and political community. I provide examples of three types of mala prohibita, the adoption of which contractarianism nicely rationalizes. The rationale for enacting mala prohibita offences begins with the recognition that committing to interact with one’s fellows on moral terms makes possible an infinite number of new activities, sources of utility, and practices that contribute to human flourishing. Cooperation makes possible new ways of experiencing the common good, but the activities constituting or enabling that good require regulation and the coordination of expectations. Institutions that serve the common good, and regulations that coordinate conduct in mutually beneficial ways, can be reinforced using criminal law, so that attacks on those institutions, or actions that culpably flout those regulations and disappoint the legitimate expectations of their participants, may be criminalized. In many cases, the activities would not be wrongful outside of the regulatory context in which they occur, but within those contexts they are wrongful and properly criminalizable. Whenever there are social practices that require regulation and conformity to norms that limit the liberty of their participants in order to produce the cooperative benefits they make possible, we may use criminal law to prohibit violations of those norms, provided doing so is necessary to secure a sufficient level of norm compliance and provided the practice secures net benefits for all its participants, even with the added risk of punishment that criminalization of conduct entails. Contractarian societies would also protect free markets, and employ measures to prevent or correct market failures, activities which again allow for the adoption of mala prohibita offences. Likewise, we may criminalize conduct aimed at undermining public institutions of the state and condemn conduct that interferes with the state’s ability to provide public goods, even though such conduct would not be wrongful had we not agreed to form a polity committed to providing mutual benefit to its members. Attending to the various ways in which forming a moral community and a liberal polity6 for mutual benefit requires the establishment of public norms and the condemnation of those who violate them makes possible new ways of justifying prohibitions on a range of malum prohibita behaviour. In assessing the legitimacy of any use of the criminal law, we should subject it to the ‘contractarian test’ developed by Gauthier; whether a given norm, social institution, or practice is morally justified depends upon whether it passes this test. 6  I assume, given the presumption that contractarians give to personal liberty, that any political and legal system eligible for adoption would be liberal in broad outline, notwithstanding internecine disputes about the precise nature of liberalism.

154

Susan Dimock

The contractarian test applied to a norm or set of norms is simply whether, were the persons within the scope of the norm or set of norms to be in a position to determine their conditions of interaction, it would then make sense for each of them, deliberating rationally, to agree on such conditions, and to include the norm or set of norms among the agreed conditions. Social institutions and practices, the opportunities and limitations, the rights and duties, that they create are thus seen as justifiable in terms of a hypothetical agreement, and indeed only in such terms. Without supposing that the appeal to hypothetical agreement is widely available for justificatory purposes, the contractarian maintains that it offers the best possible defense of a system that normatively regulates the interactions of individuals when some such system would be beneficial to everyone, but any actual system must largely pre-exist the real persons who enjoy its benefits and fall under its constraints.7

States should adopt only criminal laws that pass this test: we should demand of any proposed criminal law that it could reasonably be agreed to by members of society reasoning ex ante, as a term of subsequent cooperation. If proposed laws could be included as part of the normative structure of our society to which we could all reasonably agree were we able by universal agreement to choose that structure, they are eligible for adoption. Many norms could pass the contractarian test, of course. Many norms against serious mala in se conduct will be justified on such a test, regardless of the other normative structures and social practices found within a given society, but whether norms that prohibit mala prohibita conduct will also be justified in a particular society will often depend upon the relations between its various normative structures (e.g. family or economic organization), other branches of law (e.g. torts or contracts), and most especially, other regulatory structures.8 The justification for adoption of mala prohibita offences will typically require that we examine not just the specific offence definition, but the role of the legal prohibition within the broader norm governed practice that it protects or serves. I outline a procedure for applying the contractarian test to mala prohibita offences in Section III.

II.  The Challenge of Justifying Mala Prohibita Offences It seems that most approaches to justifying criminal prohibitions and criminal punishments will have difficulty justifying the inclusion of mala prohibita offences in penal law. A pure consequentialist of the utilitarian variety may be able to justify the disutility caused by criminalizing some conduct—the disutility caused by the restriction of freedom or by the imposition of penal sanctions—even when that conduct is itself not independently wrongful, provided doing so is an efficient means to preventing even greater harm (e.g. criminalizing the sale of burglary tools as a means to prevent burglaries). But since utilitarian theories have been rejected by 7  David Gauthier, ‘Political Contractarianism’, Journal of Political Philosophy, 5 (2) (1997), 132–3. 8  Victor Tadros, ‘Criminalization and Regulation’, in R. A. Duff, Lindsay Farmer, S. E. Marshall, Massimo Renzo, and Victor Tadros (eds.), Boundaries of the Criminal Law (Oxford: Oxford University Press, 2010).

Contractarian Criminal Law Theory and Mala Prohibita Offences

155

most modern criminal law theorists, for what seem to me to be very good reasons, they won’t concern me here. Instead, I examine the problems that mala prohibita offences create for those who believe that criminal law should prohibit only wrongful conduct and should only impose punishments that are deserved. This makes the project of much broader relevance, since most theorists agree on what we might call the ‘wrongfulness’ and the ‘desert’ constraints on criminal law.9 Although desert has long been associated with retributive theories of punishment, and the wrongfulness constraint associated with legal moralist approaches to law, the problem I address challenges anyone who accepts that we should only punish people if they deserve such treatment and only to the extent of that desert, or who accepts that wrongful actions should be the target of criminal law or that wrongdoing should serve as a side constraint on criminalization.10 Understood in this way, as a necessary but not sufficient condition of justified punishment, the desert constraint implies only a commitment to negative (weak or permissive) retributivism. Negative retributivism is widely subscribed to, even by theorists who would not describe themselves as retributivists (for example, those who develop a hybrid view, or who accept desert as placing side constraints on the achievement of instrumental values by punishing). Likewise, given the various ways in which wrongfulness can be incorporated into very different theories of criminal law, it too is a very widely held commitment. Thus if mala prohibita offences run afoul of either the wrongfulness constraint on criminalization or the desert constraint on punishment, this would have widespread repercussions for most contemporary criminal law theories. The challenge to justifying mala prohibita offences depends, then, upon the plausibility of the wrongfulness and desert constraints on criminalization and punishment. Before examining whether those constraints are compatible with the prohibition of malum prohibitum conduct, I say a bit about why we should accept those constraints and thus accept that there is a problem to solve.

A. Wrongdoing rather than harm causing as the object of criminal law There are a number of things to say in favour of the claim that only wrongful conduct should be, even in principle, criminalizable.11 I begin by defending the claim that wrongfulness is a necessary condition of legitimate criminalization. This is seen most clearly by contrasting the wrongfulness requirement with the 9  Following Husak, Overcriminalization. 10  For a very useful discussion of the varieties of legal moralism, and their relation to other principles of criminalization such as the harm principle, see R. A. Duff, Answering for Crime (Oxford: Hart, 2007), ch. 4, and ‘Toward a Modest Legal Moralism’, Criminal Law and Philosophy, 8 (2014), 237–44. 11 Among those who accept the wrongfulness constraint, there is a debate about whether all forms of moral wrongdoing are in principle subject to criminal prohibition (defended by Michael S. Moore, Placing Blame (Oxford: Oxford University Press, 1998)) or whether only public wrongs should be the object of criminalization (defended by R. A. Duff in Answering for Crime, and Husak, Overcriminalization). I agree with the latter in thinking that only public wrongs are properly criminalizable.

156

Susan Dimock

competing12 ‘harm principle’, which makes harmfulness rather than wrongfulness the necessary feature of criminalizable conduct. That wrongfulness is a necessary condition of criminalization is revealed by the fact that only wrongfully inflicted harms are even in principle criminalizable. This is reflected in the role that consent plays in removing harm-causing conduct from the realm of criminal law. Insofar as harmful conduct that is consented to by those suffering the harm is not wrongful (insofar as consent negates the wrongfulness of the conduct that produces the harm), the criminal law notes this transformation by removing what would otherwise be properly criminalizable from the realm of criminal offences. But because (or to the extent that) there is a limit on the transformative power of consent to render what would otherwise be wrongful into something permissible, we also see corresponding limits to consent’s ability to remove conduct from the purview of the law. Both the recognition of the volenti non fit injuria principle and its limits in criminal law suggest that wrongfulness, rather than harmfulness, is the necessary condition of criminalization. Consider, also, that criminal law is addressed to us as moral agents; crimes are expressions of our moral agency and are responded to as such. But whereas our agency is necessarily expressed through our wrongdoing, it is only contingently expressed through our causing of harm. That only wrongfully inflicted harms are criminalizable is reflected, too, in the requirement of mens rea; a person may cause harm without fault, but should not be able to commit a crime without fault. This is not to deny the importance of harm. One way to reflect the moral significance of harm is to make it an element of offences. But we can only make harm an offence element if we also have culpability requirements in addition to the requirement of causing harm (otherwise faultless harm causing would be criminalizable). If we want to reflect the moral significance of harm then we must adopt mens rea conditions as well. This shows that it’s wrongfully inflicted harms, and not just harms per se, that are the rightful target of criminalization, and thus that wrongfulness is a necessary condition of criminalizability. That wrongdoing, rather than harm causing, is the necessary condition of criminalization and the proper target of criminal law is reflected, as well, in the opprobrium in which strict liability offences are held by virtually all criminal law theorists, and many jurists too. Strict liability offences allow conviction for the actus reus of a crime with no proof of fault, culpability or mens rea as to one or more of its elements. In Canada, for example, the Supreme Court has declared that strict liability (‘absolute liability’ as they call it) is contrary to principles of fundamental justice if conviction of such offences exposes defendants to incarceration of any duration, and inconsistent with the Canadian Charter of Rights and Freedoms.13 Similar positions can be found in other jurisdictions and under various human rights instruments.14 The opposition to strict liability is further evidence that we 12  I explain the sense in which the wrongfulness constraint and the harm principle compete in due course in this section. 13  Reference re Motor Vehicle Act (British Columbia) s. 94(2), [1985] 2 S.C.R. 486. 14  For an overview of the issues raised by strict liability and its treatment in the UK and EU, see A. P. Simester (ed.), Appraising Strict Liability (Oxford: Oxford University Press, 2005).

Contractarian Criminal Law Theory and Mala Prohibita Offences

157

think criminal liability should be imposed only for wrongdoing; unless harm is wrongfully imposed or risked, it does not warrant a criminal response. Finally, one might adapt a significant insight from Doug Husak when he observes that certain doctrines in the ‘general part’ of criminal law cannot be made sense of without presupposing that it contains both a ‘non-trivial harm or evil’ requirement and a ‘wrongfulness’ requirement. Husak argues that our understanding of justifications in the general part of criminal law support what he calls a ‘non-trivial harm or evil’15 constraint: only conduct that is harmful or evil should be criminalized; our understanding of excuses, by contrast, he thinks supports the ‘wrongfulness’ constraint: only conduct that is in some sense wrongful should be criminalized. I think the ‘harm or evil’ requirement is subsumed within the ‘wrongfulness’ constraint, and our use of both justifications and excuses supports my position better than Husak’s. First, consider justifications. It’s unfortunate that Husak confined himself to just one kind of justification, which goes by the names of necessity, lesser-evils, or just plain justification. This is the most useful example of a justification for Husak’s purpose, which recall is to show that we cannot understand legal justifications unless we recognize a ‘non-trivial harm or evil’ constraint on criminalization. He calls this an ‘internal’ constraint because it seems dictated by features internal to criminal law itself, namely its recognition of justifications. The thought is brilliantly simple: using the lesser-evils defence requires that we compare the evil or harm done by the conduct in question with the evil or harm meant to be prevented by the criminal offence, and if the harm actually done by breaking the law is (significantly) less than the evil targeted by the criminal law itself, the defendant should be granted a justification defence and acquitted. The classic example of such a defence involves a person who engages in an act of arson, say by burning down a house and farmer’s field, in order to prevent a fire from spreading to a nearby town where it would destroy many homes and businesses. The defendant chose the lesser evil under the circumstances, and so his act of arson was all-things-considered justified. The very operation of this defence requires a comparison of harms, and so requires that we are able to specify the harm targeted by the criminal offence. Since the lesser-evils defence is entirely general, moreover, being available for almost any crime in the special part of the criminal law, we can conclude that every criminal offence specifies a harm or evil to be avoided and which can be used for comparative purposes if the defendant raises a lesser-evils defence. I said that it was unfortunate that Husak chose this as his example, not just because it seems to make his 15  Notice that Husak’s position is made ambiguous by his addition of ‘evil’ in the requirement that only ‘non-trivially harmful or evil’ conduct may be criminalized. Since he introduces the ‘non-trivial harm or evil’ constraint as a necessary condition of criminalization that is independent of and in addition to the ‘wrongfulness’ constraint, however, we must assume that whatever ‘evil’ is, conduct can be evil even if it is not wrongful. I don’t know how conduct could be evil without being wrongful, however, whereas I can readily understand the idea that conduct can be harmful without being wrongful, and so I just treat harmful and evil as interchangeable in this discussion. If ‘evil’ is necessarily also wrongful, Husak has not really offered two constraints—harm/evil and wrongful—and he and I do not really have the substantive disagreement I think we do.

158

Susan Dimock

case too easy. Rather, the lesser-evils defence is not recognized as a justification in all jurisdictions; it is not recognized at all in Canada, for example, and necessity is there treated as an excuse rather than as a justification.16 Husak’s example is thus of more limited application than he likely appreciated. For my purpose here (which, recall, is to defend the claim that wrongfulness is necessary for criminalization), however, we can see that justifications generally do support the wrongfulness constraint. Husak only considers their implications for a harm requirement, but I think they actually more strongly support the wrongfulness requirement. Consider a defence that all Anglo-American jurisdictions treat as a justification: self-defence. That self-defence is treated as a justification is puzzling if one thinks that something like the lesser-evils account provides the raison d’être of justifications. In cases of lethal self-defence, for example, you may justifiably kill any number of attackers, or even any number of legally innocent threats to your life. Both make the lesser-evils rationale for the defence problematic. We can readily identify the harm or evil at which homicide statutes are aimed—the loss of human life—but then how can the loss of more lives, even more innocent lives, be justified? Let’s take a step back. Justifications are granted in cases when the defendant’s actions, though satisfying all of the offence elements of the crime she is accused of committing—in lethal self-defence, intentionally causing the death of another human being—occur under circumstances that make that action not all-things-considered wrong. One reason why a person’s conduct might not be wrong is that it was the lesser evil, but equally it might be that she was exercising a privilege (a police officer killing in the line of duty, a private citizen acting in defence of her child). There’s no reason to think all justifications have the same basis, and it’s unlikely that self-defence and lesser-evils share a common rationale. What makes them both justifications is that the person’s conduct is not, given the circumstances, wrongful. Thus examination of justification defences actually supports the wrongfulness constraint better than the harmfulness requirement. Husak thinks our use of excuses, rather than justifications, supports wrongfulness as a requirement separate from harmfulness. Whereas persons who act with justifications do nothing wrong, those who act with excuses do act wrongfully, but they are excused because something about them or the circumstances under which they acted make it inappropriate to impute the wrongdoing to them or to blame them for it. Whereas self-defence makes the killing of another person not wrongful, driving a get-away car for criminals who threatened to break your legs unless you do so (a case of duress), or having sexual intercourse with someone you genuinely and reasonably believe is consenting but who is not (a case of mistake of fact), or improperly reporting business income on your taxes because you were given in­ correct advice by your government’s tax office (an officially induced error), or speeding in order to drive an injured family member to the hospital (a case of necessity), are wrongful acts but ones for which you should be excused. Though what you did would still be considered wrongful, the reasons why you did it (constituting the 16  Perka v R [1984] 2 S.C.R. 232.

Contractarian Criminal Law Theory and Mala Prohibita Offences

159

excusing conditions) make it inappropriate to attribute the wrongdoing to you or make you less blameworthy than a person who commits exactly the same wrong but for non-excusing reasons. To say that, in circumstances of excuse, the conduct of the accused is still wrong but she should not be blamed for that wrong, again supports the conclusion that criminal law recognizes a wrongfulness constraint. As Husak succinctly puts it, ‘If the defendant is not guilty of wrongdoing, there is nothing to excuse.’17 Thus excuses conceptually depend on wrongdoing, and since excuses also operate throughout the special part of the criminal law, we can conclude that criminal conviction requires wrongdoing. I hope at this point to have established that wrongdoing is necessary for cri­ minalization: only wrongful conduct should be criminalized. If wrongdoing is necessary for criminalization, harmfulness is, of course, not sufficient. Thus my position places in opposition at least one version of the wrongfulness constraint and the harm principle, the version that treats harm as sufficient for criminalization. The harm principle and the wrongfulness principle are, however, compatible in their weaker forms, as each sets only necessary but not sufficient conditions on crimina­lization. It seems likely that the most prominent defenders of the harm principle—John Stuart Mill and Joel Feinberg—understood it as setting only a necessary condition on the proper use of criminal law: the use of criminal law is legitimate when it is necessary to prevent harm. Such advocates of the harm principle can also insist on additional necessary conditions of its proper use, including a wrongfulness condition. This is, I take it, Husak’s position.18 If the harm principle and the wrongfulness principle are to be genuine competitors then, rather than two necessary components of a broader principle of criminalization, defenders of the wrongfulness principle will have to insist that harm is not necessary for criminalization. I have already defended the claim that harmfulness is not sufficient for criminalization (as a consequence of the claim that wrongfulness is necessary); I now defend the claim that harm isn’t necessary for criminalization either. If harm is neither necessary nor sufficient for criminalization on my view then it is genuinely incompatible with the harm principle as traditionally understood. Harm is not necessary for criminalization, since there are many kinds of conduct that are rightly prohibited by criminal law that don’t cause or risk harm to anyone. This is especially so if we must confine ourselves to harms against persons. These would be what Joel Feinberg called ‘harmless wrongs’.19 While no example is beyond controversy, we might consider cruelty to animals, desecration of human corpses, the destruction of works of art, and cases of ‘pure rape’ as examples of harmless wrongs.20 Once we consider the role of criminal law in existing societies, 17 Husak, Overcriminalization, 72. 18  Though again caution is needed. If the ‘evil’ in his ‘non-trivial harm or evil’ constraint need not involve harm, he does not accept the harm principle in even its weakest form. Because in discussions he has repeatedly said he is committed to the harm principle, to the position that only harmful conduct may be criminalized, I think we have to read the ‘evil’ constraint in a way that links evil to harm. 19  Joel Feinberg, Harmless Wrongdoing (New York: Oxford University Press, 1988). 20  John Gardner and Steven Shute, ‘The Wrongness of Rape’, in J. Horder (ed.), Oxford Essays in Jurisprudence, 4th series (Oxford: Oxford University Press, 2000), imagining a case of rape without

160

Susan Dimock

with their full panoply of regulations and legal structures, we also must recognize a greater role for criminal law, both in protecting the effective operation of governments and in enforcing the outcomes of their other regulatory and legal branches. Thus we must recognize the role of criminal law in enforcing the judgments of other legal and political institutions, such as the use of criminal law in enforcing penalties for breaches of the regulatory state, tort awards, family law outcomes, penalties for breach of contract, etc. Insofar as the state also provides services and regulations that serve the common good, interference with state institutions might also be prohibited and punished using criminal law. Thus we must move beyond the core mala in se crimes of personal violence (where the harm principle seems strongest) to crimes against the state (treason, espionage, etc.), to crimes against its institutions (counterfeiting, perjury, etc.), and to crimes against its regulatory and service activities (welfare fraud, tax evasion, etc.). Criminalizing wrongs against political and legal institutions, no less than wrongs against individual persons, can be justified, yet in many cases such wrongs may have no identifiable victims and may cause or risk no actual harm.21 I have argued that we should accept a wrongfulness, rather than a harmfulness, constraint on criminalization: harm is neither necessary nor sufficient for criminalization, whereas wrongfulness can be both. Debating the relative merits of the harm principle and the wrongfulness principle may seem a rather ‘academic’ exercise, however, given that they will surely cover much of the same ground, since many types of conduct are wrongful because they are harmful; the infliction of (avoidable) harm on others is prima facie morally wrong. Crimes at the core of criminal law, especially crimes against persons and property, rightly aim at preventing harms, and denouncing those who culpably inflict them (in attacks) or culpably risk them (in endangerments).22 Many of the most serious crimes are wrongful because they are harmful. We always have reasons to refrain from attacks, from intending others harm. Thus using criminal law against attacks is in principle always apt. We can justify imposing a prospective duty upon everyone to refrain any of the harms normally inflicted by it, because the victim does not know the rape has occurred, suffers no physical or psychological injury, etc. Gardner and Shute, however, think their example, rather than providing a counter-example to the harm principle, should encourage its reformulation. ‘It is no objection under the harm principle that a harmless action was criminalized, not even that an action with no tendency to cause harm was criminalized. It is enough to meet the demands of the harm principle that, if the action were not criminalized, that would be harmful.’ Reprinted in John Gardner, Offences and Defences (Oxford: Oxford University Press, 2007), 29. Thus they use the example of pure rape to support their suggestion that the harm principle ‘says that the criminalization of wrongs is justified only in order to prevent harm’ (30). But given the necessity that the conduct in question must be ‘non-instrumentally wrongful’ to be covered by this application of the harm principle, it cannot be used to defend the harm principle against the wrongfulness principle, as would be necessary if it were to undermine the basic point I am making in the text. Thanks to Duff for pressing me on this point. 21  Advocates of the harm principle might persist, pressing the claim that such activities risk damage to institutions upon which the wellbeing of persons depend and so their criminalization is necessary ultimately to prevent harm to persons. I worry that if such a strategy is permitted, the harm principle will now be either so capacious that it will provide no constraint on criminal law at all, or be transformed from a substantive normative claim into an unfalsifiable conceptual truth, though I won’t pursue this further here. 22  The distinction between attacks and endangerments is Duff’s in Answering for Crime.

Contractarian Criminal Law Theory and Mala Prohibita Offences

161

from attacking everyone; indeed, doing so is the first substantive rule to flow from the social contract: a mutual renunciation of violence and agreement to refrain from intentionally harming others. But where the two part company—cases of harmless wrongdoing and cases of harms innocently caused—it seems that the wrongfulness principle gives the right answer as to whether the conduct should be criminalizable. And this suggests that what is doing the work, even in cases of wrongfully inflicted harms, is the wrongdoing rather than the harm causing.

B. Desert Let’s turn now, much more briefly, to the desert constraint. Unlike the wrongfulness requirement, the desert requirement is not implied by elements found within the criminal law itself. Instead, it is implied by the fact that criminal law licenses the punishment of those found guilty of contravening its prohibitions without a valid defence. Criminal convictions license criminal punishment, the intentional imposition of moral censure and hardship (deprivation or loss) upon those who commit crimes. Such stigmatizing hardship conveys to offenders their com­munity’s disapprobation of their conduct, and expresses that they have (usually temporarily) lost the trust of their community and the right to be treated as members in good standing. Such a punitive response is appropriate only if offenders have engaged in wrongdoing for which they deserve such a response. Justified punishment must be deserved, or so I assume in what follows. We can now, finally, express the challenge that mala prohibita offences pose for criminal law theory. If the conduct prohibited by such statutes isn’t wrong prior to or independently of its legal prohibition, it seems we cannot say that it’s wrongful, and if it’s not wrongful to have engaged in it, imposing punishment for it can’t be deserved. How, then, are we to justify our modern systems of criminal law, given that they include so many such offences? I think contractarianism provides the tools we need to answer this question.

III. Contractarianism Contractarians insist that all genuinely authoritative norms (norms we ought rationally to follow) provide mutual advantage to those whom they purport to govern. ‘Advantage’ is to be understood in terms of one’s life going well; social practices are to a person’s advantage to the extent that they contribute to her life going as well as possible, to her flourishing, given the evaluations she makes about the contours of a fulfilling life.23 Social norms are justified if they would be agreed to ex ante by those bound by them, because they form part of a normative

23 David Gauthier, ‘Friends, Reasons and Morals’, in B. Verbeek (ed.), Reasons and Intentions (Aldershot: Ashgate, 2008); Michael Bratman, ‘The Interplay of Intention and Reason’, Ethics, 123 (2013), 657–72.

162

Susan Dimock

system allowing rational cooperators to achieve mutual fulfilment. They would be endorsed by deliberative rationality as constitutive of or enabling mutually beneficial cooperation. In the case of moral norms, they constrain the direct pursuit of self-interest, when that pursuit involves making others worse off; they prohibit violations of the Lockean proviso, i.e., acting in ways that benefit oneself (or third parties) by making others worse off.24 They are conditional on the willingness of others to reason as cooperators rather than utility maximizers, to pursue efficient and fair Pareto-optimal outcomes rather than best-reply equilibrium states of affairs. Though Gauthier has abandoned ‘constrained maximization’ in favour of ‘agreed Pareto-optimization’ as his preferred characterization of rational deliberation, it remains useful to think that in agreeing to moral norms we abandon the pursuit of individual interest when it would make others worse off, and adopt instead mutual benefit as our aim. Doing so is rational under conditions in which the direct pursuit of subjective interest by each would leave all worse off. In such conditions, modelled paradigmatically as prisoners’ dilemmas, the lives of each will go better if they commit to a system of norms aimed at mutual benefit, and abandon maximizing rationality in favour of optimizing rationality. Individuals must adopt mutual fulfilment as their goal, and commit to being trustworthy cooperative partners in social interaction. In agreeing to form a moral community with our fellows, we agree to interact cooperatively for mutual advantage. We agree to pursue a strategy of cooperation aimed at Pareto-optimal outcomes (that maximize the benefits realizable to us) and to distribute those benefits fairly (via the principle of maximin proportionate gain).25 The commitment to morality thus means more than just committing to act or refrain from acting in certain ways: it means adopting a new ultimate end: mutual benefit rather than self-interest; it means adopting dispositions that make us trustworthy cooperative partners; it means adopting joint strategies for choice rather than individual ones; and it means accepting agreed Pareto-optimization in place of utility maximization as our standard of rationality. As Gauthier now says, ‘we do not treat persons engaged in cooperation as maximizers who constrain their choices to yield an optimal result, but as cooperators, who seek to bring about a Pareto-optimal result whose payoffs are acceptable to all. They view cooperation as directly rational, and go on to establish the rational mode of interaction.’26 Contractarians believe people are most likely to find their lives fulfilling, in circumstances similar to ours, when they commit to cooperating on terms satisfying the contractarian test described above. Such commitment involves integrating the demands of morality into one’s conception of the good; the moral agent’s good is furthered through cooperative norms.27 Thus morality is no longer seen

24 Gauthier, Morals by Agreement and ‘Twenty-Five On’.    25  Gauthier, ‘Twenty-Five On’. 26  Gauthier, ‘Twenty-Five On’; see also Matt Matravers, Justice and Punishment (Oxford: Oxford University Press, 2000). For detailed description of the structure of contractarian arguments, and the importance of seeing that we replace assessments of the rationality of specific actions with assessment of possible deliberative procedures, see Bratman, ‘Interplay’. 27  See Matravers, Justice and Punishment, 172–5.

Contractarian Criminal Law Theory and Mala Prohibita Offences

163

as constraining the pursuit of self-interest; rather, living morally is part of what it means for one’s life to go as well as possible. Matt Matravers describes contractarian morality thus: moral cooperation ‘takes the form of part of a regulative life plan; a way of understanding oneself and one’s good. . . . It addresses agents who have an understanding of themselves as united together with others in a cooperative scheme through which each understands her good. Whilst it retains the constructivist claim that justice is a product of cooperation, it is a product of agreement between persons for whom cooperation is, as Albert Weale puts it, “a condition of individual advantage, rather than a derivative from it”.’28 This should be assumed whenever I speak of mutual benefit, advantage, or fulfilment. Which norms conduce to mutual flourishing will depend on particular and contingent circumstances of different societies. But the variation is not unlimited. Identifying those norms, the content of social morality, involves generalizing from what we know about the conditions under which human beings flourish and find fulfilment in interaction, and the kinds of lives in which they find meaning and value. While philosophers are not uniquely suited to specifying such conditions, and should be responsive to relevant empirical evidence from the social sciences and insights from the humanities, it is surely possible to identify social structures and norms that enable persons to interact in cooperative, mutually beneficial ways, and others that benefit some only at the expense of others who would not agree to bear such net losses if they were to choose the normative conditions under which all would live. Potential cooperators’ evaluations will be guided by consideration of the impact of various normative structures on their prospects for pursuing valuable lives, over the course of entire lives, recognizing that such choices will make possible new sources of fulfilment while foreclosing others. The norms of interaction will be simultaneously enabling and limiting. They define practices that expand the prospects of individuals, not only instrumentally in facilitating the attainment of independently established ends, but also intrinsically in giving rise to quite new ends. But the norms are also limiting, again not only instrumentally in constraining the way in which individuals may attain their own ends, but also intrinsically in excluding possible practices and the ends attainable through them.29

Normative systems enabling and supporting practices promising net benefit to participants will be eligible for endorsement; those imposing net losses on some will be rejected. Those normative choices will, in turn, support some values and corresponding life plans, while constraining others, and even making some ineligible for choice.30 This is the requirement of mutuality that lies at the heart of

28 Matravers, Justice and Punishment, 174, quoting Albert Weale, ‘Justice, Social Union and the Separateness of Persons’, in D. Gauthier and R. Sugden (eds.), Rationality, Justice and the Social Contract (Hemel Hempstead: Harvester Wheatsheaf, 1993). 29  Gauthier, ‘Political Contractarianism’, 132.    30  Gauthier, ‘Political Contractarianism’, 136.

164

Susan Dimock

the social contract. It rules out life plans that benefit one only at the expense of another. Mutuality requires each to eschew the rationality of utility maximization in interaction, because that strategy leads to Pareto-suboptimal outcomes for all. All can do better by committing to Pareto-optimizing rationality and then a system of norms for choice and action that enhance mutual flourishing. Thus they commit to living together on moral terms. And they evaluate existing and proposed normative structures according to the contractarian test. If existing normative systems pass the test, various normative consequences follow. First, we conclude that individuals have ‘a reason, and normally a sufficient reason, for conformity to the practices’. Further, individuals have reason to expect others to conform to justified practices, and to demand that society respond appropriately if they don’t. Gauthier says everyone has ‘a reason to demand that those who fail to do so be sanctioned in some way by society. And she has no reason to complain if she herself is sanctioned, should she fail to conform.’ Persons will rightly invoke sanctions against each other for violations of the norms inherent in social practices that enable mutually fulfilling cooperation. ‘This doctrine of justification, tying reasons in interaction to the rest of the hypothetical agreement, is the core of political contractarianism.’31 Of course, any sanctioning system must itself pass the contractarian test, as must the sanctions it imposes. The need to impose sanctions against those who violate the terms of the social contract is an important part of the justification of criminal law, and the conditions of contractarian justification can be used to determine the proper scope and limits of that practice. The result will be a contractarian theory of criminalization. If adopting norms for mutual benefit requires that we make provisions for sanctioning those who act contrary to them, moral contractarianism cannot stand alone: it must be supplemented with a theory of political contractarianism, which includes a theory of the role of criminal law in society. My suggestion for developing such a political theory begins with this thought: in agreeing to form a polity, we agree to be governed by social structures and institutions, including criminal law, necessary for mutual benefits to be realized in the face of less than perfect compliance with the moral norms of cooperation and continuing practical disagreement. We adopt the political mantle of citizen, agreeing to be guided by law as a standard of public reason rather than our own subjective reason, in conjunction with adopting the moral mantle of cooperator in committing to a moral way of life. The commitment to citizenship means more than just complying with the laws of our polity. Adopting the disposition of citizenship means committing ourselves to the common good, understood in terms of mutual fulfilment; adopting values, aims, and plans pursuable though cooperative interaction and supporting social structures through which such aims and plans are realized; supporting social conditions in which the cooperation that makes such flourishing possible on terms acceptable to each can be realized and maintained; and being committed not only to obeying 31  Gauthier, ‘Political Contractarianism’, all.

Contractarian Criminal Law Theory and Mala Prohibita Offences

165

the law, but to doing our fair share in maintaining the system of law as our system. Just as adopting Pareto-optimizing rationality, plans that are realizable through mutually beneficial cooperation, and the virtue of trustworthiness is what it means to commit to a moral community and moral way of life, so accepting the terms of citizenship structured by social practices that pass the contractarian test and the virtue of law-abidingness is what it means to commit to a political community and a liberal civic way of life. Citizens of contractarian societies accept law as providing the standard of public reason provided their legal norms and values are consistent with or serve mutual benefit, i.e. they are justified under the contractarian test.

IV. Justifying Mala in Se Offences Before turning to the task of outlining a general justificatory scheme for mala prohibita offences in criminal law, it will be useful to briefly describe how contractarians justify the prohibition of mala in se conduct. Mala in se conduct, remember, is conduct that is wrong prior to and independently of its legal prohibition (though not prior to the social contract), and the core mala in se offences are found in every modern liberal state’s criminal law. I pause to consider mala in se crimes not because they are hard to justify; indeed, any adequate theory of criminal law had better be able to justify their prohibition and punishment. Instead, my aim is to explain how contractarians justify such prohibitions. The first condition of the social contract is the reciprocal renouncing of violence and adoption of peace as our common end. Thus it is no surprise that we find prohibitions on basic forms of violence at the core of all systems of criminal law.32 Such crimes are mala in se, and are of great antiquity. Such prohibitions on violence are easily justified by contractarian reasoning. To understand the justification of enforcing other mala in se offences, however, we must attend more fully to the citizen-based approach to political issues that I advocate. The need to attend to the distinctly political aspects of contractarianism becomes apparent if we accept that not all malum in se conduct is properly criminalized. Adultery in intimate relations, many forms of dishonesty and promise breaking, and many forms of vice are wrong prior to and independently of law, yet they should not be criminalized. To reflect this fact, most liberal legal theorists distinguish between public wrongs, which are apt for criminalization, and private wrongs, which are not.33 In order to understand the contractarian justification of criminal law, we must briefly attend to how that distinction is cashed out in such theories, and how it interacts with the mala in se/mala prohibita divide. The social 32  But notice that an unconditional prohibition on force would not be a Nash equilibrium; it would be too risky, if it obligated each to refrain from violence under all circumstances. Instead, only prohibitions on force that permit self-defence and defence of others and property can be a stable focal point. On a contractarian theory of the state, states must extend self-defensive rights to citizens. 33  Except Michael Moore, who thinks all moral wrongdoing is in principle criminalizable, though he argues that there are good reasons not to actually criminalize some forms of wrongdoing. He rejects the distinction between public and private wrongs.

166

Susan Dimock

contract imposes agent-relative obligations upon those united under it. We accept duties to do or refrain from doing acts that violate the norms flowing from the contract. Those obligations are addressed to each of us, as members of the polity created by the social contract. Thus the social contract provides the content of our agent-relative obligations to our fellow citizens (and to the state when it operates as their agent). Such obligations differ from those that arise between specific parties exercising their liberty under the contract, such as obligations between intimates, contractors, and the like, which do not concern the terms of social interaction generally, or the normative structures and institutions that constitute the polity; the obligations such relationships create obtain between persons characterized, not as fellow citizens, but in some more particular way, and so are private. No contemporary theorist has done more to defend the public wrong/private wrong distinction in legal theory than Antony Duff. Public wrongs, those apt for criminalization, as Antony Duff and Sandra Marshall say from their communitarian perspective, violate core values of the community.34 They are wrongs that ‘violate a value on which the civic enterprise depends, and display a lack of the respect and concern that all citizens owe to each other as fellow citizens’.35 They are categorically wrong; their wrongness cannot be a matter of reasonable disagreement. But contractarianism better identifies which wrongs these are. Those values ‘on which the civic enterprise depends’ just are the values that must be respected if society is to be a cooperative venture for mutual advantage, because the civic enterprise just is a practice of social interaction that provides mutual benefit. Those values that are constitutive of or derived from deliberative rationality are those about which there can be no reasonable disagreement, because deliberative rationality provides the common standard of what is reasonable. Seeking mutual fulfilment through social cooperation is what it means to show proper concern and respect for fellow citizens. And the normative structure derived from our common deliberative rationality, the normative structure of social cooperation designed to achieve mutual fulfilment, specifies the content of our public values. The contractarian way of identifying public wrongs asks: which kinds of wrongdoing, if permitted, would undermine the rationality of adopting the disposition of citizenship, i.e. the commitment to pursue cooperation on terms each could accept? Public wrongdoing leaves some worse off, in ways to which they would not agree ex ante, if they were in a position to choose the norms of interaction for all. If their being so treated was not met by a formal public response of condemnation, such wrongdoing would undermine the rationality of accepting the terms of citizenship. The core mala in se offences must be treated as criminal wrongs because no society that permitted them could be construed as a polity committed to mutual advantage. Crimes against the person and personal property could not be tolerated consistently with the commitment to society being for mutual benefit. 34 R. A. Duff and S. E. Marshall, ‘Communicative Punishment and the Role of the Victim’, Criminal Justice Ethics, 23 (2) (2004), 44–6. 35  R. A. Duff, ‘Towards a Theory of Criminal Law?’, Proceedings of the Aristotelian Society, 84 (2010), 21.

Contractarian Criminal Law Theory and Mala Prohibita Offences

167

States ought to criminalize violations of the social contract that threaten optimizing cooperation and the rationality of living morally, conduct that must be prohibited in any society united for mutual advantage and supporting fair and optimal cooperative practices. Contractarians say the core mala in se offences are public wrongs because no society that formed itself for mutual advantage could permit such wrongs to be committed against any of its members without protest. Victimizing attacks on the person are public wrongs because they violate values at the core of persons’ conception of their lives going as well as possible. They are incompatible with interaction based on an understanding of ourselves as united for mutual advantage; they impose harms no one could accept if they value themselves and their own well-being. People who think their own lives are valuable would reject norms for interaction that permit the infliction of such harms on them. Thus prevention of violence is unquestionably a legitimate use of criminal law. This explains why all minimally just societies have the same offences in the core of their criminal law. Crimes of violence are also wrong because they are direct attacks on the civic enterprise itself. One of the things states must do is secure the conditions of basic trust and security that rationalize accepting the burdens of citizenship. Security of our persons and property is a precondition of the cooperative activity of living together morally. The civic enterprise is unsustainable without widespread security from violence, and states achieve this security by claiming almost complete monopolies on violence (subject to citizens retaining self-defensive rights). Acts challenging the state’s monopoly on violence are direct threats to the political structure under which cooperation is possible and rational. Mala in se are public wrongs that violate the terms of the social contract and the norms of social cooperation. If our fellows allowed us to be attacked by others, or culpably endangered by their failure to exercise the levels of care we had established as terms for our interactions, if our polity did not condemn such actions and punish their perpetrators, the rationality of being trustworthy cooperative partners would be undermined. Both crime victims and their fellows would have less reason to trust others, and the state, if certain wrongs were not responded to collectively through criminal processes. If the victimization of some is not condemned ex ante, and punished ex post, those who are victimized would have reason to doubt that society is a cooperative venture for mutual fulfilment.36

V. Justifying Mala Prohibita Offences But contractarianism provides a rationale for criminalizing considerably more than just core mala in se conduct. It would protect free markets, criminalize conduct 36 On the relationship between social trust, security, and criminal law, see Susan Dimock, ‘Retributivism and Trust’, Law and Philosophy, 16 (1997), 37; Peter Ramsay, ‘Preparation Offences, Security Interests, Political Freedom’, in R. A. Duff, Lindsay Farmer, S. E. Marshall, Massimo Renzo, and Victor Tadros (eds.), The Structures of the Criminal Law (Oxford: Oxford University Press, 2011).

168

Susan Dimock

aimed at undermining public institutions, condemn conduct that interferes with the ability of the state to provide public goods, and license prohibitions on a range of mala prohibita behaviour. Contractarians can justify a wide range of criminal prohibitions because we recognize that committing to interact with one’s fellows on moral terms makes possible an infinite number of new activities, sources of utility and practices that contribute to human flourishing. In many cases cooperation makes possible new ways of experiencing the common good, but the activities constituting or enabling that good require regulation and the coordination of expectations. Institutions that serve the common good, and regulations that coordinate conduct in mutually beneficial ways, can be reinforced using criminal law, so that attacks on those institutions, or actions that culpably flout those regulations and disappoint the legitimate expectations of their participants, may be criminalized. I discuss three kinds of mala prohibita offences in this section: offences involving market behaviour, offences against the state, and finally a broad category of hybrid offences related to the regulatory state. Though these don’t exhaust mala prohibita offences, they cover a considerable number that exist in modern penal law. If I can make the case for the legitimacy of offences within these three broad categories, I will have pointed the way to justifying the enactment of many actual offences and provided a model for assessing others.

A.  Economic crimes, or crimes against the market Though leading criminalization theorists don’t give much attention to the role of criminal law vis-à-vis market activities, contractarians would. Activities in a genuinely free and competitive market are always, necessarily, Pareto-improvements. Free market transactions make every participant to them better off, and thus require no moral constraints on their participants; this is why Gauthier called the genuinely free and competitive market a ‘morally free zone’ in Morals by Agreement: each person acting to maximize her own utility would also promote optimality.37 But we know that free markets require a number of preconditions to exist, and that markets fail in a multitude of ways. Both make room for possible state regulations, some of which may be directly enforced through criminal law and others indirectly backed up by it. Insofar as protecting the competitive market is an important state function in a contractarian polity, there’s room here for the operation of criminal law. Some of the ways we use criminal law to support free market activities involve prohibiting conduct that is malum in se, as when we criminalize various kinds of theft and otherwise protect private property. Likewise, when markets fail, producing negative externalities like pollution, we can justifiably use criminal law to encourage the internalization of such costs and punish those who impose them on others. And since markets only achieve optimal results through individual utility-maximizing interactions if fraud is eliminated, we can also justify prohibitions on misleading advertising and the like. Presumably pollution, theft, and 37 Gauthier, Morals by Agreement, ch. 4.

Contractarian Criminal Law Theory and Mala Prohibita Offences

169

deception are wrongful independently of law. But if we look, as well, to crimes of collusion, price fixing, unfair competition, and insider trading, we enter the realm of mala prohibita. If we look just at the actions prohibited under such laws, they seem not wrongful at all. What is wrong, for example, with a group of merchants agreeing to sell their products at the same price, or with someone telling a friend of an investment opportunity he is aware of because of insider information? It is only because these activities interfere with the free and competitive operation of the market, and because the costs of market failures must be internalized, that such offences can be rationalized consistently with the wrongfulness requirement. Many other offences related to facilitating or protecting free market activities will fall into the category of crimes discussed in Section V.C, because they are hybrid offences. But one final example of how criminal law may legitimately serve market activity can be mentioned here. The market, as a morally free zone, depends upon goods being private, in the sense that they are both rivalrous and excludable. Most tangible consumer goods are private: their use by one person diminishes or eliminates the opportunity of others to use them (rivalrousness) and their use can be controlled so that others are prevented from using them (excludability). Goods that are neither excludable nor rivalrous are public goods, whereas those that are rivalrous but not exclusive are common goods and those that are exclusive but not rivalrous are toll or club goods.38 Free markets are excellent vehicles for facilitating exchanges of private goods for the mutual benefit of the transactors, but fail when goods are public. Public goods tend to be undersupplied by the market, because their non-excludability allows everyone to benefit from them, even without paying for them or contributing to their production. People can free ride on the pro­duction of public goods. Various mechanisms in intellectual property law, such as copyrights and patents, artificially introduce excludability and thus transform public goods into toll or club goods, as can subjecting what would otherwise be a public good to the criminal law of theft.39 Such mechanisms can provide the inducements necessary for the continuing production of such goods, and discourage free riding. Violations of such laws may be wrongful and so properly criminalized.

B.  Crimes against the state Although modern thinking about crime seems dominated by mala in se offences against persons and property such as murder, rape, assault, and theft, this understanding of criminal offences is of fairly recent origin. Under earlier understandings, criminal law was thought of less as protecting individuals and their property than it was the public realm, the king’s peace, and the king’s property. As Lindsay Farmer reminds us, this older understanding is reflected in the seriousness with which William Blackstone ranked various crimes, in descending order from ‘crimes against god and religion, to treason and lesser felonies against the king’s 38  To borrow the labels given to various goods by Stuart Green, Thirteen Ways to Steal a Bicycle (Cambridge, Mass.: Harvard University Press, 2012), 209–10. 39 Green, Thirteen Ways to Steal a Bicycle, ch. 4.

170

Susan Dimock

prerogative, to offences against the public peace, trade, and public health, before only then coming to a discussion of homicide and other crimes against the person’.40 And even then, crimes against persons were often seen primarily as wrongs against the king, whose power (military or financial) might be diminished by damage to his subjects. Even within our modern, more individualistic framework, however, contractarians can make good sense of many crimes against the state and its institutions. Insofar as committing to live together as a polity requires the effective and efficient operation of state institutions, a wide variety of crimes against them can easily be justified. Treason, tax evasion, subversion of justice, perjury, bribery of public officials, counterfeiting, and contempt of court are examples of crimes against the state or public institutions.41 Those social or political institutions involved in the delivery of public goods can also be protected using criminal prohibitions. What makes these acts genuinely wrong is that they (aim to) undermine the ability of state institutions to function in securing the common good and protecting the interests and rights of citizens it is the duty of states to protect, and so violate the commitment to citizenship at the heart of political contractarianism. Because in assuming the mantle of citizenship we commit to a specific polity and assume duties not to undermine the operation of its central institutions, we can also explain why these duties are owed to our own state, but not to other equally just states (something other political theorists struggle to explain). Crimes against the public in this sense have been discussed extensively, yet no one seems to have noticed that a great many such crimes are actually malum prohibita, if that refers to conduct that is not wrongful independently of law. This is so because many crimes against the state, its officials, and its institutions cannot be understood or committed independently of law. It is not possible to bribe officials, be in contempt of court, counterfeit currency, commit treason, share state secrets, or evade taxes independently of the legal structures that creates public officials, courts, official currency, etc.

C.  Hybrid offences Many other offences are really hybrids between mala in se and mala prohibita. Perhaps the most famous justification of such crimes is that offered by St Thomas Aquinas. He distinguished between legal rules that could be deduced directly from the precepts of natural law and those that were mere ‘determinations’ thereof. But it must be noted that something may be derived from the natural law in two ways: first, as a conclusion from premises; secondly, by way of determination of certain generalities. The first way is like to that which, in sciences, demonstrated conclusions are drawn from the principles, while the second mode is likened to that whereby, in the arts, general forms 40  Lindsay Farmer, ‘Criminal Wrongs in Historical Perspective’, in Duff et al., Boundaries of the Criminal Law, 227, drawing on William Blackstone, Commentaries on the Laws of England (1765–9), iv. 5. This priority is still reflected to a significant degree in the Canadian Criminal Code among others. 41  cf. the historical essays in Duff et al., Boundaries of the Criminal Law.

Contractarian Criminal Law Theory and Mala Prohibita Offences

171

are particularized as to details: thus the craftsman needs to determine the general form of a house to some particular shape. Some things are therefore derived from the general principles of the natural law by way of conclusions; e.g., that ‘one must not kill’ may be derived as a conclusion from the principle that ‘one should do harm to no man (sic)’; while some are derived therefrom by way of determination, e.g., the law of nature has it that the evil-doer should be punished; but that he be punished in this or that way, is a determination of the law of nature. Accordingly, both modes of derivation are found in human law. But those things which are derived in the first way are contained in human law, not as emanating therefrom exclusively, but have some force from natural law also. But those things which are derived in the second way have no other force than that of human law.42

The under-determination of the natural law leaves room for variations within the legal codes of different communities. For while the most basic principles of all human legal codes must be derived from natural law as conclusions, and so must be the same for all communities, those human laws that are derived by determination allow societies to tailor their laws to fit their particular circumstances and needs. Thus there can be a diversity of positive laws in different communities: when human laws are enacted as particularizations from general principles, it is possible that different communities will choose different laws to give content to the general principles, just as different communities will give different particular shapes to their houses, despite the fact that houses all have a general form (walls and a roof designed to provide shelter from the elements). If we substitute ‘optimizing moral code for social cooperation’ for ‘natural law’ in Aquinas’s formulation, I think contractarians should accept his insight; the social contract is under-determined in the same way Aquinas thought natural law is. I suspect that a great many mala prohibita offences can be understood as determinations of moral requirements. Consider, for example, any number of vehicular regulations, health and safety requirements, licensing schemes, consumer protection laws, and financial regulations. If we consider the specific actions that might be required under such practices, they might well seem to be pure malum prohibitum: displaying a valid safety sticker on the windscreen of one’s automobile, driving only with a valid driver’s licence, driving only if over a stipulated age, keeping records of inspection for equipment within one’s manufacturing plant, selling alcohol only in licensed establishments or only during specified hours, requiring professionals to achieve a certain number of professional development hours per year, requiring retailers to display information concerning the origins or contents of their merchandise, prohibiting banks from selling mutual funds containing their own stock to their customers, etc. Each might seem to be a matter of moral indifference; if such actions are wrongful, it is surely only because they have been chosen as part of larger schemes aimed at conduct that is genuinely mala. Husak calls these ‘hybrid offences’, because they combine elements of both mala prohibita and mala in se. All aim to prohibit genuine malum in se conduct: to 42  St Thomas Aquinas, Summa Theologica I–II, Question 95, art. 2: Whether every human law is derived from the natural law?

172

Susan Dimock

ensure the safety of road users and reduce the risk of injury to persons and property from the use of motor vehicles by enforcing safety standards for vehicles, competency requirements for drivers, and rules of the road to reduce the risk of crashes; to ensure that employers maintain safe workplaces; to reduce the risk that minors will have access to alcohol when they may suffer physical damage by its consumption or to limit the dangers of public intoxication; to ensure that professionals maintain a high level of competence given the trust their clients must extend them in virtue of their special expertise; to protect consumers from unsafe products; to ensure that financial institutions are acting in the interests of their clients, rather than for their own benefit, when they dispense investment advice. Duff’s approach to such laws, which places them in a broader regulatory context, takes the right track, although again we differ on many important details. Reviewing modern criminal codes reveals that most hybrid offences are found within areas of law that are principally regulatory in nature, though aimed at preventing genuinely wrongful conduct: areas such as trade, traffic, and commun­ ications. Duff explains the wrongfulness of breaches of regulations aimed at securing some aspect of the common good by noting that, though such actions are not wrongful prior to their regulation, once they are regulated, acting contrary to the regulations may be wrong. This is surely true for many hybrid offences, such as traffic rules and other rules that solve coordination or collective action problems. Before regulating the flow of traffic by making a certain street one-way, or determining that we shall drive on the right side of the street, driving in a particular direction or on the left is not wrong, but violating these regulations once we have set them is wrong, because it causes unreasonable inconvenience and risk of harm to other road users. Duff thinks many mala prohibita offences can be so justified. Licensing systems of various kinds, for example, serve the common good by ensuring that our conduct is safe and assuring others that this is so (e.g. driving licences, licences for food producers or sellers, for professions requiring a level of competence if the activity is to be conducted safely). ‘[T]‌hese are regulations that we ought to obey if they help maintain the efficient workings of systems that serve the common good; we do wrong when we breach them.’43 Violations of such regulations breach our ‘civic responsibilities’ and our duties as citizens. While I agree with what Duff says here, I would frame it slightly differently, putting more emphasis on the fact that acting contrary to regulatory schemes that solve genuine coordination or collective action problems is often seriously wrongful because it disappoints the legitimate expectations of others that the regulations themselves create. In other words, there are perfectly straightforward reasons grounded in rationality that we should appeal to in justifying such laws; we don’t need to appeal, as Duff does, to a general civic responsibility all citizens share to accept limited burdens when those serve the common good. Even limited burdens must be justified, so we need to show that accepting such burdens is rational. 43 Duff, Answering for Crime, 174.

Contractarian Criminal Law Theory and Mala Prohibita Offences

173

The principal benefits of the kinds of regulatory schemes violation of which is in principle criminalizable are found in their coordination of expectations, their reduction of transaction and scrutiny costs, and the assurance they provide participants within the regulated activities that everyone will comply with the Nash equilibrium the law establishes. The reduction of scrutiny and transaction costs is perhaps clearest in the cases of licensed activities. When one needs a physician, an electrician, or a mattress retailer, the costs of finding them is reduced to virtually nothing through licensing regulations. And, equally important, once one has found the physician, electrician, or retailer, the fact that they are licensed to provide their services greatly reduces the transaction and scrutiny costs of cooperating with them. One needn’t examine their past history with other users of their services to determine whether they are competent and trustworthy; the fact that they are licensed provides that assurance. If part of such a scheme includes protections for consumers that mattresses being sold as ‘new’ are in fact made with only new materials, and the most efficient way of providing that assurance is to prohibit retailers from tearing the manufacturers’ tags off mattresses before they are sold to the end consumer, then it is wrong to tear the tag off a mattress.44 More generally, such regulations provide the kind of trust and stability of expectations that make cooperation a Nash equilibrium. On the contractarian view, criminal law is a Nash equilibrium, a focal point that persons gravitate toward as they make rational decisions regarding their own strategies for choice in light of the strategies selected by other persons. In game theory, a Nash equilibrium is a solution where each player evaluates the other players’ strategies and decides they gain no advantage by unilaterally changing strategies when all the other players keep their strategies unchanged. A Nash equilibrium functions as a kind of focal point, where participants in the game gravitate toward a particular legal norm and choose compliance as their strategy if and only if the other players in the game also choose compliance as their strategy. When a system of criminal law works, each player realizes that unilaterally choosing breach as a strategy would confer no benefit because the costs associated with that strategy are too high. So each sticks with compliance. If one player decides that a shift in strategy (breach) is in his self-interest, then the players all fall out of equilibrium. Penalties for breach are threatened to try to prevent the players from falling out of equilibrium in the first place, by ensuring that the cost of unilateral breach is too high, and those penalties are actually imposed as punishment in order to censure those who disrupt mutually beneficial equilibria by their unilateral defection from the common norms. Public wrongs of this sort must be responded to publicly to re-establish trust and provide assurance to members of society that compliance is reasonable. This justification of hybrid laws depends upon seeing such laws in the context of broader regulatory schemes that are themselves justified because they enable or facilitate mutually beneficial cooperation, or solve coordination or collective

44  Stuart Green, ‘Why it’s a Crime to Tear the Tag off a Mattress: Overcriminalization and the Moral Content of Regulatory Offences’, Emory Law Journal, 46 (1997), 1533.

174

Susan Dimock

action problems. There is another set of mala prohibita offences, however, which, although also hybrid crimes, operate somewhat differently. These even more clearly function as determinations of genuine mala, but they do so by specifying what Duff calls implicit endangerments rather than explicit endangerments. The difference is that explicit endangerment statutes explicitly identify the wrong they aim to prevent, such as dangerous or impaired driving, sexual contact with those too immature to consent, or dangerous use of firearms, while endangerments do not specify the wrong to be prevented explicitly, but target it indirectly or by proxy, such as by prohibiting driving above a certain speed or with a blood alcohol concentration above a specified amount, sexual contact with persons under a certain specified age, or removing the serial number from a firearm. Like all hybrids, such laws target genuine mala in se conduct; they are designed to prevent dangerous driving caused by excessive speed or impaired drivers, to protect those too im­mature to give valid consent to sexual activity, and to prevent dangerous uses of firearms, respectively. Given that so acting would be wrongful, it seems such uses of criminal law can be justified as determinations of genuine moral requirements. Yet Husak, among others, questions whether such a rationale succeeds. His concern, briefly, is that even if a majority of people who commit the mala prohibita offences in question are also guilty of the mala in se they aim to prevent, it will nonetheless be the case that some individuals whose conduct satisfies the conditions of the malum prohibitum offence will not, in fact, have committed the malum in se. In such cases, they will be guilty of no wrongdoing, and so their conduct cannot be justly criminalized or punished. While he readily acknowledges that if someone commits both the malum in se and the malum prohibitum, e.g. has sexual intercourse with a person who is both too immature to consent to such activity and is below the stipulated age of consent, or drives in a way that, in fact, poses an unreasonable risk of harm to other road users and at speeds exceeding the posted limit, or fails to have his equipment inspected and thereby puts his employees at unreasonable risk of injury, that person’s conduct is wrongful and may be justly punished. ‘The fundamental problem emerges, however,. . . when the conduct of the defendant is malum prohibitum without simultaneously being malum in se. Some individuals, in other words, commit a hybrid offence despite the fact that their conduct is not malum in se. Because these offences are hybrids, this outcome is inevitable.’45 Since Husak subscribes to the wrongfulness constraint on criminalization and the desert constraint on punishment, he thinks we must conclude that such hybrid offences are illegitimate and result in unjust punishments. Even if most people who commit such offences act wrongly, we know that some will not (their speeding was not dangerous, their sexual partner was mature enough to provide valid consent, their equipment was safe even though not duly inspected), and so their punishment cannot be justified. As he says: ‘A person’s act is not wrongful because it tokens a type that is wrongful when performed by the majority of agents. Only personal wrongdoing can render persons eligible for punishment.’46 45 Husak, Overcriminalization, 107.   46 Husak, Overcriminalization, 111.

Contractarian Criminal Law Theory and Mala Prohibita Offences

175

If Husak’s challenge to hybrid mala prohibita offences stands, much of contemporary criminal law is unjust, given that many laws take this form. And clearly we have to accept his basic position that not all ways of providing determinate content to mala in se wrongs are acceptable. ‘The task of line drawing is not arbitrary in the sense that any position is as good as any other; some specifications clearly are indefensible.’47 To return to Aquinas’s example, wrongdoers must be punished, but not all punishment systems are equally acceptable; imposing capital punishment or even the amputation of a limb for all crimes would, of course, be unacceptable. One obvious condition on the framing of such laws is that the determinations be made in such a way that the majority of those who commit the malum prohibita also commit the malum in se.48 We cannot place the age of consent under statutory rape laws at 25 years, or the blood alcohol level at zero, or the speed limit on highways at 30 mph, or criminalize the removal of serial numbers from all household appliances, since then most people who commit the offence would not have committed the explicit endangerment which the implicit variety ought to target and which makes the conduct wrongful. But even this does not solve the problem of justifying the punishment of those whose conduct was not wrongful. A common response to this challenge points to pragmatic considerations: we can’t do better in specifying the wrongs targeted by such laws. We use hybrids because we cannot directly prohibit the wrong. Were we to proscribe the use of unsafe equipment in workplaces, dangerous driving, impaired driving, sexual contact with persons too immature to consent, non-disclosure of material information to consumers and the like, the law would be so unacceptably vague as to be unworkable, prosecutions would be made much more difficult, and law enforcement discretion necessarily greatly expanded. Such consequentialist considerations fail to justify the compromise of principle such laws involve, however, and no one has offered principled reasons that are convincing. I think contractarianism can provide a principled justification of hybrid offences of the implicit endangerment type. First, notice that the mala in se wrongs mentioned in the previous paragraph are indeed vague. They would thus violate a number of principles of criminal law that contractarians would insist upon (and most other theorists accept on their own grounds). Laws which are so vague would violate principles of legality, most notably the requirements of fair warning or notice (citizens should be given fair warning from the content of the law of what is prohibited or required), and consistency in application of the law (different law enforcement officials should be able to guide their exercises of discretion by the content of the law, and most especially judges should be able to interpret and apply the law so that like cases are treated alike). Laws applying such vague standards, rather than bright-line rules, are deficient as a matter of principle. Secondly, law-abiding citizens should be able to use the law as a guide in their decision making. Laws that are so vague as to leave unsettled all the 47 Husak, Overcriminalization, 110. 48 As Duff, who is Husak’s direct target here, does: R. A. Duff, ‘Crime, Prohibition, and Punishment’, Journal of Applied Philosophy, 19 (2002), 102.

176

Susan Dimock

possible disputes citizens might have about whether the line between acceptable and unacceptable conduct has been crossed fail in one of law’s principal functions, which is to authoritatively settle precisely such reasonable disputes. Thirdly, laws as vague as these may actually impede the liberty of citizens more than bright-line rules would, even if the bright-line rules prohibit some conduct that is not independently wrongful. This will be so whenever citizens refrain from acting for fear of running afoul of the law in circumstances in which judges would not in, fact, find their conduct had crossed the line; vague laws have a chilling effect on legally permissible conduct. Finally, we should credit a consideration Duff raises in favour of at least some such hybrid offences, namely that they may be needed to counteract certain forms of temptation. The judgment of a person sitting in a pub wondering if she can have another pint and still drive safely home, or a person wondering whether the person with whom she wants to have sex is mature enough, may be affected by temptation or wishful thinking making her judgment less reliable than that of legislators in the cool light of day.49 This rationale may even be extended beyond the kinds of tem­ptation cases Duff considers, to include many of the economic hybrids we mentioned: employers and retailers may have powerful incentives to self-deception and rationalization when considering whether they really need to bear the cost of inspections or comply with disclosure laws, especially if they think not doing so might give them a competitive advantage. Additional considerations might be adduced in favour of determinate rules over vague standards in articulating the norms of the community, but the worry remains: not every way of specifying these norms is equally good, and some are downright unacceptable. And if we attend to actual legislative processes, we may well think our legislators are less well equipped to draw the lines in the right places than are conscientious citizens. Extraordinary tragedies often lead to demands for legal reforms, and while such tragedies might make good headlines, they almost always make bad law. At this point, then, I think it best to provide some additional guidance—beyond the requirement that the majority of those captured by the mala prohibita rules must also simultaneously commit the mala in se—as to how rational agents might go about deciding whether a candidate rule is acceptable, i.e. passes the contractarian test, keeping in mind that a number of possible options might do so. Use of mala prohibita offences raises the basic question of the proper use of criminal law, which must concern contractarians because criminal law limits the liberty of law-abiding citizens, and exposes law violators to state punishment (to the loss of many basic rights, to the diminishment of one’s status as a member in good standing in the moral community, and to public censure and stigma). The use of criminal law always involves fundamental terms of the social contract, because it involves what Rawls called ‘the basic structure’ of society. The basic structure includes those institutions that ‘distribute fundamental rights and duties 49  See Duff, Answering for Crime, §7.3; Husak accepts that a few hybrid crimes might be covered by the temptation rationale, but he doubts many can be: Overcriminalization, 106–12.

Contractarian Criminal Law Theory and Mala Prohibita Offences

177

and determine the division of advantages from social cooperation’.50 It involves the fundamental terms of the social contract, moreover, regardless of whether it is used to limit fundamental freedoms like freedom of speech or association, or trivial freedoms like the freedom to use marijuana or possess the ink used in making official currency.51 When contemplating the adoption of criminal prohibitions, contractarians invite us to weigh their benefits against their burdens. Laws that target malum in se conduct directly, as well as hybrid offences that likewise target genuine malum in se conduct indirectly or by proxy, offer as their primary benefit a reduction in malum in se conduct (according to the deterrent and other preventive effects of criminalizing them and punishing violators). The prevention of wrongdoing serves objective interpersonal security to the extent that the risk that we will suffer the wrong is thereby reduced, as well as stabilizing expectations, lowering transaction costs, and reducing burdens of individual judgment. On the other hand, the burdens of law always include the reduction in liberty experienced by the law-abiding and the risk of being punished for violating it. If the gain in security outweighs the loss of liberty and risk of punishment, the law is among the permissible options. One central consideration in deciding whether a particular legal prohibition is justified, of course, is the penalty that will be incurred by those who violate it, since part of the task, as Claire Finkelstein puts it, is to maximize ‘the marginal deterrent benefit without unduly burdening individual liberty’.52 Only if every person to be bound by the criminal law reasonably anticipates, ex ante, that they will benefit overall from its adoption will it satisfy the contractarian test for eligibility. The contractarian test of criminalization is quite demanding. The requirement that each contractor anticipate a net benefit from the use of criminal law means that the gains each anticipates in terms of having increased security against being the victim of wrongdoing (gains achieved through the deterrent and other preventive effects of the law) outweigh both the loss of liberty experienced if they obey the law and the even more serious losses they would suffer if punished for violating it. One must anticipate, ex ante, benefiting overall, regardless of whether one is punished, for the benefit requirement at the core of the contractarian test to be satisfied: ‘the agreement must be reasonably certain to increase advantage to each person under every situation he envisions once the agreement is in place.’53 This includes both situations in which one complies with the law and in which one violates it and is punished for so doing. Contractarians must consider the possibility that they will suffer the penalties authorized under their system of punishment, not because they think it likely they will break the law, but because criminal law and the system of state punishment it 50  John Rawls, A Theory of Justice (Cambridge, Mass.; Harvard University Press), 7. 51  Contra Gauthier, ‘Political Contractarianism’, 144. 52  Claire Finkelstein, ‘Punishment as Contract’, Ohio State Journal of Criminal Law, 8 (2011), 319; see also ‘A Contractarian Argument against the Death Penalty’, New York University Law Review, 81 (2006), 1238. Finkelstein applies rational choice contractarianism to punishment in these papers, but not to the more basic issue of what kinds of conduct should be criminalized to begin with. 53  Finkelstein, ‘A Contractarian Argument against the Death Penalty’, 1322.

178

Susan Dimock

creates is part of the basic structure of society. When it comes to the basic structure, the benefit principle on which contractarians insist ‘screens out gambles: the requirement is not satisfied if the agent cannot see himself as assuredly better off under the relevant rule than he would be in its absence.’54 From Locke’s insistence that we leave ‘enough and as good’ for others in appropriating goods from the commons and Rawls’s advocacy of the minimax principle for choice under uncertainty, to the precautionary principle in economics and the disaster avoidance principle in public policy, the assumption that rational actors will be risk averse when it comes to evaluating the institutions forming the basic structure of society, and so will try to ensure that even if the worst case scenario is realized they will benefit overall from that structure, is widely accepted. Applied to criminal law, the worst case is that one commits a crime and is punished for it. If one can still see oneself as having benefited from the protection such a system extends to one’s interests, given its costs, it is justified. The requirement of individual benefit from the system is what distinguishes contractarian from utilitarian justifications of punishment, even though both locate the primary benefit of the system in its deterrent effects (or preventive effects more broadly, since the law discourages norm violations and encourages norm compliance in many ways beyond just through the deterrent threat of punishment). Finkelstein again captures this advantage well. [D]‌eterrence in a utilitarian theory does not provide a justification that is addressed to individual offenders, as there need be nothing in it for them, even in the ex ante sense. For this [utilitarian] version of deterrence theory requires neither that each individual member of society regard himself as benefited nor that individual members of society consent to the deterrence scheme under which they are protected. While the traditional appeal to deterrence does restrict enhancements in punishment to instances where social welfare will increase in the aggregate, that social benefit may turn out to be unevenly distributed and hence may improve the lot of the few at the cost of the many. The benefit requirement suggests that rational contractors would reject any such gamble.55

If the institution of punishment is to leave members of society better off than they would be in its absence, it must not impose excessively harsh penalties, and it must allocate the sanctions it does impose ‘predictably, fairly, and according to principles of control and individual responsibility’.56 Thus contractarians can avoid the objections to which other deterrence theorists, especially utilitarians, fall prey, without abandoning the central role that prevention of wrongdoing plays in the justification of criminal punishment. When applied to mala in se conduct, the contractarian test is straightforward (even if dependent on many complex empirical facts). How great is our interest in protecting the good in question against wrongful interference by other persons, and how much does the criminal prohibition with its threatened penalty add to the

54  Finkelstein, ‘A Contractarian Argument against the Death Penalty’, 1317. 55  Finkelstein, ‘Punishment as Contract’, 333. 56  Finkelstein, ‘A Contractarian Argument against the Death Penalty’, 1320.

Contractarian Criminal Law Theory and Mala Prohibita Offences

179

security of that good? How greatly is our utility diminished by the loss of freedom and the threat of suffering the penalty? If the marginal gains in security outweigh the costs of the prohibition, it’s eligible for adoption. The calculation in the case of hybrid crimes is more complex than in the case of pure mala in se, however, because the marginal difference made to the estimate of enhanced security the law offers must be discounted by the percentage of non-wrongful actions that will be deterred or punished. Only if the mala in se at which such crimes are targeted are sufficiently wrongful, and a bright-line rule will be significantly more effective in deterring them than a vaguer standard would be, would such laws likely pass the contractarian test. And then violating that public norm is wrongful. Likewise, the evaluation of pure mala prohibita is more complex, because the value of such prohibitions must be determined against the background of the regulatory practice within which they are adopted. The value of that practice, and the need for criminal sanctions to provide public norms to facilitate or protect it, must figure into the assessment of whether violating the norms of the practice is a public wrong worthy of being condemned and punished through criminal law. Whereas the contractarian test can be applied directly to each discrete proposed malum in se offence, as well as to any proposed penalty for each such offence, its application to malum prohibita requires a more holistic approach, an evaluation of the complex social practice under which such conduct might become wrong, and the value of that practice for persons committed to living cooperatively together.

VI. Conclusion Criminal law can perform its functions of deterring conduct it has prohibited and declared to be publicly wrong, thus reaffirming the commitment to cooperation at the heart of the social contract, only if it actually imposes sanctions on violators. This is necessary because criminals demonstrate conclusively that moral sanctions are insufficient to induce moral conduct, and so create the very conditions for which individuals agree to create criminal law. States unwilling or unable to call offenders to account for their wrongdoing and impose legal sanctions upon them fail to govern. Recognizing that the duty to govern includes not only enforcing moral norms against the commission of mala in se, but also identifying and settling various collective action problems for the common good, makes possible new justifications for the criminalization of mala prohibita. We must understand the law as setting public reasons, and the assumption of citizenship as in part constituted by the commitment to accept the judgment of public reason in place of our individual reason.57 Just as adopting morality does not merely protect us from having our well-being set back by others, but rather makes possible new sources of utility that can only be had from cooperation, so too does forming a polity of citizens. It 57  David Gauthier, ‘Thomas Hobbes and the Contractarian Theory of Law’, Canadian Journal of Philosophy, Suppl. vol. 16 (5) (1990). The role of criminal law in setting public reasons requires deeper treatment than it receives here.

180

Susan Dimock

makes possible new practices that can be a source of individual benefit, but only on condition that the fair terms of those practices, as established by law, are complied with. It makes possible the identification of collective action problems and provides solutions to them, but only if the solution is treated as authoritative by all.58 Finally, criminal law’s role in creating reasonable expectations among persons interacting across a wide range of circumstances cannot be overstated; transaction costs are one of the greatest barriers to mutually beneficial cooperation, and to the extent that criminal law stabilizes reasonable normative expectations, it serves the common good and enables citizens to pursue sources of utility that would otherwise be simply unattainable. Contractarianism is uniquely suited to justify mala prohibita offences, as it sees social cooperation as providing new and greatly expanded opportunities for pursuing the common good; recall that the social contract makes available new ends, rather than merely instrumentally serving pre-existing ends. These additional benefits of social cooperation (and not just avoidance of unrestrained predation) can often be achieved only by the coordination of social conduct and the stabilization of legitimate expectations within the expanded set of social institutions and practices, and criminal law can help to achieve these results. Paul Robinson identifies many advantages that obtain when the outcomes of criminal justice systems are perceived as deserved within the community.59 The stigmatizing of offenders’ conduct is a powerful deterrent, but its deterrent value depends upon the stigma being seen as deserved. If the legal system is believed to give just deserts, it will more easily elicit the cooperation or at least acquiescence of those involved in its operation: legal officials, police, jurors, prosecutors, defence counsel, accused, as well as citizens as informants, witnesses, etc. If the system is not seen as giving desert, participants are more likely to try to subvert and resist its operation. Further, citizens are more likely to treat the law as authoritative (accept it as a source of public reasons), if it is seen as giving offenders what they deserve. This is an important source of respect for law generally. Finally, the perceived justice of the criminal law is crucial if it is to sustain the commitment to moral norms. ‘Criminal law, in particular, plays a central role in creating and maintaining the social consensus necessary for sustaining moral norms. In fact, in a society as diverse as ours, the criminal law may be the only society-wide mechanism that transcends cultural and ethnic differences.’60 Thus the moral credibility of criminal law is crucially important, and that credibility is a function of the degree to which it is perceived to give desert to those involved in it. ‘Thus, the criminal law’s moral credibility is essential to effective crime control, and is enhanced if the distribution of criminal liability is perceived as “doing justice”—that is, if it assigns liability and 58  See John Finnis, Natural Law and Natural Rights (Oxford: Oxford University Press, 1980). 59  Paul H. Robinson, ‘Competing Conceptions of Modern Desert: Vengeful, Deontological, and Empirical’, Cambridge Law Journal, 67 (2008), 145; ‘Criminalization Tensions: Empirical Desert, Changing Norms, and Rape Reform’, in Duff et al., Structures of the Criminal Law (2011); ‘The Role of Moral Philosophers in the Competition between Philosophical and Empirical Desert’, William & Mary Law Review: Symposium Issue, 48 (2007), 1831. 60  Robinson, ‘Criminalization Tensions’, 189.

Contractarian Criminal Law Theory and Mala Prohibita Offences

181

punishment in ways that the community perceives as consistent with its shared intuitions of justice.’61 Now it is hard to see how members of a society could object to a legal system that distributed the burdens and benefits it makes possible in ways that all could rationally have agreed to ex ante. If a legal system constrains liberty and punishes violators according to rules and procedures under which all benefit overall, its outcomes will surely be perceived as just, insofar as persons evaluating it are rational. It will deliver outcomes that are seen as deserved. Consider, in support of this contention, the main sources of perceived injustice in our current systems, the kinds of injustice that lead to the lack of support for such systems that un­dermines their efficacy in actual liberal democracies today. Surely many such objections are grounded in perceptions that the benefits and burdens of the criminal law are not distributed fairly, and in some cases are distributed so unfairly that some persons and communities do not benefit from it overall. Given the reality of overcriminalization, moreover, and the disconnect that creates between law on the books and law in the streets, and the widespread discretion law enforcement agents then must have in applying the law, any sense that the burdens of punishments are distributed fairly and predictably is undermined. The burdens and benefits are out of equilibrium both for those who suffer punishments and for those who curtail their liberty beyond the bounds necessary for social cooperation so as to remain law-abiding. Finally, given the excessive harshness of the penalties we impose, at least in North America, the balance of burdens and benefits has shifted away from overall benefit for many and has created such significant intergenerational harms that the system cannot be said to satisfy the benefit requirement. This produces all the consequences that Robinson’s work leads us to expect. I hope, however, to have shown that the mere existence of mala prohibita offences within modern criminal codes is not itself the cause of this crisis, and so we must look elsewhere for its solution.

61  Robinson, ‘Criminalization Tensions’, 190.

7 Liberty’s Constraints on What Should be Made Criminal Michael S. Moore*

I. Introduction We are now 50 years into a long-proclaimed ‘crisis of overcriminalization’.1 The latter phrase has been the academic community’s battle cry against the perceived overuse of the criminal law by legislators in Anglo-American legal systems. In the words of Sir Thomas More,2 our country is ‘planted thick with such laws,’ prohibiting or requiring at the US federal level 15,000 or more types of action. It is enough to make one wonder whether one shouldn’t consult the family solicitor *  This chapter was presented to the Criminalization Conference of the United Kingdom’s Arts and Humanities Research Council’s Criminalization Project, University of Stirling, Stirling, Scotland, 9 September 2012. It is drawn from my and Heidi Hurd’s Seminars on Liberty for East European Professors, sponsored by the Soros Foundation’s Open Society Institute Higher Education Support Program, and given at St Petersburg, Russia, 21–5 July 2008, and at Lviv, Ukraine, 25–9 July 2011. Thanks go to the organizer of those meetings, Sergei Moiseev, my co-teacher, Heidi Hurd, and to our Russian and Ukrainian colleagues for their excellent comments. 1  The current wave of crisis management is led by Douglas Husak’s Overcriminalization: The Limits of the Criminal Law (New York: Oxford University Press, 2008), reviewed by me in ‘A Tale of Two Theories’, Criminal Justice Ethics, 28 (2009), 27–48, and by Heidi Hurd, ‘Paternalism on Pain of Punishment’, Criminal Justice Ethics, 28 (2009), 49–73. Others include Dennis Baker, The Right not to be Criminalized: Demarcating Criminal Law’s Authority (Farnham: Ashgate Publishing, 2011), and A. P. Simester and Andreas von Hirsch, Crimes, Harms and Wrongs: On the Principles of Criminalization (Oxford: Hart Publishing, 2011). The discussion began in the 1960s with Herbert Packer, ‘The Crime Tariff’, American Scholar, 33 (1964), 551–7; Sanford Kadish, ‘The Crisis of Overcriminalization’, Annals of the American Academy of Political and Social Science, 374 (1965), 157–70; Herbert Packer, The Limits of the Criminal Sanction (Stanford, Calif.: Stanford University Press, 1968); Gordon Hawkins and Norval Morris, The Honest Politician’s Guide to Crime Control (Chicago: University of Chicago Press, 1973). 2  The more complete quote is, ‘This country’s planted thick with laws from coast to coast— Man’s laws, not God’s—and if you cut them down . . . d’you really think you could stand upright in the winds that would blow then?’ Robert Bolt, A Man for All Seasons, Act I (New York: Heinemann, 1967), 147. Notice that Sir Thomas More is not complaining about the size or the density of the legal forest; perhaps that was because in More’s day there were not 7,208 offences contained in 466 statutes and 37 statutory instruments as there has been recently in English criminal law. A. H. Hermann, ‘In the United Kingdom: Criminal Offenses which are not Crime’, Daily Journal, 2 February 1982.

Liberty’s Constraints on What Should be Made Criminal

183

before accepting an invitation to dinner, having a drink with friends, or taking a bath in the Jordan. In truth, of course, the sheer number of a country’s criminal prohibitions as such is no cause for alarm. No more than was the Habsburg Emperor pointing to a genuine complaint when he told Mozart that one of his compositions had ‘too many notes’.3 It all depends of course on the quality of the notes or the laws, as the case may be. Yet with the content of Anglo-American criminal law there is indeed cause for serious concern. Anglo-American criminal law has seemingly been hijacked from the kind of regulation of human affairs for which criminal law is appropriate. Every kind of social engineering seems drawn to use the criminal law as its means. Indeed, social regulation is not thought serious unless the heavy artillery of the criminal law is brought to bear on the problem being regulated. To complain of such over-criminalization presupposes some standard as to the proper level of criminalization. This is one justification for the topic for the present chapter. Yet even if we didn’t have the sense that criminal law is vastly overused, it would be worth understanding the proper limits of the criminal law. While it is a worthwhile endeavour to seek the shape and proper boundaries of any area of law, the need for such understanding is particularly acute for criminal law, independently of its apparent overuse. For criminal law physically restrains people, impoverishes them, threatens and coerces them into doing what they may not want to do, and even on occasion kills them. Law with this kind of bite wants a clear understanding of its proper limits. Such understanding might be of a singularly shallow sort. It is sometimes thought that the only limits there should be on criminal law are procedural, not substantive, limits. There are two proceduralist ideas here. One is that in de­mocratic political systems, the majority’s right to rule is so strong that the only criterion of proper criminal law is a procedural one: did the majority go through the requisite procedural hoops to get their view reflected accurately in law? When this pedigree of democratic promulgation has been met, on this view, there can be no overreaching because criminal law is and ought to regulate whatever the majority chooses it to regulate.4 The second proceduralist thought is that such majoritarianism is not unfair so long as the criminal prohibitions that result meet the formal requirements of the principles of legality and the rule of law. For then the majority’s choices are open for all to see and understand, and any who go against them assume the risk of the punishments warned of in such prohibitions.

3  Peter Schaffer, Amadeus, stage play (1929), film (1984). 4  Sir James Fitzjames Stephen’s view, in his critique of Mill’s attempt to state principled limits to the criminal law. Stephens, Liberty, Equality, Fraternity, originally published 1873 (Indianapolis: Liberty Fund edn., ed. Stuart Warner, 1993). The majoritarian sceptic is different from the particularist sceptic with whom Gerald Dworkin disputes in his ‘The Limits of the Criminal Law’, in John Deigh and David Dolinko (eds.), The Oxford Handbook of Philosophy of Criminal Law (Oxford: Oxford University Press, 2011), 4–7.

184

Michael S. Moore

This majoritarian vision both overvalues democracy and undervalues a sense of restraint that even the most ardent democrats should feel. In very general terms this sense of restraint is what Americans for over two centuries have called the ‘Madisonian Compromise’. Born of a fear of tyranny by the majority in democratic systems, the Compromise limits the output of democratic processes with substantive restraints on government power. Such constraints can take one of two legal forms, either that of enumerated powers or that of individual rights, or, as in the American case, both. In any case, the underlying vision is the same: there are some things beyond the reach of even the most democratic of governments. The restraint in question is more focused vis-à-vis criminal law than it is against law generally. That is because of the coercive nature of criminal law: it forcefully restrains those who violate it, and it threatens to so restrain all citizens subject to it. By its nature criminal law thus takes away the liberty of citizens. The restraint against it doing so is thus naturally cast as a restraint of liberty on the proper scope of criminal law. I have thus entitled this chapter, ‘Liberty’s Constraints on What Should be Made Criminal’, and my conceptualization of criminal law’s limits will be in terms of liberty. The history of using liberty as the principle or value restraining overcriminalization is not reassuring. To begin with, ‘liberty’ is one of those honorific words that mean a lot of things to a lot of different people. It is thus often unclear just what is supposed to be doing the restraining. Second, many have been sceptical that there is anything plausibly good about liberty, if liberty is taken to be freedom from state coercion; for how could there be any good in being free to do evil? Third, there is the worry that being generally at liberty would produce untrammelled anarchy.5 Liberty would thus not limit the criminal law, since having liberty in this sense eliminates criminal law, root and branch. Liberty as a concept and as a value thus needs some work if it is to be seriously proposed as defining the limits of a just criminal law. I shall thus first clarify what we should mean by liberty, if we wish to use it as marking the limits of criminal law. Secondly, we need to make plausible the idea that liberty (in the sense stipulated) has the kind of value that could sometimes outweigh the values sought to be protected by criminal law. Thirdly, we need to specify the form such competition of values takes such that liberty can serve as some kind of limit to criminal law. This last is the major task, to see how liberty in the relevant sense can mark the boundary between justified versus unjustified criminal legislation. I start with the relevant idea of liberty.

5  Edmund Burke’s reaction to the libertarian views of Locke and Rousseau. Edmund Burke, Reflections on the Revolution in France, originally published 1790, reprinted in Reflections on the French Revolution (London: J. M. Dent and Sons, 1910), 1. I discuss Burke’s non-libertarian, social conservatism in Michael S. Moore, ‘The Dead Hand of Constitutional Tradition’, Harvard Journal of Law and Public Policy, 19 (1996), 263–74. More contemporary is Robert Bork’s intellectual journey from libertarian to social conservative: The Tempting of America (New York: Macmillan, 1990).

Liberty’s Constraints on What Should be Made Criminal

185

II.  What is Liberty? As Isaiah Berlin noted years ago,6 ‘liberty’ is one of those honorific words so broadly used as to make it difficult to extract any definite reference intended by its use. Some stipulations are thus in order. The first is with reference to Berlin’s own famous ambiguity, that between positive versus negative liberty. Negative liberty is simply the absence of restraint, whereas positive liberty is the presence of something, viz., the power or ability to do what one is at liberty to do. Negative liberty is also often contrasted with the ‘liberty of the ancients’, which is the right to participate in political governance. The stipulation we need is ‘liberty’ in the sense of negative liberty; for it is liberty of this kind that is opposed to the kind of restraints constituting the criminal law. A second stipulation we need has to do with the vagueness of ‘restraint’ and thus of negative liberty. The vagueness in the idea of restraint resides in the continuity that exists between offers and threats: an offer expands one’s opportunity set, whereas a threat constitutes coercion that lessens such sets. The usual criterion for distinguishing the two lies in the baseline of entitlements of the recipient of the offer/threat:7 if I ‘threaten’ to withhold that to which you have no entitlement unless you do what I want, I have made an offer, no matter the verbal garb and no matter how coercive you experience the offer of something you may want very much; if I ‘offer’ not to do something that you are entitled that I do unless you do what I want, I have threatened you, no matter the verbal garb and no matter how welcome the threat may be to you. The vagueness lies in specifying the baseline of entitlements relative to which the threat/offer line can be drawn. The line between governments inducing behaviour by offers of conditional spending, for example, and governments coercing behaviour by threats, will thus be as vague as the baseline of entitlements vis-à-vis what is offered/threatened. A third stipulation is that it is only political liberty of relevance here. The restraints whose absence marks negative liberty will thus be the restraints distinctive of political systems, namely the restraints of legal sanctions and the threat of their imposition. A fourth and last stipulation is to focus exclusively on the restraints distinctive of criminal law, as opposed to the restraints constituted by civil sanctions. I intend negative, political liberty (hereinafter, simply ‘liberty’) to be restricted to the absence of criminal sanctions.8 I thus arrive at a notion of liberty serviceably antithetical to criminal prohibition as such. All criminal laws take away liberty in this stipulated sense. 6  Isaiah Berlin, ‘Two Concepts of Liberty’, in his Four Essays on Liberty (Oxford: Oxford University Press, 1969), 121. 7  Alan Wertheimer, Coercion (Princeton: Princeton University Press, 1987), ch. 12. 8  I thus intend a narrower notion of negative liberty than did Mill in his famous essay On Liberty, originally published 1859 (ed. David Bromwich and George Kateb, New Haven: Yale University Press, 2008), for Mill included all legal and social coercion as the enemy of liberty. I intend the narrower notion so as to explore the distinctive contours given to liberty by the theory of punishment.

186

Michael S. Moore

Yet how can liberty in this narrow, stipulated sense provide us with any normative limit on the proper reach of the criminal law? After all, liberty in this sense just is the absence of criminal coercion. Such liberty might thus be seen as only the label for such limits, but not a reason for imposing such limits, nor a value giving content to them. Whether this is so depends on whether liberty in this narrow, stipulated sense is of value. Is it good, in general, to be free of the coercion of criminal sanctions?

III.  The Goodness of Liberty There are a number of reasons for thinking that this is so. Such reasons have to do with the instrumental goodness of citizens’ being free of state coercion. I focus on the instrumental goodness of negative liberty because of the difficulties attendant on giving much content to the idea that negative liberty is an intrinsic good.9 Perhaps one could think that all coercion is intrinsically bad, state coercion included, so that its absence could thus be seen as intrinsically good.10 Yet it is more informative and less controversial to treat negative liberty as an instrumental good. Here are five suggestions as to what the more basic goods might be to which negative liberty can be a means.11 First, there is positive liberty itself. Negative liberty is one of the things necessary for positive liberty to exist. One has the ability (positive liberty) to do some action A only if the state does not prohibit A by criminal sanction. Freedom from governmental coercion is of course only one of the items needed to have such abilities—one needs physical and economic capacities, for example, as leftist critics of negative liberty are so fond of pointing out. But negative liberty is needed too, so that if positive liberty is of value then so is one of its necessary means, negative liberty. Positive liberty is itself something good. One might plausibly view it as an intrinsic good. This would be the view that enhancement of opportunity sets is always good, full stop. Alternatively, one might think that positive liberty is the kind of universal, instrumental good that John Rawls called a primary good—something so useful to the attainment of a wide variety of ends that no matter what those ends might be liberty would be good for their attainment.12 In either way, positive liberty is valuable, and so thus is one of its indispensable means, negative liberty. 9  An intrinsic good is something that is good not because it contributes to something else that is good, but good-in-itself. Every theory of value is committed to there being intrinsic values by its commitment to instrumental values, on pain of infinite regress. See Michael S. Moore, Placing Blame: A General Theory of the Criminal Law (Oxford: Oxford University Press, 1997), at 157–8, 160–2. 10  This might be the suggestion of Simester and von Hirsch: ‘No one, including the state, should coerce others without good reason . . .’,Crimes, Harms, and Wrongs, 19. Compare Samuel Freeman, ‘Criminal Liability and the Duty to Aid the Distressed’, University of Pennsylvania Law Review, 142 (1994), 1455–92, at 1484: ‘Surely there is nothing intrinsically valuable about the natural liberty to do wrong . . .’. 11  A summary of Moore, Placing Blame, 663–5, 746–8, and in Michael S. Moore, ‘Freedom’, Harvard Journal of Law and Public Policy, 29 (2005), 9–26, at 11–15. 12  John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), at 62.

Liberty’s Constraints on What Should be Made Criminal

187

Second, there is what might be called Millian autonomy (so-called to distinguish it from Kantian autonomy, which is discussed shortly). Millian autonomy is the hard-to-pin-down idea that human choice gains value when freely made. Even bad choices have value in this sense, if they are free choices. The usual metaphor is that of authorship: the autonomous person is a being who is author of his or her actions, a self-determining agent in some sense. Mill’s contrast case is a mere machine such as a steam engine: although it can do many things, it is not the author of any of them so its actions possess no value as autonomously chosen.13 It is no small matter to flesh out the ideas of agency, autonomy, and self-determination implicit in Millian autonomy. Not wanted is some metaphysically libertarian idea of free will. Even if all choices are fully determined by factors themselves unchosen by an agent, that agent’s choices can be self-determined in the relevant sense(s). The causal efficacy of one’s own second-order choices about the kind of person one wants to become is closer to the mark. Reflection about such matters and critical self-examination, as opposed to parroting views of others and unthinking adoption of a tradition, are part of what is needed to be self-determining in the relevant ways. If Millian autonomy is valuable, it is easy to see how legal coercion prevents its attainment. A coerced choice is not an autonomous choice expressive of the agent’s authorship. The threat of the law no less than the threat of a gunman can rob our actions of any of those decisional processes one rightly values as autonomous choice. The third value, Kantian autonomy, is to be distinguished from Millian autonomy. For Kant, an autonomous action is a right act done for a right reason.14 Kant’s thought is that there is value not just in doing the right act but also (and mostly) in doing it for the right reason. The emphasis is on the motives for action, not the processes of choice. Although choice may well be involved whenever an act is motivated by reasons, valuing correct reasons is quite different from valuing free choice processes. Note that in the Kantian sense of autonomy (in contrast to Mill’s sense), there is no value in ‘autonomously’ chosen wrong action, nor is there value in ‘autonomously’ chosen right action when that action is done for the wrong reason. No matter how free one’s decisional process may have been, Kantian autonomy is not achieved in such cases because the right reason does not motivate the right action.15 Again, if there is value in acting for the right reasons, it is easy to see how legal coercion can prevent the attainment of that value, for the avoidance of legal sanctions can easily supplant the more virtuous motivations for an action that might 13 Mill, On Liberty. 14  See Barbara Herman, The Practice of Moral Judgment (Cambridge, Mass.: Harvard University Press, 1993), ch. 1. 15  Joseph Raz, who uses autonomy in its Kantian sense, draws both of these inferences. Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986).

188

Michael S. Moore

otherwise have moved the agent to act. Consider, for example, gifts to the poor. If Kant is right about the locus of value residing in reasons and not just in actions, charitable giving is valuable not just because of the act of transferring wealth to those in need but also because of the benevolence motivating that act. Legally coerce such acts, and the fear of sanctions will often supplant that benevolent motivation. In such cases the virtue of benevolent giving has become the necessity of paying one’s taxes. The fourth consideration showing negative liberty to be instrumentally valuable is preference utilitarian in character. A preference utilitarian must take seriously the simple psychological truth that people generally prefer to make their own decisions free of coercion. Apart from some perhaps mythical happy slaves, people almost universally prefer their decision making to be free rather than constrained. Therefore, from this fact alone, a committed preference utilitarian must give some positive evaluation to negative liberty. Finally, there are well-known costs attendant upon the use of the criminal sanction. I refer not only to direct enforcement costs such as police, courts, and so on, although these are omnipresent costs of criminalizing behaviour and punishing it. There are also less obvious costs of criminalizing certain behaviours that are: (1) typically carried on in private and so witnessed by no one other than the participants; (2) strongly motivated so that criminalization only minimally deters the conduct in question; and (3) typically harmful to no one other than those who willingly participate. Costs of criminalizing such behaviours are well known.16 They include privacy costs in enforcement; costs in terms of disrespect for laws that predictably will be regularly ignored; opportunities for selective enforcement of laws that predictably will be underenforced; and the costs of funding organized crime by the ‘crime tariff’, namely, the artificial restriction of supply of the goods or services in question in the face of relatively inelastic demand. These costs can be significant for behaviours with the three characteristics in question, and add weight to the value of non-prohibition of such activities. These values add up to a reasonable case for leaving people free of state coercion. Yet because of their general applicability to all actions, they cannot themselves yield the content for principles marking the limits of criminal law. Rather, at most they raise a presumption in favour of liberty and against criminal legislation.17 As a presumption, their case can be overcome. How and when this presumption is overcome will give us our first take on liberty’s limits on criminal legislation.

16  These costs and the literature on them are summarized in Moore, Placing Blame, at 663–5. Most of that literature is cited in n. 1 above. 17  The presumption of liberty is discussed in Joel Feinberg, Social Philosophy (Englewood Cliffs, NJ: Prentice-Hall, 1973), 20–2. See also Douglas Husak, ‘The Presumption of Freedom’, Nous, 17 (1983), 345–62; Moore, Placing Blame, 749–50.

Liberty’s Constraints on What Should be Made Criminal

189

IV.  Proper Reasons for Overcoming the Presumption in Favour of Liberty The natural place to start in looking for good reasons with which to overcome the presumption in favour of liberty would be in the theory of punishment.18 After all, a theory of punishment is a theory of what good(s) are served by criminal law generally, and that should directly determine what are the good(s) criminal legislation should be seeking to achieve in their formulating the particular prohibitions making up the special part of the criminal law. Unfortunately the history of this subject has not followed this path. Mostly this is due to the enormous influence of John Stuart Mill’s 1859 essay, On Liberty. In that essay Mill proposed limits to the criminal law that look relatively free-standing, and this has coloured the debate since. Mill’s famous taxonomy of possible reasons for criminal legislation is often cast in terms of characteristics of the actions to be regulated by such legislation.19 These are actions: (1) harmful to persons other than the actors or those consensually joining in their actions; (2) harmful to the actor(s) themselves only; (3) not harmful to anyone but which are offensive to others; and (4) neither harmful nor offensive to anyone but which are (at least thought to be) immoral to perform. Mill’s famous harm principle asserts that only actions harmful to others may be prohibited by the criminal law; actions merely harmful to the actor himself, or only offensive or immoral, may not be prohibited. Mill presented his harm principle as ‘one simple principle’.20 Yet it is not clear that it is either singular or simple.21 One could defend Mill’s own view that it is both by regarding the content of the principle thus: it is necessary and sufficient for the justification of criminal legislation that it prohibits actions harmful to others. Such a principle could then be seen to have three corollaries: since harm to others is necessary, then nothing else can be sufficient to justify a criminal prohibition; specifically, prevention of harm to self, offence to others, or immorality, could not be a sufficient justification. Yet Mill himself doesn’t really present his anti-paternalist, anti-offence, and anti-moralist principles as mere corollaries to the basic harm principle. They seem to have independent normative force for him. Take the anti-paternalist corollary, 18  Doug Husak and I have long agreed about this foundational point. Compare Moore, Placing Blame, 70–1, and Husak, Overcriminalization, 57 (‘principles about the justification of punishment have important implications for criminalization’). 19  Although Joel Feinberg formulated his harm, offence, paternalist, and moralist principles in terms of good reasons for use of state coercion (Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), 26–7), his magisterial four-volume treatise, The Moral Limits of the Criminal Law, is popularly subtitled in terms of the actions that may or may not be coerced by those principles. Feinberg, Harm to Others; Feinberg, Offense to Others (New York: Oxford University Press, 1985); Feinberg, Harm to Self (New York: Oxford University Press, 1986); Feinberg, Harmless Wrongdoing (New York: Oxford University Press, 1988). 20 Mill, On Liberty. 21  As Gerald Dworkin points out in his ‘Paternalism’, in Richard Wasserstrom (ed.), Morality and the Law (Belmont, Calif.: Wadsworth, 1971), 107.

190

Michael S. Moore

for example. Coercing people with the criminal law for no better reason than the state judges what is good and what is bad for them, contrary to their own best judgments, seems independently offensive to Mill’s sense of liberty. It is not hard to see Mill’s point here: it is one thing to tell people that they cannot do what they may want to do because what they want to do will harm the interests of others. It is another to tell them that their own best judgments about what they should want for themselves is defective. The latter, paternalistic judgment offends Millian autonomy much more than does the former, harm-to-others-based judgment. A like independent force seems to be reflected in Mill’s rejection of offence and immorality as justifications for criminal legislation. However we interpret the harm principle and its three negative corollaries, the principle is usually discussed as a free-standing political principle. Yet the principle cries out for some explanation as to why it is that only actions harmful to others may be prohibited by the criminal law. Why are actions harmful to the actors themselves, offensive to non-actors, or just plain immoral, also not eligible for criminal prohibition? The beginning of wisdom here is to see that Mill derived his harm principle’s definition of the sphere of actions immune from state coercion, from a deeper insight about the proper reasons for criminal prohibition. His position was that the only legitimate reason to use the coercion of the criminal law is to prevent harms to people other than the actor(s) whose behaviour is being regulated. Reasons of paternalism, offence, or moralism are improper reasons—either because they do not aim at preventing harm to others, or perhaps, independently so. I call this the motivational principle behind Mill’s harm principle.22 According to this background motivational principle, legislators should aim only at preventing harm to others in their criminal legislation; they should not aim at preventing harms to the actors themselves, offence, or immorality. It is from these proper and improper reasons for criminalizing behaviour that Mill derives the characteristics of actions that may or may not be criminalized. If this focus on the background motivational principle is correct, then surely the only place to seek to justify this principle is in the theory of punishment. For surely legislators should aim in their criminal legislation at whatever criminal law itself is aimed at. What criminal law in general is good for should tightly control what good(s) legislators should seek in their criminal legislation. Mill was of course a utilitarian in his theory of punishment. He believed, like utilitarians generally, that punishment is an evil because it decreases welfare (translated as pleasure, happiness, preference satisfaction, etc.). As such, it can only be justified by producing something sufficiently increasing of welfare that on balance, some net social gain in welfare is achieved. The welfare gain that pun­ ishment achieves might be thought to be the prevention of non-consensually imposed harm. Therefore legislators should aim in their criminal legislation at the prevention of harm to others, and the rational way to do that is by prohibiting all and only actions harmful to others. 22 Moore, Placing Blame, 750–4, 767–9.

Liberty’s Constraints on What Should be Made Criminal

191

It is far from obvious that this move—from the utilitarian theory of punishment to the harm principle’s limit on criminal legislation—works. For one thing, it requires the assumption that the harms to self prevented by paternalistically motivated punishment will always be less than the harms of threatening and inflicting that punishment itself (for if this were not so then prohibition of such actions could be justified on utilitarian grounds). Mills backs this assumption with the supposed epistemic superiority of each individual vis-à-vis the state in judging what is harmful to him,23 but that epistemic backing itself has long been questioned.24 For another thing, to support the inference one must assume that the ‘soft’ harm of psychic offence (prevented by punishment of offensive actions) can never outweigh the harms imposed by the threatening and the inflicting of those punishments; some careful dance through various ideas of welfare (pleasure, happiness, preference-satisfaction) would have to be done to make this assumption very plausible. Whether Mill’s theory of criminal legislation can in fact be derived from his theory of punishment is thus somewhat in doubt. But surely that kind of showing is the right move if one is generally a utilitarian about punishment and is a Millian liberal about the proper motives for criminal legislation. What if one’s theory of punishment is retributive rather than utilitarian—what then becomes of the harm principle as a limit on criminal legislation? The answer is that the harm principle changes quite a bit, so much so that it would scarce be recognizable to Mill or to latter-day Millians. A retributive theory of punishment demands that criminal punishment be imposed on all and only those who deserve it.25 Such desert in turn is cashed out in terms of two moral properties, wrongdoing and culpability: one’s degree of desert is determined by the seriousness of the wrong done, and the culpability with which that wrong is done.26 Retributivism so construed leads to legal moralism—the theory that all and only moral wrongs should be prohibited by the criminal law—as its theory of legislation. It does this via two other assumptions.27 First, one must assume that a principle of legality is good independently of the good of retributivism. Such a principle requires prospective, clear, non-contradictory, legislative rules before punishments may be exacted. Such a principle arguably side-constrains the attainment of retributive justice, allowing punishments giving such justice to be given only if the morally wrong act was legally prohibited beforehand. Such an assumption thus requires that, prima facie, all moral wrongs be prohibited by the criminal law, for otherwise wrongdoers doing those wrongs could not (under the principle of legality) receive their just deserts. 23 Mill, On Liberty. 24 See e.g. Dworkin, ‘Paternalism’; H. L. A. Hart, Law, Liberty, and Morality (Stanford, Calif.: Stanford University Press, 1963). 25 Moore, Placing Blame, chs 2–4. 26 Moore, Placing Blame, part II. 27 Moore, Placing Blame, 70–3, ch. 16. My argument connecting the retributivist theory of punishment to the legal moralist theory of legislation is critically examined in David Brink, ‘Retributivism and Legal Moralism’, Ratio Juris, 25 (4) (2012).

192

Michael S. Moore

The second assumption is that law as such does not obligate citizen obedience, not even prima facie. This means that the legislative enactment of a legal prohibition cannot make (morally) wrong an act not morally wrong before. The upshot is that citizens do no wrong in violating the criminal law, unless that law is reflective of an antecedently existing moral wrong. Since retributivism bars punishment of the innocent as much as it requires punishment of the guilty, and because one is innocent if he does nothing morally wrong, therefore retributivism bans the prohibition of acts that are not morally wrong. Retributivism with these two assumptions yields what I shall call the ‘moral wrong principle’: criminal legislation must exclusively aim at preventing or punishing moral wrongs, and this it can do by prohibiting all and only those behaviours that are in fact morally wrong. It is fruitful to compare the moral wrong principle of the retributivist with the harm principle of the utilitarian. They each purport to give a necessary and sufficient condition of justified criminal legislation at which all legislators should aim, one in terms of harm to others and the other in terms of moral wrongs. It could turn out that the two principles are extensionally equivalent in the laws that they justify; if all moral wrongs consisted of actions harming others, and if all actions harmful to others constituted moral wrongs, then the two principles would justify exactly the same legislation. In fact, however, there is only a large overlap in the two notions, not a perfect congruence. There are wrongless harms, such as competitive injuries in capitalist economies; and there are harmless wrongs, such as cruelty to animals, abuse of a corpse, extinction of a species.28 So while largely coextensive, the two principles diverge in these two classes of cases. The principles also diverge in their modes of reasoning to Mill’s three corollary principles. Whether paternalistically motivated legislation is permissible according to the moral wrong principle depends on the question of whether we are ever obligated not to cause harm to ourselves. While there are certainly diverging views about this in ethics, one plausible view denies that there are duties to self, and thus denies that there can be wrongdoing involved in actions harming the actor only. On this view of moral wrongdoing, the moral wrong principle can be as condemning of paternalistically motivated legislation as can the harm principle. Of course, if the Millian liberal has independent grounds for condemning paternalistically motivated legislation—grounds independent of the absence of a motivation of preventing harm to others—there is nothing preventing the legal moralist from adopting those independent grounds too. If paternalism independently insults liberty (because it insults one of the main values behind liberty, Millian autonomy), as I for one think plausible, then a retributivist can condemn paternalism on this ground too, as well as can any Millian. With regard to the prevention of offence as a reason for criminal legislation, how the moral wrong principle applies depends on what one thinks of the moral status of causing offence to others. Surely sometimes intentionally causing serious 28  Described in some detail in Feinberg, Harmless Wrongdoing, 20–5.

Liberty’s Constraints on What Should be Made Criminal

193

psychic distress to others for no good reason is morally wrong, in which case it may be prohibited by a legal moralist legislator. Notice that latter-day Millians such as Joel Feinberg reach this same result by assuming that such serious psychic distress can be a greater harm than the punishment needed to prevent it.29 Here too the two principles need not diverge in their justification of (some) punishment/ prevention-of-offence-motivated criminal legislation. With regard to Mill’s third corollary—prohibiting behaviour because it is immoral—the two principles seem to diverge dramatically. What the moral wrong principle requires—prohibiting all and only morally wrongful behaviour because it is morally wrong—the harm principle seemingly condemns. Yet all is not as it seems here. I have elsewhere speculated that Mill’s animus against morals-motivated legislation seems largely to be directed at legislative reliance on conventionally accepted moral shibboleths, such as that abhorring polygamy.30 Indeed, now it seems to me that that is all Mill’s third corollary can be directed against. As a dedicated utilitarian, Mill had to see his harm principle as an implication of his utilitarianism generally. Utilitarianism generally was both a principle of legislation and a (indeed, the) principle of morals.31 Mill must have thought that the touchstone of his harm principle was also the touchstone of morality. Behaviour harmful to others was not only the only behaviour prohibitable by the criminal law, it was also the only behaviour that was morally wrong (according to utilitarian ethics). Because of this, behaviours not harmful to others could not really be morally wrong—they could only be thought to be morally wrong by irrational people who did not see the truth of utilitarian ethics. When Mill thus rails against behaviour being criminalized because it is morally wrong, his real target is against criminalization of behaviour that is incorrectly believed to be morally wrong by many or most people. His animus, in other words, is directed at moral error being crammed down people’s throats (via criminal prohibition) for no better reason than that the majority subscribes to such error. If erroneous moral belief is Mill’s real target here, the retributivist/legal moralist can heartily join in its condemnation. Retributivist justice is only achieved by the punishment of morally wrongful behaviour. If behaviour isn’t really morally wrong, it is only thought to be wrong by some sizeable percentage of the population, only the appearance of retributive justice is achieved by punishment of such behaviour. Retributivism, as I have long urged,32 enforces real morality, not conventional moral beliefs. So Mill and the legal moralist can happily agree both in their rejection of conventionalist morality as the basis for criminalization and in their acceptance of real morality as that basis; while they significantly disagree as to the basic shape and the content of real morality.

29 Feinberg, Offense to Others. 30  Michael S. Moore, ‘Four Reflections on Law and Morality’, William & Mary Law Review, 48 (2007), 1523–69, at 1545. 31  As Bentham proclaimed in his An Introduction to the Principle of Morals and Legislation, originally published in 1789 (New York: Hafner, 1948). 32 Moore, Placing Blame, 177–80.

194

Michael S. Moore

The most widely held theory of punishment is neither purely utilitarian nor purely retributive in nature. Rather, it is a mixed theory that requires both that some net social gain and some retributive justice be produced by punishment in order for that punishment to be justified.33 For completeness, consider briefly what theory of legislation such a mixed theory of punishment implies. It implies what I shall call the ‘wrongful harm’ principle. On this principle, criminal legislation is rightly motivated only insofar as it aims both to prevent future harms to others and to achieve punishment of the deserving. Actions may thus be prohibited only if they are both acts causing harm to others and acts that are morally wrong to do—‘wrongful harms’, for short. The wrongful harm principle has many contemporary adherents,34 no surprise given the popularity of the mixed theory of punishment on which it depends. Notice that the principle shares with the moral wrong principle the limiting of criminal prohibitions to those acts that are morally wrong. Thus, acts harmful to others but not morally wrong to do are insulated from criminal prohibition by both principles, in contrast to the harm principle. Similarly, the wrongful harm principle can also condemn paternalistically motivated legislation, on the Millian ground that there is no harm to others sought to be presented, on the legal moralist ground that (since there are no duties to self ) there are no moral wrongs to be punished, or on the ground that paternalistically motivated legislation independently insults liberty and (Millian) autonomy. Similarly, the wrongful harm principle can follow the Millian and the legal moralist in (wholly or partly) condemning offence-motivated legislation, and legislation motivated by erroneous but popular moral beliefs.

V.  First Definition of that Liberty of Action Protected against State Coercion We are now in position to see two closely related rights to liberty that mark one of the hallmarks of a just criminal law. I shall state such rights in terms of my own retributivist theory of punishment. From what has been said before, it should not be difficult for utilitarians and mixed theorists about punishment to modify my statement of these two rights to fit the implications of their own theories. The first is what I have long called the derived right to liberty.35 (It is ‘derived’ because the right is derivative of a more basic legislative duty to enact legislation for right reasons.) It is the right of each citizen to be free of improperly motivated criminal legislation. According to the moral wrong principle, this translates into a right to be free of criminal legislation motivated by anything except prevention 33  On mixed theories of punishment, see Moore, Placing Blame, 92–4. 34  See e.g. Dennis Baker, The Right not to be Criminalized; Simester and von Hirsch, Crimes, Harms, and Wrongs. Arguably included here is Doug Husak, for his internal principles limiting criminalization include harm, evil, wrongs, and desert. Overcriminalization, ch. 2. 35 Moore, Placing Blame, 750–62.

Liberty’s Constraints on What Should be Made Criminal

195

and punishment of wrongdoing. Legislation aimed at social betterment in any other way is forbidden by such a principle. The second right is the right to be free of state coercion (however motivated) that prohibits behaviour that is in fact morally innocuous (even if the legislature thinks that it is immoral). The right here protects against state coercion for any reason; even rightly motivated legislation (that is, legislation that aims to prohibit moral wrongdoing but misses its target by prohibiting morally innocuous behaviour) can run afoul of this second right. The limitations on criminal legislation protected by these two rights do not come to the same thing. There are two possible kinds of divergence between the two. First, imagine that some behaviour such as suicide is morally wrong, but that the legislature prohibits it not for that reason but in order to preserve something good for the actor that the actor himself has chosen to reject. Even though the prohibition (ex hypothesi) is of morally wrongful behaviour, it was not enacted for that reason but for a paternalistic reason, and is thus beyond the first limitation on criminal legislation. Second, imagine that some behaviour is not wrongful, such as oral sex, but the legislature believes that it is and prohibits it for that reason. Now the legislative reason for the prohibition is permissible, but the prohibition is not because it prohibits behaviour that is not in fact wrongful. On the retributivist view (here assumed), criminal law exists in order to exact punishment of culpable wrongdoers; since this does not include practitioners of oral sex, such a law is thus beyond the second limitation on criminal legislation. At a minimum, then, we have a liberty right to a criminal law that aims at moral wrongdoing, and another liberty right to a criminal law that in fact hits its only proper target, viz., behaviour that is actually morally wrong to do. Just how much liberty these rights bequeath to us depends mightily on how rich one imagines morality to be. This is a large question to which I can hardly do justice here. In very general terms, there are a number of lines to be drawn and defended. First, within the category of things people ought (or have reason) to do, we must distinguish moral reasons (and moral ‘oughts’) from practical reason generally. For if morality were equated with all of practical reason, the wrong and wrongful harm principles would exclude nothing from the ambit of proper criminal prohibition. Needed here are distinctions both in terms of the force of a reason, and in terms of its content. Moral reasons are both more stringent in their force, and more other-directed in their content.36 Second, within the class of moral reasons we need to distinguish the aretaic from the deontic, virtues from duties.37 For wrongdoing is breach of obligation, not absence of virtue. I have elsewhere38 taxonomized aretaic categories between 36  For some thoughts in these directions, see Joseph Raz, Practical Reasons and Norms (Oxford: Oxford University Press, 1975). 37  Not an easy line to draw, given the plausibility of thinking that there are aretaic duties, albeit perhaps imperfect duties (in Mill’s sense). See Heidi Hurd, ‘Duties beyond the Call of Duty’, Annual Review of Law and Ethics, 6 (1998), 3–39. 38 Michael Moore, ‘Liberty and Supererogation’, Annual Review of Law and Ethics, 6 (1998), 111–43.

196

Michael S. Moore

non-obligatory acts that are: (1) supererogatory (an act would be morally heroic to do but if it is not done there is nothing morally untoward about the person who fails to do it); (2) suberogatory (an act would be morally odious to omit to do but there is no moral heroism in the doing of it); (3) quasi-erogatory (an act would be morally heroic to do and morally odious not to do); and indifferent (an act is neither good to do nor bad to omit to do). I have also taxonomized obligations between those that are most stringent (agent-relative, categorical, deontological), and those that are as a class less stringent (consequentialist duties, such as the duty of rescue).39 The moral wrong and wrongful harm principles as I would construe them allow and prima facie require that both stringent and non-stringent moral obligations be criminalized: they forbid the criminalization of the indifferent, the suberogatory, the quasi-erogatory, and the absence of the supererogatory. Some defend this on possibility lines: ‘you can’t coerce virtue without destroying it.’ My own defence is retributive: no one deserves to be punished for their aretaic failures.40 Desert is made up of culpable wrongdoing, not aretaic failure. These general lines of course do not give much content to the moral wrong/ wrongful harm principles. Such more detailed content can only come from a detailed taxonomy of the moral obligations we have. On my view we do have what are sometimes called ‘public’ obligations, such as the obligation to support just institutions, the obligation to seek and to support the solutions to certain coordination games, the obligation not to free ride on public goods and to solve prisoner’s dilemmas and games of chicken, etc. This allows for criminalization of a wide variety of instrumental wrongs as malum prohibitum crimes.41 On the other hand, one of the areas popularly regarded as the subject of detailed moral obligations, sexual conduct, strikes me as belonging to the morally indifferent category, not in any other aretaic category and certainly not in the category of moral obligations. It diminishes morality to think that it obligates us to practise sex one way rather than another. Indeed, it strikes me that morality is indifferent about what organ we insert into what orifice of what gender of what species. These are at most matters of taste and of aesthetics, but not of morality. The upshot for one with my views of morality, is that these two rights protect much of what we do from criminalization. They also condemn large portions of existing criminal codes, aimed as those codes are at morally permissible behaviour. Still, large as the protection is, one should regard these two rights as but the safe harbours of liberty. Within such harbours there is no balancing to be done, no trade-offs to be made. Liberty should be absolute against these two forms of illegitimate legislative intrusion. Can we say the same about the absence of liberty on the other side of these lines? That is, if legislation aims at prohibiting moral wrongdoing, and in fact what it prohibits is morally wrong, is it necessarily true that there is no principled 39 Moore, Causation and Responsibility, 37–8. 40  Moore, ‘Liberty and Supererogation’, 135–42; Moore, Placing Blame, 584–7, 758–9. 41 Moore, Placing Blame, 760–1.

Liberty’s Constraints on What Should be Made Criminal

197

objection to make against such legislation? Or do our rights to liberty encompass some right to be free of state coercion in our doing of the wrong thing? Do we, in short, sometimes have the right to be wrong? If so, then while liberty is secure when we do no wrong, our liberty does not end where our obligations begin. That is the possibility I want next to explore.

VI.  Second Definition of that Liberty of Action Protected against State Coercion: Our Liberty to Act Wrongly According to the harm and the wrongful harm principles, we of course have liberty to do some morally wrong acts, namely, any such acts as are not harmful to others. But I am interested in illustrating the retributivist view about such things. How, for a retributivist and holder of the moral wrong principle, can it be true that we have a right to be free of state coercion for any morally wrongful actions? In such cases the state admittedly has in fact good reason to punish and it is motivated by that reason, so how can it be nonetheless wrong for the state to proceed? I shall shortly outline several possibilities here. But first we need to dispose of a conceptual objection which would, if true, rule out all of these possibilities in one swoop. This is the objection that it is contradictory to think that we can have moral rights to do morally wrong actions. The basis for this objection is pretty straightforward: a wrong act is an act we were obligated not to do; if we were obligated not to do that act, we were not permitted to do it; if we were not permitted to do it, we had no right to do it. We can sidestep this otherwise potent objection by noting that the liberty right in question is not literally a right to do a wrongful action. Rather, the right we need is a right against the state coercing us out of doing a wrongful action. There is no contradiction in the possibility that it would be more wrong of a state to coerce me out of doing some action A, than it is wrongful of me to do A. Indeed, such ‘two-level’ arguments abound in the literature on liberty. For example: ‘It is wrong for a woman to obtain a late-term abortion, but it would be more wrong for the state to coerce her out of that choice.’

A. The criminalizability of only some kinds of wrongs and the liberty to do the others, the ‘non-grievance’ or ‘private’ wrongs Because many share my sense that we must be at liberty to do some wrongful actions, various theorists have sought to capture this possibility with some idea of non-prohibitable moral wrongs. The intended implication of all wrongs being divided into two classes is that one of those classes of wrongs is not a fit subject for criminalization. The difference the distinction between two classes of wrongs is supposed to make is thus clear enough; just what the difference is that is supposed to make this difference is less clear.

198

Michael S. Moore

Consider two well-known possibilities. One is Joel Feinberg’s division of all moral wrongs into two classes: grievance wrongs and non-grievance wrongs.42 A grievance wrong is a wrong to someone, often a correlative rights holder but in any case someone to whom one owed the duty not to do the act in question and one who has standing to complain about the violation when it occurs. Grievance wrongs almost always (always for Feinberg?) harm the person to whom they are done. Whereas non-grievance wrongs are acts that morality forbids but not because anyone has a grievance; if they are done such wrongs violate obligations not based on rights, obligations that are not owed to anyone,43 obligations breach of which harms no one. These are ‘free-floating’ obligations, prohibiting the causing of ‘free-floating’ evils.44 Feinberg’s use of this distinction is both careful and honest. Roughly, breach of grievance morality—where there is wrongful harm to a non-consenting, identifiable victim who thus has a ‘grievance’—is the law’s business whereas breach of non-grievance morality is not. Honestly enough, and against his own apparent inclination, he recognizes that even breach of non-grievance morality provides a reason for state prohibition and punishment.45 Yet Feinberg’s view was that such a reason is always so weak that it loses out to the liberty costs attendant upon criminalization, viz., the values that back the presumption in favour of liberty.46 A second proposal for cabining the kinds of moral wrongs suitable for criminalization is that of Sandra Marshall, Tony Duff,47 and Doug Husak.48 This proposal is based on a distinction between wrongs that are public and those that are private. As the names suggests, only public wrongs are suitably criminalized; private wrongs are not the fit subject for criminalization, although other forms of social or even legal sanctioning may be appropriate. In Marshall, Duff’s, and Husak’s hands, the distinction between public and private at least initially has to do with the standing of the state to complain of violations, as opposed to the exclusive standing of the individuals wronged to complain.49 That might seem to suggest that Feinberg’s non-grievance morality is the proper basis for criminal legislation—for there, only the state could complain—but that is not their intent. Non-grievance wrongs (what Husak calls ‘wrongful conduct that does not wrong an identifiable victim’50) are already off the table. Their public/ private distinction is within grievance wrongs, separating such wrongs into two camps, the private wrongs where only the victim wronged has standing to complain, and public wrongs, where the state (also or exclusively) has such standing. 42 Feinberg, Harmless Wrongdoing, 18–19. 43  Whether all moral duties must be relational—owed to someone—is an old debate in ethics. Compare G. E. Moore, Principia Ethica (Cambridge: Cambridge University Press, 1903), s. 50, with Henry Sidgwick, The Methods of Ethics (London: Macmillan, 1907), Bk I, ch. IX, s. 4. 44  Feinberg’s terminology in Harmless Wrongdoing. 45 Feinberg, Harmless Wrongdoing, 20. 46 Feinberg, Harmless Wrongdoing, 20, 67, 79–80, 154, 164. 47  S. E. Marshall and R. A. Duff, ‘Criminalization and Sharing Wrongs’, Canadian Journal of Law and Jurisprudence, 11 (1998), 7–22; Husak, Overcriminalization. 48 Husak, Overcriminalization, 135–9. 49 Husak, Overcriminalization, 136. 50 Husak, Overcriminalization, 136 n. 57.

Liberty’s Constraints on What Should be Made Criminal

199

Needed of course is some basis for saying when the state does or does not have standing to complain. Unhelpfully we are only told that ‘some wrongs done to individual members of a community should also be understood as wrongs done to the community itself, a community defined by shared values and interests’.51 These public wrongs are thus done not only to the individuals wronged, but also are wrongs done ‘to the shared values and interests of communities’.52 I fail to see any criterion here for the public/private distinction, a distinction that is supposed to mark the proper limits of criminal legislation. Communities can and do share values and interests about all kinds of human behaviour, incest between adults as much as murder, picking one’s nose in public as much as theft, thinking lustful thoughts as much as fraud. Communities also can and do regard these shared values as constitutive of their being moral communities. If there is a line here, I don’t see it. Moreover, recall that in an earlier generation Lord Devlin too urged that the state could protect what he called a ‘public morality’.53 Upon examination such ‘public morality’ was no more than the most fervent beliefs of a people, marked by the intensity of their disgust and not by any criterion about the content of their moral beliefs. The Marshall/Duff/Husak line is in danger of returning us to the conventionalist and relativist ethics that made Devlin’s brand of legal moralism so distasteful. No self-respecting retributivist should find there to be any justice achieved by prohibiting and then punishing behaviours that are only conventionally regarded as wrong, even when those conventional moral beliefs are so deeply held that they are (and are conventionally regarded as being) essential to the society’s moral sense of itself. Retributive justice is achieved only when actual wrongs are prohibited and punished, whatever the sociology might be of the particular society in which such wrongs are done. I do better with Doug Husak’s examples than with his criterion. He tells us that many torts (the non-intentional ones at least) and all breaches of promise should not be criminalized.54 Violence against others, economic frauds, and failure to pay taxes may be criminalized.55 Whatever the criterion of public versus private might be, Husak is surely persuasive in classifying the latter as public and the former as private. Yet this reveals something I have long thought to be true, which is that there is only a quantitative distinction here, not a qualitative one. Start with Joel Feinberg’s ‘grudging concession to moralism’,56 which is to concede that ‘the prevention of an evil, any kind of evil at all, is a reason for criminal legislation . . .’.57 About this, surely Feinberg is right. The world is a better place if any evil is not brought into existence and any wrong is not done. Add the retributivist’s distinctive insight that punishing the wrongful causing of an evil, any kind of evil and any kind of wrongful 51 Husak, Overcriminalization, 136. 52 Husak, Overcriminalization, 137. 53  Patrick Devlin, The Enforcement of Morality (Oxford: Oxford University Press, 1965). 54 Husak, Overcriminalization, 137. 55 Husak, Overcriminalization, 138–9. 56 Feinberg, Harmless Wrongdoing, 20. 57 Feinberg, Harmless Wrongdoing, 20.

200

Michael S. Moore

causing at all, is also a reason for criminal legislation, and the prospects for making any qualitative distinction here—Feinberg’s as well as Marshall’s, Duff’s, and Husak’s—become dim. The correct line is quantitative: the goodness of preventing and punishing minor moral wrongs is . . . well, minor, small potatoes, morally speaking. Breaches of promise and negligent torts are, by my light, only minor moral wrongs (despite the ridiculous emphasis on promise keeping in many traditional ethics texts).58 As Feinberg pointed out, the minor goodness of such prohibition may at least sometimes be outweighed by the standing case for not prohibiting behaviour, i.e. the standing case for liberty. It is this possibility which I next shall explore.

B.  Minor wrongs and the costs of prohibiting and punishing them Being unable to make sense of any public/private wrong distinction here, and being unable to make moral sense of any qualitative distinction here, grievance/ non-grievance included, I seek to accommodate the common intuition that not all moral wrongs should be criminalized in a different way. Suppose we substitute a quantitative distinction for the qualitative distinctions just explored. With minor moral wrongs, the retributive justice achieved by punishing them is correspondingly small. A minor wrong unpunished is less of an offence to retributive justice than a major wrong unpunished. This raises the possibility that the small good of punishing minor wrongs could be outweighed by other goods achieved by not punishing such minor wrongs. We already have a list of the latter kinds of goods in the five values standing behind the presumption in favour of liberty, values that give that presumption moral weight. Each of these values argues against criminalization, and if the only good of criminalization is the minor retributive justice achieved by the punishment of minor wrongs, surely sometimes such values win out and forbid criminalization. There are two ways this could work. The simplest way would be for the five values standing behind the presumption of liberty to speak uniformly (across all kinds of actions) in the weight they give in favour of not criminalizing. Then the line between wrongs that may be punished by criminal law and those that may not would be drawn simply by the degree of wrongdoing instantiated by different kinds of action. The presumption in favour of liberty, in such a case, would operate like a fixed threshold of wrongfulness below which liberty is to be found. To know whether the state could criminalize any action, one would only need to know the degree of wrong such kind of action typically instantiates. This invariance in the strength of the presumption of liberty was Feinberg’s seeming view of the matter, reflected in his judgment that the value of liberty always outweighs the goodness achieved by punishing and preventing the minor moral wrongs constituted by non-grievance wrongdoing.

58  Exploding the conventional pieties about promise keeping seems to be the major point in Heidi Hurd’s ‘Promises, Schomises’, forthcoming.

Liberty’s Constraints on What Should be Made Criminal

201

The more intriguing but more complicated possibility is that the weight of the values standing behind the presumption in favour of liberty vary depending on the kind of action involved. Then the threshold of wrongfulness (below which liberty is to be found) will vary, depending on the kind of action and how strongly the presumption-of-liberty values speak against criminalization of that kind of action. I shall examine this latter possibility with regard to each of the five values earlier surveyed.

1.  Positive liberty Do some criminal law rules take more (positive) liberty than others? That is, do some coercive restraints diminish the opportunity sets of citizens more than others? Even if we restrict our gaze to negative prohibitions, surely this is so. Prohibiting citizens from exiting their homes except for a few hours every day (curfews, martial law lockdowns) diminishes opportunities more than prohibiting rape, arson, etc. If we enlarge our gaze to include positive requirements in addition to negative prohibitions, further differentiation appears. It is often and correctly said that one positive requirement at a given point in time diminishes one’s opportunity set as much as many negative prohibitions.59 The idea being that multi-tasking is usually not possible, so that having to do action A now precludes an actor from now doing acts B to Z. If all an actor could do at a given time is acts A or B or . . . Z, then requiring A is equivalent to prohibiting B, C, D . . . Z. Our positive duties in morality are typically less stringent than their negative counterparts.60 Our positive duty to prevent the death of strangers, for example, is less stringent than our negative duty not to kill strangers. The wrong of violating such positive duties is thus considerably less than the wrong of violating the corresponding negative duties. This lesser wrong done in violating positive duties, together with the greater (positive) liberty taken to prevent such wrong, give some credence to the libertarian insistence that criminal law should by and large not criminalize positive moral duties. Positive duties inter se also differ considerably in the positive liberty that their enforcement consumes. My opportunity sets are more diminished by imposing a legal duty to care for a child through maturity than by imposing a legal duty of rescuing strangers if the isolated and momentary occasion arises.

2.  Millian autonomy The goodness of Millian autonomy will by itself protect wrongful actions from criminalization whenever it is more important that we choose freely than that we 59  By me, among others. Michael S. Moore, Act and Crime: The Implications of the Philosophy of Action for the Criminal Law (Oxford: Clarendon Press, 1993), 87; Moore, Placing Blame, 277–83. 60 Moore, Act and Crime, 58; Michael S. Moore, Causation and Responsibility: An Essay in Law, Morals and Metaphysics (Oxford: Oxford University Press, 2009), 55–9.

202

Michael S. Moore

choose rightly. The importance of our choosing rightly of course varies with the kind of choice involved and with the stringency of the obligation determining the rightness of that choice. Does the importance of choosing freely also vary with the kind of choice involved? I shall argue shortly that even with respect to some serious moral wrongs, the goodness of free choice swamps the goodness of right choice. But even restricted to minor moral wrongs, surely there is some variation in the importance of protecting free choice by not coercing the right choice. It seems less important that we be free to choose to refrain from participating in exploitative, commercial sex, for example, than it is that we be free to choose whether to fulfil our filial duties to visit our ageing parents.

3.  Kantian autonomy The goodness of Kantian autonomy will by itself protect wrongful actions from criminalization wherever it is more important that an act be done for the right reason than that it be done rightly (even for the wrong reason, say, coercion). The same will be true of our refraining: whenever it is more important that we refrain from doing a certain action for the right reason, than that we refrain, period, for any reason, Kantian autonomy by itself protects our choice from state coercion. Sometimes these balances sound plausible, and sometimes they do not, even when we confine our focus to merely minor moral wrongs. If there are duties of beneficence, as Kant believed, Kant sounds persuasive in urging that the state stay its (coercive) hand in requiring the fulfilment of these duties; for the danger that compliance with coercion will supplant genuine beneficence as the motive for giving seems both real and morally undesirable. Whereas our duties to keep our promises, where there is no significant reliance by the persons to whom we made them, seem different; while there may be some gain to a society if people kept their promises because of their reverence of the moral law, that they keep their promises for whatever reason seems a more important good so that if the state can achieve that good through its motivationally skewing coercion, it should do so.

4.  Desire for free choice My armchair sense of the pattern of human desire is that people desire to be free to make some choices much more than others. As the school busing disputes in America in the 1970s illustrated, many parents have a strong desire to be free to choose which school to which to send their children (recognizing that some of their choices might violate their moral duty to provide a good education for their children, or their duty to provide equal educational opportunity to other people’s children). Whereas few seem to desire strongly that they be free of state coercion in whether to keep their promises, not inconvenience their neighbours with snow-covered sidewalks, etc.

Liberty’s Constraints on What Should be Made Criminal

203

5.  The evils of un- or under-enforced laws As we saw from the literature of the 1960s on over-criminalization,61 criminal laws that are rarely or never enforced impose costs on any legal system. These are principally the costs of a lessened respect for law, of the potential for selective, arbitrary, and unequal enforcement, and of the potential for extortion from officials with discretion whether to enforce such laws. The degree of under-enforcement (and thus the degree of these evils being realized) surely varies with the kinds of actions prohibited. When criminal law is used to prohibit actions that are: (a) widespread and ubiquitous; (b) strongly motivated so that such actions will continue to be done even if prohibited; (c) unclear in a populace’s mind as to their moral status; (d) clear but divided in the populace’s mind about their moral status; or (e) clearly wrong in a populace’s mind but not popularly thought properly punishable by the criminal law (so that criminal prohibition is just ‘for show’)—when criminal law prohibits these kinds of actions, it invites under-enforcement and its attendant evils. Less so for actions without these characteristics.

6.  Literal enforcement costs While legislative prohibition of actions may only require minimal state resources, actual enforcement and punishment of violations of those prohibitions requires more. These enforcement/punishment costs include not just dollars and the opportunity costs such dollars represent, but also privacy costs. These will predictably vary somewhat across different kinds of actions. Actions that are typically done with no non-participant witnesses—victimless crimes, crimes done in secrecy—will incur greater such enforcement costs. Sexual behaviours form prime examples, of course, if any of those actions were to be regarded as even somewhat immoral to start with.

7. Enhanced profitability to those willing to engage in crime (the ‘crime tariff’) It is a well-known fact that criminalizing certain behaviours will not stamp them out but only make it more profitable to suppliers of the goods and services making such behaviours possible. Laws against prostitution and against recreational drug use are standard examples of this. Such artificial restriction of supply to those willing to break the law of course creates something quite bad, namely, well-financed criminal organizations. Such organizations can in turn cause all sorts of bad things—witness Columbia’s demand on the United States some years ago for reparations for the destruction of Columbia’s government, caused, it was said, by America’s criminalization of recreational drug use. 61  See n. 1.

204

Michael S. Moore

The behaviours whose criminalization will produce these ill effects are behaviours for which there is a relatively inelastic demand curve amongst the populace: they are strongly motivated toward the behaviours, so that even steep increases in the cost of engaging in the behaviour have little effect on the amount of it that goes on. Some behaviours exhibit this motivational salience—drug usage and prostitution again—but others do not. Again, the evil here—monopoly prices funding organized crime—varies enormously by the kinds of actions prohibited by criminal law. We end up, then, with a complicated calculus about the liberty to do minor wrongs. To be protected from criminalization on this basis, an action must be a sufficiently minor wrong that the good of its punishment and prevention is outweighed by the costs the state incurs in achieving such punishment and prevention. What makes the calculation so complicated is that the characteristics of actions making them costly to criminalize are so various: degree of opportunity set diminishment, omissions rather than actions, importance of Millian autonomy, importance of Kantian autonomy, degree of desire to be free of state coercion, degree of division or of clarity in popular opinion about moral status, commonality of the behaviour, depth of motivation, secrecy in execution, all have a place in this calculation. Still, the barrier made up of these many ingredients is a plausible interpretation of the liberty each of us has to do some minor moral wrongs free of state coercion.

C. Serious moral wrongs and the costs to moral agency in prohibiting them From the previous section’s discussion we have seen that our liberty does not end where our obligations begin. We are at liberty to be free of state coercion even for some actions that constitute minor moral wrongs. Even a relatively weak presumption of liberty gives us this much. Yet it is intuitive that the presumption operates more strongly than this, protecting even behaviours that are seriously wrong from being criminalized. Several examples support this intuition.62 One is the woman’s liberty to have an abortion. One might think that women do serious wrong in aborting their foetuses. The extreme view is that abortion is murder because the foetus is a person. But less extreme views will also illustrate the point: even if foetuses are not persons, they could still be regarded as serious moral interests that it is seriously wrong to harm. Despite such judgments about the wrongness of abortion, one might nonetheless think that women have the right to do this great wrong. The thought would be that it is an even worse wrong for the state to coerce the woman’s choice here than is any wrong that choice could do. Another example is suicide. It is plausible to think that when children or others depend on the actor, it is a serious wrong to suicide. Yet even accepting this, I find equally plausible the view that each of us has the right to do this wrongful act. Our

62  These are culled from Moore, Placing Blame, 763–4.

Liberty’s Constraints on What Should be Made Criminal

205

life is our own, and it would be a greater wrong for the state to dictate its ending than it is to end life wrongfully. A third example is liberty of speech. The child’s slogan, ‘sticks and stones may break your bones but words can never harm you,’ is surely false. Acts of speech can do great harm, as when one reveals private and embarrassing facts about another whose precariously reconstructed life is then destroyed. Such acts, when done for no better reason than to sell magazines, strike me as seriously wrongful; yet one might think it would be worse for the state to compel these wrongful actors not to speak. Lastly, consider parental decisions about how to raise and educate their own children. Bad parenting (well short of physical abuse) can cripple children for life, and is a serious wrong. Yet many believe parents are within their rights to be harmful in myriad and unsubtle ways, in the sense that it would be worse for the state to coerce better parenting. These examples all suggest that liberty can extend to some choices no matter how seriously immoral may be the wrongful exercise of that choice. Here too it is more important that one choose freely than that one choose rightly, even when ‘choosing rightly’ is a high stakes matter, morally speaking. The key to understanding the nature of those choices where freedom is crucial, lies in Mill’s idea of autonomy. Recall that Mill likened human beings who were not self-determining in some sense to mere machines like steam engines. Choices of such beings lack the value of the choices of free peoples. What Mill saw is that uncoerced choice at key points is necessary for any of us to develop into beings whose subsequent choices can have value. Aristotle said that every choice determines who we are. Perhaps, but surely not all choices are created equal in this respect. My choice for an extra dessert no doubt contributes its little bit to the constitution of my gluttonous character. Yet it compares poorly in the degree of its contribution to that character, with the choice about what skills and talents I should develop, what education I should seek, what career I should undertake, what forms of friendship, love, and sexual intimacy I should indulge, whether I should marry, who and when I should marry, whether I should have children, how my body is treated while I am alive, what is done with my body after I am dead, and how and when I die. A creature denied all of these latter kinds of choices could rightly complain that he never got to live his own life, never got to choose what kind of person he would be, never got to be the artist whose best creation was himself. How forming of self different kinds of choices are is surely a matter of degree. That this is so does not mean that the latter kinds of choices are not on the upper end on the scale of importance that they not be coerced—they are. It is so important that choices on this upper end of the scale not be coerced, that even serious immorality should be risked to allow such choices to be made freely. My own sense is that this last right to be free of state coercion is continuous with the right to do minor wrongs discussed in the last section. Not only is there no sharp line, there is not even a lumpiness in slope that would justify separating these as two rights to liberty. There is only one right to do the wrong thing, with the

206

Michael S. Moore

content of the right varying by the degree of immorality of wrongful exercises of the choice protected by the right, and varying by the degree of Millian autonomy, positive liberty, Kantian autonomy, etc., taken by coercion of that choice.

VII.  Constitutionalizing the Rights to Liberty The right way of looking at what has just been done is as part of a theory of legislative role in a democracy. Just as a theory of adjudication can be seen as being a theory of the judicial role, so can a theory of legislation be seen as a theory of the legislative rule. As such it is a theory whose audience consists of individual legislators (and their academic critics) thinking through how they ought to do their job. As a political and moral theory, nothing has been said about remedies for violations of the strictures of such a theory. (One remedy, of course, would be to vote out of office any legislator overstepping his or her role, as here defined.) But my concern hitherto has been confined to describing the moral constraints on criminalizing behaviour, leaving aside as we generally do in ethics what remedies there might be against legislators who overstep their roles. In this final section, however, I want to consider legal restraints on legislative action, building on the moral constraints above described. In a constitutional democracy with judicial review as in America, this translates into the question of constitutionalizing the rights to liberty described above. I shall approach this legal question in two steps. First, how protected are these rights to liberty in already existing US constitutional law? And second, how ought that law to be changed to protect such rights better than they are already protected?

A.  American constitutional law on liberty America’s is a written constitution, so most constitutional rights are thought to require a textual location. With enumerated rights such as those contained in the first eight amendments to the Constitution, such textual location is unproblematic. But general rights to liberty are nowhere enumerated in the US Constitution as such, so the question of their textual basis is quite contentious in American constitutional law. The problem is exacerbated by three peculiarities of American constitutional history. The first is the emasculation of the natural textual home for these rights, the federal privileges and immunities clause of the Fourteenth Amendment. The US Supreme Court’s five–four decision in the Slaughter House Cases of 1873 limited those privileges and immunities to those that were essential to national citizenship in a federal union, such as the right to travel interstate.63 It must have shocked the drafters of that clause to learn that the privileges and immunities of national citizenship was not the great charter of unenumerated rights and liberty that they 63  The Slaughter-House Cases, 83 US (16 Wall.) 36 (1873).

Liberty’s Constraints on What Should be Made Criminal

207

had intended. The result, from 1873 to today, is that liberty is protected under the due process clauses of the Fifth (versus the federal government) and Fourteenth (versus state government) Amendments. And it is admittedly a bit of a linguistic stretch to take the operative language of those amendments—‘no person shall be deprived of life, liberty, or property except by due process of law’—to protect substantive rights to liberty. Justice Scalia is but the most recent of those expressing scepticism about ‘substantive due process’: [This statute] undoubtedly imposes constraints on liberty. So do laws prohibiting prostitution, recreational use of heroin, and for that matter, working more than 60 hours per week in a bakery. But there is no right to ‘liberty’ under the Due Process Clause, though today’s opinion repeatedly makes that claim. The Fourteenth Amendment expressly allows States to deprive their citizens of ‘liberty,’ so long as ‘due process of law’ is provided.64

The second historical contingency is the substitution of ‘privacy’ for ‘liberty’ as the name of the right protected in some of the US Supreme Court’s most recent and best-known liberty-protecting opinions.65 It is understandable why this was done—it was the court’s attempt to distance itself from the repudiated Lochner era protection of economic liberties. But the result is a label, ‘privacy’, that poorly fits the true nature of the rights protected.66 Furthermore, it reinforced the idea of such a right has no home in the text of the US Constitution: The Court talks about a constitutional ‘right of privacy’ as though there is some constitutional provision or provisions forbidding any law ever to be passed which might abridge the ‘privacy’ of individuals. But there is not . . . . For these reasons I get nowhere in this case by talk about a constitutional ‘right of privacy’ as an emanation from one or more constitutional provisions. I like my privacy as well as the next one, but I am nevertheless compelled to admit that government has a right to invade it unless prohibited by some specific constitutional provision.67

Thirdly, the textual basis of the right to liberty was not aided by Justice Douglas’s wild and unseemly scramble for any textual pigeonhole in a storm in his opinion in Griswold, the leading modern case on liberty. Douglas wrote that the right to liberty (‘privacy’) was to be found in the ‘penumbras formed by emanations’ by no less than five amendments, not including the Fourteenth. As much fun as it is to bamboozle constitutional law students in America with these three historical anomalies, in truth the rights to liberty have a good textual home in the language and the intent behind the Constitution, specifically the privileges and immunities clause of the Fourteenth Amendment. Debate about this is in any case only of parochial interest to American constitutional historians. 64  Lawrence v Texas, 539 US 558, 592 (2002) (Scalia, J., dissenting). 65 Notably Griswold v Connecticut, 381 US 479 (1965). This miscategorization has forced later members of the court to distinguish ‘decisional privacy’ (better known as ‘liberty’) from ‘area privacy’, as in Blackman’s opinion in Bowers v Hardwick, 478 US 186, 203–5 (1986). 66  True privacy is better captured as lack of information about one, lack of attention to one, and lack of access to direct attention to one. Ruth Gavison, ‘Privacy and the Limits of Law’, Yale Law Journal, 89 (1980), 421–71. 67  Griswold v Connecticut, 381 US 479 (1965) (Black, J., dissenting).

208

Michael S. Moore

More interesting is the extent to which the structure of the rights to liberty as I have described it is to be found in the decisional law of American courts, particularly that of the US Supreme Court. Consider first the derived right to liberty, the right to be free of improperly motivated coercive legislation. Justice Peckham’s opinion for the court in Lochner seems to constitutionalize some such right rather openly.68 Peckham nicely distinguishes justifying reasons from motivating reasons, then opines that the proper enquiry is into the actual reasons that motivated the legislature. The state might later claim that the 60 hour per week limit on the hours bakers could work was justified by the health of workers and/or by the wholesomeness of the bread that non-tired bakers can bake. Yet such justifying reasons Peckham rejects as mere pretexts and evasions: When assertions such as we have adverted to become necessary in order to give, if possible, a plausible foundation for the contention that the law is a ‘health law,’ it gives rise to at least a suspicion that there was some other motive dominating the legislature than the purpose to subserve the public health or welfare. . . . It is impossible for us to shut our eyes to the fact that many of the laws of this character, while passed under what is claimed to be the police power for the purpose of protecting the public health or welfare, are, in reality, passed from other motives . . . .The purpose of a statute must be determined from the natural and legal effect of the language employed . . . and not from [its] proclaimed purpose.69

Peckham then divines a paternalistic purpose to New York’s hours limitation, and sees none of the characteristics of disabled persons that could justify such paternalistic state concern: There is no reasonable ground for interfering with the liberty of person or the right of free contract, by determining the hours of labor, on the occupation of a baker. There is no contention that bakers as a class are not equal in intelligence and capacity to men in other trades or manual occupations, or that they are not able to assert their rights and care for themselves without the protecting arm of the state, interfering with their independence of judgment and of action. They are in no sense wards of the state.70

It appears that like Mill before him Peckham found paternalistically motivated coercive legislation to be one of the main enemies of liberty. Unlike Mill, but like my imagined retributionist-legal moralist legislator, Peckham elsewhere approves of coercive legislation motivated by the prevention and punishment of moral wrongdoing.71 Some more contemporary members of the court might appear to disagree with this. Justice Blackman, for example, speaking for the court in Roe v Wade, shunted aside any argument that the abortion prohibition could be justified as a means to discourage immoral sex, with the comment that ‘no court or commentator has taken the argument seriously . . .’72 More pointedly, Blackman in his dissent in Bowers v Hardwick (which later became the court’s view in Lawrence v Texas) rejected the idea that ‘the fact that the acts [described in the Georgia anti-sodomy statute] “for hundreds of years, if not

68  Lochner v New York, 198 US 45 (1905). 70  198 US at 57. 71  198 US at 53.

69  198 US at 62–4. 72  Roe v Wade, 410 US 113, 148 (1973).

Liberty’s Constraints on What Should be Made Criminal

209

thousands, have been uniformly condemned as immoral” is a sufficient reason to permit a state to ban them today . . .’73 Religious morality particularly came in for condemnation as an acceptable motive for coercive legislation: The legitimacy of secular legislation depends instead on whether the State can advance some justification for its law beyond its conformity to religious doctrine. . . . A State can no more punish private behavior because of religious intolerance than it can punish such behavior because of racial animus.74

Justice Stevens too, speaking eventually for a majority of the court, held that ‘the fact that the governing majority in a State has traditionally viewed a particular practice as immoral is not a sufficient reason for upholding a law prohibiting the practice . . .’75 Notice that like Mill before them, Justices Blackman, Stevens, and Kennedy need not be construed to be directing their fire at legal moralist motivations for criminal legislation as such; rather, like Mill, their complaint is with the con­ ventionalist moral beliefs of the majority being treated as proper reasons for coercive legislation. Indeed, likening religious doctrine to racial animus might well be thought to express the view that on the matters before the court the moral views being rejected are not just conventional but are also incorrect. Like Mill these justices’ rejection of these moral views would then be based on the substantive moral judgment that the popular, traditional moral judgments are in error. Understandably judges have a hard time saying this openly; but it is more honest, is it not, to see that their real beef is not with the kind of reason proferred by the state for coercion (to enforce morality); it is rather with the correctness of the morality the state would enforce? Not all of the current court’s justices, of course, share these views of Justices Stevens, Kennedy, and Blackman about the justifiability of criminalizing behaviour for no other reason than that it is immoral. Justice Scalia, for example. In Lawrence, Scalia conceded that Texas in its anti-gay sex laws was trying ‘to further the belief of its citizens that certain forms of sexual behaviour are “immoral and unacceptable,” Bowers—the same interest furthered by criminal laws against fornication, bigamy, adultery, adult incest, bestiality, and obscenity’.76 Such morals-motivated criminal prohibition Scalia thinks to be entirely proper. Scalia could think this either because he thinks that these moral views are correct, making him a legal moralist; more likely it is because Scalia thinks the majority (as reflected in a long-standing moral tradition) has the right to impose its views through coercive legislation, be those views right or wrong. The latter view is neither Millian nor legal moralist in its limits on the proper reasons that may motivate state coercion; it is a return to a kind of J. F. Stephen majoritarianism, with Scalia’s long-standing tradition representing for him the long-term majority will. 73  Bowers v Hardwick, 478 US 186, 210 (1986) (Blackman, J., dissenting). 74  Bowers v Hardwick, 478 US 186, 210 (1986) (Blackman, J., dissenting). 75  Bowers v Hardwick, 478 US 186, 216 (1986) (Stevens, J., dissenting). 76  Lawrence v Texas, 539 US 558, 599 (2002) (Scalia, J., dissenting).

210

Michael S. Moore

Some aspects of the derived right to liberty are thus part of American constitutional law, although the content of that right is very much up for debate between Millians like Stevens, Blackman, and Kennedy, legal moralists like Peckham, and social conservatives/majoritarians like Burger and Scalia. The same cannot be said about the substantive right to be free of state coercion with respect to acts that are in fact neither harmful to others nor wrongful, the second right to liberty I distinguished earlier. The sad truth is that the American Supreme Court has never come close to adopting the harm principle, the moral wrong principle, or the wrongful harm principle, as marking the boundary of justifiable state coercion. It has not even had the issue raised explicitly, as it was before the Canadian Supreme Court in 2003.77 Dennis Baker purports to find some support for the harm principle in the US Supreme Court’s proportionality decisions under the Eighth Amendment,78 but I fear that that is mostly creative reading and wishful thinking. With regard to the right to be free of state coercion with respect to minor moral wrongs, and the right to make even seriously immoral decisions for oneself if they are decisions constitutive of self, American constitutional law has drawn some peculiarly rigid lines. This is the well-known system of two-tiered review, one tier for ordinary liberty interests and the other for ‘fundamental’ liberty interests. Ordinary liberty interests receive only rational basis review, asking only if the state’s coercion has some conceivable justification. Fundamental liberty interests, by contrast, can only be coerced if the state has a compelling state interest, with narrowly tailored means, for doing so. The court has protected as fundamental liberty interests: the choice whether and whom to marry,79 the choice of whether to become a parent (both at the time of conception by use of birth control devices,80 and after by use of abortion),81 the right to educate one’s children in private schools of one’s choice82 and in languages of one’s choice,83 the choice of medical treatments,84 the decision whether to allow oneself to die (but not the decision to kill oneself, either alone or with physician assistance).85 State laws coercing these choices for anything less than the most compelling of reasons are unconstitutional, even recognizing that such choices may be made in immoral ways. By contrast, one’s choice about medical use of marijuana, one’s choice to work at grinding lenses and manufacturing eye glass frames,86 and about all other choices, are ‘non-fundamental’ and can be coerced by the state for very little reason.

77  Regina v Malmo-Levine [2003] SCC 74. 78 Baker, The Right not to be Criminalized, 85–8. 79  Loving v Virginia, 388 US 1 (1967); Zablockie v Redhail, 434 US 374, 391 (1978) (Stewart, J., concurring). 80  Griswold v Connecticut, 381 US 479 (1965). 81  Roe v Wade, 410 US 113 (1973); Planned Parenthood of S.E. Pennsylvania v Casey, 505 US 833 (1992). 82  Pierce v Society of Sisters, 268 US 510 (1925). 83  Meyer v Nebraska, 262 US 390 (1923). 84  Cruzan v Director, Missouri Dept. of Health, 497 US 261 (1990). 85  Washington v Glucksberg, 521 US 702 (1997). 86  Williamson v Lee Optical, 348 US 483 (1955), the court’s leading rational basis review standard for ordinary liberties.

Liberty’s Constraints on What Should be Made Criminal

211

The court has never agreed on any criterion as to what marks a liberty interest as fundamental. Liberals such as Blackman, Stevens, and Kennedy adopt some kind of self-defining choice criterion.87 Conservatives such as Rehnquist resign themselves to the fact that ‘the outlines of the ‘liberty’ specifically protected by the Fourteenth Amendment’ have ‘never been fully clarified’ and is ‘perhaps not capable of being fully clarified’; such conservatives content themselves with a list of ‘concrete examples involving fundamental rights found to be deeply rooted in our legal tradition.’88 In other words, a list of precedents with no common thread or rationale as to why they were correctly decided, and thus no criterion for extending them, except for the conventionalist thought that all such rights have traditionally been regarded as important.

B.  Changes in US constitutional law so as better to protect liberty Whether US constitutional law should be changed so as to better protect liberty depends in part on some larger issues that we cannot hope to resolve here. These are the issues about whether judicial review with respect to the rights-conferring clauses of the US Constitution better protects rights than does unreviewed and unreviewable legislative determination of the rights of citizens. My own view, defended elsewhere, is that there is a weak but tenable case for the idea that societies get better rights protection with judicial review under a written constitution than without it.89 If that is so generally, it should be so with respect to the liberty rights described herein. Needed then would be judicially enforced liberty rights that hew closely to the moral rights to liberty described earlier. Specifically: 1. The derived right to liberty should be explicitly recognized as a separate liberty right. 2. The content of that right should be based on either the wrong or the wrongful harm principles, principles that condemn much paternalistically and offence-motivated coercive legislation and that requires that the legislature aim to prevent wrongful harms and to punish wrongs in its decisions as to what to criminalize. 3. The substantive right closely related to the derived right should also be explicitly and separately recognized. 4. Its content should also be in terms of the wrong and wrongful harm principles, so that behaviours that in reality are not wrongful may not be criminalized even when a legislature thinks that they are wrongful and seeks to stamp them out for that admittedly proper reason.

87  Bowers v Hardwick, 410 US 186, 205–6, 217–18 (Blackmon, J., and Stevens, J., dissenting). 88  Washington v Glucksberg, 521 US 702, 722 (1997). 89 Michael S. Moore, ‘Justifying the Natural Law Theory of Constitutional Interpretation’, Fordham Law Review, 69 (2001), 2087–117. This writes up my half of a formal debate with Jeremy Waldron on the Rights-Based Justification of Judicial Review, held at the University of Pennsylvania, 1995. For Waldron’s side of that debate, see Jeremy Waldron, Law and Disagreement (Oxford: Oxford University Press, 1999).

212

Michael S. Moore

5. The right to do some morally wrongful actions should be explicitly recognized. 6. Such a right should not be given two radically different tiers of protection; rather, the degree of liberty-worthiness of actions should be seen as the smooth continuum that it is. This should result in more protection than is currently afforded for ordinary liberties, and about as much protection as is currently afforded for what are now called fundamental liberties.90 7. The criteria for determining where on the scale of liberty-worthiness an action falls should be made explicit. This includes the criterion for what is high on that scale, the self-defining choices now separately protected as ‘fundamental liberties’. Given the US Supreme Court’s reluctance since the New Deal to develop ambitious constitutional theories of liberty such as this one, it is best not to be too optimistic in waiting to see any of these changes taking place in the United States itself. Perhaps elsewhere?

90  In defence of a constitutional presumption in favour of liberty with some teeth, see Randy Barnett, ‘Scrutiny Land’, Michigan Law Review, 106 (2008), 1479–500; Barnett, Restoring the Lost Constitution: The Presumption of Liberty (Princeton: Princeton University Press, 2004).

8 Polygamy A Novel Test for a Theory of Criminalization Douglas Husak*

I.  How to Test a Theory of Criminalization We philosophers earn our reputations by providing abstract arguments for our theories. Sometimes we become so obsessed with these arguments that we forget why we bother to produce theories in the first place. A theory has little value if it only tells us what we already know. The real test of a theory consists in its potential to shed light on issues about which we are uncertain. Theories of criminalization are no exception. My own theory of criminalization consists in a number of constraints that a purported crime must satisfy before it should be included within a penal code.1 I supported each of these constraints with one or more arguments. Needless to say, not all philosophers who have responded to my efforts find each of these arguments to be persuasive—even when they agree with the constraint the argument was designed to support.2 Alternatively, they contest the necessity of a supposed constraint by proposing a clever counter-example—a case in which I should agree that criminalization is warranted even though a constraint is violated. At this time, however, I propose to move the debate beyond these familiar tactics. A different strategy for evaluating a theory is to examine its application to particular cases. Any respectable theory of criminalization can be expected to assure us that murder and rape should be prohibited and that acts of generosity should be permitted. But the crucial test of a theory lies in its ability to offer insights on matters about which we are undecided. Polygamy is one such topic.3 In this chapter I propose to apply my set of constraints to the question of whether the state should punish polygamy. Let me be *  Generous support for this chapter was provided by a Fellowship from the Straus Institute for the Advanced Study of Law and Justice, New York University. Thanks to Kenneth Simons for some helpful suggestions. 1  Douglas Husak, Overcriminalization (Oxford: Oxford University Press, 2008). 2  See e.g. Gideon Yaffe, ‘Harmfulness, Wrongfulness, Lesser Evils and Risk-Creation: A Comment on Douglas Husak’s Overcriminalization’, Jerusalem Review of Legal Studies, 1 (2010), 35. 3  At least three other examples might have been chosen. For instance, we may be unsure whether the criminal law should punish persons who know they are infected with HIV but have sexual relations

214

Douglas Husak

clear about how this particular exercise amounts to a test of my theory. I do not believe that it qualifies as a test because it reaches a judgment that conforms to our intuitions. Admittedly, this way of testing theories can be quite useful. Still, this methodology has limited value. It assumes that we already know or at least have strong antecedent beliefs about whether a proscription is legitimate. One of the reasons I choose my particular example is that I doubt that the informed citizenry or audience of legal philosophers has especially strong intuitions about polygamy or about the justifiability of banning it. Perhaps committed liberals will reflexively adopt a laissez-faire attitude about the family arrangements the criminal law should allow. Unless I misread public and academic opinion, however, I detect no groundswell of support to decriminalize polygamy—even among liberals. Still, I may be mistaken about the views of my colleagues and fellow citizens. At least I can attest that I approached this exercise without any strong intuitions about the legitimacy of penal sanctions against polygamists. If we do not have firm antecedent beliefs about the outcome we should accept, how can we decide whether the answer a theory offers is correct? The short reply is that we cannot. For present purposes, the test of a theory of criminalization I envision does not consist in its ability to generate a clear solution. Instead, the nature of this test is less direct. Like any other statute, a criminal ban on polygamy is justified only if it satisfies the several constraints that comprise my theory. At the end of the day, I strongly doubt that the offence in question does satisfy each of these constraints. I do not try to conceal my scepticism. Still, my theory is corroborated if it asks the right questions in deciding whether polygamy statutes are legitimate. My theory may not always provide straightforward answers to these questions, but it must at least pose the issues that we agree must be addressed if an answer is to be reached. Thus my theory identifies the kinds of positions one would have to defend in order to support punitive sanctions. No one—and certainly not a philosopher—should purport to be too confident about how to resolve the complex issues on which an outcome depends. Examining particular examples from a theoretical perspective is equally valuable for a different reason: it helps us to identify the issues that are not important from the standpoint of criminalization. The voluminous scholarly literature about polygamy raises a number of concerns that I do not believe should be given any weight in decisions about punishment. It is imperative to avoid misunderstanding on this point. To say that the concerns I will mention should have no weight in the criminalization decision is not to say that these concerns are irrelevant for any and all purposes. Nor is it to say that these concerns are irrelevant for any and all normative purposes. In fact, I believe that many of the moral concerns that have loomed large in existing scholarship about polygamy are absolutely crucial for such with partners to whom they fail to disclose their status. For a thoughtful approach, see Margo Kaplan, ‘Restoring Reason to HIV-Exposure Laws’, Indiana Law Review, 87 (2012), 1517. Or we may be uncertain about the criminalization of such practices as incest. See Vera Bergelson, ‘Vice is Nice, but Incest is Best: The Problem of a Moral Taboo’, Criminal Law and Philosophy, 7 (2013), 43. Finally, proscriptions of prostitution involve many considerations parallel to those discussed here. See Peter de Marnefee, Liberalism and Prostitution (New York: Oxford University Press, 2010).

Polygamy

215

matters as (1) how polygamy should be regulated by non-criminal law; (2) whether polygamous relations should be encouraged or discouraged by the state and by private institutions, or recognized at all; (3) what attitudes thoughtful and reflective persons should adopt toward polygamy; and (4) whether a given individual should elect to become polygamous. Clearly, each of these issues is significant, and may even be more significant than the issues I will address here. But these matters need not be important for decisions about criminalization. In other words, not all significant moral considerations about a type of conduct are material to the question of whether to criminalize it. Among the main benefits of a particular theory is its ability to contrast the interesting normative questions about a type of conduct that are material to criminalization from those that are not. We should avoid the view (which I believe to be widespread) that any consideration that bears on the moral status of a type of behaviour must be relevant to the judgment about whether or not to punish it. Legal moralism (in each of its many variations) is perhaps the main rival to my theory of criminalization,4 and I suspect that many philosophers who adhere to this school of thought will disagree on the foregoing point. If principled questions of criminalization are to be resolved solely by settling on the morality of a type of behaviour, it is tempting to believe that any consideration that bears on its moral status must also bear on the determination of whether or not to proscribe it. According to this train of thought, judgments about criminalization should reflect all-things-considered judgments of morality. And all-things-considered judgments, after all, consider all things. My exercise might be construed as (yet another) reason to reject legal moralism as so construed. I will seek to identify several normative questions about polygamy that, however interesting and important they may be for other purposes, are not material to decisions about criminalization according to the theory I defend. Highlighting potential disagreement about the relevance of these considerations is an instructive way to understand and assess competing theories of criminalization. In other words, normative considerations that are material to criminalization decisions on some theories are irrelevant on others. The supposed contrast between public and private wrongs offers an excellent way to illustrate my point. According to my theory, the categorization of conduct as a private wrong entails that its wrongfulness is not pertinent to criminalization at all. Unfortunately, the distinction between public and private wrongs is murky and elusive—a contrast that, as we will see, raises special issues for the criminalization of offences against the family. Of course, any respectable theorist will find the resources to oppose punishing most or all of the wrongs I categorize as private.5 Still, the structure of our opposition to punishment frequently differs.6 We should not require countervailing 4  Legal moralism has many variants. See Michael S. Moore, ‘A Tale of Two Theories’, Criminal Justice Ethics, 28 (2009), 27; and R. A. Duff, ‘Towards a Modest Legal Moralism’, Criminal Law and Philosophy, 8 (2014), 217. 5  See Moore, ‘A Tale of Two Theories’ (n. 4), 27. 6  See Douglas Husak, ‘Convergent Ends, Divergent Means: A Response to my Critics’, Criminal Justice Ethics, 28 (2009), 119.

216

Douglas Husak

reasons not to punish private wrongs in order to defeat the principled case in favour of criminalization. Instead, private wrongs give us no reason to criminalize in the first place; there is nothing to defeat.7 As I will explain, empirical facts about polygamy have a special stature in my theory. A theory of criminalization should tell us which facts matter and why. I believe that empirical facts about penal sanctions are important largely because institutions of criminal justice are created by the state at enormous cost to taxpayers. Citizens should support the enactment and enforcement of a given law only if it promotes a substantial state interest. In applying this constraint, we need to ascertain not only what substantial state interest (if any) is at stake in prohibiting polygamy, but also whether this prohibition actually furthers that interest. This determination raises empirical controversies which legal philosophers are not especially equipped to resolve. To my mind, it is a merit of my theory that these controversies are relevant to the criminalization decision. I indicated that I will examine polygamy partly because I (and hopefully many other commentators) lack clear intuitions about the justifiability of prohibiting it. An additional reason to focus on polygamy is that criminal theorists have said remarkably little about it.8 They have tended to concentrate almost entirely on core crimes where a theory is less likely to be useful. In this domain, a theory merely reinforces what we already know. When criminal theorists have ventured beyond the core, they have not been inclined to stray too far. The extension of a theory of criminalization to conduct pertaining to marriage and the family breaks relatively new ground among many legal philosophers.9 Criminal theorists are only beginning to explore some of these issues. Of course, the very novelty of this enquiry helps to explain why intuitions about polygamy are not more fully developed. Although few philosophers of criminal law in the United States seem to have thought long or hard about this topic, it is not merely of academic interest. In a recent complex decision that ‘weighed heavily on the court for many months’, a federal judge in Utah has ruled unconstitutional some of the key parts of the state bigamy law that criminalizes polygamous cohabitation.10 But the ruling stopped short of legalizing polygamous marriages or invalidating the entirety of Utah’s 7  See R. A. Duff, Answering for Crime (Oxford: Hart Pub. Co., 2007), 140–6; and Duff, ‘Towards a Modest Legal Moralism’ (n. 4). 8  Some scholars would be surprised by my assertion. According to one commentator, ‘debates about decriminalizing same-sex sodomy and legalizing same-sex intimacy have generated a veritable cottage industry on the law’s stance toward polygamy’. Adrienne D. Davis, ‘Regulating Polygamy: Intimacy, Default Rules, and Bargaining for Equality’, Columbia Law Review, 110 (2010), 1955, 1981. But much of the literature in this supposed ‘cottage industry’ focuses on the ‘slippery-slope’, that is, whether decisions in cases such as Lawrence v Texas inevitably lead to the legalization of polygamy. Apart from a handful of scholars who concentrate more narrowly on the family, the best-known legal philosophers who specialize in criminal theory have said almost nothing about polygamy. 9  Some recent interesting theoretical work on polygamy has taken place in the broader context of deciding whether the criminal justice system should place benefits or burdens on individuals depending on their family status. See Dan Markel, Jennifer M. Collins, and Ethan J. Leib, Privilege or Punish: Criminal Justice and the Challenge of Family Ties (New York: Oxford University Press, 2009). Much of the analysis these authors apply to criminal laws that affect families I apply to criminal laws more generally. 10  Brown v Buhman, US Dist Ct., Case No. 2:11-cv-0652-CW (2013).

Polygamy

217

bigamy statute. In an even more detailed opinion, Canadian courts have wrestled with the question of whether §293 of its Criminal Code, its ban on polygamy,11 is consistent with its Charter of Rights and Freedoms.12 The approach Canadian courts take to such issues is governed by the famous Oakes test.13 This test is roughly analogous (although not identical) to that taken by applying several of the constraints in my theory of criminalization.14 Whether or not proscriptions of polygamy raise constitutional questions in the United States, the recent developments in Utah and Canada demonstrate the need to approach this real-world problem by reference to the principles contained in a theory of criminalization. My final point about how to test a theory of criminalization is among the most important, even though it is ignored with surprising frequency. To decide whether a statute passes or fails a test of criminalization obviously requires the existence of a (real or hypothetical) statute. We should not attempt to evaluate whether something we loosely understand as polygamy should be the legitimate target of the penal sanction unless we have a concrete instance of a law before us.15 On many occasions, the devil is in the details. Of course, some normative problems might be overcome by making improvements in the content of the statute with which we begin. Soon I will suggest a few such improvements.16 But serious normative analysis necessarily starts with a specific statute. With what statute should we begin? After all, bans on polygamy have a long history. Parliament decreed in 1604 that a married person who entered into an unlawful marriage with another person committed a felony. The Morrill Act, adopted in 1862, punished polygamy in the Territories of the United States. Since such arrangements remain criminalized in each of the United States, it is hard to know which particular law to select as representative. Here, as in most other contexts, I borrow from the influential Model Penal Code. Among the most 11  Section 293(1) of the Criminal Code, RSC 1985, c. C-46, states: ‘Everyone who (a) practises or enters into or in any manner agrees or consents to practise or enter into (i) any form of polygamy, or (ii) any kind of conjugal union with more than one person at the same time, whether or not it is by law recognized as a binding form of marriage; or (b) celebrates, assists or is a party to a rite, ceremony, contract or consent that purports to sanction a relationship mentioned in subparagraph (a)(i) or (ii), is guilty of an indictable offence and liable to imprisonment for a term not exceeding five years.’ Curiously, polygamy itself is not defined in the Criminal Code. 12 The Canadian Charter of Rights and Freedoms guarantees given rights and freedoms subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. 13  R v Oakes, [1986] 1 S.C.R. 105. As recently summarized in Hutterian Brethren (2009 SCC 37 (2009) 2 S.C.A. 567), the Oakes test requires the purpose for which the limitation of liberty is imposed to be pressing and substantial. If this condition is satisfied, the test further demands proportionality between the means and the goal. This latter condition, in turn, requires the limitation to be rationally connected to its purpose, to minimally impair the Charter right, and to be proportionate in its effect. 14 In Reference re: Section 293 of the Criminal Code of Canada, 2011 BCSC 1588, the highest court of British Columbia held that the ban on polygamy is a proportionate response to preventing the harmful discrimination women suffer in polygamous unions. 15  Polygamy is the generic name for marriages with multiple spouses. It has two variants: Polygyny (one husband with multiple wives) is far more common than polyandry (one wife with multiple husbands). It is hard to see why a sensible statute would allow one but not the other of these variants. 16 For one of the more radical reform suggestions, see Greg Strauss, ‘Is Polygamy Inherently Unequal?’, Ethics, 122 (2012), 516.

218

Douglas Husak

important innovations of the Code’s treatment of offences against the family is its separation of the crime of bigamy from that of polygamy. A married defendant commits the misdemeanour of bigamy ‘if he contracts or purports to contract another ­marriage’.17 A defendant commits the felony of polygamy ‘if he marries or cohabits with more than one spouse at a time in purported exercise of the right of plural marriage’.18 Apart from the numbers that can be involved (‘bi’ as opposed to ‘poly’), the difference between these two offences is not altogether obvious. In most cases, the two spouses of the bigamist are unaware of each other, and the defendant is guilty of deserting or failing to support his prior spouse as well as defrauding his new one.19 But nothing in the text of the statute requires this result; it is legally possible that all parties could be aware of, and even consent to, the subsequent marriage of the bigamist. One difference between these offences stands out. As written, the crime of polygamy, unlike that of bigamy, requires the defendant to act ‘in purported ex­ercise of the right of plural marriage’.20 The bigamist, I assume, typically acts furtively, fully aware that he lacks the legal right to remarry. By contrast, according to the Commentaries of the Code, the polygamist is ‘engaged in open defiance of the law by purporting to engage in plural marriage under claim of right’.21 Although this basis for distinguishing the crime of bigamy from that of polygamy is relatively clear, its normative significance is not. The polygamist, at least, is honest about what he does. Why should his defiance be important—especially when his claim of right is likely to involve religious grounds? Yet the importance of defiance is prominently codified in the text of this statute; the Commentaries cite it to justify a harsher sentence for polygamy, grading it as a felony and bigamy as a misdemeanour.22 Thus if a man purports to have several wives without alleging a claim of right, hoping that no one will notice his unusual family arrangement, it is legally impossible for his behaviour to constitute the crime of polygamy and at most would amount to the less serious offence of bigamy. It is hard to think of another example in which the penal law attaches comparable significance to a defiant attitude on the part of the defendant.23 Henceforth I will assume that a statutory proscription of polygamy could delete this curious element, even though this assumption blurs the contrast between bigamy and polygamy. With these preliminaries behind us, I turn to the fundamental question of how a ban on polygamy fares when measured by the several constraints in my theory of criminalization. As I have cautioned, the outcome of this exercise is not altogether clear. Still, states must ultimately decide whether to ban polygamy or not; they 17  Model Penal Code, §230.1(1). The statute lists a number of exceptions, that is, cases in which a married person does not commit bigamy even though he contracts another marriage. In particular, the Code diverges from common law in allowing a limited defence of mistake. 18  Model Penal Code, §230.1(2). 19  See Glanville Williams, ‘Bigamy and the Third Marriage’, Modern Law Review, 13 (1950), 417. 20  Model Penal Code, §230.1(2). 21  Commentaries to §230.1, p. 393. 22  ‘[R]‌etention of felony sanctions for polygamy is justified by the open defiance of strongly held community ideals of family organization.’ Commentaries to §230.1, p. 393. 23  For a discussion of the role of defiance in the substantive criminal law generally, see Jenny E. Carroll, ‘The Resistance Defense’, Alabama Law Review, 64 (2013), 588.

Polygamy

219

do not have the luxury available to the academic philosopher who may reserve judgment pending further reflection. Since I believe the onus should be on the state to make the case for criminalization with no presumption in favour of the status quo, I tentatively conclude that polygamy should be decriminalized. I am less interested, however, in establishing this particular result than in describing the approach I believe a theory of criminalization should take. Most of all, I hope that this exercise helps us to understand how to test a theory of criminalization—and that my own theory will emerge as plausible.

II.  Internal Constraints and Polygamy My theory of criminalization consists in seven constraints, four of which I loosely describe as internal, and three of which I call external. No statute should be accepted as part of the penal law unless it satisfies each of these seven constraints. As we will see, the justifiability of polygamy statutes is jeopardized by several of these constraints. Again, however, the goal of this brief overview is not to defend a definitive position, but rather to show how I believe a debate about the criminalization of polygamy (or any other putative offence) should be structured. I begin by assessing polygamy bans against the four internal constraints in my theory before proceeding to the remaining three external constraints. The first of my constraints demands that conduct not be punished unless it is wrongful. Even apart from debates about how wrongfulness should be explicated, it might seem obvious that this condition is satisfied.24 Although public opinion is changing rapidly, more than 90 per cent of Americans viewed polygamy as immoral in 2004.25 What is needed, however, is an argument to support this judgment. Why might polygamy be wrongful? Admittedly, sincere disagreement about whether a type of conduct is wrongful may be difficult to resolve—a problem that often arises when assessing mala prohibita offences.26 I doubt, however, that the best answer to this question construes polygamy as an instance of malum prohibitum. As long as we remain on secular grounds, I believe that the best case for regarding polygamy as wrongful categorizes it as an instance of conduct that is wrongful because of the harms it causes (or risks). That is, polygamy is not plausibly construed as wrongful in the same way as, say, a broken deathbed promise is wrongful—that is, wrongful apart from its causal consequences. If polygamy is wrongful, its wrongfulness consists not in its inherent properties but rather in the harms it causes.27 24 See the exchange between Antje du Bois-Pedain, ‘The Wrongfulness Constraint in Criminalisation’, Criminal Law and Philosophy, 8 (2014), 149; and J. R. Edwards and A. P. Simester, ‘Wrongfulness and Prohibitions’, Criminal Law and Philosophy, 8 (2014), 171. 25  See Markel, Collins, and Leib, Privilege or Punish (n. 9), 135. 26  See Douglas Husak, ‘Malum Prohibitum and Retributivism’, in Douglas Husak, The Philosophy of Criminal Law: Selected Essays (Oxford: Oxford University Press, 2010), 410. 27  Many and perhaps most wrongs, I believe, are wrong because of the harms they cause (or risk), and would not be wrong otherwise. Some wrongs, however, are wrong quite apart from their causal consequences.

220

Douglas Husak

If I am correct about how to think about the supposed wrongfulness of polygamy, nearly all of the hard work in determining whether a polygamy proscription satisfies the wrongfulness constraint in my theory of criminalization involves determining whether it satisfies the second of these necessary conditions: the harm constraint. This constraint is satisfied, I believe, by showing that the conduct to be proscribed causes harm or the unacceptable risk of harm.28 Although the very existence of a harm constraint in a theory of criminalization has been contested,29 most of the critical discussion about the harm constraint involves its alleged vacuity. Without an account of what harm is, commentators who favour the enactment of a given penal law will be able to construe any undesirable state of affairs as harmful and thus as satisfying the second constraint in my theory of criminalization.30 A substantive analysis of harm is needed to provide content to this constraint and to rescue it from the charge of triviality. In the clearest cases, the bad consequence that a person suffers through human agency should not be regarded as a harm unless his rights are implicated.31 Ideally, anyone who contends that a bad effect qualifies as harmful should be prepared to identify the right(s) that is infringed or violated by the conduct that causes the effect in question. Of course, disagreement about whether conduct implicates a right may be just as controversial as disagreement about whether it is wrongful. Since no particular theory of rights has gained a consensus among philosophers, a rights-based analysis of harm is no guarantee that the charge of vacuity will be met. Fortunately, this obstacle is not very formidable in the present context. It is relatively clear, I hope, that many of the negative effects allegedly caused by polygamy do not implicate rights, hence probably are not harms, and therefore do not contribute to the case for criminalization under the harm constraint—although they may be relevant for a variety of other normative purposes in assessing polygamy, and may even be relevant to the case for criminalization when some other constraint is applied. Conduct that is bad and produces social disutility need not implicate rights and thus need not be harmful. This conclusion is important. In my judgment, the failure to distinguish disutility from harm is perhaps the most serious mistake committed by scholars who write about the penal law but lack expertise in criminal law theory. These commentators tend to treat ‘bad effect’ and ‘harm’ as though they were synonyms. Even sophisticated criminal theorists are susceptible to a similar mistake. Each bad consequence caused by a kind of conduct presumably is relevant to its moral status. Legal moralists, as we have seen, have a hard time showing that any of these bad 28  This way to conceptualize the harm constraint has been contested. According to some legal philosophers, the harm constraint is satisfied when (so-called) harm would result from the failure to criminalize. See John Gardner and Stephen Shute, ‘The Wrongness of Rape’, in Jeremy Horder (ed.), Oxford Essays in Jurisprudence, 4th Series (Oxford: Oxford University Press, 2000), 193. 29  See Arthur Ripstein, ‘Beyond the Harm Principle’, Philosophy & Public Affairs, 34 (1999), 235. 30 See Bernard Harcourt, ‘The Collapse of the Harm Principle’, Journal of Criminal Law & Criminology, 90 (1999), 109. 31 See the analysis of harm presented in Joel Feinberg: Harm to Others (New York: Oxford University Press, 1984), especially 34.

Polygamy

221

consequences are immaterial to the case for criminalization. As I have indicated, the appropriate test of a theory of criminalization is that it asks the right questions and does not demand answers to the wrong questions. Before turning to polygamy in particular, it may be instructive to briefly show how the foregoing mistake is committed in other contexts. Drug offences— my favourite example of over-criminalization—will serve as my illustration. Commentators who argue that drug users should continue to be punished frequently describe a number of bad consequences allegedly caused by the ingestion of illicit substances. Many of these bad consequences seem unabashedly paternalistic, involving the health or welfare of the drug user himself. To cite just one of several examples, the use of drugs such as marijuana is frequently said to produce what has been called the amotivational syndrome.32 These commentators proceed to call this bad consequence a harm, which brings its significance squarely within the parameters of the harm constraint. Bolstered by the claim that the drug in question causes harm, a presumptive case for criminalization is made to seem plausible. In my judgment, however, the weak link in this argument—quite apart from obvious worries about the empirical evidence—is the unsupported equation of disutility with harm. If my analysis is correct, conduct that decreases motivation is difficult to conceptualize as harmful unless it implicates rights. Whose rights could possibly be infringed or violated by conduct that lessens the motivation of the agent who engages in it? This allegation about illicit substances, at least, probably does not describe a genuine harm.33 A similar observation can be made about much of the scholarly opposition to polygamy. Consider, for example, the litany of evils described by Maura Strassberg, one of the most thoughtful scholars to express scepticism about the decriminalization of polygamy (and of polgyny in particular). Strassberg contends ‘polgyny not only fails to produce critical building blocks of liberal democracy, such as autonomous individuality, robust public and private spheres, and affirmative reconciliation of individuality and social existence, but promotes a despotic state populated by subjects rather than citizens.’34 Although one may doubt that Strassberg is correct on empirical grounds, there can be no dispute about whether the state has ample reason to be concerned about polygamy if she is correct about its social consequences. Even if these evils are real, however, I deny that they support the case for criminalization. This litany does not provide a basis for criminalization under the harm constraint, although it almost certainly is relevant for other purposes,

32  See the discussion in Mitch Earleywine: Understanding Marijuana: A New Look at the Scientific Evidence (Oxford: Oxford University Press, 2002), 197–209. 33  In fact, the justifiability of all criminal paternalism is called into question by a requirement that conduct must implicate a right before it may be criminalized. Paternalism is almost certainly a legitimate justification for law outside the criminal domain, but usually provides a poor rationale for punishment. For critical thoughts about paternalism and my theory of criminalization, see Heidi M. Hurd, ‘Paternalism on Pain of Punishment’, Criminal Justice Ethics, 28 (2009), 49. 34  Maura Strassberg, ‘The Crime of Polygamy’, Temple Policy & Civil Rights Law Review, 12 (2003), 353, 356. These allegations elaborate on an earlier article by Maura I. Strassberg, ‘The Challenge of Postmodern Polygamy: Considering Polyamory’, Capital University Law Review, 31 (2003), 439.

222

Douglas Husak

including many that pertain to law. Again, it is hard to see whose rights are infringed or violated, for example, when the state allows a practice that ‘fails to produce critical building blocks of liberal democracy’.35 After all, a great many other contemporary practices are vulnerable to this criticism as well. Should we really accept the proposition that the state should punish citizens who engage in activities that fail to build liberal democracies? Whatever one thinks about perfectionist arguments in political philosophy generally—and I tend to sympathize with them— they provide a far better rationale for non-penal than for penal legislation.36 Suppose, however, that we concede that at least some of the bad consequences alleged to be caused by polygamy violate rights. The Attorney General of Utah claims that statutory rape, incest, unlawful sexual conduct with a minor, child abuse, and cohabitant abuse are products of polygamy.37 The Attorney General of British Columbia maintains that injuries to vulnerable groups are prevented by polygamy proscriptions; they are said to protect women and children from commodification and exploitation.38 Since all or most of these consequences amount to harms—as surely is the case under any plausible theory of rights—must we then admit that polygamy satisfies the first two constraints in my theory of criminalization? That is, must we then admit that polygamy is harmful, and is wrongful because of the harms it causes? Not at all. Three challenging sets of issues remain. In my judgment, the importance (and the difficulty) of these issues reveal the point at which the empirical social sciences are likely to be most relevant to questions about criminal law and criminalization.39 The first is the most straightforward, so I deal with it briefly. These allegations all posit that polygamy causes specified harms. In order to assess these claims empirically, we must construct a baseline of comparison.40 Dyadic relationships are the obvious choice for a control group. In other words, polygamy could be said to cause the above harms if they occur less frequently in the baseline of dyadic marriage. Some social scientists are dubious, however, that men in polygamous marriages are more likely to be guilty of neglect or abuse than those in dyadic relationships.41 We must not idealize the traditional family and suppose it to be

35  Strassberg, ‘Crime’, 356. 36  Generally, see Thomas Hurka, Perfectionism (New York: Oxford University Press, 1993); and Jonathan Quong, Liberalism without Perfection (Oxford: Oxford University Press, 2011). 37 The allegations are found in Kathleen Tracy, The Secret Story of Polygamy (Naperville, Ill.: Sourcebooks, Inc., 2002), 188–9. 38  Brief of the Attorney General of British Columbia in Reference re: Section 293 of the Criminal Code of Canada (n. 14). 39  Paul Robinson has dedicated the latter part of his career trying to establish the relevance of empirical findings to questions about crime and punishment. He contends that lay perceptions of desert must be reflected in the penal law in order to motivate the voluntary compliance on which nonauthoritarian systems of criminal law depend. In my judgment, the empirical issues about the causal connection between conduct and harm that I explore here are an even more important place to locate the significance of social science to penal theory. 40  See Jonathan Schaffer, ‘Contrastive Causation in the Law’, Legal Theory, 16 (2011), 259. 41  See Janet Bennion, Women of Principle: Female Networking in Contemporary Mormon Polygyny (Oxford: Oxford University Press, 1998), especially 154.

Polygamy

223

immune from many of these maladies in order to paint an exaggerated picture of the harms caused by polygamy.42 The second set of issues arises once we realize that polygamy need not cause any of the consequences I concede to be harms. Some polygamous relationships, it must be admitted, are consensual, supportive, loving, and contribute to the flourishing of the parties involved. At most, polygamy is wrongful because it poses an unacceptable risk that the foregoing harms will occur. The harm constraint must be construed to allow the enactment of a penal law not only when it causes harm, but also when it causes an unacceptable risk of harm. The anticipatory offences of attempt, conspiracy, and solicitation are a legitimate exercise of the penal sanction because (or insofar as) the conduct they proscribe creates an unacceptable risk of harm.43 I take very seriously the effort to defend polygamy statutes as anticipatory offences; much of the battle about its criminalization is likely to be fought on this ground. Does polygamy cause an unacceptable risk of any of the consequences I concede to be harms? Further empirical data are needed to answer this question. The data that are needed must be fairly precise; conduct should not be banned under the harm constraint unless it poses a substantial risk of harm. Unfortunately, judgments about whether given risks qualify as substantial are vague and im­precise; reasonable minds differ about where lines should be drawn.44 What percentage of polygamous families is guilty of child or spousal abuse, for example? If the probability is sufficiently low, it becomes hard to argue that the risks are substantial, and a proscription of polygamy cannot be justified as an anticipatory offence. A third and final set of challenges is as follows. One must always ask whether the litany of evils offered by commentators who favour the criminalization of polygamy is caused by polygamy itself. Two alternative possibilities must be confronted. First, many of these consequences may be due to the particular and contingent manner in which polygamy happens to be practised.45 For obvious reasons, the bulk of the empirical studies on the effects of polygamy have been conducted in Mormon societies in the United States. The cultural background of such societies may contribute mightily to the perceived effects of polygamy. Research on the consequences of polygamy in non-Mormon societies is limited but urgently needed. Second, many of these bad consequences may be more properly attributed to the criminalization of the practice than to the practice itself.46 We cannot decide whether a given effect is caused by polygamy as opposed to its proscription unless we determine whether it 42  For a discussion of some of the abuses in the family tolerated by the criminal law, see Wayne Logan, ‘Criminal Law Sanctuaries’, Harvard Civil Rights—Civil Liberties Law Review, 38 (2003), 321. 43  It is not easy to show that factually impossible attempts, for example, pose an unacceptable risk of harm. See Douglas Husak, ‘Repaying the Scholar’s Compliment’, Jerusalem Review of Legal Studies, 1 (2010), 48. 44  See the empirically informed discussion of the risks of HIV transmission by infected persons in Kaplan, ‘Restoring Reason to HIV-Exposure Laws’ (n. 3). 45  The highest court of British Columbia in Canada concluded ‘that polygamy is inevitably associated with sundry harms, and that these harms are not simply isolated . . . but inhere in the institution itself.’ Reference re: Section 293 of the Criminal Code of Canada (n. 14), para. 1343. 46  See Shayna M. Sigman, ‘Everything Lawyers Know about Polygamy is Wrong’, Cornell Journal of Law and Public Policy, 16 (2006), 101.

224

Douglas Husak

would persist through a regime of decriminalization.47 To make this determination, we must have some idea of what polygamous relationships would be like if the ban were lifted. I hope it is clear a legal regimen at least as complex as that which governs dyadic marriage would need to be created. Admittedly, arguments for decriminalization do not always bother to describe the regulations that should be applied to a type of conduct previously proscribed. Often it is thought to be sufficient to defend the conclusion that a kind of activity should not be prohibited by the criminal law. Yet this perspective is incomplete if we hope to determine whether given harms are caused by the proscription of a practice rather than by the practice itself. I have conceded arguendo that polygamy is widely associated with many kinds of harm. If mechanisms to reduce these consequences cannot be implemented in the real world, it becomes more plausible to believe that the conduct itself—rather than its criminalization—is harmful (and thus wrongful). The foregoing worries are especially trenchant in evaluating proscriptions of polygamy. With a handful of exceptions,48 almost all commentators admit that polygamous marriages are highly disadvantageous to women in a variety of ways.49 As we have seen, many (but not all) of the criticisms cite consequences that almost certainly should be countenanced as harms. Women in polygamous relationships are vulnerable to abuse and highly subject to exploitation.50 Even if force and fraud were eradicated, it is tempting to believe that no set of reforms could ameliorate these problems. As Sigman explains, ‘what makes American polygamy dangerous is the extent to which it requires polygamous communities to game the social demographics and economics to create artificial characteristics that are favorable to polygamy in order to sustain itself.’51 Whether these evils would persist in a real-world regime of decriminalization is probably the hardest single question which theorists must confront. But we should not automatically suppose that these problems are intrinsic to polygamous relationships and cannot possibly be reduced. Here, as in other contexts, decriminalization might create opportunities to implement a regulatory scheme that could be expected to improve the predicament of women in polygamous relationships. Progress toward achieving this goal has been made in a thoughtful article by Adrienne Davis.52 She argues that many of the disadvantages to women in

47  For an argument that polygamy is inherently inegalitarian, see Thomas Brooks, ‘The Problem with Polygamy’, Philosophical Topics, 37 (2009), 109. An additional reason to reach this conclusion is defended by Strauss, ‘Is Polygamy Inherently Unequal?’ (n. 16). 48  Notable exceptions include Gary S. Becker, A Treatise on the Family (Cambridge, Mass.: Harvard University Press, 1991); and Richard A. Posner, Sex and Reason (Cambridge, Mass.: Harvard University Press, 1992). 49 See Maura I. Strassberg, ‘Distinctions of Form or Substance: Monogamy, Polygamy, and Same-Sex Marriage’, North Carolina Law Review, 75 (1997), 1501. 50  One might well be suspicious of paternalistic arguments designed to protect women from their ill-advised choices. See Jill Elaine Hasday, ‘Protecting Them from Themselves: The Persistence of Mutual Benefits Arguments for Sex and Race Inequality’, New York University Law Review, 84 (2009), 1464. See also de Marnefee, Liberalism and Prostitution (n. 3). 51  Sigman, ‘Everything Lawyers Know about Polygamy is Wrong’ (n. 46), 166–7. 52  Davis, ‘Regulating Polygamy’ (n. 8). See also Strauss, ‘Is Polygamy Inherently Unequal?’ (n. 16).

Polygamy

225

polygamous marriages could be reduced by applying better default rules—rules that identify legal outcomes for which the parties have made no provision ex ante. Because parties are rarely inclined to reach agreements in intimate relationships— to make prenuptial agreements, for example—default rules are especially important in family law.53 Since multiple parties are involved, Davis looks to commercial partnership law for insights about how to design default rules for newly decriminalized polygamous relationships.54 These rules would radically alter the way polygamy is currently practised in many communities. Specifically, Davis proposes that plural marriages should borrow the unanimity default rule from commercial partnership law. As a result, subsequent spouses could not be added to a marriage without the express approval of all prior members. In addition, existing spouses could not be expelled without unanimous consent. Finally, a no-fault rule would govern decisions to unilaterally exit a polygamous marriage without dissolving the relationship among the members who chose to remain.55 These recommendations hardly answer the myriad regulatory questions about polygamy. Many issues remain. Would the relatives of polygamist spouses automatically qualify for privileges under immigration law, for example? Would un­limited transfers of wealth to each spouse be exempted from the taxes ordinarily levied on gifts or inheritances? Although the specific recommendations advanced by Davis would not solve all of the worries about polygamous relationships, they help to show that many of the more serious problems could be improved by a regime of legal regulation that cannot be imposed as long as the practice remains underground. They would go a long way toward promoting the ideals of egalitarianism and fairness to which the contemporary law of dyadic marriage aspires (but achieves imperfectly). At the very least, Davis’s proposals demonstrate that we are not inevitably saddled with all of the horrors presently associated with polygamy if we opt for decriminalization. Careful legal regulations could improve the plight of polygamists.56 We have an additional reason to move beyond the mere decision to decriminalize by identifying and evaluating the set of regulations to be imposed on the newly lawful practice of polygamy. The criminal law already allows fornication, and no serious criminal theorist recommends that the ancient ban be reinstated. As a result, a de facto institution of polygamy is permitted unless the participants attempt to formalize their arrangement with a marriage. Thus the debate about polygamy cannot evade the question of recognition; criminal behaviour is 53  See Brian Bix: ‘Bargaining in the Shadow of Love: The Enforcement of Premarital Agreements and How We Think about Marriage’, William & Mary Law Review, 40 (1998), 145. 54 Many commentators seem to think that other unconventional lifestyle arrangements—such as gay marriage—offer insights about regulatory reform. But Davis is persuasive in showing ‘from a regulatory perspective, and the question of what stake the law has in how intimates treat each other, the challenges posed by dyadic marriage, gay or straight, and plural marriage differ substantially’. See Davis, ‘Regulating Polygamy’ (n. 8), 1987. 55  Davis, ‘Regulating Polygamy’ (n. 8), 1998–2032. 56  For a second thoughtful attempt to show how careful regulations of polygamy might improve the plight of women, see Michele Alexandre, ‘Big Love: Is Feminist Polygamy an Oxymoron or a True Possibility?’, Hastings Women’s Law Journal, 18 (2007), 3.

226

Douglas Husak

distinguished from its non-criminal counterpart precisely by the set of regulations in place. In sum, questions of recognition cannot be evaded if we hope to ensure that the newly permitted practice causes less disutility than its illegal predecessor.57 I now move beyond harm and causation to the third internal constraint in my theory of criminalization: penal sanctions must be deserved. Although this constraint overlaps considerably with the wrongfulness and harm constraints, the overlap does not make the desert constraint redundant. Some persons do not deserve to be punished even though their wrongful conduct causes harm. Many private wrongs offer good illustrations. But the most obvious examples of this phenomenon involve excuses. We tend to think of excuses as highly individualized, but a great many polygamists proffer general claims on behalf of their arrangements that tend to evoke sympathy and respect in other contexts. In particular, polygamists often seek to be excused from liability on religious grounds. Although this issue was said to be resolved by the Supreme Court in 1878,58 the reasoning in that decision is hardly unassailable.59 As we have seen, the defiance expressed by a claim of right is a material element of the crime of polygamy in the Model Penal Code. Of course, the polygamist is unlikely to assert a claim of legal right to plural marriage; any right he alleges is probably derived from his religious convictions. Even if these religious convictions should not constitute a complete defence to the charge of polygamy, it is counterintuitive to construe them as an aggravating factor that renders a defendant eligible for a greater quantum of punishment. The fourth and final internal constraint in my theory of criminalization differs from its predecessors: the state is required to bear the burden of proof in demonstrating that a given statute should be included within its penal code. In other words, a particular law must be shown to be justified; it is not presumed to be justified simply because it is on the books. The real issue to be decided is not whether polygamy or any other conduct should be decriminalized; instead, the case in favour of criminalization must be made. Whether a ban on polygamy satisfies this fourth constraint depends mostly on how high this burden of proof is set—yet another general matter on which commentators can be expected to differ. But the burden should be substantial; after all, my theory is alleged to be minimalist. Clearly, the foregoing points establish reasonable doubts about whether polygamy statutes satisfy each of the earlier internal constraints. In view of these reasonable doubts, these statutes probably fail to satisfy the fourth constraint as well.60 57  Because recognition is crucial to the issue before us, it is unlikely that the alleged wrongs of bigamy should be regarded as private wrongs that fail to trigger a case for criminalization. The public nature of the conduct provides the draftsmen of the Model Penal Code with a rationale for creating a separate offence. The Commentaries to §230.1 indicate (at 377) that an independent crime is needed because, unlike casual and private adultery, bigamy ‘amounts to a public affront and special provocation to the first spouse’. Even though this affront may be conventional, it is hard to doubt its reality. 58  Reynolds v US, 98 US 145 (1878). 59 See Keith E. Sealing, ‘Polygamists out of the Closet: Statutory and State Constitutional Prohibitions against Polygamy are Unconstitutional under the Free Exercise Clause’, Georgia State University Law Review, 17 (2001), 691. 60  As one scholar concludes, ‘the full case has not been made for criminalization of polygamy’. Sigman, ‘Everything Lawyers Know about Polygamy is Wrong’ (n. 46), 106.

Polygamy

227

III.  External Constraints and Polygamy I turn now to the question of whether polygamy statutes conform to what I call the external constraints in a theory of criminalization. The external constraints in my theory are designed to give citizens in a democratic state good reason to employ the criminal justice system to cope with a problem even though many other demands on scarce taxpayer resources compete for their attention. I borrow these conditions from tests of intermediate scrutiny developed by the Supreme Court in the context of protecting rights as valuable as those implicated by punishment. Three hurdles must be crossed before these criteria are satisfied. First, the state must have a substantial interest in whatever objective the statute is designed to achieve. Second, the law must directly advance that interest. Third, the statute must be no more extensive than necessary to achieve its objective. Much of the ground covered by these constraints is redundant with the analysis offered thus far. As we will see, however, the application of these constraints suggests further reason to doubt that statutes against polygamy are a legitimate exercise of the penal sanction. To apply the first of these constraints, we must identify the state interest in banning a practice such as polygamy. In some contexts, this task can be extraordinarily difficult. It may be hard to specify the state interest a statute is designed to serve, let alone to determine whether that interest is legitimate and substantial. Drug offences, I believe, illustrate this difficulty perfectly. Frequently, however, the problem of identifying the state interest of a statute simply replicates that of identifying the harm(s) caused by the conduct proscribed. I conceded that some of the bad consequences alleged to be caused by polygamy are harms, and no one should doubt that the state has a substantial interest in preventing them. In fact, however, efforts to identify the state interest in banning polygamy may require more work than my analysis suggests thus far. It is plausible to suppose that a thorough account of the state interest that underlies the proscription of polygamy cannot be provided without addressing the broad question of the role of law in general, and of the criminal law in particular, in governing intimate relationships.61 Although it is almost inconceivable that the state should get out of the business of recognizing marriage altogether,62 we still need to enquire why it should privilege certain kinds of intimate relationships at the expense of others.63 This is a huge topic I obviously cannot discuss here—partly because it involves nothing less than a theory of the liberal state.64 61  This topic has generated a mountain of literature. For recent contributions, see Lori Watson and Christie Hartley, ‘Political Liberalism, Marriage, and the Family’, Law and Philosophy, 31 (2012), 185; and David Archard: The Family: A Liberal Defense (New York: Palgrave Macmillan, 2010). See also Markel, Collins, and Leib, Privilege or Punish (n. 9). 62  But see Andrew F. March, ‘Is There a Right to Polygamy? Marriage, Equality and Subsidizing Families in Liberal Public Justification’, Journal of Moral Philosophy, 8 (2011), 246. 63  According to one study, current law recognizes ‘more than 1,100 rights, responsibilities, prerogatives, duties, entitlements, tax breaks, and tax obligations for married couples’. Jason Kuznicki, ‘Marriage against the State: Toward a New View of Civil Marriage’, Cato Institute Policy Analysis, 671 (12 January 2011), 2. 64  But see Martha Albertson Fineman, The Neutered Mother, the Sexual Family and Other Twentieth Century Tragedies (London: Routledge, 1995).

228

Douglas Husak

As a matter of history, however, it is relatively clear that the ban on polygamy originated in the enforcement of conventional morality and the discouragement of a practice deemed to be offensive to Christian conceptions of marriage. The legitimacy of this objective, of course, has been discredited by legal philosophers for a long time, and recently has been rejected by the Supreme Court in such cases as Lawrence v Texas.65 But why should history matter to our enquiry? To answer this question, we must decide whether to apply this first external constraint by using the historical objective of the law, the stated objective of the law, the possible or conceivable objective of the law, or the objective actually held by legislators at the present time. I have no easy answer to this question. Nonetheless, we cannot get off the ground in applying the external constraints of my theory of criminalization unless we are willing to tackle it. Indeed, the application of any theory of criminalization that evaluates statutes depending on their rationale or objective must surmount this hurdle and find a way to specify the interest a given law is designed to serve. The second of the external constraints in my theory of criminalization requires a determination of whether the law directly advances the interest it is intended to promote. This enquiry has two parts. First, criminalization must be reasonably effective; it must make the existing problem better. Second, criminalization must not be counterproductive; it should not create new problems greater than those it solves. To apply either part of this enquiry, we need empirical evidence that given laws actually help to achieve their objective without producing too many pernicious side effects. Legal philosophers are understandably reluctant to make these determinations; they belong more narrowly to the purview of the social sciences. Again, however, this enquiry is unavoidable if a statute is to be deemed justifiable. I produced a number of reasons to be sceptical that drug proscriptions succeed in furthering their objective,66 and many of the same doubts can be raised about proscriptions of polygamy. We should not simply assert that penal sanctions deter, and thus are bound to induce individuals to commit fewer acts of polygamy (and thus fewer of the harms of polygamy) if we continue to punish it. Confidence about deterrence is problematic in a great many contexts,67 and might be especially misplaced here. Persons who engage in conduct for religious reasons are not easily dissuaded. Although it might be easier to deter those persons whose motives are not religious, it is hard to gauge the long-term extent to which individuals would choose to enter into polygamous relationships for non-religious reasons if the ban were lifted. I detect no pent-up demand among secular citizens that is curtailed by the prohibition of polygamy.68 Still, recent changes in social attitudes about gay 65  539 US 558 (2003). 66 Husak, Overcriminalization (n. 1), 146–9. 67 For example, see Raymond Paternoster: ‘How Much Do We Really Know about Criminal Deterrence?’, Journal of Criminal Law & Criminology, 100 (2010), 765. 68  The Attorneys General of Canada alleged that ‘not only do reason and logic suggest that the incidence of polygamy would increase in a non-trivial way if it were decriminalized, but that considerable evidence exists regarding this point. For example, the evolutionary psychology evidence supports the notion that polygyny is an advantageous mating strategy for men who are able to afford it.’ See Reference re: Section 293 of the Criminal Code of Canada (n. 14), para. 1290.

Polygamy

229

marriage indicate that customs and mores about marriage and family are capable of evolving rapidly; it would be rash to predict that polygamy would forever remain far outside the mainstream if the statute were repealed. The existing law may not achieve much deterrence for an additional reason: it is easy to evade. The letter of the law is circumvented by polygamists who simply neglect to enter into a civil marriage with a second, third, or fourth spouse. This possibility is not merely speculative. To avoid the obvious flouting of the law, polygamous men often legally marry only their first wife and have a religious wedding ceremony for those who follow.69 This evasive device not only calls deterrence into question, but also indicates that the prohibition of polygamy may be counterproductive. If a civil union does not take place, potential spouses lack the opportunity to express their reservations about the pending marriage before a neutral and disinterested authority, thus exacerbating some of the legitimate worries about polygamy as it tends to be practised in the United States. More generally, it is disingenuous to criticize polygamists for such matters as ‘failing to create the building blocks of liberal democracy’ when the very law they violate marginalizes their communities and encourages their insularity and seclusion from the rest of society.70 Here, as elsewhere, most of the real harms associated with polygamy are more difficult to detect and prevent when perpetrators remain furtive. My third and final external constraint requires the challenged offence to be no more extensive than necessary to achieve its objective. This condition creates a strong presumption against overinclusive penal statutes. Again, this constraint cannot be applied unless we are confident about what the objective of polygamy laws is; only then can we begin to decide whether a given statute is vulnerable to the charge of overinclusion. The best way to ensure that a more narrowly tailored law would further the substantial state objective without bringing innocent conduct within its scope is to do the hard work of proposing an alternative offence and showing how it would be preferable to the status quo. As I have indicated, impressive progress in this direction has been made by Davis and others.71 I have conceded arguendo that the percentage of cases of polygamy resulting in child or spousal abuse is high. Even so, a sweeping ban on polygamy would seem to be overinclusive if it is intended to prevent these harms. Spouses and children present separate problems if we focus on actual statutory texts. A polygamy statute may punish both the man and the women in a polyandrous relationship. It is peculiar (and perhaps even incoherent) to suppose that these statutes are designed to prevent harms to women while simultaneously subjecting these same women to penal sanctions. I doubt that one and the same person can be both perpetrator and victim of a given offence.72 Thus I will suppose either that polygamy statutes should be altered to protect women rather than to punish them, and/or that our 69 See Irwin Altman and Joseph Ginat, Polygamous Families in Contemporary Society (Cambridge: Cambridge University Press, 1996), especially 131–2. 70  See Strassberg, ‘The Crime of Polygamy’ (n. 34), 356. 71  Davis, ‘Regulating Polygamy’ (n. 8); Strauss, ‘Is Polygamy Inherently Unequal?’ (n. 16). 72  In some contexts, this rule is known as the Tyrell principle.

230

Douglas Husak

focus is confined to the abuse of children. Even here, worries about overinclusion arise; the state may lack a persuasive rationale to criminalize all polygamous arrangements in order to reach the smaller subset of cases that victimize children. Unless this feat simply cannot be accomplished, the state should craft a more narrowly tailored statute to punish only those types of case in which harms are likely to result. Despite its centrality to my theory of overcriminalization, allegations of overbreadth are sometimes dismissed as relatively unimportant in the real world.73 Prosecutorial discretion, we are assured, will prevent enforcement of statutes in circumstances in which liability would be unwarranted. Polygamy is no exception. Utah government officials are aware of thousands of polygamist families in the state and regularly interact with such families as part of a ‘Safety Net’ programme.74 And the Attorney General of British Columbia expressly admits that ‘simple polygamy’—described as ‘polygamy without some direct harm to the participants or others, such as children’—would not be prosecuted. He continues: ‘Experience and logic both suggest that a polygamy investigation could never even result in charges without some serious aggravating factors.’75 Prosecutorial discretion clearly helps to minimize the injustice otherwise caused by overinclusive statutes. Nonetheless, reliance on the good judgment of prosecutors is no substitute for getting our laws right in the first place.76 Allegations of overbreadth raise issues of fair notice and respect for legality—two of the most serious problems caused by the phenomenon of overcriminalization. Prosecutorial discretion, however wisely it may be used, provides a second-best solution to these problems. Overcriminalization often results from recriminalizing behaviour already proscribed. All jurisdictions have separate statutes to proscribe most or all of the harms said to be caused by polygamy: underage sex, sexual assault, and the trafficking of persons, for example. Failing to create an additional offence of polygamy to punish these harms has the enormous advantage of retarding overcriminalization while simultaneously avoiding overinclusion. Thus, to the extent that a prohibition of polygamy is supported for these reasons, a distinct offence of polygamy may not be needed.77 Admittedly, however, the fact that the worst consequences of polygamy are already punished is not a decisive argument against a separate statute. The same argument could be pressed against any anticipatory offence, since the harms these offences anticipate are independently banned. If polygamy creates a substantial risk of these harms, there may be no principled objection to enacting an offence of

73  Even those commentators who tend to favour polygamy prohibitions are worried about problems of overinclusion. See Strassberg, ‘The Crime of Polygamy’ (n. 34), 370. 74  This point was conceded in Brown v Buhman (n. 10), 6. 75  Brief of the Attorney General of British Columbia, Reference re: Section 293 of the Criminal Code of Canada (n. 14), para. 102 of opening statement (2011). 76 See Josh Bowers, ‘Legal Guilt, Normative Innocence, and the Equitable Decision not to Prosecute’, Columbia Law Review, 110 (2010), 1655. 77  The same is true of bigamy, since the Code separately prohibits non-support (§230.5) and inducing a woman to have intercourse in the mistaken belief that the defendant is her husband (§213.1).

Polygamy

231

polygamy—assuming, of course, that the other constraints in my theory of criminalization are satisfied.78

IV. Conclusion According to my theory of criminalization, a thorough debate about the justifiability of criminal laws against polygamy must address each of the following topics. Is polygamy wrongful? Is its wrongfulness due solely to the harms it causes? If not, why not? If so, what are these harms? Are they clearly harms, or simply undesirable consequences that polygamy causes? What is the probability that these harms will occur? Does polygamy really cause these harms, or is it more plausible to regard them as a product of polygamy proscriptions? Are these harms caused by the peculiar way polygamy is practised in the United States, or are they likely to be detected in any time and place in which polygamy is found? Is it realistic to suppose that polygamy could be regulated to minimize the occurrence of these harms? What kinds of regulation should be imposed? Do perpetrators deserve punishment in light of their religious motivations for entering into polygamous relationships? Does the state have a substantial interest in banning polygamy? What is this interest, and how should it be identified? Is there empirical evidence that the ban on polygamy actually furthers this state interest in the real world? Might the prohibition be counterproductive? Is the law overinclusive, and is it realistic to suppose that a more narrowly tailored statute can be drafted and enforced? Finally, do reservations about the foregoing matters satisfy the burden of proof that all arguments in favour of criminalization must meet? Clearly, each of these issues contains its own complexities. I have not attempted to hazard an ultimate resolution of the question of whether punishment for polygamy survives the intense scrutiny my minimalist theory requires. Obviously, I am sceptical; I have spent far more time critiquing than defending the rationale for polygamy laws. But I believe that my theory of criminalization is corroborated if it asks the right questions in assessing such laws. By this standard, I hope that my theory emerges as plausible. A better theory would have to demonstrate that some of the questions I have raised are not relevant, or that some of the questions I have not raised are relevant to the criminalization decision.

78  I provide a more detailed discussion of the conditions that must be satisfied if an anticipatory offence is to be justified in Husak, Overcriminalization (n. 1), 159–77. In particular, perpetrators must act with some degree of culpability with respect to the ultimate harm these offences anticipate.

9 Civil Peace and Criminalization Anthony Bottoms*

The principal concern of this chapter is to discuss in some detail a cluster of issues that are raised by two claims in the late Neil MacCormick’s seminal book, Institutions of Law. First, MacCormick states that ‘the most basic demand citizens ought to make of criminal law is that it contribute to securing the conditions of civility and social peace’.1 Then, secondly, a more explicitly ethical claim is made: ‘what is wrong about committing crimes is that it involves some wilful form of behaviour that violates the conditions of civil peace.’2 These claims are made in c­ hapter 12 of Institutions of Law, and that chapter ends with a helpful restatement of the author’s purpose. The book is not, MacCormick emphasizes, intended to be ‘an essay in . . . critical moral philosophy’. Rather, it is ‘an interpretative account of the most essential functions of law, including criminal law, within a contemporary constitutional state’,3 conducted within what he describes elsewhere as a ‘legal-institutional’ theoretical framework.4 Any review of MacCormick’s work needs to remain sensitive to this purpose. Shortly after the publication of Institutions of Law, a gathering was held at the University of Edinburgh at which—as the editors of the published proceedings later put it—‘a group of legal theorists got together to celebrate, but also critically discuss’ the book with its author.5 From the point of view of scholars of the criminal law, one of the most interesting features of this gathering (hereafter ‘the Edinburgh symposium’) is that no fewer than three contributors (Nils Jareborg, Victor Tadros, and Magnus Ulväng) were critical of MacCormick’s ‘civil peace’ thesis. All these authors, in different ways, expressed the view that, as Jareborg succinctly put the matter, while the notion of crimes as violations of civil peace ‘is very *  I am extremely grateful to Antje du Bois-Pedain for several very helpful discussions on earlier drafts of this chapter. Thanks also to Antony Duff and to Nina Peršak for penetrating comments on a close-to-final draft. Of course, none of these colleagues is to be held responsible for the arguments put forward. 1  N. MacCormick, Institutions of Law: An Essay in Legal Theory (Oxford: Oxford University Press, 2007), 221. 2 MacCormick, Institutions of Law (n. 1), 216. 3 MacCormick, Institutions of Law (n. 1), 221. 4  N. MacCormick, ‘Concluding for Institutionalism’, in M. Del Mar and Z. Bankowski (eds.), Law as Institutional Normative Order (Aldershot: Ashgate, 2009), 190. 5  Del Mar and Bankowski (eds.), Law as Institutional Normative Order (n. 4), 1.

Civil Peace and Criminalization

233

important in discussing why we have a system of criminal justice, or why we do not dismantle the criminal justice system’, nevertheless ‘it is of little or no help in deciding where to draw the line between crime or non-crime, that is, in discussing why some type of act or omission should or should not be criminalised’.6 The most extensive of these three critiques was that by Magnus Ulväng, for whom, within the context of ‘a modern . . . institutionalised state’, the characterization of criminal law as seeking to ensure civil peace is ‘too simple’; rather, the criminal law should be seen as ‘perhaps first and foremost an institution for distributing instrumental censure within a regulated practice of sentencing’.7 In consequence, he argued, we need to supplement MacCormick’s account with ‘a normative theory of criminal law that would enable us to distinguish . . . what ought to be criminalised from what ought not to be criminalised’. When we consider that issue, the ‘moral justification must come first’; and ‘the concept of civil peace is too nebulous to provide any guidance’.8 MacCormick’s claims, and the sceptical responses to them of respected commentators, nicely set the stage for the wide-ranging discussion of civil peace and criminalization that I shall attempt in this chapter. This will begin with a description of the main features of MacCormick’s argument, after which I shall offer a conceptual and explanatory discussion of the concept of social order, which is the term normally used by sociologists to encompass what MacCormick calls ‘civil peace’. Then in the third and fourth sections of the chapter I shall consider how, if at all, civil peace relates to the justification of criminalization. In this discussion, particular attention is given to two important recent contributions to debates on criminalization, namely Marshall and Duff’s ‘crimes as public wrongs thesis’,9 and the ‘harm and wrongfulness’ approach of liberal theorists Simester and von Hirsch.10 Additionally, and in part to address the issue of the alleged ‘nebulousness’ of the concept of civil peace, an empirical example of criminalization will be considered, namely the Road Safety Act of 1967, which first introduced into UK law the offence of driving a motor vehicle when being above a specified blood-alcohol level. In the final substantive section of the chapter, I shall briefly address two broader theoretical issues, namely: (i) the concern of liberal theorists that ‘civil peace’ might turn out to be a vehicle for the importation into criminal law of an unjustifiable social authoritarianism; and (ii) the possibility that a first-order ethic of reciprocity might be a helpful foundation for future theorization. 6  N. Jareborg, ‘A Comment on Personality and Corporate Crime’, in Del Mar and Bankowski (eds.), Law as Institutional Normative Order (n. 4), 125. 7  N. Ulväng, ‘Criminal Law and Civil Peace’, in Del Mar and Bankowski (eds.), Law as Institutional Normative Order (n. 4), 133. 8  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 141. 9  S. E. Marshall and R. A. Duff, ‘Criminalization and Sharing Wrongs’, Canadian Journal of Law and Jurisprudence, 11 (1998), 7; R. A. Duff, Answering for Crime (Oxford: Hart, 2007), ch. 6; R. A. Duff and S. E. Marshall, ‘Public and Private Wrongs’, in J. Chalmers, F. Leverick, and L. Farmer (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010). 10  A. Simester and A. von Hirsch, Crimes, Harms and Wrongs: On the Principles of Criminalisation (Oxford: Hart, 2011).

234

Anthony Bottoms

In conclusion, I will argue that MacCormick’s first claim (as set out above) is correct, but that the second requires revision. However, in the last instance I am less concerned with the specifics of MacCormick’s claims than in first, establishing that—contrary to the views of the Edinburgh symposiasts—the concept of civil peace is by no means irrelevant to decisions about criminalization; and secondly, opening up a discussion about how that concept can best be used—in conjunction with other concepts—in future debates about criminalization.

I.  MacCormick on Criminal Law In the conclusion to ­chapter 12 of Institutions of Law, Neil MacCormick says: [T]‌he most basic demand citizens ought to make of criminal law is that it contribute to securing the conditions of security and civil peace, thus sustaining civil society. This, however, cannot be a function of the criminal law alone, for without social justice the conditions of solidarity and civility are not capable of being achieved. Criminal law and public law have to work in tandem, and cannot work at all to those ends without background social conditions that the law does not establish, but which it can damage.11

It is clear from this passage, when it is quoted in full, that in enunciating his claim about the ‘most basic’ function of the criminal law MacCormick was working (at least in large part) within a sociological framework, with a special focus on social order.12 That point is more explicitly spelt out earlier in the chapter: Human life is always, as Aristotle observed, social. We cannot live in complete isolation, and the full flourishing of human potentialities depends on our coexisting in relatively extensive social groups, or groups of groups. . . . Civil society exists to the extent that peaceful interaction prevails among persons who trust each other to observe a set of legally-established constraints, mainly on a voluntary basis, but with coercive state institutions in the background, as everyone is aware. Resort to violent self-help can then be effectively excluded to a very considerable extent.13

We should note here the emphasis on trust between citizens, a point developed by MacCormick a little later when he says that ‘peaceful relations among persons who can trust relative strangers to avoid violating their persons or their property are the fundamental conditions of civility’.14 For him, the existence of the criminal law 11 MacCormick, Institutions of Law (n. 1), 221. 12 That might seem surprising, given that MacCormick’s primary intellectual training was in law and in philosophy. However, as previously noted, in his later work he consciously adopted a ‘legal-institutional’ theoretical stance, and he once commented that such an approach aims ‘to theorise institutions within jurisprudence in the hope of creating a serious bridge across two cognate disciplines, and constructively linking legal and social theory’: MacCormick, ‘Concluding for Institutionalism’ (n. 4), 190. Moreover, as MacCormick’s close friend William Twining said of him in an interview: ‘he was an excellent philosopher, but unlike some contemporary analytical jurists his conception of philosophy was in the spirit of the Scottish Enlightenment rather than a more narrowly conceived form of conceptual analysis’ (interview with William Twining by Raymundo Gama, 2010, 17; see ). 13 MacCormick, Institutions of Law (n. 1), 208, emphasis added. 14 MacCormick, Institutions of Law (n. 1), 216.

Civil Peace and Criminalization

235

‘in the background’ is a vital (but certainly not the only) contributor to desirable social conditions of civility and trust. As an example of what MacCormick is saying in such passages, consider a modern road traffic system.15 A casual observer of such a system will quickly note that there seems to exist a significant degree of trust among drivers that each will follow some basic procedures, such as driving on a particular side of the road, operating an understood system of priorities at junctions and roundabouts, and so on. A slightly deeper investigation will reveal that for most drivers most of the time, all this is accomplished on a semi-automatic basis, with little conscious attention to formal rules. Sometimes, too, drivers will behave with obvious politeness, signalling to another driver or to a pedestrian to take priority over themselves. But behind these clear manifestations of trust and civility are laws (criminal laws) which prohibit, on pain of state punishment, activities such as careless or dangerous driving, failure to stop at traffic lights, and the driving of unsafe vehicles. Clearly, the dominant motif of these legal prohibitions is a dual focus on (i) an effective road transport system, which allows each driver to pursue his/her own journey with reasonable speed, coupled with (ii) a concern for the safety of all road users, including other drivers and pedestrians. It is, I think, not fanciful to characterize these dual objectives as a framework for securing ‘civil peace’ on the highway. Indeed, the whole system can be regarded as a kind of liberal polity—that is, as a sensible way of regulating the behaviour of thousands of individuals each pursuing his or her own ends, peacefully coexisting within what MacCormick called a ‘relatively extensive social group’ of road users. In Neil MacCormick’s jurisprudence, as enunciated in Institutions of Law, his claims about the criminal law must be set within the framework of his more fundamental thesis that ‘law is institutional normative order’.16 Every word in this formulation is important. A normative order exists when social relations are patterned in accordance with an understood set of social norms applying to that social group (for example, in a family’s traditions about how Christmas Day is to be celebrated).17 An institutional normative order is a normative order that is not purely informal, but is underpinned by some kind of formal social-structural mechanism, as in those commercial outlets where (to borrow MacCormick’s own example) the norm of queueing has been institutionalized by requiring those who wish to receive the service to take a numbered ticket, indicating their position in the queue.18 A road traffic system can then be identified as a significantly more complex institutional normative order than a numbered-ticket queue, while a developed state legal system is an even more elaborate example of such an order.

15  This is a particularly telling example, given that most road users (unlike, say, members of a residential community) are complete strangers to one another. 16 MacCormick, Institutions of Law (n. 1), 11, emphasis added. 17  As MacCormick, Institutions of Law (n. 1), 18 emphasizes, the norms only have to be ‘understood’ for a normative order to exist; hence there can be normative order ‘wherever implicit norms are in fact largely observed and respected’. 18 MacCormick, Institutions of Law (n. 1), 1–2 and ch. 2.

236

Anthony Bottoms

In a modern legal system, there will be various different kinds of law, each fulfilling discrete purposes within the total normative system (see, for example, MacCormick’s comment, in the quotation given above, about the need for the criminal law and public law to work ‘in tandem’). Moreover, these various formal and institutional strands of the overall normative order sit alongside, and interact with, the informal social order of the particular society (see MacCormick’s comment that criminal law ‘cannot work at all . . . without background social conditions that the law does not establish’). In his response to the papers given at the Edinburgh symposium, Neil MacCormick specifically mentioned a second feature in his analysis of criminal law, separate from but linked to his claims about civil peace; the implication of the comment was that his critics had tended to neglect this aspect of his thought. This second feature he described as the ‘thesis of the “moral substratum” of criminal law’;19 and in Institutions of Law it was stated as a claim that ‘there is always some kind of moral substratum beneath the surface of the positively established criminal law anywhere’.20 In developing this claim, MacCormick refers to the traditional criminal lawyer’s distinction between mala in se (acts that are ‘wrong in themselves’) and mala prohibita (acts that are ‘wrong because prohibited’). As to the former: No intelligible moral code gives blanket permission for killing, maiming, raping, assaulting, cheating, deceiving, housebreaking against, or stealing from other people. No intelligible criminal code fails to include such items in its catalogue of most serious crimes. To punish people for other deeds while condoning such actions would be radically incoherent.21

The case for including mala prohibita in the ‘moral substratum’ is of course less straightforward. However, according to MacCormick it is possible to identify ‘with some clarity’ the characteristics of those mala prohibita that are justified prohibitions (as opposed to those that are morally inappropriate). Justified prohibitions: . . . look to some state of affairs which is considered either seriously disadvantageous to the state or civil society and the citizens or members thereof, or to have significant value for them. The circumstances are such that no individual would have sufficient reason to act in ways that counter the disadvantage or procure the positive advantage save in the context of a relevant, universal and reasonably well-enforced prohibition.22

Examples of justified mala prohibita would therefore include, MacCormick suggests, ‘most legislation concerning road traffic’,23 as well as prohibitions designed to prevent environmental damage or to promote safety at work. The moral substratum in such prohibitions is clear. 19  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 192. 20 MacCormick, Institutions of Law (n. 1), 212. 21 MacCormick, Institutions of Law (n. 1), 212. 22 MacCormick, Institutions of Law (n. 1), 213. 23 MacCormick, Institutions of Law (n. 1), 213. But MacCormick also argues that some road traffic offences, such as dangerous driving and drink driving, ‘belong among mala in se’.

Civil Peace and Criminalization

237

II.  Social Order and ‘Law Jobs’ Having established the main features of MacCormick’s theorization about criminal law, it is time to develop the discussion in a number of directions. I shall begin with the concept of social order, which is a central topic within sociology, and which, while apparently somewhat broader than MacCormick’s concept of civil peace, has many similarities with it.24 The American sociologist Dennis Wrong once made the following bold claim: The problem of order is . . . a genuinely transhistorical problem rooted in inescapable conflict between the interests and desires of individuals and the requirements of society: to wit, the pacification of violent strife among men and the secure establishment of co-operative social relations making possible the pursuit of collective goals.25

My experience of using this quotation in seminars is that it is sometimes met with an initial scepticism. Everyone knows that human societies vary hugely in their social patterning—consider, for example, the dominant characteristics of social order in contemporary Colombia, Japan, Saudi Arabia, and Sweden.26 Given such diversity, can it sensibly be claimed that there exists a ‘genuinely transhistorical’ (that is, ‘universal’) problem of order? The answer, I firmly believe, is in the affirmative, but to establish this we will need to consider some fundamental facts about the human condition. Let us begin with two such facts. First, anthropologists report that ‘wherever we encounter them, . . . humans are invariably social, or better stated societal beings’;27 that is to say, we are a species that, in pre-modern times, always lived in groups and societies, and never as separate individuals. A second fundamental feature of human existence is what might be described as human reflexivity. John Polkinghorne, a senior British physicist, once commented in an interview that ‘one of the most astonishing things . . . in the history of the universe . . . is . . . the coming to be of human life and human consciousness here on earth. Something really new emerged . . . with that, because we are able to be aware of ourselves, we are able to understand the world.’28 In other words, humans are self-conscious beings with a capacity to reflect on their own situation—a capacity that is not shared by the rest of the animal kingdom.

24  To anticipate a later point, social order includes elements of both ‘regularity’ and ‘rule’; ‘civil peace’ seems to refer primarily to the latter, and especially to the creation and maintenance of peaceful and trusting relations among citizens. 25  D. Wrong, The Problem of Order: What Unites and Divides Society (New York: Free Press, 1994), 36. 26  Anthropological studies have also clearly demonstrated similar variations in pre-modern societies: see S. Roberts, Order and Dispute: An Introduction to Legal Anthropology (Harmondsworth: Penguin Books, 1979). 27  M. Fortes, Rules and the Emergence of Society (London: Royal Anthropological Institute of Great Britain and Ireland Occasional Paper No. 39, 1983), 1. 28  John Polkinghorne, video interview entitled ‘The Cosmic Process and Human Awareness’, available online at (accessed October 2012).

238

Anthony Bottoms

What are the implications for social life of these two fundamental features of the human condition? In seeking an answer to this question, a sensible first step is to consider what the evidence on the processes of human evolution can teach us; and fortunately, this evidence has been recently summarized by a distinguished ethologist, Robert Hinde. On Hinde’s account, human beings have propensities to behave both with selfish assertiveness and in ways that are ‘co-operative, kind, loving, and so on’ (which he calls prosociality).29 Both of these propensities can be seen in evolutionary terms as the result of natural and cultural selection. Selfish assertiveness can be accounted for as the direct result of natural selection (‘the selfish gene’). The phenomenon of social cooperation has to be divided into ‘within-kin group’ and ‘non-kin’ prosociality. The second of these is the harder to explain in evolutionary terms, but a probable explanation focuses on the fact that very frequently humans were living in a group that was in competition with other groups: Early in human history, individual survival depended on membership of a cohesive group, and even today we function better if living in a harmonious society, where individuals co-operate with each other, [rather] than in one composed of selfish individuals. It is reasonable to suppose that early human groups competed with each other (either directly, or by being better at acquiring limited resources), and that groups with a higher proportion of individuals who behaved prosocially to, and co-operatively with, other in-group members tended to do better than those containing many selfishly assertive individuals.30

So we have human groups composed of individuals with evolved propensities for both selfish assertiveness and cooperation. But those individuals also have the capacity for reflection, which includes the capacity to reflect on whether other members of the community will act cooperatively or selfishly—in short, on whether they can be trusted to act cooperatively. Accordingly, this line of enquiry seems to have confirmed MacCormick’s suggestion that trust is of great importance for social peace; or, as John Dunn once memorably put the matter, ‘the creation and sustaining of trust remains, and will always remain, an indispensable human contrivance for coping with the freedom’ of other people around us.31 Thus, this brief excursus into the fields of anthropology and ethology has rather quickly confirmed Dennis Wrong’s insight that there is indeed a ‘genuinely transhistorical’ problem of social order, with a potentially ever-present tension between the self-assertive desires of some individuals and the ‘requirements of society’, including ‘the secure establishment of co-operative social relations making possible the pursuit of collective goals’.32 A further implication of this research literature is that social order (or ‘civil peace’) will tend often to be relatively fragile, given the constant tension between selfish assertiveness and cooperation. This fragility suggests a need for the development of social-structural mechanisms to encourage 29  R. A. Hinde, Bending the Rules: Morality in the Modern World (Oxford: Oxford University Press, 2007), 9. 30 Hinde, Bending the Rules (n. 29), 35–6. See also H. Gintis, J. Heinrich, S. Bowles, R. Boyd, and E. Fehr, ‘Strong Reciprocity and the Roots of Human Morality’, Social Justice Research, 21 (2008), 241. 31 J. Dunn, ‘“Trust” in the Politics of John Locke’, in Rethinking Modern Political Theory (Cambridge: Cambridge University Press, 1985), 54. 32 Wrong, The Problem of Order (n. 25), 36.

Civil Peace and Criminalization

239

cooperation, and—so far as possible—to limit selfish assertiveness to its healthy and creative forms. At this point we must journey a little further with Dennis Wrong, who helpfully distinguishes two elements within the concept of ‘social order’, namely regularity and rule. Wrong reminds us that ‘ “order” means regularity, predictability and system as opposed to randomness, chance and chaos’.33 Moreover, the social-scientific evidence is clear that a degree of regularity or routine (leading to expectations of ‘how things will be tomorrow’) is vital to what Anthony Giddens has termed humans’ sense of ‘ontological security’: we need a degree of stability in our everyday expectations, or we could not pursue our goals.34 But expectations, as such, carry no necessary normative connotations: I expect that the Football Association Cup Final will be played in London next May, but this expectation has no ethical significance for me. Sometimes, however, expectations are perceived by actors sensitized to them as requiring some action, either positive (‘you should act in this way’) or negative (‘you should refrain from doing this’). At that point, we have a norm, and we have moved from ‘regularity’ to ‘rule’. These informal (and eventually also formal) rules have an important dual function: they help to establish regularity, and at the same time they are normatively geared to the encouragement of cooperation and the discouragement of selfish assertiveness. All this explains why social anthropologists have found that all known human groups have systems of rules. Indeed, the late Meyer Fortes, a distinguished anthropologist, went so far as to argue that ‘the capacity and the need to have, to make to follow and to enforce rules are of cardinal importance for human social existence . . . for without rules there can be neither society nor culture’.35 It is against this background that we can more fully appreciate a distinctive feature of MacCormick’s jurisprudence, namely his deep awareness of the pervasiveness of norms in human social life. Perhaps the clearest manifestation of this, expressed on the very first page of Institutions of Law, is his claim that ‘one of the funda­mental aspects of our nature is that we human beings are norm-users’.36 It is, accordingly, primarily from the perspective of the norm-user that he views social and legal systems. Hence, if one considers: . . . possible human conduct . . . where people interact on the basis of each other’s behavioural orientation, one can understand how each thinks others ought to, and possibly will, behave. From this also, we can construct an idea of ‘orderliness’ in conduct as distinct from disorder, orderliness that results from reciprocal observance of essentially similar beliefs and dispositions of different persons.37 33 Wrong, The Problem of Order (n. 25), 37, emphasis in original. 34  See A. Giddens, The Constitution of Society (Cambridge: Polity Press, 1984), 60–4, citing the extreme degree of unpredictability of the regimes in Nazi concentration camps, so that ‘the disruption and the deliberately sustained attack upon the ordinary routines of life produce[d]‌a high degree of anxiety, a “stripping away” of the socialized responses associated with . . . a predictable framework of social life’ (at 63). 35 Fortes, Rules and the Emergence of Society (n. 27), 6. 36 MacCormick, Institutions of Law (n. 1), 1. 37  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 188–9.

240

Anthony Bottoms

This concept of ‘reciprocal observance’, creating ‘orderliness’, is therefore central to MacCormick’s jurisprudence, and I shall return later to some wider issues concerning reciprocity. But, as we have seen, humans not only have an evolved capacity for cooperation, they also have an evolved capacity for selfish actions. That means that the rules and coercive threats of the criminal law are, in MacCormick’s view, also a necessary (but not sufficient) element in the production of orderliness in contemporary societies. An important implication of MacCormick’s emphasis on the pervasiveness of norms—though he did not himself fully develop this point—is that legislators and law enforcement personnel need always to be aware of the normative context into which their criminal-law-based intervention is inserted. This impli­ cation has been supported empirically by several research studies. For example, in social-psychological experiments in the Netherlands, it has been shown that people are more likely to behave in a prosocial manner if others around them are also doing so—even if the character of the two prosocial actions is different.38 Again, in a famous series of randomized controlled trials relating to domestic violence, led by Lawrence Sherman,39 police officers in various US cities were required to deal with (non-life-threatening) calls for assistance by a randomly predetermined response of either arrest of the domestic partner, or something short of arrest. The results showed that the same intervention (arrest) led to different results in different kinds of community with different normative structures.40 Such variations of effect are not yet fully understood. But a theoretical concept that has been derived from work of this kind is of great interest in the context of MacCormick’s theorization; this is the concept of ‘control signals’, one of a trinity of linked ideas developed by Martin Innes, the others being ‘signal crimes’ and ‘signal disorders’.41 Rooted in the sociological theory of symbolic interactionism,42 the central insight behind this conceptualization is that certain acts communicate (‘send signals’) to the general public, in a way that may either hinder or promote the general sense of order in a given social context; for example, an abduction of a child from a public park will send an immediate ‘signal’ about the dangers of such locations. As regards ‘control signals’, the implication is that relevant authorities 38  K. Keiser, The Spreading of Disorder, Proefschrift, University of Groningen. (Available online at .) 39  L. W. Sherman, Policing Domestic Violence (New York: Free Press, 1992). 40  Three cities (Minneapolis, Colorado Springs, and Miami) showed a deterrent effect for arrests, but in three others (Omaha, Charlotte, and Milwaukee) arrests produced a ‘backfiring’ effect—that is, arrestees had more subsequent violence than controls. At a community level, the strongest single difference between ‘deterrence’ and ‘backfiring’ cities was that in the former 36 per cent of all suspects were black, but in ‘backfiring’ cities that proportion rose to 63 per cent (Sherman, Policing Domestic Violence, 18). Thus, using arrest on a domestic violence suspect seemed to produce better effects in cities that were less socially disadvantaged. 41  ‘A control signal is an act of formal or informal social control that functions to communicate a message about the presence or absence of security mechanisms’: M. Innes, S. Hayden, T. Lowe, H. MacKenzie, C. Roberts, and L. Twyman, Signal Crimes and Reassurance Policing (Guildford: University of Surrey, 2004), viii. 42 On symbolic interactionism see H. Joas and W. Knöbl, Social Theory: Twenty Introductory Lectures (Cambridge: Cambridge University Press, 2009), ch. 6.

Civil Peace and Criminalization

241

can sometimes successfully send a ‘signal’ indicating that matters are under control, thus calming the responses of people in the situation. A well-documented example of this process has been reported from Chicago, where senior managers at the University of Chicago became concerned by the potential damage that might be done to the university’s longer-term prospects by the high crime rate that had developed in the local area surrounding the campus. In response, university managers not only invested heavily in the local urban infrastructure (including obtaining substantial Federal urban renewal funds), they also introduced certain initiatives that directly addressed ‘safety in public space’ issues—such as a private security force (with direct links to the Chicago Police Department), 24-hour ‘safety buses’, and emergency telephones. This whole package of measures seemed to send a strong ‘control signal’; for while crimes such as burglary remained high, after the improvements the area was perceived by the residents as safe, and property values rapidly appreciated—neither of which phenomena are normally found in high crime areas.43 Drawing on this Chicago research, and on our own much smaller study in four sub-areas of Sheffield, Andrew Wilson and I subsequently concluded that the weight of the sociological evidence pointed to the possibility that positive social change can, under the right conditions, flow from ‘action by the authorities which allows citizens to participate in their communities more fully, and to exercise informal social control more effectively’.44 In principle, therefore, it seems clear that an act of criminalization, backed by congruent social practices, can act as an effective ‘control signal’ (although, of course, this sociological conclusion is not in itself a justification for criminalization). Relatively few contemporary legal theorists have shared MacCormick’s interest in incorporating this kind of deep awareness of sociological realities into their philosophies of law.45 But one legal theorist of an earlier generation, Karl Llewellyn, certainly did so, and we can still learn from some features of his work. In his famous 1940 article, ‘The Normative, the Legal and the Law-Jobs’, Llewellyn postulated that every human group, small or large, formal or informal, needs to formulate ways of dealing with some basic problems if it is to remain stable as a group.46 Because these processes are of the sort that in developed organizations are part of a legal system, Llewellyn referred to them as ‘law-jobs’, although he knew very well that in pre-modern societies and contemporary informal groups they are not described as ‘legal’. Four ‘law-jobs’ are, he suggested, of universal relevance for

43  R. P. Taub, D. G. Taylor, and J. D. Dunham, Paths of Neighborhood Change: Race and Crime in Urban America (Chicago: University of Chicago Press, 1984), 19–23 (crime and housing data) and 96–102 (area description). 44  A. E. Bottoms and A. Wilson, ‘Civil Renewal, Control Signals and Neighbourhood Safety’, in T. Brannan, P. John, and G. Stoker (eds.), Re-Energizing Citizenship: Strategies for Civil Renewal (Houndmills: Palgrave Macmillan, 2007). 45 But see W. Twining, General Jurisprudence: Understanding Law from a Global Perspective (Cambridge: Cambridge University Press, 2009); B. Z. Tamanaha, A General Jurisprudence of Law and Society (Oxford: Oxford University Press, 2001). 46  K. Llewellyn, ‘The Normative, the Legal and the Law-Jobs: The Problem of Juristic Method’, Yale Law Journal, 49 (1940), 1355.

242

Anthony Bottoms

the functioning of human groups;47 these are set out below, with examples that I have added from my running example of a road traffic system. Preventive Channelling and Re-channelling. The term ‘preventive channelling’ refers to the ‘ordering’ mechanisms within any group, intended to help the group to achieve its primary purposes (e.g. growing crops; educating students), and to do so with a minimal level of conflict: in other words, social arrangements designed to help people to go about their daily business with a minimum of trouble. As may be deduced from Fortes’s comments, quoted above, societal rules—both informal and formal—have ‘preventive channelling’ as a primary objective. Accordingly, in a road traffic context we find rules about the need to reach a certain standard of competence before one may drive alone; the designation of some streets as open only for one-way traffic, with a rule prohibiting driving in the other direction; and so on. But preventive channelling can also be achieved by means other than rules—for example, the replacement of a single carriageway with a dual carriageway on a particular stretch of road will normally have the doubly desirable effect of both cutting average journey times and reducing the accident rate. Where there has been a major disruption to normal routines (for example, a multi-vehicle fatal crash on a motorway) processes of preventive re-channelling are quickly set in place (e.g. temporary road closure, diversions, working to clear the affected carriageway) to ensure that life proceeds as normally as possible during the disruption; that the disruption is as short as possible; and perhaps in due time that modifications (e.g. smoothing out a curve in the road) are made to prevent a recurrence. The Disposition of Trouble-Cases. However good may be a group or society’s preventive channelling, sooner or later some sort of incident or ‘trouble’ will arise (e.g. a collision between two vehicles with each driver blaming the other). In such circumstances, groups/societies need some method of resolving that particular incident—although such methods can and do vary hugely, even in pre-modern societies.48 In contemporary societies, ‘disposition of the trouble case’ can of course vary from an informal resolution of the ‘trouble’ (for example, between neighbours, or between two drivers in a minor road collision) to prosecution and imprisonment. In other words, the use of the criminal law is one possibility for dealing with certain kinds of trouble-case; but other possibilities also exist, and societies may differ in the way that a particular kind of incident is responded to. That opens the way, of course, for normative debates about the optimum way of handling a given type of trouble-case. ‘The Say’. In all groups, it is necessary to reach some resolution of the question: who has ‘the say’—that is, who makes the rules, and who decides how ‘trouble-cases’ are to be dealt with? This is what Llewellyn describes as the

47  Llewellyn, ‘The Normative, the Legal and the Law-Jobs’ (n. 46), 1373. Later, Llewellyn also discussed ‘the fifth law-job’, which he called the job of ‘juristic method’. This arises only after the emergence within a group of ‘the perceptibly “legal” as a body of ways, people and ideology’, and it focuses on ‘the ways of handling “legal” tools to law-job ends, and of the ongoing upkeep and improvement of both ways and tools’: Llewellyn, ‘The Normative, the Legal and the Law-Jobs’, 1392, emphasis in original. 48 Roberts, Order and Dispute (n. 26). 

Civil Peace and Criminalization

243

‘constitutional law’ of human groups, although of course deciding ‘who has the say’ does not have to be a formal process. (For example, if a group of students have decided to share a house together for an academic year, they will have no formal constitution, but they will still need to decide who has ‘the say’ in setting house rules and settling disputes.) ‘Net Drive’. This fourth law-job is described by Llewellyn as ‘the Whither of the net Totality’,49 by which he means the overall policy or ideology that is shaping (or is helping to shape) the way that the first three law-jobs are actually accomplished in a given society or group. Thus, for example, some societies might place more emphasis on speed, and others on safety, in the ‘Net Drive’ of their road traffic laws. More fundamentally, those Muslim countries that base their criminal code to a large extent on the traditions of the Shari’a obviously work with a very different Net Drive than that typically encountered in secular liberal-democratic states.50 It is therefore important to make clear that in the remainder of this chapter the focus is limited to the latter type of state. Of course, for criminal lawyers the main interest of Llewellyn’s typology rests on the first and second of these ‘law-jobs’, although the others are also of relevance (as we shall see later in relation to ‘Net Drive’). In Llewellyn’s terms, in a developed society the formal provisions of the criminal law and the daily ­routines of policing, together with the symbolic effects of these ‘control signals’, can be seen as institutionalized ways of encouraging ‘preventive channelling’. Clearly also, ‘the disposition of the trouble case’ is one of the central functions of various agencies of the criminal justice system—police, prosecutors, courts, prisons. MacCormick does not mention Llewellyn in his chapter on criminal law, but the law-jobs analysis helps us to grasp more concretely what MacCormick had in mind in claiming that the criminal law is above all to be seen as a way of pro­ mo­ting ‘civil peace’.51 However, as both Llewellyn and MacCormick fully recognized, there are many other social mechanisms that contribute to social peace—a matter that must be constantly borne in mind in conducting analyses of the functioning of the criminal law.

49  Llewellyn, ‘The Normative, the Legal and the Law-Jobs’ (n. 46), 1387. 50  ‘Muslims . . . see in the [Shari’a] an all-embracing sacred law whose source is God’s will and whose purpose is to guide humanity to fulfilment in life. Its content is understood, classically, as governing and determining all areas of life—personal, communal, social, civil, political and cultic’: K. Cragg and R. M. Speight, The House of Islam (3rd edn.; Belmont, Calif.: Wadsworth, 1988), 43. For a discussion of Shari’a law and jurisprudence and its application in Saudi Arabia and Iran, see S. M. Shahidullah, Comparative Criminal Justice Systems (Burlington, Mass.: Jones and Bartlett, 2012), chs 9 and 10. 51 William Twining has argued that, although incomplete as an analytical theory of law, the law-jobs theory can and does provide a ‘useful heuristic for analysing the internal ordering of institutions and groups’ (Twining, General Jurisprudence (n. 45), 104). He also makes the more unusual claim that the law-jobs theory is ‘not falsifiable’ (Twining, General Jurisprudence, 107). In my view, this claim is theoretically incorrect but actually correct. How so? It is theoretically incorrect because it is conceptually possible to imagine—for example—a society in which everyone follows the rules for preventive channelling and therefore no trouble-cases ever arise. However, given the actual characteristics of humans as they have evolved (see earlier discussion), the theory will never be falsified.

244

Anthony Bottoms

III.  Issues Relating to Criminalization: Setting the Scene We are now in a position to begin to address the principal criticism raised against MacCormick’s ‘civil peace’ thesis by the three contributors to the Edinburgh symposium; namely, that ‘civil peace’ is of little value as a criterion for considering which acts should, and which should not, be classified as criminal. To discuss this issue, I will give the greatest attention to the fullest of the three sceptical accounts, that by Magnus Ulväng.52 Before doing so, however, it seems important to consider how a legal-institutional theoretical framework might address the core concern of criminalization theory, namely (in Jareborg’s words) ‘where to draw the line between crime and non-crime’.53

A.  A legal-institutional approach to criminalization In his response to the papers in the Edinburgh symposium, Neil MacCormick explicitly took issue with those who argue that ‘the vital underpinning of any attempt to produce a comprehensive understanding of law must be a grand theory of justice’, and that ‘substituting for this an explanation or analysis of the concept of institutional legal order is opting for what is boring and rather obvious’.54 Although conceding that analyses of law and legal systems should ‘ideally [be] guided by some attractive overarching conception of legal justice’, MacCormick insisted that such analyses need also to take into account law’s ‘institutional aspect’; moreover, he claimed, no one has yet shown ‘that the explanatory (or, in the weak sense, ‘sociological’) elements in the explanation of [state law] are redundant or subordinate to the ideal element’.55 If this is correct, how should one study this ‘institutional aspect’ within jurisprudential analyses? In an essay written in 1990, MacCormick outlined a methodological approach that he called ‘rational reconstruction’,56 and at the Edinburgh symposium Victor Tadros shrewdly suggested that this method seems to have guided MacCormick in writing Institutions of Law, although this was not made explicit in that volume.57 So what exactly is ‘rational reconstruction’? According to MacCormick, a doctrinal legal scholar: is confronted by a vast body of material . . . [which] may seem confused or disorderly, partly or potentially conflicting, gappy in places. . . . The task of scholarship . . . is then to take these selected items and put them back together, to reconstruct them in a way that makes them comprehensible. . . . This requires explanatory principles . . . . In legal scholarship . . . rational 52  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7). 53  Jareborg, ‘A Comment on Personality and Corporate Crime’ (n. 6), 125. 54  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 187–8. 55  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 189–90. 56  N. MacCormick, ‘Reconstruction after Deconstruction: A Response to CLS’, Oxford Journal of Legal Studies, 10 (1990), 539. 57 V. Tadros, ‘Institutions and Aims’, in Del Mar and Bankowski (eds.), Law as Institutional Normative Order (n. 4), 84.

Civil Peace and Criminalization

245

reconstruction means the production of clear and systematic statements of legal doctrine . . . . [This] is sometimes dismissed as mere transcription . . . but this is a total misunderstanding . . . the great works of legal dogmatics, starting with Gaius’ Institutes, are among the higher achievements of human intelligence.58

Of course, in carrying out such analyses of legal texts, with a view to synthesis, the researcher may sometimes discern values or patterns underpinning particular practices or decisions even where these have not been overtly stated. When that occurs, the method of ‘rational reconstruction’ bears at least some resemblance to Foucault’s ‘archaeological’ method of ‘uncovering . . . discursive and nondiscursive practices’—which he famously used to try to explain, among other things, how in European societies it had become ‘possible, indeed natural, to speak of surveillance, re-education and training—words from the military and scholastic vocabulary—in the context of judicial punishment’.59 Tadros has criticized the method of rational reconstruction because, in his view, it ‘assumes that distinct elements of the institutions of law rest on distinctly valuable practices’.60 However, this view seems to be mistaken (as, indeed, the analogy with Foucault’s archaeological method suggests). Properly conducted, in legal scholarship rational reconstruction aims to bring out into the open all aspects of the working of given institutions—including both their explicit values and their implicit values, as embodied in practices. Once these issues have been clarified, one can then debate normatively whether the discerned rules, institutions, practices, and values are admirable or not. Of particular interest will be those rational reconstructions that yield insights not previously utilized within the scholarly field in question.61 How might this approach be relevant to debates about criminalization? Writing on this subject has predominantly been conducted within a framework of straightforwardly normative discourse, but it would seem that ‘rational reconstruction’ would require a subtly different approach. To illustrate how this might work, let me offer an example—that of the ‘public interest test’ for prosecutors, utilized in most common law jurisdictions.62 In a famous statement in the UK Parliament in 1951, the then Attorney General for England and Wales, Sir Hartley Shawcross, said that it ‘has never been the 58  MacCormick, ‘Reconstruction after Deconstruction: A Response to CLS’ (n. 56), 556–7. 59 T. Flynn, ‘Foucault’s Mapping of History’, in G. Gutting (ed.), The Cambridge Companion to Foucault (Cambridge: Cambridge University Press, 1994), 35. The reference is, of course, to M. Foucault, Discipline and Punish: The Birth of the Prison (London: Allen Lane, 1977). As Flynn points out, Discipline and Punish includes aspects of what Foucault describes as genealogical as well as archaeological method, but we need not pursue these niceties here. 60  Tadros, ‘Institutions and Aims’ (n. 57), 84. 61  MacCormick did not envisage rational reconstruction as a method to be used in field research. However, there seems to be no reason in principle why this should not be an appropriate extension of the original concept. Natalia Vibla of the Cambridge Institute of Criminology has been pursuing research of this kind by interviewing Swedish judges about the principles they apply when sentencing a single offender for multiple offences. 62  Some civil law jurisdictions alternatively use what is described as the ‘legality principle’, whereby every offence should be prosecuted if a prosecutor considers that she or he has sufficient evidence to secure a conviction.

246

Anthony Bottoms

case in this country—I hope it never will be—that suspected criminal offences must automatically be the subject of prosecution’; rather, even if it is clear that an offence has been committed, the prosecutor must ask him or herself whether it ‘appears that the offence or the circumstances of its commission is or are of such a character that a prosecution . . . is required in the public interest’.63 Since the establishment of the Crown Prosecution Service (CPS) in England and Wales in 1986, prosecutors’ decisions have been guided by an official Code for Crown Prosecutors,64 and successive editions of this Code have always stipulated that the criterion for prosecution is a dual one—each case prosecuted must pass both an ‘evidential sufficiency’ test and a ‘public interest’ test. If, therefore, the Code (or other formal CPS guideline)65 indicates that a particular type of case might regularly not be prosecuted despite the presence of sufficient evidence to secure a conviction, in legal-institutional terms this has to count as ‘non-crime’ rather than ‘crime’ (to use Jareborg’s dichotomy) notwithstanding that it is defined as ‘crime’ by the formal criminal law. Thus, applying careful ‘rational reconstruction’ to the kinds of cases that are recommended or not recommended for prosecution in the Code (or other CPS guideline) might give us some hints as to underlying institutional and cultural views about what is appropriately dealt with as ‘criminal’, and these might (but will not necessarily) yield some valuable ethical insights. In the present instance, one such insight might be embedded within the simple fact that any prosecution is required to be ‘in the public interest’—a choice of phrase that is worth careful attention. (Why, precisely, is the ‘public interest’ relevant?) In addition, within the latest version of the English Code for Crown Prosecutors, there are a number of more detailed points of guidance that might merit ethical reflection: for example, that prosecution is more appropriate where the victim ‘was at the time a person serving the public’, or where there existed ‘a position of trust or authority . . . between suspect and victim’; and, in the case of suspects under 18, an instruction that prosecutors should consider ‘whether a prosecution is likely to have an adverse impact on [the accused’s] future prospects that is disproportionate to the seriousness of the offending’.66 I shall return to these points later.

B.  Ulväng on MacCormick We turn now to Ulväng’s analysis of MacCormick’s chapter on the criminal law. At an early point in this analysis, Ulväng states that ‘MacCormick’s reasoning initially seems indisputable, but it rests on the assumption that Locke is right in his postulation that there is a “universal right to punish” ’.67 This seems at first sight to

63  HC Deb. vol. 483, col. 681 (29 January 1951). 64  Promulgated under the Prosecution of Offences Act 1985, s. 10. 65  Such as that promulgated in respect of assisted suicide in response to a request by the Appellate Committee of the House of Lords: see Policy for Prosecutors in Respect of Cases of Encouraging or Assisting Suicide (Crown Prosecution Service, 2010). 66  Code for Crown Prosecutors (7th edn.; Crown Prosecution Service 2013), para. 4.12(c) and (d). 67  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 131.

Civil Peace and Criminalization

247

be a rather abstruse point, but in my view it lies at the heart of the disagreement between Ulväng and MacCormick, so it requires careful consideration. MacCormick’s comment on this point is as follows: Whatever else is credible or otherwise in Locke’s depiction of the ‘state of nature’, he is surely convincing on the topic of a universal right to punish. If everyone were in the position of having to defend her or his own rights against all comers, and having to take her/ his own steps to secure some remedy from, or to punish, anyone who invaded their rights, the prospects for civil peace would be slim. . . . The comparatively modern development of a specialized body of criminal law, backed up by organized police forces . . . has been a condition of a progressively greater civility in civil society.68

It is evident from this passage that MacCormick does not consider that there will be adequate ‘civility’ in any ‘state of nature’ where everyone has a ‘right to punish’. Therefore, although he says that Locke’s view of the universal right to punish is ‘convincing’ (though he does not make clear whether he means descriptively or normatively convincing), he certainly does not normatively endorse the social conditions that will result if there is a full exercise of such ‘rights’. Ulväng’s response is that he is ‘reluctant to accept’ the universal right to punish as a normative claim.69 More surprisingly, and by contrast to MacCormick, he takes this concept out of the framework of the ‘state of nature’ and applies it to contemporary societies: ‘[j]‌ust because we live in Scotland or Sweden, we should never assume that we have a universal right to punish . . . such a starting-point can be very dangerous’.70 Presumably, the assumption here is that if there is a ‘universal right to punish’ in the state of nature, then this is taken over by the state in contemporary societies, and the state could in principle use this right in a dangerously authoritarian fashion. This interpretative disagreement reveals, I believe, some fundamentally different assumptions about the nature of the criminal law. Thus, Ulväng considers that criminal law is ‘purely repressive’,71 and it also becomes clear that he is deeply sceptical about state punishment as a social institution. In his words: [The criminal law] does not seek to solve conflicts or promote welfare or social values. The repressive character of criminal law and punishment gives its practice a negative character, and it is acknowledged by many criminologists and criminal lawyers that punishment is not a particularly efficient way of solving conflicts or maintaining peace and stability. On the contrary, punishment (and especially imprisonment) actually causes instabilities in societies, and it is open to dispute whether it is a fair method for upholding a fair society.72

MacCormick’s view is, as we have seen, very different. For him, the criminal law is not ‘purely repressive’, nor does it have only a ‘negative character’ (though it certainly has repressive and negative elements). For MacCormick (as for Llewellyn) 68 MacCormick, Institutions of Law (n. 1), 207. 69  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 131. 70  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 141. 71  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 134. 72  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 134.

248

Anthony Bottoms

use of the criminal law can be a way (though not the only way) of solving conflicts; additionally, through its indirect normative work as a ‘control signal’, it can be, and has historically been, a way of promoting civility.73 By contrast, Ulväng considers that ‘punishment causes instabilities in societies’, and there is little suggestion that it might bring any social benefits. Sociologically, it seems clear that MacCormick’s wider view of the criminal law has the greater empirical and theoretical support, and is therefore to be preferred. Ulväng’s approach, however, has the merit of reminding us that to criminalize an action involves punishing (at least some of) those who commit that act, and we therefore need always to be sure that such punishment is morally justified. Fortunately, it is possible to incorporate these important insights within MacCormick’s broader theorization. Turning to the detail of Ulväng’s paper, an important feature of his analysis is a claim that theorists of the criminal law need always to maintain a ‘clear distinction’ between three crucial questions: ‘(i) why do we punish (at all)?; (ii) what do we (ought we to) punish?; (iii) how are we to punish (in terms of rationales for sentencing)?’ These questions, Ulväng comments, ‘reveal a multi-layered normative structure, and different types of reasons are relevant on each level’.74 As his argument develops (in sections 2 and 3 of his paper), it becomes clear that, for him, ‘civil peace’ is a partial (but only a partial) answer to the first question, but it does not help us with the second. The reasons for these conclusions need brief elaboration. As regards the first question (‘why do we punish at all?’), Ulväng agrees that MacCormick is ‘certainly correct’ to stress the importance of ‘peaceful civilian coexistence’ as an ‘overall aim for the organisation of society’, and that achieving this will require, amongst other things, ‘coercive measures such as punishment’. But he also emphasizes that ‘the justification for criminal law is much more complicated than simply concentrating on a single- though necessary—aim’. Perhaps, therefore, the concept of ‘civil peace’ might be regarded as ‘an “umbrella concept” which could serve as a necessary [but not sufficient] ulterior condition’ for criminal legislation. Even so, Ulväng is concerned that without clear ‘substance’, the concept of civil peace could become vacuous, and ‘thus subject to possible exploitation by either unlimited utilitarianism or legal moralism’.75 This is therefore at most a 73 MacCormick, Institutions of Law (n. 1), 207 suggests that ‘the comparatively modern development of a specialized body of criminal law . . . has been a condition of a progressively greater civility in civil society’, thus helping to overcome the social problems evident from historical accounts of ‘blood-feuds in clan based societies, and . . . “frontier justice” during the westward expansion of the United States or during the wild days of the Anglo-Scottish border’. Other research suggests that the political legitimacy of the criminal law has been of central importance in achieving effects of this kind. Studies of the long-term decline in homicide rates over several centuries in Europe and the United States have concluded that this decline cannot be fully explained either by ‘economic development [or] the rise in the coercive power of the state’; rather, ‘it seems that on both sides of the Atlantic change in the legitimacy of the political and social order was a major force behind change in the likelihood that men would fight each other in public spaces’: M. Eisner and A. Nivette, ‘Does Low Legitimacy Cause Crime? A Review of the Evidence’, in J. Tankebe and A. Liebling (eds.), Legitimacy and Criminal Justice (Oxford: Oxford University Press, 2013), 310. 74  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 132 (emphasis in original). 75  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 136.

Civil Peace and Criminalization

249

partial endorsement of the relevance of ‘civil peace’ to debates about the purpose of the criminal law, coupled with an important warning about the dangers of socially authoritarian applications of this concept. As regards the second question (‘what ought we to punish?’), given Ulväng’s general view of the nature of criminal law (see above), it is not surprising that, in his opinion, when we are considering issues of criminalization ‘we ought to begin with the normative element and ask why we think it is appropriate to intentionally inflict pain or suffering on someone’.76 It follows both that the criminal law should be used sparingly, and that a ‘certain amount of “moral wrongness” becomes a necessary condition of criminalisation’.77 When making decisions about criminalization, policymakers should bear in mind that the criminal law is ‘perhaps first and foremost an institution for distributing instrumental censure within a regulated practice of sentencing’.78 Within such a framework, Ulväng’s contention is that the concept of ‘civil peace’ has little purchase. Although acts of mala in se such as murder or rape do indeed violate civil peace, this ‘tells us little about why such deeds are blameworthy’;79 and as regards mala prohibita the concept of civil peace is even less helpful. In the final sentence of his paper, Ulväng also suggests that ‘the concept of civil peace is too nebulous to provide any guidance’ in the context of specific debates about criminalization in a given legal system.80 These are strong criticisms; we shall have to examine them in the sections to follow.

IV.  Criminalization and Civil Peace: Core Issues This section takes forward the discussion initiated in Section III.B. I shall initially focus on Ulväng’s positive answer to the second of his three core questions (‘what ought we to punish?’), and then on his proposed ‘multi-layered normative structure’. An alternative approach to criminalization (the so-called ‘crimes as public wrongs’ thesis) will subsequently be considered; and finally I shall offer an empirical example of criminalization processes.

A.  Criteria for criminalization I: wrongfulness and harm To convict a citizen of a crime is to impose a degree of moral censure, so the concept of what Ulväng calls ‘moral wrongness’ is prominent not only in his analysis, but in most contemporary theories of criminalization, including those of Duff and of Simester and von Hirsch.81 (It is also clearly implicit within MacCormick’s ‘thesis of the moral substratum of criminal law’). Nevertheless, moral wrongness 76  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 141. 77  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 132. 78  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 133. 79  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 137. 80  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 141. 81 Duff, Answering for Crime (n. 9); Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10).

250

Anthony Bottoms

is not, within a liberal-democratic state, a sufficient ground for criminalization. To use Ulväng’s examples, wrong acts that are rightly not criminalized in such a state include ‘committing adultery . . . [or] misusing power to degrade or humiliate employees’.82 It follows from the above that we need an approach to criminalization that incorporates the criterion of ‘moral wrongness’, but which also provides clear normative grounds for distinguishing between ‘crimes’ and ‘other wrongs’. For Ulväng, the appropriate demarcation line is the well-known ‘harm principle’: ‘blamewor­ thiness is connected with harming a protected interest, and the justification for censure is derived from a moral principle that we ought not to harm others’.83 Thus, Ulväng’s essay prefigures, in skeletal form, the later and more fully argued view of Simester and von Hirsch that there are two key criteria for a theory of criminalization, namely moral wrongfulness and harm.84 However, Ulväng’s particular formulation (see previous paragraph) seems only to push the problem back one stage: for one can legitimately ask: ‘what constitutes a “protected interest” for this purpose, and why does a shoplifter violate such an interest while a cheating husband whose actions financially damage his wife does not?’ Perhaps this issue is soluble within the framework of a ‘harm and blameworthiness’ criminalization theory, though not all theorists think this is possible.85 I shall not pursue that topic here, but I will return to the ‘adulterer vs. shoplifter’ puzzle when considering the ‘crimes as public wrongs’ thesis (see Section IV.C). There is however one further, and vitally important, point to be made in relation to the harm principle. As Lindsay Farmer points out elsewhere in this volume, this principle is frequently discussed on the basis that the envisaged harm is simply a harm to an individual legal person.86 Yet, on closer examination, in the case of property offences a subtle social dimension has to be introduced into the argument. This can be shown by reference to Simester and von Hirsch’s illuminating discussion of the hypothetical case of D, who takes and keeps an old shirt that V, its owner, had intended to throw away, but had not yet actually binned. As a preliminary point, Simester and von Hirsch emphasize that the concept of property is ‘quintessentially legal . . . I don’t need the law to know that this arm is mine, but I do need the law to know that this is my table’.87 It is this legal-institutional dimension of property rights that determines how the case of the shirt is to be treated: [E]‌ven if misappropriating V’s old clothes is, in itself, harmless, the Harm Principle lies in the background. No doubt V’s material amenities are unaffected. But the property regime 82  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 138. 83  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 137. 84  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10). These authors also, following Joel Feinberg (The Moral Limits of the Criminal Law, 4 vols. (Oxford: Oxford University Press, 1984–8)) identify a separate ‘Offence Principle’, but Simester and von Hirsch additionally propose a ‘harm’ dimension within crimes justifiably identified through the Offence Principle. 85 Duff, Answering for Crime (n. 9), 136. 86 L. Farmer, ‘Criminal Law as an Institution: Rethinking Theoretical Approaches to Criminalization’, this volume. 87  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 41.

Civil Peace and Criminalization

251

supplies V with something else: the dominion over his belongings, including the shirt. It is he, not D, who may decide when an item has outlived its usefulness for him. . . . This possibility of dominion is itself valuable, and enhances the well-being of owners. Hence theft of the shirt is wrongful. . . . The connection to harm in such cases is indirect: absent some reason for treating D’s case as special (e.g. in circumstances of necessity or emergency), the widespread perpetration of such conduct could ultimately harm the operation of the [property] regime itself. 88

Two crucial conceptual moves are made in this passage. First, it becomes clear that the interest of V that is harmed is not the material loss of the shirt, but rather the concept of ‘dominion’ offered by the property law regime, which would be compromised if the law were to say that there is in this instance no theft. Secondly, in the final sentence of the passage issues of ‘civil peace’ are directly raised; it is suggested that if the property regime is not upheld, this could have adverse social consequences (i.e. there will be a diminution in civil peace).89 These intrigiung points require brief elaboration. As regards the first point (the concept of dominion), what is being referred to here is what is known to sociologists as social capital. ‘Social capital’ is a term used to refer to the intangible resources or assets that are available to an individual through the effective functioning of a social system, or a local group or community.90 For example, a family moving to a house in an area with what criminologists call ‘high collective efficacy’91 can expect (other things being equal) to enjoy a lower crime victimization rate than if they had moved to an area with lower collective efficacy.92 In the present case, the assumption is that the property law regime provides valuable social capital to people in a given society. As Simester and von Hirsch put it, this regime ‘serves our well-being; it provides a reliable means by which we can pursue a good life, through the voluntary acquisition, use and exchange of resources’;93 in this sphere of life, it therefore provides the regularity and rule that we previously identified as necessarily implicated in social order. Hence, a deliberate breach of this dominion is itself an indirect harm, even though 88  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 42 (emphasis added). 89  Although the authors also rightly note that the property regime itself requires a higher-level moral and political justification: Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 43. 90  Social capital is therefore to be distinguished from ‘human capital’, which refers to the skills available to a given group through the training and/or experience of individuals within the group. For an overview of research on social capital see D. Halpern, Social Capital (Cambridge: Polity Press, 2005). 91  ‘Neighbourhood collective efficacy’ is a concept with two main components: first, a willingness by local residents to intervene for the common good in specified situations (for example, to stop children spray-painting graffiti on a wall); and secondly, background conditions of working trust and mutual support (without which interventions are less likely). See R. J. Sampson, ‘Collective Efficacy Theory’, in F. T. Cullen, J. P. Wright, and K. R. Blevins (eds.), Taking Stock: The Status of Criminological Theory (New Brunswick, NJ: Transaction Publishers, 2006). 92  Although the specific concept of ‘social capital’ is of recent origin, the phenomenon it is describing is not. See e.g. the previously quoted suggestion by Robert Hinde (Bending the Rules (n. 29), 36) that, in prehistory, ‘groups with a higher proportion of individuals who behaved prosocially to, and co-operatively with, other in-group members tended to do better than those containing many selfishly assertive individuals’. 93  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 41.

252

Anthony Bottoms

V suffers no actual loss. If this argument is correct, then—contrary to the views of the Edinburgh symposiasts—a specific piece of potential criminalization (based on indirect harm) owes its justification directly to considerations relating to civil peace.94

B.  A multi-layered structure for understanding criminalization Ulväng is well aware that there is an illustrious precedent in criminal law theory for a multi-layered normative structure of the kind that he proposes95—namely H. L. A. Hart’s theory of punishment, with its distinction between the General Justifying Aim of punishment, and the further questions ‘how much should we punish?’ and ‘who should we punish?’96 Famously, Hart adopted a utilitarian justification as the General Justifying Aim, but a non-utilitarian justification for the ‘who’ question, based on the values of fairness and liberty. Ulväng similarly argues, in respect of criminalization theory, that ‘different types of reasons are relevant at each level’.97 This approach, however, has some potential difficulties. As John Tasioulas has argued, if ‘proponents of hybrid theories can offer no coherent rationale for the principles they have combined . . . apart from the fact that they have been explicitly manufactured to yield results more attuned to our settled moral convictions’, then ‘the theory arrived at [may be] little better than an ad hoc compromise among radically disparate concerns’.98 In light of Tasioulas’s critique, the key question therefore becomes: within criminalization theory, if different types of reason apply at the different levels, as Ulväng suggests, can these disparate concerns be coherently related? To answer this question, I shall again turn primarily to the theory of Simester and von Hirsch, which as we have seen can be regarded as similar to (though more developed than) that of Ulväng as regards its core principles of criminalization.99 A key feature of these authors’ discussion is that it partially alters the framework of the criminalization debate from that proposed by Ulväng. Consistently with his emphasis on the ‘negative’ and ‘repressive’ character of criminal law, the three questions within Ulväng’s ‘multi-layered normative structure’ all concern punishment; but for Simester and von Hirsch, that is an inappropriate sole focus. According to them, in discussing criminalization we need to move back a stage, because the act of criminalization itself does not per se involve punishment, but is an ex ante 94  The term ‘potential criminalization’ (rather than simply ‘criminalization’) is used to acknowledge the possibility that there might be additional ‘mediating considerations and constraints’ (Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), ch. 11) that should cause policymakers to draw back from criminalization in cases of this kind. See also N. Peršak, Criminalising Harmful Conduct (New York: Springer, 2007), 92–4, on the ideal of a ‘three-step criminalisation process’, the last step of which would be a ‘pragmatics filter’ concerned with issues such as the financial and social costs of criminalization, and the practical feasibility of enforcement. 95  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7) , 132 n. 96  H. L. A. Hart, Punishment and Responsibility (Oxford: Oxford University Press, 1968), ch. 1. 97  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 132. 98  J. Tasioulas, ‘Punishment and Repentance’, Philosophy, 81 (2006), 279 at 281. 99  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10).

Civil Peace and Criminalization

253

(before the act) ‘declaration that designated conduct is, so far as the state is concerned, wrongful and should not be done’.100 That ex ante stage is then followed by the two ex post (after the act) stages of conviction and punishment. In developing their argument, Simester and von Hirsch show that they are more favourable than is Ulväng to the inclusion of the concept of civil peace within debates on criminalization. Thus, they claim that there are some explicit links between ‘the principal reasons for the state’s existence’ and its ‘involvement in legal regulation’ (both criminal and civil).101 Indeed, they continue: . . . one may even describe it as a defining role of the state to facilitate peaceable co-existence among citizens, and to safeguard the basic means by which citizens can live good lives. A paradigm mechanism for doing so is through the issuance of prohibitory rules, with sanctions attached as a disincentive to the behaviour.102

This formulation for the ex ante stage is clearly very close to that of MacCormick. However, Simester and von Hirsch also very strongly affirm the role of what Ulväng calls ‘moral wrongness’ in relation to the two ex post stages of criminalization.103 This use of different criteria at the different stages therefore potentially exposes them to Tasioulas’s critique. How do they deal with this? According to Simester and von Hirsch, the ex ante declaration within a process of criminalization may be read also as a ‘conditional threat’, which treats its addressees as rational agents, able to ‘evaluate the conditionally threatened consequence [and] to decide for themselves upon compliance’ or otherwise.104 This form of ‘rational coercion’ necessarily operates ex ante, which is ‘why it is mistaken to focus (as retributivists tend to do) upon the imposition of punishment when seeking to justify the coercive nature of the criminal law’.105 However, retributivism (desert) is, in the view of Simester and von Hirsch, certainly not irrelevant when it comes to the ex post stage of punishment. Moreover, for them the fact that it is deserved punishment that is threatened in the ex ante stage of rational coercion is crucial to ‘the moral legitimacy of the act of coercion’, because it means that while the rational coercion ‘is still a threat, it is a moral threat’.106 Thus, a coherent moral justification for a multi-layered normative structure is indeed possible. Separation between the ex ante and the ex post stages creates moral space for a hybrid theory because it allows other rationales, in addition to punishment-based ones, to be considered at the ex ante stage of criminalization; but the various stages 100  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 5. Simester and von Hirsch’s definition of ‘crimes’ is restricted to formal promulgations in criminal codes, or in statutes or leading cases. However, the ex ante theorization they have developed can equally be applied to formally published prosecutorial guidelines, which in this chapter I have included within a legal-institutional approach to criminalization. 101  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 17. 102  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 17. 103 Simester and von Hirsch, Crimes, Harms and Wrongs, passim. See also A. von Hirsch and A. Ashworth, Proportionate Sentencing: Exploring the Principles (Oxford: Oxford University Press, 2004). 104  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 8. 105  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 8. Note that this critique applies also to Ulväng’s punishment-focused discussion of criminalization, as previously described. 106  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 8.

254

Anthony Bottoms

are linked in a principled way, thus avoiding the danger of what Tasioulas called an ‘ad hoc compromise’.107 The ex ante/ex post theorization developed by Simester and von Hirsch is also interesting because it helps us to understand more clearly what typically happens when a fresh criminal offence is suggested (usually by a politician, a review group, or an interest group). Such a proposal will usually be made because of some perceived ‘gap’ in the law, leading to a deficiency in the adequate regulation of a particular sphere of social life—for example, the control of dangerous dogs, or internet security, or anti-social behaviour in local neighbourhoods. The initial demand will typically be for what Simester and von Hirsch call a ‘declaration that designated conduct is, so far as the state is concerned, wrongful and should not be done’108—i.e., the ex ante stage of criminalization. This then allows scope for a debate about some of the issues that Ulväng rightly raises, such as whether it is actually appropriate to visit state punishment on those who commit such an act.109 A key point to note is that issues of social order or ‘civil peace’ are naturally part of this discussion, even if at the end of the day they are overridden by other considerations; moreover, the terms of such a debate are subtly different from, and more specific than, Jareborg’s suggestion that questions of civil peace are relevant only to the question ‘why we have a system of criminal justice at all’.110 This will hopefully become clearer when we consider an empirical example of criminalization later in this section.

C.  Criteria for criminalization II: crime as public wrong Let us now return to a puzzle raised earlier in this section: why, within a liberal-democratic state, do theorists of the criminal law consider that adultery causing financial loss is ‘not the criminal law’s business’ (to adopt a phrase f­requently used by Duff),111 while shoplifting or theft from the person is—notwithstanding that in a particular case the emotional and financial harm suffered by the wronged spouse might be significantly greater than that experienced by the victim of the theft? In my view, the most convincing answer to this problem argues that (absent additional complications such as an assault within the marital dispute) adultery is a private matter between the spouses, while theft is a public matter involving the community at large, and therefore (in the modern context) the state.

107  Tasioulas, ‘Punishment and Repentance’ (n. 98), 281. 108  Simester and von Hirsch, Crimes, Harms and Wrongs (n. 10), 5. 109  Debates of this kind have been prominent in the sphere of neighbourhood anti-social behaviour. Such behaviour frequently causes anxiety among residents, but whether those who commit such acts deserve to be convicted of a crime is a contested and difficult matter. For a discussion see: A. E. Bottoms, ‘Incivilities, Offence and Social Order in Residential Communities’, in A. von Hirsch and A. Simester (eds.), Incivilities: Regulating Offensive Behaviour (Oxford: Hart, 2006). 110  Jareborg, ‘A Comment on Personality and Corporate Crime’ (n. 6), 125. 111  As pointed out by M. M. Dempsey, ‘Public Wrongs and the “Criminal Law’s Business”: When Victims won’t Share’, in R. Cruft, M. H. Kramer, and M. R. Reiff (eds.), Crime, Punishment and Responsibility: The Jurisprudence of Antony Duff (Oxford: Oxford University Press, 2011).

Civil Peace and Criminalization

255

In its contemporary guise, the ‘crimes as public wrongs’ thesis originated in a 1998 paper by Marshall and Duff.112 In a more recent restatement of this thesis, however, Duff113 usefully reminds us that the general concept of crime as public wrong was centrally raised in the writing of the great eighteenth-century English jurist William Blackstone. Blackstone distinguished between private wrongs, which infringe ‘the civil rights which belong to individuals, considered merely as individuals’, and crimes, which in his view ‘are breach and violation of the public rights and duties, due to the whole community, considered as community, in its social aggregate capacity’.114 There is, however, a potential difficulty with this kind of approach. As Marshall and Duff point out, what most people immediately think of as archetypical ‘crimes’ are serious acts against individuals—examples would be murder, rape, or the burglary of a dwelling house.115 Since such crimes are committed against individuals or households, why are they ‘public wrongs’? The answer, they argue, cannot be because such actions: ‘threaten the social order’ or . . . cause ‘social volatility’; or because such attackers must be punished in order to incapacitate them from further such attacks or in order to deter other potential attackers . . . [for] such accounts . . . seem to subordinate the individual victim (a concern for their good, or for the wrong done to them) to some supposedly larger social good.116

Accordingly, and importantly, accounts of this kind significantly misstate the nature of the wrongfulness of the conduct, which consists above all in ‘the wrong [the offender] does to the individual victim’.117 However, Marshall and Duff argue that one can avoid this error, and still retain the concept of ‘crime as public wrong’, by saying that an act such as a serious assault is rightly classified as a crime ‘because it properly concerns the public, i.e. the polity as a whole’.118 Interestingly, MacCormick offered a very similar formulation, though without specific reference to the Marshall/Duff thesis and largely en passant: ‘a “public wrong” is a wrong that it is right for the public to be concerned about . . . [it] is not one that is done to the public’.119 But why do acts such as rape or residential burglary ‘properly concern’ the public as a whole? For Marshall and Duff, the answer to this question is ultimately communitarian: ‘the wrong does not cease to be [the victim’s] wrong, but it is also “our” wrong insofar as we identify ourselves with her . . . the attack on this individual victim is also an attack on us—on her as a member of the group and on us as fellow members’.120 But of course liberal theorists are notoriously suspicious of 112  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9). 113 Duff, Answering for Crime (n. 9), 51–2. 114  W. Blackstone, Commentaries on the Laws of England, vol. iv (Oxford: Clarendon Press, 1769), 5; emphasis added. 115  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9). 116  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9), 12. 117  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9), 12. 118 Duff, Answering for Crime (n. 9), 141, emphasis added. 119 MacCormick, Institutions of Law (n. 1), 212, emphasis in original. The specific sentence cited begins with the words ‘on this view’. 120  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9), 19–20.

256

Anthony Bottoms

communitarian claims, so it will be best to postpone comment on this claim until we have discussed liberal theory in Section V.A. Before proceeding further, we need to address a recent critique of the Marshall/ Duff thesis by Ambrose Lee.121 Among other things, Lee focuses attention on the non-trivial number of acts (such as assault and dishonest appropriation of another’s property) that are defined in most legal systems as both a crime and a tort or delict. How, he asks, can the same act be simultaneously something that it is ‘right for the public to be concerned about’ (when treated as a crime) and something that does not fall within that definition (when it is treated as a tort)? Pursuing this point, Lee argues that all torts, and probably breaches of contracts, are matters that it is ‘right for the public to be concerned about’, which is why the state makes available publicly funded facilities (the civil courts) for plaintiffs to pursue their alleged grievances. In light of this problem, however, Lee does not suggest that we should abandon talk of ‘crimes as public wrongs’, but instead argues that we should understand them not as ‘wrongs that the public should be concerned about’, but as ‘wrongs that should be punished by the state’.122 This is, he admits, in one sense ‘circular and unhelpful’.123 According to Lee, however, it has the merit of drawing attention to a rather different set of considerations, namely that: what is distinctive about criminal law is . . . the kind of response that it affords to wrongs, the role that the state should play in it, and the legal process that is appropriate to it; and all these are just as important as the nature and the kinds of wrongs that concern the public, when we are thinking about the nature of criminalization.124

Although Lee does not make this point (nor does he cite MacCormick), this conclusion has an interestingly ‘legal-institutional’ and sociological character. It is not, however, wholly original. For example, Marshall and Duff rightly emphasize the considerable conceptual significance of what non-institutionalists might regard as (in MacCormick’s words) some ‘boring and rather obvious’ facts,125 namely that ‘a “criminal model” [of procedure] puts the community (the state) in charge’, with a prosecution being ‘brought by Regina, the people or the State’.126 When considering the first element of Lee’s argument (leading to the claim that torts as well as crimes are public wrongs), it is instructive to look more closely at Blackstone’s text. In a dispute over the ownership of a field, Blackstone tells us, ‘it is immaterial to the public which of us is in possession of the land’; but— [T]‌reason, murder and robbery are properly ranked among crimes since, beside the injury done to individuals, they strike at the very being of society, which cannot possibly subsist, where actions of this sort are suffered to escape with impunity. In all cases the crime

121  A. Y. K. Lee, ‘Public Wrongs and the Criminal Law’, Criminal Law and Philosophy (published online May 2013). 122  Lee, ‘Public Wrongs and the Criminal Law’ (n. 121). 123  Lee, ‘Public Wrongs and the Criminal Law’ (n. 121). 124  Lee, ‘Public Wrongs and the Criminal Law’ (n. 121). 125  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 188. 126  Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n. 9), 15.

Civil Peace and Criminalization

257

includes an injury; every public offence is also a private wrong, and somewhat more; it affects the individual, and it likewise affects the community.127

In the context of the present discussion, what is of special interest about this statement is that Blackstone introduced a strong linkage between what in modern parlance would be called the concept of ‘crime as public wrong’ and the idea that the criminal law can help to promote civil peace and social order. (Note the claims that acts of the kind cited ‘strike at the very being of society’ and that social order ‘cannot possibly subsist’ if they are allowed to flourish). Given such conceptual underpinnings, Blackstone was in no doubt that the same act can be simultaneously a private wrong against an individual (who has been significantly harmed and wronged) and a public wrong against the community (since if left unchecked it will tend to weaken social order). This argument seems at first blush, however, to take us back to the beginning of this discussion. Does not Blackstone’s formulation, by emphasizing the threats to social order posed by major crimes, run the risk of misstating the nature of the wrong committed, and thus of downplaying the harm and hurt caused to the victim? Certainly, that will be the result if the claim that ‘every public offence is also a private wrong’ is read as functionally equivalent to the statement that ‘every public offence (= crime) is also a private wrong (= tort)’. But suppose we read it differently? Suppose we say (as, indeed, the wording of the first sentence of the above quotation might encourage us to say) that a criminal trial involves both a recognition of the wrong done to the victim, and a community concern and response, particularly because of the threat to social order posed by this kind of act? If we read a little further, we find that such an interpretation is congruent with Blackstone’s underlying philosophy of ‘wrongs’: Upon the whole we may observe that in taking cognizance of all wrongs, or unlawful acts, the law has a double view: viz., not only to redress the party injured . . . but also to secure to the public the benefit of society, by preventing or punishing every breach and violation of those laws which the sovereign power has thought proper to establish for the government and tranquillity of the whole.128

In further pursuit of these complex issues, it is perhaps instructive to consider an example from a precursor of a modern legal system. The social context to be described was one where, in sociological terms, the distinction between the individual and the collectivity was far less apparent than it is in modern liberal-democratic states; yet—as we shall see—when it came to responses to wrongdoing, that differentiation was of considerable importance.129 127 Blackstone, Commentaries on the Laws of England (n. 114), iv. 5. 128 Blackstone, Commentaries on the Laws of England (n. 114), iv. 7. Despite this philosophical underpinning, however, Blackstone’s text preserves the problematic procedural separation between public/criminal and private/tortious wrongs. 129  This example is introduced to assist with the conceptual argument. It is not claimed that all pre-modern social groups adopt similar dispute-resolution processes, for they do not: see Roberts, Order and Dispute, also A. E. Bottoms, ‘Some Sociological Reflections on Restorative Justice’, in A. von Hirsch et al. (eds.), Restorative Justice and Criminal Justice (Oxford: Hart, 2003).

258

Anthony Bottoms

The example is taken from an anthropological account by Joan Ryan of the Dogrib hunter-gatherer tribal group, who are part of the Dene people living in northwest Canada, close to the Arctic Circle.130 Social life for the Dogrib was organized into a number of local camps, each of which moved on from time to time; and each camp had a k’aowo (an assistant to the overall tribal leader), who was assigned a primary role in keeping order. (In Llewellyn’s terms, the k’aowo had a special ‘say’ within the camp.) If a member of the community did something that was considered to be a moderately serious wrong act, such as taking an animal from a trap set by another person, the k’aowo would speak some ‘harsh words’ to the perpetrator, and would ask him or her ‘to acknowledge the theft and to return the fur (or another of equal value) to the person from whom it had been stolen’.131 But if the perpetrator refused to comply with the k’aowo’s request in such a case, or if the act committed by the perpetrator seemed to the k’aowo to be so serious that he should not be dealing with it on his own, then the perpetrator was called before a ‘circle’, consisting of the whole Dogrib tribal group, presided over by the tribal chief (the yabahti) and senior men and women. After the offender had admitted guilt, ‘harsh words’ were spoken by the leaders, which usually included ‘reference to the harm done to individuals and/or the group’.132 The gathering then collectively determined the appropriate response. Banishment from the tribe was a possible, but very rare, sanction.133 More usually, attention turned to issues of apology, compensation, and the restoration of harmony. But although all this took place in a group context (and, clearly, in a highly pressurized social setting) the fact that an individual member of the tribe had been wronged was never forgotten: for example, after the decision about the response had been finalized, ‘the group demands that [the offender] face the victim, that restitution be made, that reconciliation start’.134 If we reflect on this account from the perspective of the ‘crime as public wrongs’ thesis, three points seem to be of special significance. First, as indicated by Marshall/Duff and by Lee, procedures are important; and in this community there were two distinct modes of ‘handling the trouble-case’—an everyday system of informal justice, facilitated by the camp k’aowo, but also a rarer and more formal procedure, of a quasi-criminal kind,135 involving the whole community.136 130 J. Ryan, Doing Things the Right Way: Dene Traditional Justice in Lac La Martre, NWT (Calgary: University of Calgary Press and Arctic Institute of North America, 1995). 131 Ryan, Doing Things the Right Way (n. 130), 34. 132 Ryan, Doing Things the Right Way (n. 130), 57. 133 Given the harsh environmental conditions in which the Dogrib lived (close to the Arctic Circle), banishment was effectively a death sentence. 134 Ryan, Doing Things the Right Way (n. 130), 91. 135 Ryan, Doing Things the Right Way (n. 130), at 90–1 points out that there are a number of similarities—though not, of course, total similarity—between the traditional Dogrib circle procedures and the modern Canadian criminal trial. 136  There is an interesting parallel here with early Roman Law. The Roman Law name for crimes was delicta publica; and apparently, the reason for this was not, as later jurists supposed, the ‘specially strong tendency of crimes to injure the public’, but rather ‘the fact that in early Rome all charges of crimes were tried by the public itself, i.e. by the whole Roman people assembled in comitia centuriata’ (C. S. Kenny, Outlines of Criminal Law (15th edn. rev. G. G. Phillips; Cambridge: Cambridge University Press, 1936), 5–6, emphasis in original).

Civil Peace and Criminalization

259

Clearly, this more formal ‘circle’ procedure was of great symbolic significance within the tribe’s social order. Secondly, within the circle procedure the Dogrib seem to have managed to devise a tribunal that successfully focused attention on (in Blackstone’s terms) both ‘public offence’ (including ‘civil peace’) and private wrong. In principle, it would seem that the criminal laws and procedures of a modern nation state ought to be able to achieve a similar result—though there is no doubt that in practice they have often failed to do so.137 Thirdly, it is noteworthy that one of the two reasons for resorting to the circle procedure was that the authority of the camp k’aowo had been challenged, which clearly suggests that the rarely used circle procedure was, at least in part, utilized as a kind of ‘longstop’ device to assist with the maintenance of local social order in the camps. There is therefore here a striking parallel with MacCormick’s understanding of contemporary criminal law as an ultimate guardian of trust and reciprocal relationships in everyday social relationships. The above account, following Blackstone, therefore suggests that there is a significant degree of convergence between the ‘public wrongs thesis’ and the ‘civil peace’ thesis. But is this convergence total? To examine this issue, it is useful to reflect—in the spirit of rational reconstruction—on the provisions of the English Code for Crown Prosecutors, as previously described. The first point to note is the routine and frequently unexamined use of the term ‘the public interest test’. Whatever the precise guidance given under that rubric (and in England this has changed from time to time in the quarter-century since the foundation of the CPS), from a legal-institutional point of view the very use of the term ‘public interest’ clearly suggests that some version of the ‘common good’ is deeply embedded in this legal system’s thinking about what should be criminalized. Accordingly, it is not too surprising that the current version of the Code identifies, for example, more serious offences and suspects with higher degrees of culpability as more likely to be prosecuted, for these can be regarded as the strongest violations of everyday trust and reciprocal relations.138 What is more interesting from a theoretical point of view are the previously noted suggestions in the Code that alleged offences against public servants, and offences involving breach of trust, are more likely to be prosecuted, because here we are more clearly moving into the territory of political sociology. (That is to say: (i) there is an assumed need for the criminal law to protect especially those who are serving the common good, as well as (ii) a need for the law to act against those who commit wrongs that involve breaches of the trust relations that are so vital to a functioning society.) In all these respects one can therefore discern, in the provisions of the Code, elements of thought that are closely linked to the concepts of both ‘public wrongs’ and ‘civil peace’. 137  Most notably, by the all-too-frequent neglect of the interests of the victim in the accusatorial procedures used in common law jurisdictions. See A. E. Bottoms and J. V. Roberts (eds.), Hearing the Victim (London: Routledge, 2011), especially chapters by Shapland and Hall; Reeves and Dunn; Roberts and Erez; and Bottoms. 138  Code for Crown Prosecutors (n. 66), para. 4.12(a) and (b).

260

Anthony Bottoms

Very importantly, however, the concept of the ‘public interest’ is, in the Code, not confined to issues relating to civil peace, and this is well exemplified in a further Code provision, as previously cited, namely that in the case of suspects who are juveniles prosecutors should always consider ‘whether a prosecution is likely to have an adverse impact on [the accused’s] future prospects that is disproportionate to the seriousness of the offending’. This provision is clearly concerned with the state’s responsibility for the appropriate development of young citizens, and is therefore more closely attuned to the conceptual framework of ‘virtue ethics’139 than of ‘civil peace’. Accordingly, its theoretical significance lies in the exemplification of the fact that the ‘public wrongs/public interest’ theoretical approach to criminalization, while closely related to the concept of ‘civil peace’, is not simply coterminous with it. Is it possible to be more precise about the content of the ‘public’ dimension of ‘public wrongs’? The answer, I believe, is both negative and affirmative. It is negative in the sense that it is not possible to provide definitive, universally acceptable answers, even within the relatively restricted framework of a secular, liberal-democratic state. But more positively, we can here call in aid Llewellyn’s concept of the ‘Net Drive’, which—it will be recalled—he defined, using characteristically unorthodox language, as ‘the Whither of the net Totality’.140 The ‘public’ element in ‘public wrongs’ might therefore be characterized as something that the ‘Totality’ (that is, in contemporary societies, the state) is able to defend as an ex ante conditional threat that could lead to punishment, within principles that can be coherently articulated and justified (that is, the defensible ‘Whither’ of state policies). Such a formulation is, perhaps, simply another way of saying, with Duff, that a crime is a public wrong that ‘properly concerns the public, i.e. the polity as a whole’;141 but the explicit linkage to Llewellyn’s theorization makes clearer that what ‘properly concerns’ the public needs to be carefully articulated in defensible principles of political philosophy. We should note also that what counts as a ‘public wrong’ is to an extent socially determined; that is, altered social conditions can and do create changes in the ‘Whither of the net Totality’, as (for example) sociologists of punishment have clearly demonstrated.142 A classic example of this point concerns acts of rape within marriage. Until remarkably recently, such acts were not deemed to be criminal in English law, relying on the historic view that the marriage bed was a private sphere outwith the purview of the state, and that in consenting to marriage the wife consented to all acts of intercourse. Of course, with the advent of a different kind of state polity in which the concept of human rights is prominent, this historic rule had to change, and did change.143 Again, therefore, we can see that the concept of a ‘public wrong’ necessarily contains a political dimension. 139  R. Hursthouse, On Virtue Ethics (Oxford: Oxford University Press, 1999). 140  Llewellyn, ‘The Normative, the Legal and the Law-Jobs’ (n. 46), 1387. 141 Duff, Answering for Crime (n. 9), 141. 142 See e.g. D. Garland, Punishment and Welfare (Aldershot: Gower, 1985); D. Garland, The Culture of Control (Oxford: Oxford University Press, 2001). 143  R v R [1992] 1 AC 599 (HL).

Civil Peace and Criminalization

261

D.  A case study of criminalization To conclude this section, I will offer a brief empirical case study which, I hope, might serve two purposes: first, to illustrate some of the matters to which Neil MacCormick sought to draw attention concerning how the criminal law can intersect with, and help to establish and sustain, everyday relationships of trust; and secondly, to act as something akin to a test case through which we might examine Ulväng’s claim that ‘civil peace’ is too nebulous a concept to be utilized in debates on criminalization. The case study concerns a new offence, created by the United Kingdom Parliament in the Road Safety Act 1967, of driving a vehicle with a blood-alcohol concentration above 80mg/100ml. Prior to 1967, English criminal law did not proscribe a fixed alcohol limit when driving; instead, the relevant offence consisted of being in charge of a motor vehicle when ‘unfit to drive’, with a further provision that a person ‘shall be taken to be unfit to drive if his ability to drive properly is for the time being impaired’.144 However, juries could be very sympathetic to drivers charged with this offence; in a 1964 London study, a third of drivers acquitted by juries on a charge of ‘unfitness to drive’ were shown by urine analyses to have had a blood-alcohol concentration in excess of 200mg/100ml.145 The principal purpose of the new offence introduced by the 1967 Act was to reduce the toll of death and injuries on the roads caused by drink-driving. This can be read as simple ‘harm reduction’, but it can also be read as having a wider purpose, namely to increase the sense of safety among drivers, thus enhancing ‘peaceful relations among persons who can trust relative strangers to avoid violating their persons’ (in this case by irresponsible driving).146 As suggested by the theorization of Simester and von Hirsch, the new offence created a ‘conditional threat’ intended to deter drink-driving—not least because conviction for the offence led, save in special circumstances, to a mandatory minimum 12 months’ disqualification from driving—a level of penalty that is still high by international standards.147 However, despite this clear evidence of intended deterrence in the creation of this new offence, the government White Paper that preceded the legislation was explicit in articulating the view that the ultimate goal was broader and deeper than simple deterrence. As the White Paper put it: The prevention of road accidents caused by drink is largely a social problem. Many people have grown accustomed to drive to and from places where they consume alcohol and . . . the problem will grow unless steps are taken to solve it. If the deaths and casualties caused by alcohol are to be prevented, social habits must change.148

144  R. Light, Criminalizing the Drink-Driver (Aldershot: Dartmouth, 1994), 46. 145 Light, Criminalizing the Drink-Driver (n. 144), 60. 146 MacCormick, Institutions of Law (n. 1), 216. 147  House of Commons Transport Committee, Drink and Drug Driving Law (First Report, Session 2010–11), para. 33. 148  Ministry of Transport, Road Safety Legislation 1965–6 (Cmnd 2859, 1965), para. 15 (emphasis added).

262

Anthony Bottoms

Accordingly, using MacCormick’s terminology, it was hoped that this legislation would help to create a greater degree of ‘civility’ in this sphere of social life, through a change in social habits. When the 1967 Act was implemented, there was an initial sharp drop in alcohol-related road accidents, as drivers feared being caught when above the new limit; but they soon realized that enforcement levels were not high, so the accident level crept up again.149 The longer-term effects of the legislation are, however, more interesting; they are summarized in Table 9.1. In short, since 1980150 there has been a dramatic reduction in drink and drive casualties, even in absolute terms (in round terms, from 21 per day killed or seriously injured in 1980 to 4 per day in 2010). The reduction is even more pronounced when the increased use of the roads since 1980 is taken into account (see final two columns of the table). What has caused this reduction? One view, recently expressed by the House of Commons Transport Committee, is that ‘the success of Great Britain’s drink driving policy has been largely attributable to the deterrent effect of the . . . 12-month mandatory disqualification period’.151 This is, however, unlikely, for two linked reasons: first, because the clear and consistent evidence from deterrence research is that a high probability of being caught is much more effective in producing deterrent effects than is the severity of the sentence;152 and secondly, because, in Britain, the police enforcement levels for drink driving are much lower than they are in most European countries.153 It is much more likely, especially given the scale of the Table 9.1  Reported drink and drive casualties in Great Britain, 1980–2010 Year

Killed

1980 1,040 1990 760 2000 530 2010 240

Serious injury

Slight injury

Total drink-drive casualties

Total motor Total drink-drive vehicle journeys casualties per billion km (billion km)

6,810 4,090 2,540 1,240

19,380 15,550 14,990 8,210

27,220 20,400 18,060 9,690

271.9 410.8 467.1 495.6

100.1 49.7 38.7 16.6

Source: Department for Transport Statistics Table RASS1001 (for drink and drive casualties). The data on total motor vehicle journeys has been added from other Department for Transport statistics. The data are for the area known as ‘Great Britain’ (that is, England, Scotland, and Wales, but not Northern Ireland); this matches the scope of the application of the Road Safety Act 1967.

149 Light, Criminalizing the Drink-Driver (n. 144), 77. 150  Data on road casualties specifically attributable to drinking and driving, as shown in Table 9.1, were not collected until 1979. 151  House of Commons Transport Committee, Drink and Drug Driving Law (First Report, Session 2010–11), para. 41. 152  A. von Hirsch, A. E. Bottoms, E. Burney, and P.-O. Wikström, Criminal Deterrence and Sentence Severity (Oxford: Hart 1999); A. N. Doob and C. Webster, ‘Sentence Severity and Crime: Accepting the Null Hypothesis’, Crime and Justice: A Review of Research, 30 (2003), 143. 153  House of Commons Transport Committee, Drink and Drug Driving Law (First Report, Session 2010–11), para. 45. In Great Britain, 2 per cent of drivers were breath-tested during a one-year period; many other countries had substantially higher rates (e.g. the Netherlands 20 per cent; France 30 per cent; Sweden 45 per cent).

Civil Peace and Criminalization

263

reduction in casualties, that the effects shown in Table 9.1 have been achieved by the ‘change in social habits’ that the 1965 White Paper had hoped for, and which have been encouraged by various public educational initiatives, although of course always with the criminal law in the background. It also seems clear that, over time, there has been a gradual change in people’s moral evaluation of the act of taking the wheel after drinking alcohol. For example, in the British Social Attitudes Survey of 2012, 68 per cent of respondents said they ‘strongly agreed’, and a further 19 per cent said they ‘agreed’ with the statement that a person who ‘has drunk any alcohol they should not drive’;154 this is clearly a major change from attitudes at the time of the 1967 Act. Today, therefore, nearly half a century after the 1967 Act, no one contemplates going back to the pre-1967 law. Indeed, the United Kingdom is now one of only two countries in the European Union to allow driving with a blood-alcohol concentration of up to 80mg/100ml (the other being Malta)—no other country has a limit above 50mg/100ml. It is therefore not surprising that there has been a recent official proposal to reduce the British level to 50mg/100ml; and careful analysis of this proposal and its reception will hopefully throw further light on the actual social processes of intended criminalization. The proposal was made by a legal academic and former Vice-Chancellor of Oxford University, Sir Peter North QC, in an independent review of the drink and drug driving laws for the government’s Department for Transport.155 The main factors that North cited in support of the proposal were (in headline terms): (i) that it ‘would undoubtedly save a significant number of lives’; (ii) that there was evidence of ‘very considerable public support’ for a lower limit; (iii) that alleged adverse consequences for the drinks and hospitality industry had been ‘exaggerated’ in the press, and the industry ‘could do more to protect itself ’ by, for example, promoting ‘designated driver’ schemes; and (iv) that a reduction of the offence to a level of 20mg/100ml (as in some European countries) would be ‘a step too far’, risking a loss of legitimacy by the public in the drink-drive legislation.156 However, in its response the government was critical of some aspects of North’s reasoning.157 A major feature of the government’s approach was to argue that, while lowering the limit would indeed save some lives, relatively few drink-driving fatalities are in the marginal range of 50–80mg/100ml; indeed, ‘most . . . are well above the [existing] limit. . . . it is these drinkers who need to change their behaviour radically to get major casualty savings’.158 Therefore, the government suggested, those who drink and drive at these dangerously 154 Department for Transport, British Social Attitudes Survey 2012: Public Attitudes Towards Transport (Department for Transport, 2013), 22–3 (emphasis added). 155 Sir Peter North, Report of the Review of Drink and Drug Driving Law (Department for Transport, 2010). 156 North, Report of the Review of Drink and Drug Driving Law (n. 155). All quotations are taken from the executive summary of the report, at 6–7. 157  The Government Response to the Reports by Sir Peter North CBE QC and the Transport Select Committee on Drink and Drug Driving, Cm 8050 (The Stationery Office, 2011). 158  The Government Response to the Reports (n. 157), 8.

264

Anthony Bottoms

intoxicated levels present a significant enforcement problem, which needs to be better addressed. It follows that if the drink-drive legislative limit were lowered, then (given limited resources) the police would be faced with ‘a tricky choice— how do they focus on the dangerous people if they have to catch [a new group of ] lower-level drinkers too?’159 Additionally, if there were a new law, the lower-level drinkers might drink less, ‘especially . . . when they are out’, which could have ‘a substantial impact on the businesses they patronise’, offset by only a small reduction in casualties.160 Overall, the government concluded that its strategy was ‘to help the police to focus on the most dangerous people’, and ‘it did not believe that ‘lowering the [drink-driving] limit is consistent with this approach’.161 Sir Peter North’s proposal was therefore rejected. This is not an essay on the drink-drive laws, so I shall not comment substantively on these contrasting views. What is of central interest for present purposes is the nature of the arguments used by the two sides, which include questions of harm to road users; comparisons with standards of criminalization in other European states; where to focus scarce criminal justice resources; public support and legitimacy; and the impact on the hospitality industry. All but the last of these properly come within the framework of ‘civil peace’, and they are all, in different ways, elements of a debate about the optimum kind of achievable social order (or ‘civility’) for a society like ours. These identified elements are complex, and sometimes conflicting; but they cannot, I think, fairly be described as ‘nebulous’. What is apparent from this discussion, however, is that ‘civil peace’ is indeed an ‘umbrella concept’, though in a somewhat different sense from that intended by Ulväng.162 Within that generic concept, there are several different strands (for example: harm; enforcement priorities; and public legitimacy), and mature debates about criminalization will recognize and consider these different strands, and their interrelationship. Such debates will also, with MacCormick, recognize that civil peace ‘cannot be a function of the criminal law alone’;163 law always interacts with other aspects of a given society as it seeks to promote social order.

V.  Criminalization and Civil Peace: Wider Theoretical Issues In this final section, I shall address two wider theoretical issues that have been hinted at, but not adequately addressed, in preceding sections. As will be seen, the aim is more to sketch some parameters for future debates than to provide definitive answers. 159  The Government Response to the Reports (n. 157), 15. 160  The Government Response to the Reports (n. 157), 16. 161  The Government Response to the Reports (n. 157), 18. 162  Ulväng, ‘Criminal Law and Civil Peace’ (n. 7), 136; see also text to n. 75 above. 163 MacCormick, Institutions of Law (n. 1), 221.

Civil Peace and Criminalization

265

A.  Liberal theory, social authoritarianism, and civil peace In the course of the discussion of the Marshall/Duff ‘public wrongs’ thesis (Section IV.C), these authors’ explicit communitarianism was mentioned, as was the deep suspicion of communitarian claims shown by many liberal authors. In the present context, any full consideration of the liberal-communitarian debate is clearly inappropriate, but it is both necessary and important to acknowledge the core concerns of liberal theorists, since these concerns raise issues that are highly germane to any attempt to connect criminal law to ‘civil peace’. A good starting point in this respect is the celebrated debate in the early 1960s between Lord Devlin and H. L. A. Hart on ‘the enforcement of morals’.164 Looking back on this controversy 40 years later, Hart’s biographer, Nicola Lacey, suggested that his contribution to this debate: remains . . . the resounding late twentieth-century statement of principled liberal social policy, articulating a vision of a social democratic state which should use the criminal law sparingly in the interests of individual liberty. [Hart] was particularly eloquent in calling attention to the misery which can be caused by the enforcement of morality: particularly in the area of sexuality, where law uses coercion to enforce standards which may conflict with people’s deepest feelings, he denounced the legal enforcement of morality [where no harm is caused to another] as a form of cruelty.165

Thus, Hart decisively rejected Devlin’s claim that ‘the enforcement of a society’s morality is necessary to preserve social stability’;166 and he was surely right to do so, because that kind of attempt to ‘preserve social stability’ is essentially a form of social authoritarianism, potentially stifling the freedom of peaceable individuals to develop as they choose. Nor is this line of criticism restricted to the sphere of personal morality—other forms of social authoritarianism, such as restrictions on free speech, can be (and are) advocated in the name of ‘social stability’ or ‘civil peace’, and they are equally suspect. (Indeed, phrases such as ‘the need to maintain social order’ are the stock-in- trade of repressive rulers everywhere.) Accordingly, any attempt to advocate criminal legislation even in part on the grounds of ‘civil peace’ must—as Ulväng rightly noted—take these issues very seriously. The standard way of dealing with this issue, adopted by Hart and before him by Mill, is to restrict the scope of the criminal law to acts involving harm to another person, where that harm also involves a wrong (e.g. it was caused not by mere accident, but by a person acting culpably). In the context of the debates considered in this chapter, this is the approach adopted by Ulväng, and more fully by Simester and von Hirsch, and it is clearly an approach with much merit. However, Lacey was right to say, in her appraisal of Hart’s liberal theory, that the theory is ‘not without its flaws’, particularly because ‘the all-important limiting condition—the specification of what counts as harm [for the purposes of criminalization]—is not 164  P. Devlin, The Enforcement of Morals (Oxford: Oxford University Press, 1960); H. L. A. Hart, Law, Liberty and Morality (Oxford: Oxford University Press, 1963). 165  N. Lacey, A Life of HLA Hart: The Nightmare and the Noble Dream (Oxford: Oxford University Press 2004), 259. 166 Lacey, A Life of HLA Hart (n. 165), 259.

266

Anthony Bottoms

self-defining’.167 (We have thus returned here to examples such as the emotional and financial harm to a partner that may be caused by a cheating spouse). Duff’s solution to this problem, based on the public wrongs thesis, is that ‘the Harm Principle needs to be turned into a Public Harm Principle’.168 We therefore have a complicated scenario in which the public wrongs thesis, linked as it necessarily is to conceptions of ‘civil peace’ (see earlier section) needs the Harm Principle to alleviate potential social-authoritarian deficiencies, while the Harm Principle needs a version of the ‘public wrongs thesis’ to prevent it from straying inappropriately into areas of private life that are ‘not the criminal law’s business’. There are clearly complex issues here that require more extensive discussion than can be attempted in this chapter. However, on the basis of the arguments presented here, it will hopefully be apparent that a full appreciation of the linkages between the criminal law and social order, based on MacCormick’s central insight that ‘law is institutional normative order’, must be part of that discussion, not only in relation to the criminal law in general, but also in relation to specific offences.

B.  Reciprocity as a conceptual framework As a final theoretical thought, it seems appropriate to return to a concept that I earlier described as central to MacCormick’s jurisprudence, namely that of the ‘orderliness that results from reciprocal observance of essentially similar beliefs and dispositions of different persons’.169 Since Neil MacCormick’s death, the publication of David Johnston’s Brief History of Justice has created a renewed interest in the concept of reciprocity. Johnston argues that ‘justice’ is, ultimately, not a word that can be defined any way one wishes, because ‘we must respect the [human] intuitions that are fun­damental to the sense of justice’.170 His historical analysis and, he suggests, some contemporary empirical research,171 support the view that ‘a sensibility for reciprocity in relations among human beings is integral to that [intuitive] sense’,172 notwithstanding that in recent centuries some of the most influential formal theories of justice (including that of John Rawls173) have neglected that sensibility. The Oxford philosopher David Wiggins, in his anything-but-introductory ‘twelve lectures on the philosophy of morality’ has also refocused attention on similar issues, in a discussion of what he calls ‘a first-order ethic of solidarity and reciprocity’.174 At the beginning of his chapter on this theme, Wiggins places an epigraph by Philippa Foot: It has been suggested [by the author] that one criterion for a good moral system is that it should be possible [just by virtue of what it gives and takes] to demand [some recognizant 167 Lacey, A Life of HLA Hart (n. 165), 259. 168 Duff, Answering for Crime (n. 9), 136. 169  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 188–9. 170  D. Johnston, A Brief History of Justice (Oxford: Wiley-Blackwell, 2011), 226. 171  Johnston cites J. Heinrich et al., ‘In Search of Homo Economicus: Behavioral Experiments in 15 Small-Scale Societies’, Economics and Social Behavior, 91 (2001), 73. 172 Johnston, A Brief History of Justice (n. 170), 226. 173  J. Rawls, A Theory of Justice (Oxford: Oxford University Press, 1972). 174  D. Wiggins, Ethics: Twelve Lectures on the Philosophy of Morality (Harmondsworth: Penguin Books, 2006), 220.

Civil Peace and Criminalization

267

return] from every individual because of the good the system renders to him. . . . It has also to be such that anyone can conform to it and still live well enough in the ordinary, nonmoral, sense.175

This is therefore a first-order ethic of reciprocity, derived ultimately from the ineradicably social nature of human life. In the same paper, and also quoted by Wiggins, Foot moves from a principle of reciprocity to one of ‘a kind of solidarity between human beings, as if there is some sense in which no one is to come out against one of his fellow men’.176 Wiggins uses Foot’s suggestions in a complex argument which cannot be discussed here, but in which the following comment seems pivotal: We should see Foot’s [suggestions] as spelling out one non-negotiable corollary of the transition that human beings make from mutual recognition under the aspect of personal beings to a social morality that requires personal beings to live out that recognition in a solidarity that is all of a piece with the recognition that requires each personal being to recognize the other one as participating in that morality, and equally requires the other one to recognize him or her as a participant in it.177

This is a very complex single sentence, but one that repays careful study. From the perspective of the issues raised in this chapter, one crucial phrase refers to humans’ ‘mutual recognition under the aspect of personal beings’. These words take us back to one of the starting points of this discussion—namely, the fact that human beings are both social animals and creatures with a unique capacity for self-reflection; and that, in consequence, the ‘creation and sustaining of trust remains, and will always remain, an indispensable human contrivance for coping with the freedom’ of other people.178 Reciprocity seems a very sensible foundation for attempts to build that trust. But any shrewd social leader knows very well that (given that humans also have an evolved capacity for selfish assertiveness) some individuals will probably try to break out of what they see as the constraints of reciprocity and claim certain advantages for themselves (for example, a larger share of resources), thus requiring the group to take coercive action to protect its more vulnerable members. If the above account has any merit, then the concept of reciprocity may have much to commend it as a conceptual foundation for MacCormick’s legal-institutional approach to criminal law theory, which as we have seen tries to link the formal proscriptions of the criminal law with informal normative practices in the everyday production of reciprocally generated social order. Perhaps, also, Philippa Foot’s language of reciprocity, leading on to solidarity, might be a framework within which liberals and communitarian criminal law theorists might unite. This might 175 P. Foot, ‘Morality, Action and Outcome’, in T. Honderich (ed.), Morality and Objectivity (London: Routledge, 1985), 36. The quotation is given in the form in which it appears in Wiggins, Ethics (n. 174), 230; Wiggins obtained permission from Professor Foot to make some small adaptations to the original text. 176  Foot, ‘Morality, Action and Outcome’ (n. 175), 86; Wiggins, Ethics (n. 174), 241, emphasis in original. 177 Wiggins, Ethics (n. 174), 245, emphasis in original. 178  Dunn, ‘ “Trust” in the Politics of John Locke’ (n. 31), 54.

268

Anthony Bottoms

result, for example, in some modification of Marshall and Duff’s communitarian foundation for the ‘public wrongs’ thesis (expressed as: ‘the attack on this individual victim is also an attack on us—on her as a member of the group and on us as fellow members’) to an alternative formulation based on reciprocity. This might take something like the following form: ‘the attack on this individual is one that must be of concern to the state as well as the individual, since to promote reciprocal trust—the foundation of social order—the state must espouse a philosophy of the mutual recognition of persons, and this individual’s personhood has not been properly respected’.

VI. Conclusion At the beginning of this chapter, two central claims by Neil MacCormick were set out, and they must now, finally, be revisited. The first claim, which is about citizens’ expectations of the criminal law, is that ‘the most basic demand citizens ought to make of criminal law is that it contribute to securing the conditions of civility and social peace’.179 It is clear that MacCormick expected this claim to be evaluated within the framework of the ‘legal-institutional’ theoretical approach that he espoused, including the complex interrelations between the formal doctrines of the law; the practical working of the criminal justice system; and day-to-day relationships among ordinary people within civil society. In line with MacCormick’s hypothesis, the arguments of this chapter suggest that ‘securing the conditions of civility and social peace’ is indeed a vital function of the criminal law. Moreover, and contrary to the claims of various contributors to the Edinburgh symposium, considerations of ‘civility and social peace’ are by no means irrelevant to debates about the criminalization of specific acts. That is not to say, of course, that ‘civil peace’ is the only relevant variable in such debates; as we have seen, amongst other things issues of harm and of whether the relevant act is in an appropriate sense a ‘public wrong’ are also crucial. But MacCormick did not claim that ‘civil peace’ operated alone within criminal law theory; only that it is ‘the most basic demand’ that citizens ought to make of their criminal law and their criminal justice system. In this claim, he was surely right. MacCormick’s second claim was more straightforwardly ethical: ‘what is wrong about committing crimes is that it involves some wilful form of behaviour that violates the conditions of civil peace’.180 The scope of this claim seems incautiously wide, and it is open to the objection that it potentially permits socially authoritarian criminal legislation. The claim therefore requires revision.181 That 179 MacCormick, Institutions of Law (n. 1), 221. 180 MacCormick, Institutions of Law (n. 1), 216. 181  See further the chapter by Lindsay Farmer, ‘Criminal Law as an Institution’, elsewhere in this volume, arguing that MacCormick’s account of criminal law is underdeveloped, but nevertheless rich in significant pointers that can be used by those of us who follow after him.

Civil Peace and Criminalization

269

revision would not jettison the claim that violating the conditions of civil peace is an important element in what is wrong with (at least most) just criminal legislation; but it would also have to take account of other central features of such legislation, in particular the requirements of harm, and of wrongfulness that is appropriately regarded as a ‘public wrong’. In the introduction to this chapter, I indicated that my purpose is, in the final analysis, less to address the specifics of MacCormick’s claims than to establish that ‘civil peace’ is by no means irrelevant to debates about the criminalization or otherwise of specific acts; and also to try to open a discussion about how that concept can be used—in conjunction with other concepts—in future debates about criminalization. If I have been successful in these aims, other debates must follow—not least about the methodology for taking forward this discussion.182 But for now it seems best to leave the final words to Neil MacCormick. ‘Criminal law’, he claimed, ‘is always and inevitably expressive, or perhaps it would be better to say constitutive, of a prevailing social morality adopted and enforced by the state.’183 That is not to say that this law, or the social morality that underpins it, is always admirable, as indeed MacCormick’s preceding paragraph (about pre-1861 slavery laws in the United States) makes very clear. But if MacCormick’s broad claim is right, as I believe it is, then the literature on criminalization, without abandoning either its normative orientation or its philosophical sophistication, needs to take greater account than it mostly has done heretofore of work in social and political science, both empirical and theoretical. Such an approach would be a fitting tribute to MacCormick, whose legal-institutionalism was an attempt (among many other things) to create ‘a serious bridge across two cognate disciplines, and constructively [link] legal and social theory’.184

182 For an important contribution to this debate, published just as this chapter was going to press, see N. Lacey, ‘The Rule of Law and the Political Economy of Criminalisation: An Agenda for Research’, Punishment and Society, 15 (2013), 349. Lacey argues for an explicitly ‘institutional approach to criminalisation’, which would ‘[draw] on historical and comparative methodologies, and on the resources of several disciplines including law, sociology and political science’. The aim would be to lay the intellectual foundations for ‘a broad political economy of criminalisation within modern societies committed to the Rechsstaat/rule of law’ (at 349). 183 MacCormick, Institutions of Law (n. 1), 211. 184  MacCormick, ‘Concluding for Institutionalism’ (n. 4), 190.

10 Marginality, Ethnicity, and Penality A Bourdieusian Perspective on Criminalization Loïc Wacquant*

Prefatory Note This chapter draws an analytic map of the research programme pursued across my three books Urban Outcasts (2008), Punishing the Poor (2009), and Deadly Symbiosis: Race and the Rise of the Penal State (2015), in an effort to relate urban marginality, ethnicity as public (dis)honour, and penalization as a modality of government action entailing the conjoint deployment of legitimate material and symbolic violence: physical force and legal stigma. To effect this triadic connection is to invite criminologists and penologists to alter their baseline analytics in two ways. The first entails broadening their empirical purview beyond the weary couple of ‘crime and punishment’ to encompass changing forms of urban dispossession and concurrent modalities of state-crafting. For the stunning and unforeseen resurgence of the prison and its extensions in advanced society at century’s turn cannot be explained by stagnant or sagging rates of offending (or even increasing rates, for that matter): it partakes of a broader revamping of the state and its relation to the poor. The second alteration is to shift their explanatory focus from criminalization—the multistep process whereby the authorities come to stipulate that certain conditions or conducts constitute crimes—to penalization as a socially selective response to lower-class offending and its proximate social correlates, namely, destitution and disrepute.1 For ‘governing through crime’ does not mechanically and automatically imply the expansion of punishment; it can just as well lead to the enlargement and elevation of the social or medical missions of the state, as it did in the period of Fordist-Keynesian

* This text is a compressed and clarified version of my opening keynote to the conference ‘Marginalité, pénalité et division ethnique dans la ville à l’ère du néolibéralisme triomphant: journée d’études autour de Loïc Wacquant’, organized at the Université Libre de Bruxelles on 15 October 2010. The full version, along with a response to my critics, appears as a symposium in Ethnic & Racial Studies (vol. 37, no. 10, Fall 2014, Symposium). 1  J. Irwin, The Jail: Managing the Underclass (Berkeley: University of California Press, 1980).

Marginality, Ethnicity, and Penality

271

consolidation.2 It is precisely this contingent political connection between crime and penal sanction that we must put at the epicentre of our enquiries. The theories of Pierre Bourdieu provide muscular and flexible tools to effect this double move, and thus to reformulate root and branch the question of criminalization as state action. In the trilogy, I disentangle the triangular nexus of class fragmentation, ethnic division, and state-crafting in the polarizing city at century’s turn to explain the political production, socio-spatial distribution, and punitive management of marginality through the marriage of disciplinary social policy and neutralizing criminal justice. In this chapter, I explicate how I adapted and extended key notions from Bourdieu as well as coined new ones (territorial stigmatization and advanced marginality, punitive containment and liberal paternalism, hyperincarceration and negative sociodicy) as tools for the comparative sociology of the unfinished genesis of the post-industrial precariat, the penal regulation of poverty in the age of diffusing social insecurity, and the building of the neo-liberal Leviathan. Bringing the study of the contemporary permutations of class, race and immigration, and the state into a single framework shows how the racialization, penalization, and depoliticization of the urban turbulences associated with advanced marginality reinforce one another in Western Europe as in the United States. It confirms that punishment is not just a key index of social solidarity, as Durkheim proposed over a century ago in The Division of Labor in Society, but also a core capacity and key site for staging the sovereignty of the state as classifying and stratifying agency. And it reveals the deep kinship between race and judicial sanction as kindred forms of official dishonour that converge in the constitution of public outcasts. * This chapter aims to activate communication among three clusters of researchers who usually do not encounter one another and therefore do not talk to each other, or do so too rarely and from a distance. In the first corner, we have people who study class fragmentation in the city in the wake of the crumbling of the traditional working class which issued from the Fordist and Keynesian era (that is, roughly the long century running from 1880 to 1980) under the press of deindustrialization, the rise of mass unemployment, and the diffusion of labour precarity, at the intersection of what Robert Castel puts under the notion of the ‘erosion of the wage-earning society’3 and Manuel Castells calls ‘the black holes’ of urban development in ‘the information age’.4 These researchers are concerned with employment and labour market trends and with their polarizing and ramifying impacts on social and spatial structures—leading in particular, at the bottom of the ladder of classes and places, to the unfinished genesis of the post-industrial precariat in the urban periphery at the dawn of the twenty-first century. But they scarcely engage 2  See e.g. J. Gilbert, A Cycle of Outrage: America’s Reaction to the Juvenile Delinquent in the 1950s (Berkeley: University of California Press, 1988). 3  R. Castel, Les Métamorphoses de la question sociale. Une chronique du salariat (Paris: Fayard, 1996). 4 M. Castells, End of Millennium: The Information Age: Economy, Society, and Culture, vol. iii (Oxford: Basil Blackwell, 2000).

272

Loïc Wacquant

in sustained discussion with their colleagues who, in the second corner, are studying the foundations, forms, and implications of ethnic cleavages. Grounded in ethnoracial classifications in the United States (that is, in the institutionalization of ‘race’ as denegated ethnicity), in ethno-national classifications in the European Union (to wit, the ‘national/foreigner’ cleavage), and in a varying mix of the two in Latin America and a good part of Africa, (re)activated by immigration and by the cultural differences of which migration can be the carrier, ethnic division is nonetheless essential to grasping the formation and deformation of classes. And conversely: how can one not see that those who are designated—indeed, defamed—across Europe as ‘immigrants’ are foreigners of post-colonial origins and lower-class extraction while others, of upper-class standing, are ‘expats’, whom everyone wants to attract and not drive out? And how can one ignore that the collective perception one has of them, their modalities of incorporation, their capacity for collective action, in sum, their fate, depend a great deal on their social position and trajectory, and therefore on shifts in the class structure in which they become ensconced? This domain of enquiry, which is experiencing an unprecedented boom across Europe, fuelled by the fear of immigration and by the political and media fad over ‘diversity’, has grown largely autonomous (under the impetus of American-style ethnic studies programmes) and increasingly distant from—even opposed to—class analysis. Thus an artificial alternative has crystallized, which summons us to make a disjunctive choice between class and ethnicity, to grant analytical preference and political priority to either ‘the social question’ or ‘the racial question’—I am thinking here, in the case of France, of Pap Ndiaye’s resounding study La Condition noire (2008),5 which aspires to found ‘black studies à la French’, which, in my view, is a double mistake, theoretical and practical, and of the book edited by the Fassin brothers, De la question sociale à la question raciale? (2006), which speaks volumes about the drift of the progressive ‘common sense’ of the moment.6 Now, it is abundantly evident, as Max Weber emphasized a century ago, that these two modalities of ‘social closure’ (Schließung), based respectively on the distribution of material and symbolic powers, are profoundly imbricated and must necessarily be theorized together.7 Finally, in the third corner, studiously isolated from the other two, we have a group of criminologists and assorted specialists in criminal justice issues. They 5  La Condition noire. Essai sur une minorité française (Paris: Calmann-Lévy, 2008). 6  P. Ndiaye, La Condition noire; D. Fassin and E. Fassin (eds.), De la question sociale à la question raciale? Représenter la société française (Paris: La Découverte, 2006). 7  M. Weber, Economy and Society (Berkeley: University of California Press, [1922] 1978). I argued this point a long time ago, in the course of a reinterpretation of the political and scientific controversy stirred up in the United States by the masterwork of my Chicago mentor, William Julius Wilson, as well as in an article calling for the elaboration of an ‘analytic of racial domination’ escaping the logic of trial which construes racialization as one among many competing modalities of the fabrication of collectives. See W. J. Wilson, The Declining Significance of Race: Blacks and Changing American Institutions (2nd edn.; Chicago: University of Chicago Press, [1978] 1980); L. Wacquant, ‘The Puzzle of Race and Class in American Society and Social Science’, Benjamin E. Mays Monograph Series 2 (1) (Fall 1989), 7–20; L. Wacquant, ‘For an Analytic of Racial Domination’, Political Power and Social Theory, 11 (1997), 221–34.

Marginality, Ethnicity, and Penality

273

burrow away with zeal the closed perimeter of the ‘crime and punishment’ duet, which is historically constitutive of their discipline and continually reinforced by political and bureaucratic demand. Hence, they pay hardly any attention (not enough for my taste, in any case) to shifts in class structure and formation, the deepening of inequalities and the broad revamping of urban poverty, on the one hand, and to the dynamic, and historically variable, impact of ethnic divisions on the other (except under the narrow and limiting rubric of discrimination and disparity, typically conflated). In so doing, they deprive themselves of the means to grasp the contemporary evolution of penal policies. As Bronislaw Geremek showed in his master work La Potence ou la pitié, since the invention of prison and the emergence of modern states in the West at the close of the sixteenth century, these policies have aimed less at reducing crime than at curbing urban marginality.8 Better yet, penal policy and social policy are but the two flanks of the same politics of poverty in the city—in the double sense of power struggle and public action. Finally, always and everywhere, the vector of penality strikes preferentially at categories situated at the bottom of both the order of classes and gradations of honour. It is therefore crucial to connect criminal justice to marginality in its double dimension, material and symbolic, as well as to the other state programmes that purport to regulate ‘problem’ populations and territories. I hope that my contribution here can help us overcome the isolation and even mutual ignorance in which the explorers of these three thematic regions hold one another, so that we may set in motion a dialogue among students of urban relegation as a product of class restructuring, of the reverberations of ethnicity, and of the transformations of the state in its different components targeted at dispossessed and dishonoured populations—first among these being its penal arm (the police, courts, jail, prison, juvenile facilities, and their extensions). If there is a key argument that I want to lay out, it is that we urgently need to link these three areas of inquiry and get the corresponding disciplines to work together: urban sociology and economic analysis, the anthropology and the political science of ethnicity, and criminology and social work, with diagonal input from geography to help us capture the spatial dimension of their mutual imbrications. At the epicentre of this analytic space emerges the figure of a ‘Centaur state’, liberal at the top and punitive at the bottom, which flouts democratic ideals by its very anatomy as by its modus operandi.

I I propose to begin by sketching a rough analytic cartography of the research programme I have pursued over the past two decades at the crossroads of these three thematics, a programme of which my books Urban Outcasts, Punishing the Poor, 8  La Potence ou la pitié. L’Europe et les pauvres du Moyen Âge à nos jours (Paris: Gallimard, 1978) (English trans. Poverty: A History; Oxford: Basil Blackwell, 1994).

274

Loïc Wacquant

and Deadly Symbiosis are both the product and summation. These books form a trilogy that probes the triangle of urban transformations with class, ethnicity, and state as its vertices and paves the way for a properly sociological (re)conceptualization of neo-liberalism. So much to say that they gain from being read together, sequentially or concurrently, insofar as they complement and bolster one another to outline in fine a model of the reconfiguration of the nexus of state, market, and citizenship at century’s start, and a model that one can hope to generalize by means of reasoned transpositions across borders. This revisit is an opportunity to draw up a provisional and compact balance-sheet of these inquiries and to specify their stakes, but also to signpost how I adapted key notions from Pierre Bourdieu (social space, bureaucratic field, symbolic power) to clarify categories left hazy (such as that of the ghetto) and to forge new concepts with which to dissect the emergence of the urban precariat and its punitive management by the neo-liberal Leviathan. Each volume of this trilogy shines light on one side of the ‘class–race–state’ triangle and probes the impact of the third vertex on the relationship between the other two.9 And each book builds on the other(s) as both empirical backdrop and theoretical stepping-stone. 1. Urban Outcasts diagnoses the rise of advanced marginality in the city, in the wake of the collapse of the black ghetto on the American side and the dissolution of working-class territories in Western Europe, along the ‘class–race’ axis angled by state structures and policies. 2. Punishing the Poor charts the invention and deployment of punitive containment as a technique for governing problem areas and populations along the ‘class–state’ axis stamped by ethnoracial or ethnonational divisions. 3. Deadly Symbiosis disentangles the relationship of reciprocal imbrication between penalization and racialization as kindred forms of dishonour and reveals how class inequality intersects and inflects the ‘state–ethnicity’ axis. Each of these books labours its own problematic and can therefore be read separately. But the arguments that link them together extend beyond each to make a broader contribution, first to a comparative sociology of the regulation of poverty and the (de)formation of the post-industrial precariat and, secondly, to a historical anthropology of the neo-liberal Leviathan.10 They offer a way to rethink neo-liberalism as a transnational political project, a veritable ‘revolution from above’ that cannot be reduced to the naked empire of the market (as both its opponents and its advocates

9  I use the term ‘race’ in the sense of denegated ethnicity: a principle of classification and stratification stipulating a gradation of honour (declensed according to ancestry, phenotype, or some other sociocultural characteristic mobilized for the purpose of social closure, cf. L. Wacquant, ‘For an Analytic of Racial Domination’, Political Power and Social Theory, 11 (1997), 221–34.) that purports to be based in nature; or else a paradoxical variety of ethnicity that claims to not be ethnic—a claim that, infeliciter, sociologists endorse every time they carelessly invoke the duet ‘race and ethnicity’ that anchors ethnoracial common sense in English-speaking countries. 10  L. Wacquant, ‘Three Steps to a Historical Anthropology of Actually Existing Neoliberalism’, Social Anthropology, 20 (1) (2012), 66–79.

Marginality, Ethnicity, and Penality

275

would have it) but necessarily encompasses the institutional means required to bring this empire into being: namely, disciplinary social policy (encapsulated by the notion of workfare) and the diligent expansion of the penal system (which I christen prisonfare), without forgetting the trope of individual responsibility that acts as the cultural glue binding these three aforementioned components together.11 I briefly summarize the key arguments made in each book before highlighting their common theoretical foundations and their interconnected implications. 1. The political production of advanced marginality: The first book, Urban Outcasts: A Comparative Sociology of Advanced Marginality, elucidates the nexus of class and race in the districts of dispossession or bas-quartiers of the post-industrial metropolis in its phase of socio-spatial polarization.12 I describe the sudden implosion of the black American ghetto after the acme of the civil rights movement and attribute it to the turnaround of local and federal policies after the mid-1970s—a multisided shift that David Harvey captures well as a move ‘from the managerial city to the entrepreneurial city’,13 but one which assumed a particularly virulent form in the United States as it also partook of a sweeping racial backlash. This policy turnaround accelerated the historic transition from the communal ghetto, confining all blacks in a reserved space that both entrapped and protected them, to the hyperghetto, a territory of desolation that now contains only the unstable fractions of the African-American working class, exposed to all manners of insecurity (economic, social, criminal, sanitary, housing, etc.) by the unravelling of the web of parallel institutions that characterizes the ghetto in its full-fledged form.14 I then contrast this sudden crumbling with the slow decomposition of working-class territories in the European Union during the era of deindustrialization. I show that urban relegation obeys different logics on the two continents: in the United States, it is determined by ethnicity, modulated by class position after the 1960s, and aggravated by the state; in France and its neighbouring countries, it is rooted in class inequality, inflected by ethnicity (for which read: post-colonial immigration), and partially deflected by public action. It follows that, far from drifting towards the socio-spatial type of the ghetto as instrument of ethnic closure,15 the dispossessed districts of European cities are moving away from it on all dimensions, so much so that one can characterize them as anti-ghettos.16

11 L. Wacquant, ‘Crafting the Neoliberal State: Workfare, Prisonfare and Social Insecurity’, Sociological Forum, 25 (2) (2010), 197–220. 12  L. Wacquant, Urban Outcasts: A Comparative Sociology of Advanced Marginality (Cambridge: Polity Press, 2008). 13 D. Harvey, ‘From Managerialism to Entrepreneurialism: The Transformation in Urban Governance in Late Capitalism’, Geografiska Annaler, 71 (1) (1989), 3–17. 14 L. Wacquant, ‘Les deux visages du ghetto: construire un concept sociologique’, Actes de la recherche en sciences socials, 160 (2005), 4–21. 15  L. Wacquant, ‘A Janus-Faced Institution of Ethnoracial Closure: A Sociological Specification of the Ghetto’, in R. Hutchison and B. Haynes (eds.), The Ghetto: Contemporary Global Issues and Controversies (Boulder, Colo.: Westview Press, 2011), 1–31. 16  The predicament of lower-class post-colonial immigrants across Europe is that they suffer from the symbolic taint spread by the panic discourse of ‘ghettoization’, which overtly designates them as a threat to national cohesion in every society, without garnering the ‘paradoxical benefits’ of actual

276

Loïc Wacquant [neo-liberalism] Left hand “workfare” Punishing the Poor

STATE Right hand “prisonfare” [prison]

Deadly Symbiosis

CITY [hyperghetto anti-ghetto] CLASS (market)

Urban Outcasts

RACE (ethnicity)

(BODY) Body and Soul

Figure 10.1  The ‘fatal triangle’ of the urban precariat

I thus refute the fashionable thesis of a transatlantic convergence of dispossessed districts on the pattern of the African-American ghetto and instead point to the emergence, on both sides of the Atlantic, of a new regime of poverty in the city, fuelled by the fragmentation of wage labour, the retrenchment of social protection, and territorial stigmatization. I conclude that the state plays a pivotal role in the social as well as the spatial production and distribution of urban marginality: the fate of the post-industrial precariat turns out to be economically underdetermined and politically overdetermined. This is true in the United States no less than in Europe—yet another nick in what the historian and jurist Michael Novak has nicely called ‘the myth of the “weak” American state’.17 So much to say that we must urgently place government structures and policies back at the heart of the sociology of the city (where Max Weber had properly put it18) hanging over and bearing on the dyadic relationships between class and ethnicity at the foot of the spatial structure, as shown in Figure 10.1. 2. The punitive management of poverty as component of neo-liberalism: How will the state react to and handle this advanced marginality that, paradoxically, it has fostered and entrenched at the point of confluence of the policies of economic ‘deregulation’ and social protection cutbacks? And how, in turn, will the normalization and intensification of social insecurity in territories of urban relegation contribute to redrawing the perimeter, programmes, and priorities of public authority (I use this expression on purpose)? The two-way relationship between class transformation and state re-engineering in its social and penal missions are the topic of ghettoization (L. Wacquant, ‘Designing Urban Seclusion in the 21st Century’, Perspecta: The Yale Architectural Journal, 43 (2010), 165–78), among them the primitive accumulation of social, economic, and cultural capital in a separate life-sphere liable to give them a shared collective identity and an increased capacity for collective action, in the political field in particular. 17  M. J. Novak, ‘The Myth of the “Weak” American State’, American Historical Review, 113 (3) (2008), 752–72. 18  M. Weber, The City, ed. Don Martindale (New York: The Free Press [1921] (1958)).

Marginality, Ethnicity, and Penality

277

the second book, Punishing the Poor: The Neoliberal Government of Social Insecurity, which covers the left-hand side of the ‘deadly triangle’ determining the destiny of the urban precariat.19 State managers could have ‘socialized’ this emerging form of poverty, by checking the collective mechanisms that feed it, or ‘medicalized’ its individual symptoms; they opted instead for another route, that of penalization. Thus was invented in the United States a new politics and policy of management of urban marginality married with restrictive social policy—through the supersession of protective welfare by mandatory workfare, whereby assistance becomes conditional on orienting oneself toward degraded employment—and expansive penal policy— intensified by the concurrent drift from rehabilitation to neutralization as operant philosophy of punishment and centred on declining and derelict urban areas (the US hyperghetto, dilapidating working-class banlieues in France, ‘sink estates’ in the UK, krottenwijk in the Netherlands, etc.) delivered to public vituperation by the discourse of territorial stigmatization in the dualizing metropolis. This policy contraption will then spread and mutate through a process of ‘treasonous translation’ across national borders, in accordance with the make-up of social space and the configuration of the political-administrative field particular to each receiving country.20 Punishing the Poor effects three breaks to roll out three major arguments. The first break consists of decoupling crime from punishment so as to establish that the irruption of the penal state, and thus the great comeback of the prison (which had been declared moribund and destined to disappear in short order around 1975),21

19  L. Wacquant, Punishing the Poor: The Neoliberal Government of Social Insecurity (Durham, NC, and London: Duke University Press, 2009). 20  Those who would doubt the relevance of the US workfare regime to non-Anglo-Saxon countries should consult I. Lødemel and H. Trickey’s book, neatly entitled ‘An Offer You Can’t Refuse’: Workfare in International Perspective (Bristol: Policy Press, 2001). Already over a decade ago, it documented the generalized drift of social policies from the rights to the obligations of recipients, the multiplication of administrative restrictions on access, and the contractualization of support, as well as the introduction of mandatory work programmes in six European Union countries. In his meticulous review of two decades of programmes of ‘social welfare activation’, Jean-Claude Barbier warns against sweeping generalizations and stresses cross-national as well as intra-national variations in architecture and outcomes; but he concedes that, aside from fostering ‘cost-containment’, these programmes partake of ‘a deep ideological transformation’ that has fostered everywhere ‘a new “moral and political logic” articulated to a moralizing discourse of “rights and duties” ’ (‘Le Workfare et l’activation de la protection sociale, vingt ans après: beaucoup de bruit pour rien? Contribution à un bilan qui reste à faire’, Lien social et politiques, 61 (2009), 23–36 at 30). For a broader discussion of the political-economic roots and variants of the ‘workfare state’, see J. Peck, Workfare States (New York: The Guilford Press, 2001). 21 When Michel Foucault (1975) published Surveiller et punir (translated two years later as Discipline and Punish), the international consensus among analysts of the penal scene was that the prison was an obsolete and discredited institution. Confinement was unanimously viewed as a relic of a bygone age of punishment fated to be supplanted by alternative and intermediate sanctions in the ‘community’ (this was the peak of the so-called ‘anti-institutional’ movement in psychiatry and of mobilization in favour of ‘decarceration’ in penology). Foucault himself stressed that ‘the specificity of the prison and its role as seal are losing their raison d’être’ with the diffusion of carceral disciplines ‘through the entire thickness of the social body’ and the proliferation of agencies entrusted with ‘wielding a power of normalization’: Surveiller et punir. Naissance de la prison (Paris: Gallimard, 1975;

278

Loïc Wacquant

is a response not to criminal insecurity but to the social insecurity spawned by the precarization of wage labour and to the ethnic anxiety generated by the destabilization of established hierarchies of honour (correlative of the collapse of the black ghetto in the United States and of the settlement of immigrant populations and advances in supranational integration in the European Union). The second break is to encompass in one and the same model the turnabout of penal policy and the permutations of social policy that are customarily kept separate, in both governmental and scholarly visions. For these two policies are mutually imbricated: they are aimed at the same populations caught in the cracks and ditches of the polarized socio-spatial structure; they deploy the same techniques (case files, surveillance, denigration, and graduated sanctions) and obey the same moral philosophy of behaviourist individualism; and the panoptic and disciplinary objectives of the former tend to contaminate the latter. To effect this integration, I enlist Bourdieu’s concept of ‘bureaucratic field’, which leads me to revise Piven and Cloward’s classic thesis on ‘regulating the poor’ through welfare: henceforth, the Left hand and the Right hand of the state join together to effect the ‘double punitive regulation’ of the unstable fractions of the post-industrial proletariat.22 The third rupture resides in ending the sterile confrontation between the advocates of economic approaches inspired by Marx and Engels, who construe criminal justice as an instrument of class compulsion, deployed in linked relation with swings in the labour market, and the culturalist approaches derived from Émile Durkheim, according to which punishment is a language that helps to draw boundaries, revivify social solidarity, and express the shared sentiments which found the civic community. It suffices, thanks to the concept of bureaucratic field, to bring together the material and the symbolic moments of any public policy to realize that penality can perfectly well fulfil both the functions of control and communication either simultaneously or successively, and thus operate in concert in the expressive and the instrumental registers. Indeed, one of the distinctive traits of neo-liberal penality is its teratological accentuation of its mission of figurative extirpation of danger and pollution from the social body, even at the cost of reducing rational crime control, as illustrated by the hysterical revamping of sex offenders sentencing and supervision in most advanced societies. I conclude Punishing the Poor by contrasting my model of penalization as political technique for managing urban marginality with Michel Foucault’s characterization of the ‘disciplinary society’, David Garland’s thesis of the emergence of the ‘culture of control’, and the vision of neo-liberal policy propounded by English trans. Discipline and Punish: The Birth of the Prison (New York: Vintage 1977), 358, 354, 359. Since then, against all expectations, the incarceration rate has boomed practically everywhere even as the overall incidence of crime (and especially violent crime) has stagnated or declined: it has increased fivefold in the United States and doubled in France, Italy, and England; it has quadrupled in the Netherlands and Portugal and increased sixfold in Spain. 22 See P. Bourdieu, ‘Esprits d’État: Genèse et structure du champ bureaucratique’, Actes de la recherche en sciences sociales, 96–7 (1993), 49–62; (English trans. ‘Rethinking the State: On the Genesis and Structure of the Bureaucratic Field’, Sociological Theory, 12 (1994), 1–19); F. F. Piven and R. A. Cloward, Regulating the Poor: The Functions of Public Welfare (New York: Vintage, 1993).

Marginality, Ethnicity, and Penality

279

David Harvey.23 In doing so, I demonstrate that the expansion and glorification of the penal arm of the state (centred on the prison in the United States and led by the police in the European Union) is not an anomalous deviation from, or a cor­ruption of, neo-liberalism but, on the contrary, is one of its core constituent components. Just as at the end of the sixteenth century, the nascent modern state innovated conjointly poor relief and penal confinement to stem the flow of tramps and beggars then invading the trading cities of Northern Europe,24 so at the close of the twentieth century the neo-liberal state bolstered and redeployed its policing, judicial, and carceral apparatus to stem the disorders caused by the diffusion of social insecurity at the bottom of the ladder of classes and places, and staged the garish spectacle of law-and-order pornography to reaffirm the authority of a government wanting in legitimacy due to having forsaken its established duties of social and economic protection. 3. The transformative synergy between racialization and penalization: The crescendo of advanced marginality and the turn toward its punitive containment have both been powerfully stimulated and inflected by ethnic division, rooted in the ‘black/white’ opposition in the United States and centred on the ‘national/postcolonial foreigner’ schism in Western Europe (with certain categories, such as the Roma, treated as quasi-foreigners even in their home countries). This inflection operates indirectly, through the bisectrix of the ‘class–race–state’ triangle shown in Figure 10.1 above,25 but also directly, through the two-way relationship between race-making and state-crafting. This relationship is figured by the right-hand side of the triangle and covered by the third book, Deadly Symbiosis: Race and the Rise of the Penal State.26 The synergistic connection between ethnoracial cleavage and the development of the penal state is the most difficult issue of this plank of investigations, both to raise and to resolve, and this for several reasons.27 First, the study of racial domination is a conceptual quagmire and a sector of social research where political posturing and moral ranting too often take precedence over analytical rigour and the quality of empirical materials.28 Next, the probability of slipping into the logic of the trial, which is the sworn enemy of sociological reasoning, already very high when one deals with the slippery and loaded notion of ‘racism’, is redoubled in 23 Foucault, Surveiller et punir (n. 21); D. Garland, The Culture of Control: Crime and Social Order in Contemporary Society (Chicago: University of Chicago Press, 2001); D. Harvey, A Brief History of Neoliberalism (New York: Oxford University Press, 2005). 24  C. Lis and H. Soly, Poverty and Capitalism in Pre-industrial Europe (London: Harvester Press, 1979); G. Rusche and O. Kirchheimer, Punishment and Social Structure (rev. edn.; New Brunswick, NJ: Transaction Publishers, 2003). 25  It is also dealt with briefly in c­ hapter 7 of Punishing the Poor, ‘The Prison as Surrogate Ghetto’. 26  Cambridge: Polity Press, 2015. 27  The concept of synergy (descended from the Greek syn, together, and ergon, work) conveys very well the idea that racialization and penalization operate in unison to produce state outcasts, in the manner of two symbolic organs acting together upon the functioning of the social body. When Émile Littré inserted it into his Dictionnaire de la langue française (1872–7), he traced the notion to physiology and defined it as ‘cooperative action or effort between various organs, various muscles. The association of several organs to accomplish a function’. 28  Wacquant, ‘Analytic of Racial Domination’ (n. 7).

280

Loïc Wacquant

the case at hand as we are tackling an institution, criminal justice, whose official mission is precisely to render judgments of culpability. Third, to understand the contemporary connection between race and public power, one must go back four centuries, to the founding of the American colony that would become the United States, without falling into the trap of making the present the inert and ineluctable ‘legacy’ of a shameful past that remains to be expiated. Finally, since ethnoracial division is not a thing but an activity (and a symbolic activity at that, a relation objectified and embodied), it is not congealed and constant; it evolves by fits and starts throughout history, precisely as a function of the operative mode of the state as paramount symbolic power. Deadly Symbiosis shows how ethnoracial cleavage lubricates and intensifies penalization and how, in turn, the rise of the penal state moulds race as a modality of classification and stratification, by associating blackness with devious dangerousness and by splitting the African-American population along a judicial gradient.29 The demonstration proceeds in three stages which take the reader to three continents. In the first stage, I reconstitute the historical chain of the four ‘peculiar institutions’ that have worked successively to define and confine blacks throughout the history of the United States:30 slavery from 1619 to 1865; the regime of racial terrorism in the South known as ‘Jim Crow’ from the 1890s to 1965; the ghetto of the Fordist metropolis in the North from 1915 to 1968; and finally the hybrid constellation born of the mutual interpenetration of the hyperghetto and the hypertrophic carceral system. I establish that the stupendous inflation in the confinement of lower-class blacks since 1973 (the black bourgeoisie has both suppported and benefited from that same penal expansion, which suffices to invalidate the counter-evangelical thesis of the coming of ‘The New Jim Crow’) resulted from the collapse of the ghetto as ethnic container and the subsequent deployment of the penal net in and around its remnants. This carceral mesh was strengthened by two convergent series of changes which, on the one hand, have ‘prisonized’ the ghetto and, on the other, have ‘ghettoized’ the prison, such that a triple relationship of functional surrogacy, structural homology, and cultural syncretism has coalesced between them.31 The symbiosis between the hyperghetto and the prison perpetuates the socio-economic marginality and the symbolic taint of the black urban subproletariat; and it revamps the meaning of ‘race’ and reshapes citizenship by secreting a racialized public culture of denigration of criminals.

29  L. Wacquant, ‘Race as Civic Felony’, International Social Science Journal, 181 (2005), 127–42. 30  Recall that the social and legal assignation to the category ‘black’ in the United States relies on genealogical descent from a slave imported from Africa and not on physical appearance, and that it magically ‘erases’ ethnoracial mixture (which concerns the vast majority of persons deemed black) by strict application of the principle of ‘hypodescent’ according to which the offspring of a mixed union belong to the category considered inferior. This symbolic configuration, which prefigures the extreme spatial and social isolation of African Americans in their society, is virtually unique in the world. See J. F. Davis, Who Is Black? One Nation’s Definition (University Park, Pa.: Pennsylvania State University Press, 1991). 31  L. Wacquant, ‘Deadly Symbiosis: When Ghetto and Prison Meet and Mesh’, Punishment & Society, 3 (2001), 95–133.

Marginality, Ethnicity, and Penality

281

I then expand this model to encompass the massive over-incarceration of post-colonial immigrants in the European Union, which turns out to be steeper in most member states than the over-incarceration of black Americans across the Atlantic—a revealing yet little-known fact that is either overlooked or denied by continental criminologists.32 The selective targeting and preferential confinement of foreigners issued from the West’s former empires take the two complementary forms of internal and external ‘transportation’, carceral expurgation and geographic expulsion (dramatized by the bureaucratic-cum-journalistic ceremony of the ‘charter flight’). These are complemented by the rapid development of a vast network of detention camps reserved for irregular migrants and by aggressive policies of detection and exclusion that incite informality among those migrants and normalize the ‘misrule of law’ across the continent as well as export it to sending countries via the ‘externalization’ of programmes of immigration and asylum control.33 All these measures aim to trumpet the fortitude of the authorities and to reaffirm the boundary between ‘them’ and a European ‘us’ that is painfully crystallizing.34 The penalization, racialization and depoliticization of urban turbulences associated with advanced marginality thus proceed apace and reinforce one another in a circular nexus on the European continent as in the United States. The same logic is at work in Latin America, which is where I last take the reader in order to scrutinize the militarization of poverty in the Brazilian metropolis as revelator of the deep logic of penalization.35 In a context of extreme inequalities and rampant street violence backed by a patrimonial state that tolerates routine judicial discrimination by both class and colour and unfettered police brutality, and considering the appalling conditions of confinement, to impose punitive containment on the residents of declining favelas and degraded conjuntos is tantamount to treating them as enemies of the nation. And it is guaranteed to fuel 32  L. Wacquant, ‘“Enemies of the Wholesome Part of the Nation”: Postcolonial Migrants in the Prisons of Europe’, Sociologie (Amsterdam), 1 (2005), 31–51. 33  D. Broeders and G. Engbersen, ‘The Fight against Illegal Migration: Identification Policies and Immigrants’ Counterstrategies’, American Behavioral Scientist, 50 (2007), 1592–609; B. Ryan and V. Mitsilegas (eds.), Extraterritorial Immigration Control (The Hague: Martinus Nijhoff Publishers, 2010). 34  The infamous speech delivered by Nicolas Sarkozy in Grenoble in July 2010 offers a hyperbolic as well as outlandish illustration of this logic of symbolic segmentation and vilification through penalization. Concerned to restore his blown credibility on the issue of public safety with a view to the 2012 presidential elections, the French head of state officially declared ‘war on traffickers and offenders’ and announced the appointment of a tough police chief to the post of local prefect. He directly linked undesirable foreigners to criminality (even though the incident that prompted his speech involved only French citizens); he singled them out for the full wrath of the state and prescribed enhanced and overtly discriminatory sanctions by the justice system (proposing, in addition to mandatory minimum sentences, to strip of their citizenship ‘French nationals naturalized for less than 10 years’ if they are convicted of acts of violence towards the police—a measure in direct violation of the French constitution and European conventions). And he launched a police campaign to dismantle ‘illegal Roma camps’ and to expel their residents en masse, aiming to rack up numbers of arrests and provide video footage for the evening TV news. This flash of law-and-order pornomania earned France the vigorous diplomatic protests of Romania and Bulgaria, official remonstrations and threat of sanctions from the European Union, and wide international reprobation (from the Vatican, the UN, etc.). 35  L. Wacquant, ‘The Militarization of Urban Marginality: Lessons from the Brazilian Metropolis’, International Political Sociology, 2 (2008), 56–74.

282

Loïc Wacquant

disrespect for the law and the routine abuse, as well as runaway expansion, of penal power, which one can indeed observe across South America in response to the conjoint rise of inequality and festering of marginality.36 This Brazilian excursus confirms that the vector of penalization always aims highly selectively, striking as a matter of structural priority those categories doubly subordinated in the material order of class and in the symbolic order of honour.

II I come now to the theoretical inspiration for my work, which is not always clearly perceived by my readers (or else only faintly or elliptically), even as it provides the key to the overall intelligibility of a set of investigations which, without it, might seem rather dispersed if not disconnected. In order to disentangle the triangular connections between class restructuring, ethnoracial division, and state-crafting in the era of triumphant neo-liberalism, I have adapted several concepts developed by Pierre Bourdieu (1997) and put them to work on new fronts—marginality, ethnicity, penality—from the micro-level of individual aspirations and interpersonal relationships in everyday life to the meso-level of social strategies and urban constellations to the macrosociological level of state forms (see Figure 10.2).37 Symbolic power is ‘the power to constitute the given by enunciating it, to make people see and believe, to confirm or transform the vision of the world, and thereby action upon the world, and thus the world itself’.38 It illuminates marginality as social liminality (translating alternately into civic invisibility or hypervisibility), penality as state abjection, and racialization as cognitively based violence. More broadly, it symbolic power (protective pole)

STATE E bureaucratic field

(disciplinary pole)

social space CITY

CLASS (market)

habitus

RACE (ethnicity)

Figure 10.2  The underlying theoretical architecture 36  M. M. Müller, ‘The Rise of the Penal State in Latin America’, Contemporary Justice Review, 15 (2012), 57–76. 37  See P. Bourdieu, Méditations pascaliennes (Paris: Seuil, 1997; English trans. Pascalian Meditations (Cambridge: Polity Press 1999)). 38  P. Bourdieu, Langage and Symbolic Power, ed. John Thompson (Cambridge: Polity Press, 1991), 170.

Marginality, Ethnicity, and Penality

283

exposes how public policies contribute to producing urban reality through their activities of official classification and categorization. (One example in France is the invention of the notion of ‘sensitive neighbourhood’ and the nefarious effects it has induced, not only upon the behaviour of state bureaucrats, the media and firms, but also among residents of the areas thus denigrated as well as among their neighbours). Bureaucratic field refers to the concentration of physical force, economic capital, cultural capital, and symbolic capital (entailing, in particular, the monopolization of judicial power) that ‘constitutes the state as holder of a sort of meta-capital’ enabling it to impact on the architecture and functioning of the various ‘fields’ that make up a differentiated society.39 It designates the web of administrative agencies that both collaborate to enforce official identities and compete to regulate social activities and enact public authority. Bureaucratic field puts the spotlight on the distribution (or not) of public goods and enables us to link together social policies and penal policies, to detect their relationships of functional substitution or of colonization, and thus to reconstruct their convergent evolution as the product of struggles about and within the state, pitting its protective (feminine) pole and its disciplinary (masculine) pole, over the definition and treatment of the ‘social problems’ of which neighbourhoods of relegation are both the crucible and the point of fixation. Social space is the multidimensional ‘structure of juxtaposition of social positions’, characterized by their ‘mutual externality’, relative distance (close or far), and rank ordering (above, below, between), arrayed along the two fundamental coordinates given by the overall volume of capital agents possess in its different forms and by the composition of their assets, that is, ‘the relative weight’ of ‘the most efficient principles of differentiation’ that are economic and cultural capital.40 As ‘the invisible reality’, irreducible to observable interactions, that ‘organizes the practices and representations of agents’, social space helps us identify and map out the distribution of the efficient resources that determine life chances at different levels in the urban hierarchy, and then to probe correspondences—or, indeed, disjunctures—between the symbolic, social, and physical structures of the city.41 Habitus, defined as the system of socially constituted ‘schemata of perception, appreciation and action enabling us to effect the acts of practical knowledge’ that guide us in the social world,42 pushes us to reintroduce into the analysis the carnal experience of agents—and marginality, racialization, and incarceration are nothing if not bodily constraint, manifested most intensely intus et in cute. It helps us attend to ‘the psychosomatic action, wielded often through emotion and suffering’, through which people internalize social conditionings and social limits, such that the arbitrariness of institutions gets erased and their verdicts are accepted.43 It 39  Bourdieu, ‘Esprits d’État’ (n. 22), 52. 40  P. Bourdieu, Practical Reasons: On the Theory of Action (Cambridge: Polity Press, 1994), 20–2. 41 Bourdieu, Practical Reasons (n. 40), 25. 42 Bourdieu, Méditations pascaliennes (n. 37), 200. 43 Bourdieu, Méditations pascaliennes (n. 37), 205. It is revealing that Bourdieu evokes the pivotal passage of Franz Kafka’s 1914 short story In the Penal Colony (New York: Penguin, 2011), in which

284

Loïc Wacquant

invites us to trace empirically, rather than simply postulate, how social structures are retranslated into lived realities, as they become sedimented into socialized organisms in the form of dispositions towards action and expression. Such dispositions tend to validate and reproduce or, on the contrary, to challenge and transform, the institutions that produced them, depending on whether their conformation agrees with or diverges from the patterning of the institutions they encounter. There is, moreover, a relation of logical entailment and a two-way chain of causality running among these different levels (suggested by Figure 10.2):44 symbolic power imprints itself on social space by granting authority and orienting the distribution of efficient resources to the different relevant categories of agents. The bureaucratic field validates or amends this distribution by setting the mutual ‘exchange rate’ between the various forms of capital they possess. In other words, we cannot understand the organization of urban hierarchies, including whether and how powerfully they get ethnicized, without putting into our explanatory equation the state as a stratifying and classifying agency. In turn, the structure of social space becomes objectified in the built environment (think segregated residential neighbourhoods and the differential distribution of amenities across districts) and embodied in the cognitive, affective, and conative categories that steer the practical strategies of agents in everyday life, in their social circles, on the labour market, in their dealings with public institutions (police staff, welfare offices, housing and fiscal authorities, etc.), and therefore shape their subjective relationship to the state (which is part and parcel of the objective reality of that same state). The causal chain can then be retraced back from the bottom up: habitus propels the lines of action that reaffirm or alter the structures of social space, and the collective meshing of these lines in turn reinforces or challenges the perimeter, programmes, and priorities of the state and its categorizations. It is this conceptual gearing which articulates the ethnography of boxing presented in my book Body and Soul to the institutional comparison that organizes Urban Outcasts.45 In my eyes, these books are the two sides of a single investigation into the structure and experience of marginality (as indicated at the bottom of Figure 10.1), approached from two opposite but complementary angles: Body and Soul delivers a carnal anthropology of a bodily craft in the ghetto, a sort of phenomenological cross-section, from the standpoint of the ‘signifying agent’ dear to the pragmatists, embedded in an ordinary slice of life seen from within and from below, while Urban Outcasts lays out an analytic and comparative the sentence of the condemned is carved onto his body by a torture machine as a grotesque variation on what he calls the ‘cruel mnemotechnics’ through which groups naturalize the arbitrary that founds them. This scene puts us at the point where the material-cum-symbolic spear of the penal state encounters and pierces through the body of the offender in an official act of radical desecration resulting in physical annihilation: the citizen shall exist only within the historical ambit of the law. 44  For a fuller discussion of the internal relationships between these different concepts, which stresses the barycentric place of symbolic capital in its various incarnations, see P. Bourdieu and L. Wacquant, An Invitation to Reflexive Sociology (Chicago: University of Chicago Press, 1992). 45  L. Wacquant, Body and Soul: Notebooks of an Apprentice Boxer (New York: Oxford University Press, 2004).

Marginality, Ethnicity, and Penality

285

macrosociology of the ghetto, constructed from without and from above the lived world it frames.46 I use these notions as so many theoretical levers to machine concepts that help me to detect new forms of urban marginality, to identify state activities directed at producing it upstream and treating it downstream, and thence for sizing up emerging vectors of inequality in the dualizing metropolis in the age of diffusing social insecurity (see Figure 10.3). Thus, in Urban Outcasts, I lean against the notion of social space to introduce the triad of ghetto/hyperghetto/anti-ghetto and to dissect the changing socio-spatial constellations that contain the dispossessed and dishonoured populations trapped at the bottom of the ladder of places that make up the city.47 Marrying Bourdieu’s theory of symbolic power with Goffman’s analysis of the management of ‘spoiled identities’, I coin the concept of territorial stigmatization to reveal how, through the mediation of cognitive mechanisms operating at multiple enmeshed levels, the spatial denigration of neighbourhoods of relegation affects the subjectivity and the social ties of their residents as well as the state policies that mould them.48 In keeping with the precepts of Bachelard’s epistemology, staging sovereignty

SYMBOLIC POWER

liberal -paternalism

STATE

“prisonfare”

BUREAUCRATIC FIELD

punitive containment territorial stigmatization advanced marginality

SOCIAL SPACE ghetto

CLASS (market)

race = civic crime negative sociodicy judicial citizenship penal segmentation hyperincarceration

CITY hyperghetto antighetto HABITUS

RACE (ethnicity)

Figure 10.3  The main concepts developed 46  A detailed examination of the life strategies of a ‘hustler’ in the predatory street economy (L. Wacquant, ‘Inside the Zone: The Social Art of the Hustler in the Black American Ghetto’, Theory, Culture & Society, 15 ([1992] 1998), 1–36) and of the normative twist and practical stretch that the hyperghetto imposes on marriage (L. Wacquant, ‘Un mariage dans le ghetto’, Actes de la recherche en sciences socials, 113 (1996), 63–84) are two of the multiple points of junction between these two levels and modes of analysis: in both of those case studies, my chief field informants were also boxers. Likewise, the extended judicial enmeshment of my best friend and ‘ring buddy’ at the Woodlawn Boys Club across two decades provided me with a live analyser of the relationships between marginality and penality in biographical time and at the microsociological scale. 47 Wacquant, Urban Outcasts (n. 12). See also ‘Urban Desolation and Symbolic Denigration in the Hyperghetto’, Social Psychology Quarterly, 20 (3) (2010), 1–5. 48 E. Goffman, Stigma: Notes on the Management of Spoiled Identity (Englewood Cliffs, NJ: Prentice-Hall, 1964). This concept has since been developed theoretically and extended empirically across three continents (L. Wacquant, ‘Territorial Stigmatization in the Age of Advanced Marginality’, Thesis Eleven, 91 (2007), 66–77; Wacquant, ‘Urban Desolation and Symbolic Denigration in the Hyperghetto’ (n. 47); Wacquant, ‘Designing Urban Seclusion in the 21st Century’ (n. 16)). See also

286

Loïc Wacquant

I develop an ideal-typical characterization of the new regime of advanced marginality (thus called because it is not residual, cyclical, or transitional but organically linked to the most advanced sectors of the contemporary political economy, and notably to the financialization of capital) which supplies a precise analytic grid for international comparison. In Punishing The Poor and a suite of articles derived from it, I elaborate the notion of prisonfare by conceptual analogy with that of ‘welfare’, to designate the lattice of policies—encompassing categories, bureaucratic agencies, action programmes, and justificatory discourses—that purport to resolve urban ills by activating the judicial arm of the state rather than its social and human services.49 I suggest that punitive containment is a generalized technique for governing marginalized categories that can take the form of assignation to a dispossessed district or endless circulation through penal circuits (police, court, jail, prison, and their organizational tentacles: probation, parole, criminal justice databases, etc.). I describe the ascendant policy contraption, which relies on the double regulation of the poor through disciplinary ‘workfare’ and neutralizing ‘prisonfare’, as ‘liberal-paternalist’ since it applies the doctrine of laissez-faire et laissez-passer at the top of the class structure to the holders of economic and cultural capital, but turns out to be intrusive and supervisory at the bottom, when it comes to curbing the social turbulences generated by the normalization of social insecurity and the deepening of inequalities. This contraption partakes of the erection of a Centaur state that presents a radically different profile at the two ends of the scale of classes and places, in violation of the democratic norm mandating that all citizens be treated in the same manner. Its rulers use the ‘war on crime’ (which is not a war) as a bureaucratic theatre geared to reaffirming their authority and to staging the ‘sovereignty’ of the state at the very moment when this sovereignty is being breached by the unbridled mobility of capital and by juridical-economic integration into supranational political ensembles. In Deadly Symbiosis, I propose to replace the seductive but misleading notion of ‘mass incarceration’, which currently frames and constricts civic and scientific debates on prison and society in the United States (I used it myself, rather unthinkingly in my publications prior to 2006), by the more refined concept of hyperincarceration, in order to stress the extreme selectivity of penalization according to class position, ethnic membership or civic status, and place of residence—a selectivity which is a constitutive feature (and not an incidental attribute) of the policy of punitive management of poverty).50 I recount that punishment is not just a direct indicator of solidarity and core political capacity for the state, as Émile Durkheim asserted over a century ago in the investigations carried out within the frame of the international and interdisciplinary network , and the selective bibliography compiled by Tom Slater, Virgílio Pereira, and Loïc Wacquant for the special issue of Environment & Planning A on the theme of ‘Territorial Stigmatization in Action’, 46 (6) (June 2014). 49  L. Wacquant, ‘Class, Race and Hyperincarceration in Revanchist America’, Daedalus, 139 (3) (2010), 74–90; L. Wacquant, ‘Prisoner Reentry as Myth and Ceremony’, Dialectical Anthropology, 34 (4) (2010), 604–20; L. Wacquant, ‘The Wedding of Workfare and Prisonfare Revisited’, Social Justice, 38 (1–2) (2011), 203–21. 50  Wacquant, ‘The Wedding of Workfare and Prisonfare Revisited’ (n. 49), 218–19.

Marginality, Ethnicity, and Penality

287

De la division du travail social:51 it is also the paradigm of public dishonour, inflicted as a sanction for individual moral, and thus civic, ‘demerit’. This leads me to characterize penality as an operator of negative sociodicy: through its ordinary functioning more so than through the glare of the scandals that it alternately unleashes and appeases,52 criminal justice produces an institutional justification for the misfortune of the precariat at the bottom of the social scale, a justification that echoes the positive sociodicy of the good fortune of the dominant effected by the distribution of credentials from elite universities on the basis of academic ‘merit’ at the top of that same scale.53 Penal sanctions and their official recording in judicial files or ‘rap sheets’ (casier judiciaire in France, Führungszeugnis in Germany, strafblad in the Netherlands, etc.) operate in the manner of ‘reverse degrees’: they publicly attest to the individual unworthiness of their bearers and incite the routine curtailment of their life chances, as revealed by the amputation of the social and marital ties, housing options, employment opportunities, and earnings of ‘ex-cons’ in nearly every advanced country. It suffices, then, to construct ‘race as civic felony’54 to detect the deep kinship— which is much more than a similarity or an affinity, even an ‘elective affinity’ à la Weber—between racialization and penalization: both entail an amputation of social being that is validated by the supreme symbolic authority. Racial categorization and judicial sanction produce state outcasts, who are all the more diminished as these are more closely conjugated.

IV I apologize if I have been allusive when I should have been didactic, and vice versa, but to cover my subject while remaining brief I have had to simplify my reasoning and to compress my arguments. Nonetheless, I hope that these rudiments of analytic cartography will enable you to better understand and, especially, to link together the three works discussed. The empirical progress effected and the conceptual novelties proposed in each book are directly dependent upon those made in the other two; the whole is more than the sum of the parts that each corresponding group of readers tends to autonomize according to the focus of their subfield.55 One example: I would not have detected the subterranean link between 51  The Division of Labor in Society (London: Palgrave Macmillan, 2007). 52  A. Garapon and D. Salas, Les nouvelles sorcières de Salem. Leçons d’Outreau (Paris: Seuil, 2006). 53  P. Bourdieu, The State Nobility: Elite Schools in the Field of Power (Cambridge: Polity Press, 1998). I adapt here the duality of ‘theodicy’ proposed by Max Weber in his ‘Social Psychology of the World Religions’, in H. H. Gerth and C. W. Mills (eds.), From Max Weber: Essays in Sociology (New York: Oxford University Press, [1915] 1946, 267–301), which contrasts doctrines that validate ‘the external and inner interests of all ruling men’ (Theodizee des Glückes) with doctrines that legitimize and rationalize the suffering of ‘socially oppressed strata’ (Theodizee des Leidens). 54  Wacquant, ‘Race as Civic Felony’ (n. 29). 55  It is revealing that the contributions to the symposia devoted to Urban Outcasts (by City in 2008, International Journal of Urban and Regional Research, Revue française de sociologie, and Pensar in 2009, and Urban Geography in 2010) and to Punishing the Poor (organized by the British Journal of Criminology, Theoretical Criminology, Punishment & Society, Critical Sociology, and Studies in Law, Politics & Society, Criminology & Justice Review, The Howard Journal of Criminal Justice, Amerikastudien,

288

Loïc Wacquant

penalization and racialization as kindred forms of state infamia if I had not first theorized territorial stigmatization as one of the distinctive properties of advanced marginality, and then discerned the functional and structural parallelism between the hyperghetto and the prison. I should make it clear, by way of coda and reassurance, that I did not sit down, back around 1990, with the extravagant project of writing a trilogy in mind. It is the unplanned unfolding of my investigations, the empirical advances (and repeated retreats) it permitted as well as the theoretical problems it made emerge (or vanish) that have taken me, over the years, from one to another vertex of the triangle Class–Ethnicity–State; and it is unforeseen existential connections that have propelled me along the sides that tie them to one another.56 At the start, there was the shock—inseparably emotional and intellectual— that I experienced in the face of the gruesome urban and human desolation of the vestiges of the South Side, whose lunar landscape stretched away, literally, from my doorsteps when I landed in Chicago. This shock pushed me to enter into the boxing gym construed as an observation post from which I took up the question of the coupling of ‘race and class’ in the American metropolis and set about reconstructing the notion of ghetto from the ground up, in opposition to the gaze from afar and from above that dominates the national sociology on the topic.57 In response to the irruption of the panic discourse on the alleged ‘ghettoization’ of working-class districts in France and its ensuing diffusion across Europe, I enriched my historical perspective by adding a comparative axis. This comparison highlights the role of the state in the production of marginality, a role that is pivotal yet different on the two sides of the Atlantic. Then, magnetized by the craft of the boxer, I drew up the life stories of my gym buddies and discovered that nearly all of them had gone through prison or jail gates: if I wanted to map out the space of possibilities open to them—or, as the case may be, closed to them—I imperatively had to bring the carceral institution into my sociological line of sight. It was then I realized that the bulimic growth of the American penal system since 1973 is perfectly concomitant with and complementary to the organized atrophy of public aid and its disciplinary reconversion into a springboard toward Prohistoria, and Revista española de sociología) reproduce the established separation between disciplines (with, broadly, urban geography and sociology on one side and criminology on the other, while social work and political science are conspicuous by their absence), and deal exclusively with only one of these two books while omitting the other. The collective book edited by Peter Squires and John Lea (Criminalisation and Advanced Marginality: Critically Exploring the Work of Loïc Wacquant (Bristol: Policy Press, 2012)) is a rare attempt to connect the schema of advanced marginality to my analysis of the penal state, but at the price of neglecting the racialization–penalization axis. The tome assembled by González Sánchez does cover carnality, marginality, and penality, but its contributors typically stay within one of these rubrics rather than connect all three (Ignacio González Sánchez (ed.), Teoría social, marginalidad urbana y Estado penal. Aproximaciones al trabajo de Loïc Wacquant. Madrid: Dykinson, 2012). 56  See L. Wacquant, ‘The Body, the Ghetto and the Penal State’, Qualitative Sociology, 32 (1) (2009), 101–29, for a fuller discussion. 57  Wacquant, ‘Three Pernicious Premises in the Study of the American Ghetto’ (n. 9).

Marginality, Ethnicity, and Penality

289

precarious employment. The historical revisit of the invention of prison in the sixteenth century subsequently confirmed the organic link that has joined poor relief and penal confinement ever since their origin, and it provides a structural basis for the empirical intuition of their functional complementarity. Meanwhile, in Les Prisons de la misère (Prisons of Poverty) I charted the planetary diffusion of the policing strategy and trope of ‘zero-tolerance’, spearhead of the penalization of poverty in the polarizing city. I showed that it operates in the wake of the ‘deregulation’ of deskilled work and of the conversion of welfare into workfare: in sum, it partakes of the building of the neo-liberal Leviathan.58 At each stage, ethnoracial division serves as a catalyst or multiplier: it accentuates the fragmentation of wage labour by segmenting workers and pitting them against one another; it facilitates welfare retrenchment and the deployment of the penal apparatus, as it is much easier to toughen up policies directed at welfare recipients and criminals when the latter are perceived as civic ‘outsiders’, congenitally tainted and terminally incorrigible, opposed in every respect to ‘established’ citizens.59 But, above all, racial branding turns out to be similar in nature to penal punishment: they are two twin manifestations of state dishonour. Thus, without ever setting out to do so, I have come to practise a kind of eccentric (some might say quirky) sociology of political power, since in the end I find myself confronted with the question of the state as material and symbolic agency, and dragged reluctantly into theoretical and comparative debates on the nature of neo-liberalism and the contribution of penality to its advent.60 The ‘deadly triangle’ that decides the fate of the urban precariat is an ex post schema that emerged gradually as I progressed in the investigations of which I recapped the main lines in this chapter. This explains the fact that the three books that synthesize them were published late (with a lag of nearly a decade, on average, from the data production phase) and also out of order: I had to rethink them and to rewrite them several times over so as to better separate and link them at the same time. This analytic configuration is also what gives more strength and weight to each of them. This chapter is then an invitation to a generative and transversal reading, not for the aesthete pleasure of breaking with academic conventions, but so that we may collectively draw out the full empirical and theoretical benefits of connecting

58  L. Wacquant, Prisons of Poverty (rev. and enlarged edn.; Minneapolis: University of Minnesota Press, 2009); and Wacquant, “The Global Firestorm of Law and Order: On Neoliberalism and Punishment’, Thesis Eleven, 122 (Spring 2014), 72–88. 59  To invoke a dichotomy dear to N. Elias and J. L. Scottson, The Established and the Outsiders (London: Sage, 1994). 60 A Bourdieusian approach in terms of the ‘rightward tilting of the bureaucratic field’ (itself caught up in the drift of the field of power towards the economic pole) allows me to chart a via media between the two dominant and symmetrically mutilated models of neo-liberalism as ‘market rule’ or ‘governmentality’ inspired by Marx and Foucault respectively (see Wacquant, ‘Three Steps to a Historical Anthropology of Actually Existing Neoliberalism’ (n. 10), and the seven responses to this thesis in subsequent issues of the same journal).

290

Loïc Wacquant

the themes I have discussed. I shall therefore conclude with this analytic cri du coeur: scholars of urban marginality, scholars of ethnicity, and scholars of penality, unite. You have nothing to lose but your intellectual chains! And you have a world of scientific discoveries to gain as well as a wealth of practical recommendations to interject into the public debate.

11 ‘It Isn’t Just About You’ Victims of Crime, their Associated Duties, and Public Wrongs S. E. Marshall*

A Welsh woman jailed for retracting ‘truthful’ rape claims against her alleged violent husband is to challenge her conviction at the court of appeal in London after a legal team offered to represent her for free. The 28-year-old mother, known as Sarah, wants to clear her name after being released from jail by the lord chief justice, Lord Judge, last month. She had served 18 days of an eight-month sentence for perverting the course of justice. . . . Judge ruled that the judiciary had a duty to show a ‘broad measure of compassion for a woman who had already been victimised’. But he left her with a criminal record when he replaced her custodial sentence with a community sentence and a two-year supervision order. “I’ve done nothing wrong and yet I’m the one being punished,” said Sarah this week from her sister’s home in Powys. . . . The case against her—described as ‘a miscarriage of justice’ by campaigners— was not that she lied about the rapes, but about her claim that they had never happened. When he set her free from jail, Judge said Sarah’s original sentence ‘had to be assessed on the basis that she had perverted the course of justice by falsely retracting a true allegation that her husband had indeed raped her’.1

In what follows I hope to show that the claim that ‘Sarah’ had ‘done nothing wrong’ is mistaken, and that a crucial part of the wrong that she did is precisely a function of being a victim. Victims, I will suggest, have duties, as well as rights. Furthermore, one implication of my argument will be that victims have duties to those who have committed wrongs against them. Thus the wrong done in such a case as the one outlined here is not simply that of ‘perverting the course of justice’, *  There are many people who have patiently indulged me in discussions of different version of this chapter. Those discussions have been enormously helpful and I owe them all my thanks. 1  The Guardian, 18 December 2010; (emphasis mine). The husband in the case had been originally charged with six counts of rape on the basis of the woman’s accusation.

292

S. E. Marshall

or of perjury—though it is both of those—but that, paradoxical (or outrageous)2 as it might seem, in lying in the very particular way that she did, ‘Sarah’ wronged her husband. These arguments, and my account of the particular kind of wrongs at stake here, are grounded in an account of crimes as ‘public wrongs’ which had a much earlier outing in two papers co-authored with R. A. Duff.3 What I say in this chapter might, I hope, serve as one way of fleshing out that idea. It will be a further question (one that I cannot discuss in detail here, though I will say a little more about it later) whether her commission of that wrong could be justified or excused: but my central argument will be that she did commit a wrong (that she had ‘done something wrong’), for which a justification or excuse is then needed. There are no doubt many features of this case that would be worth discussing: the problematic nature of the investigation and prosecution of rape and domestic violence cases being the one that many would think the most significant and pressing. These are not, however, directly my concern in this chapter. Indeed, for the purposes of my argument it could just as well be a case of fraud, burglary, or wounding. There are, however, three features of the particular case to be highlighted by way of introducing the topic of this chapter. First, we need to be clear that ‘Sarah’ is appealing against the conviction itself, not simply against the sentence. The actual sentence was the subject of an earlier appeal, as a result of which the original sentence was reduced, as indicated in the brief newspaper report above.4 The point is that she wishes to ‘clear her name’, and the claim made both by ‘Sarah’ herself, and by those campaigning on her behalf, is that she has ‘done nothing wrong’, although she does not deny that she made false claims to the effect that there had been no rapes. Thus, secondly, we should note that she now claims that her original accusations were true. It is very important to be clear about this, since many commentators on this particular case have treated it as if it were a rather different kind of case: namely that of a woman being punished for making false allegations of rape. This leads to the third feature I want to highlight: the case against ‘Sarah’ depends upon the truth of her claim that she was raped—a claim to which the prosecution agreed there was at least some substance. Indeed, at the Appeal Court the lawyer acting for the crown conceded that the crown ‘unreservedly accepts the factual background to this case. It’s plain that the woman was subjected to a lengthy period of domestic violence and it’s implicit in the plea that she was the victim of rape.’5 It is significant then that the accusation against her husband remains in

2  My experience is that this is indeed how the idea strikes many people, at least at first sight, if not at second and third sight. 3  See S. E. Marshall and R. A. Duff, ‘Criminalisation and Sharing Wrongs’, Canadian Journal of Jurisprudence, 11 (1998), 7; S. E. Marshall and R. A. Duff, ‘Public and Private Wrongs’, in James Chalmers et al. (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010), 70. 4  The Appeal Court dismissed the appeal against conviction on 13 March 2012: R v A [2012] 2 Cr App R 8. The woman now seeks to take her case to the Supreme Court. 5  The Guardian, 13 March 2012; . See also R v A [2012] 2 Cr App R 8.

Victims of Crime

293

the public domain, even though the charges did not, and now will not, get to trial (since the husband was formally acquitted after she withdrew her claims of rape).6 Moreover, despite the husband’s acquittal, that accusation still has a certain legal status, since its truth was publicly accepted not only by the prosecution, as noted above, but also by the Court of Appeal itself: the conduct for which ‘Sarah’ was convicted, said Lord Judge, ‘involved repeated assertions which led directly to the acquittal of the man who had committed rape on more than one occasion’.7 I note, though only in passing for the moment, that nowhere in the reporting of the case has the presumption of innocence been mentioned, let alone invoked, although the Appeal Court does note that ‘[n]‌evertheless it is only fair to the appellant’s husband to record that he has consistently denied the allegations and has not had any opportunity to publicly challenge or refute them’.8 The more relevant point to note here, however, is that both ‘Sarah’ and those who support her argue that she is the victim, and that, at least by implication, being a victim undermines the case against her.9 The question we should ask here then is: what exactly does being a ‘victim’ have to do with it? Are we to suppose that being a victim is, just by itself, some sort of excuse or even a justification for lying? Or is it that being a victim of this particular sort of crime is special? Suppose we change the case to one of fraud say, keeping everything else in the example in place, would the appeal to victimhood have the same force? Now such questions are obviously, in part at any rate, rhetorical, but they provide a starting point for the main arguments of this chapter.

I.  Roles and Duties One thing to keep clearly in view is that ‘Sarah’s’ claim is that she is the victim of a crime and not simply the victim of wrongdoing or misfortune. Thus her claim and the arguments surrounding it get part of their sense from being located within a particular institution. Law, of which criminal law is a part, is a political institution, through which political communities express the decisions of the community regarding the way people should behave, the entitlements they have etc., and the views about these matters, however sound and however popular, which do not have the imprimatur of the community. By and large the law represents those standards which are considered in that community as expressing the decisions of that community.10 Attention needs to be paid, therefore, to the roles and status of the participants in that institution and recently philosophers, notably John Gardner, Malcolm 6  It is not a negligible matter that the husband was originally charged and released on conditional bail. 7  R v A [2012] 2 Cr App R. 8, at para. 58.    8  R v A, at para. 3. 9  So, Professor Clare McGlynn was quoted as saying, ‘Where is the recognition that the woman is the victim here?’: The Guardian, 23 November 2010; . 10 Joseph Raz, Between Authority and Interpretation (Oxford: Oxford University Press, 2009), 101–2.

294

S. E. Marshall

Thorburn, and Kimberly Brownlee,11 have in different ways addressed questions about the roles and responsibilities of officials. My general interest is in the normative structure of the roles of other citizens, who share with officials responsibility for and in the institution of criminal law: victims, witnesses, bystanders, and offenders. This chapter concerns only the victim’s role in the criminal process, and the implications of that role for the possible criminalization of the victim’s conduct. Clearly, a full theory of criminalization will need to say a great deal more about the interrelations between the victim’s role and the other roles mentioned above, not to mention the interrelation of all these roles with those of officials, than I discuss here. One way of characterizing the role of officials in the criminal law and its ­ processes—the ‘citizens in uniform’ conception—is usefully captured by John Gardner: It can be expressed as an admonition. ‘Don’t think that when you step into your official role (your “uniform”) you stop being yourself and can abdicate responsibility in your capacity as an ordinary member of the public (a “citizen”) for the things that you do. You still answer to the law as yourself, and you can’t hide behind your public role when you do it.’12

One might say that the account offered in my chapter characterizes ordinary citizens as ‘officials in plain clothes’—with the admonition that we cannot abdicate our responsibilities in the criminal process and hide behind our role as ‘ordinary citizens’, since all the roles in the criminal process are ones which we inhabit as citizens.13 I take roles to be structured by norms, and to be active rather than merely passive: filling a role, that is to say, involves undertaking a range of actions—actions either directly required by the relevant norms, or serving the goals specified by those norms. Furthermore, roles bring with them what can be called, following Wenar,14 ‘duty-based desires’. Thus, a parent qua parent wants whatever will help them to fulfil that role, even if their children’s flourishing is not central to their own well-being beyond their role as parent.15 This should not be understood, however, as a claim about the individual psychology of role bearers. It is, rather, a structural claim: This attribution of attitudes to role-bearers is not based on perceived regularities, or on insights into individuals’ psychological states. Rather, it follows from our understanding of the roles themselves. Consider the goalie. Part of what it is to be a goalie is to try to keep 11 See Kimberley Brownlee, ‘Responsibilities of Criminal Justice Officials’, Journal of Applied Philosophy, 27 (2010), 123; Malcolm Thorburn, ‘Criminal Law as Public Law’, in R. A. Duff and S. P. Green, Philosophical Foundations of Criminal Law (Oxford: Oxford University Press, 2011), 21; John Gardner, ‘Criminals in Uniform’, in R. A. Duff, Lindsay Farmer, S. E. Marshall, Massimo Renzo, and Victor Tadros (eds.), The Constitution of the Criminal Law (Oxford: Oxford University Press, 2013), 97. 12  Gardner, ‘Criminals in Uniform’ (n. 11), at 98. 13  This raises obvious questions about the privatization of various parts of the criminal justice process which I cannot pursue here but see e.g. Alon Harel, ‘Why Only the State May Inflict Criminal Sanctions: The Argument from Moral Burdens’, Cardozo Law Review, 28 (2006–7), 2629; Avihay Dorfman and Alon Harel, ‘The Case against Privatization’, Philosophy and Public Affairs, 41 (2013), 69–102. 14  Leif Wenar, ‘The Nature of Claim-Rights’, Ethics, 123 (2013), 202. 15  Wenar, ‘The Nature of Claim-Rights’ (n. 14), 210.

Victims of Crime

295

the ball out of one’s team’s goal. We do not learn this through surveys showing that 99 or even 100 percent of goalies have this common aim. Rather, an individual who does not try to keep the ball out of his team’s goal is not playing goalie. To play goalie is to try to stop the other team from scoring, so we attribute to goalies the corresponding desire. We can say either that ‘someone playing goalie has reason to want to stop the other team scoring’, or that ‘a goalie wants to stop the other team scoring’, or that ‘goalies qua goalies want to stop the other team scoring’. The three are equivalent: the reason comes with the role, and the role comes with the desire.16

To put it another way: the role bearer has an interest in being able to fulfil the role, i.e. to carry out the duties of that role. Just as officials have a role-based right to perform their duties, so citizens have a right to carry out theirs. That in turn will reflect on the role duties of others—especially, but not exclusively, the officials—to enable the role bearer to carry out that role and not to impede them. This, then, is an account that provides a foundation for my arguments, which are not just about victims’ duties but also about their rights, though it will not necessarily generate all the rights that are sometimes claimed for victims. Central to my argument is the claim that we should see victims, along with such other participants as witnesses, defendants, and jurors, as role bearers in the criminal justice system; and that their role should be understood not merely in passive terms of what may or must be done to or for them, but in active terms of their responsibilities and duties: I hope to make that claim more persuasive in what follows. How then should we characterize the role of the victim of crime in terms of the duties that partly constitute it? Consider the case of ‘Sarah’ above. As far as the case against her goes, the core of the wrongdoing for which she was prosecuted—perverting the course of justice—and for which she was convicted, was the lying nature of her retraction of her complaint against her husband. The case is complicated by the fact that she initially faced two (mutually incom­patible) accusations of lying. First, when she not only withdrew her original complaint that her husband had raped her, but ‘then proceeded to assert and reassert that her complaint had been false’, proceedings against him were stopped, and she was charged with perverting the course of justice by making a false complaint of rape. Then, when after discussing her position with her counsel and solicitor ‘she reasserted the truth of the original complaint’, she faced a further indictment alleging that she had perverted the course of justice by ‘ma[king] and pursu[ing] a false retraction’:17 she pleaded ‘not guilty’ to the first indictment, and was acquitted when the prosecution offered no evidence; it was the second indictment (to which she pleaded guilty) that became the focus of the prosecution and her subsequent appeals. Suppose instead that she had merely withdrawn her complaint, surely she would then have done no wrong? There might be all sorts of reasons why someone would choose to withdraw a complaint, or just not make a complaint at all, without any implication that the complaint was false. Surely the most that their role demands

16  Wenar, ‘The Nature of Claim-Rights’ (n. 14), 215.

17  R v A 2 Cr App R 8, para. 2.

296

S. E. Marshall

of victims is that they, like all other participants in the legal process,18 engage in it honestly and do not stand in the way of others performing their proper duties. However, while not making a complaint (perhaps) does not amount to ‘standing in the way’, withdrawing what I know to be a true charge might well amount to standing in the way of police and prosecutors who are trying to carry out their duties; though this will undoubtedly depend on our conception of the officials’ duties. On one view the role might be no more than to investigate only, but not all, allegations that are made to them. On a more expansive view the role might be to investigate all, or most, potential criminal wrongdoing, however it might come to their attention. (I do not tackle these issues here: they do, however, speak to my point that the roles which form the legal process are interdependent.) Still, whether a victim chooses to participate in the legal process in the first place is surely up to them. A particular kind of focus on victims’ rights might, indeed, bolster this thought. Such appeals to ‘victims’ rights’ have increasingly focused on demands for greater participation by victims in the judicial processes of criminal law and punishment;19 but the right to participate is typically portrayed as the right to participate if they wish to do so, rather than as a right to participate because it is their duty to do so. We can see here a similarity with some versions of Restorative Justice, which are inspired by Nils Christie’s early claim that, as he somewhat (over-)dramatically put it, the criminal law ‘steals’ our conflicts: taking away the ‘conflict’ which he claimed is the real nature of what we call ‘crime’ from those whom it directly concerns and transferring it to the professional world of the law.20 In this way those we call the ‘victim’ and the ‘offender’ are deprived of the chance to deal with the ‘conflict’ themselves. The solution would thus be to put matters back into the hands of those whose concern crimes (or rather ‘conflicts’) properly are. The only role then for any formal process would be to facilitate the discussion between the parties. Whether this kind of position could generate a duty on the part of victims to take part in the process of conflict resolution is perhaps arguable, although that seems implausible at first sight. Now, the connection between Restorative Justice and the more familiar conceptions of ‘victims’ rights’ should not be overplayed: one very marked difference would certainly be that part of some classical Restorative Justice arguments was that the very concept of ‘crime’ should be rejected and that what need to be addressed are simply ‘harms’ or ‘conflicts’. The outcome to be achieved was the ‘repair’ of a broken relationship between the parties—‘punishment’ was rejected as barbaric and pointless. However, the development of ‘victims’ rights’ does not necessarily involve any such attempt at reconceptualization. Quite to the contrary, appeals to ‘victims’ rights’ more commonly involve a demand for greater participation by victims in the criminal processes of investigation, prosecution, trial, and 18  Note that I use the term ‘legal process’ very broadly here to cover more than just court proceedings. 19  Some of the potentially perverse consequences of some of these demands are captured by Marcus D. Dubber, Victims in the War on Crime (New York: NYU Press, 2002), 214. 20  Nils Christie, ‘Conflicts as Property’, British Journal of Criminology, 17 (1977), 1.

Victims of Crime

297

sentencing. So, from this perspective, ‘Sarah’s’ claim to have ‘done nothing wrong’ is best understood as an assertion of her right to take the case forward or not, since it is essentially her conflict. What business is it of anyone else whether she chooses to make the allegations or not, whether she requires her husband to be subject to punishment through the criminal law or not? Maybe a choice not to request a criminal prosecution would be seen by others as a foolish choice, but there would be nothing that should count as any kind of wrongdoing. Perhaps a better way for her to proceed, and one which she might even have preferred had the option been available to her, would have been to sue her husband for compensation for the harms done,21 and to secure a rapid divorce with an order preventing any attempt on his part to contact her. I have argued elsewhere that a more developed version of such a conception of the criminal law makes it more like a civil process.22 In doing so it does not adequately account for the idea of ‘criminal wrongdoing’ and for the process which puts the political community in charge of responding to such wrongs. It does not capture the sense in which, in the criminal law context, the relevant wrongdoing precisely is ‘others’ business’. Of course, it is not meant to do so, being rather a reformist account that proposes change and the elimination of much that currently falls under the concept of crime. Similarly inadequate is what one might think of as the directly opposing view: that the criminal law is one of the central and legitimate forms of coercive state power, since it is the state which is, as it were, the ‘victim’ of crime; it is its laws which are attacked. On such a view crimes, even those involving individual victims, are to be understood as attacks on the state. (I call this ‘the statist view’, and recognize, of course, that I have given a singularly crude version of it here.) On the statist view the individual victims stand as the occasions for crime, and thus any role that they might have in the process will be only such as is necessary to fulfil the state’s interest.23 I do not mean to suggest, by putting the matter in this way, that the state is anything other than a legitimate body, or that it has no right to exercise coercive power, though some states may indeed be illegitimate, in which case they have no right to exercise coercive power.24 It does not follow either from the statist view that victims should be kept out of the criminal process—there might, for instance, still be room for victim impact statements as aspects of the punishment, so that 21  Or indeed to sue him as a matter of ‘civil recourse’ for the wrongs that he did to her—the key point about civil recourse being that it is for the wronged victim to decide whether or not to pursue such redress: see John C. P. Goldberg and Benjamin C. Zipursky, ‘Torts as Wrongs’, Texas Law Review, 88 (2010), 917. 22  See S. E. Marshall and R. A. Duff, ‘Criminalisation and Sharing Wrongs’, Canadian Journal of Jurisprudence, 11 (1998), 7; ‘Public and Private Wrongs’, in James Chalmers et al. (eds.), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh: Edinburgh University Press, 2010), 70. 23  We should note that the statist view might focus more on harms than on wrongs, although I do not think this would make a difference to the substance of my argument at this particular point. 24  The argument about what grounds the legitimacy of state power is, of course, one of the deep questions of political theory but not one I address here. All that is necessary to my argument here is that there is a clear enough distinction between ‘the state’ and what I, along with others, call a ‘political community’.

298

S. E. Marshall

the offender is required to hear the statement, the better to grasp the full nature of their actions. The important difference lies in the reasons for such inclusion: on the statist view, these reasons are instrumental reasons which focus on the way in which the participation of the victim serves the state’s interest. One advantage, however, that the statist view has over the civilizing account above is that it takes wrongdoing to be at the core of the characterization of crime, and gives some content to the idea that the criminal law is concerned with ‘public’ rather than ‘private’ wrongs. It is in the light of the idea that criminal law is concerned with ‘public wrongs’ that not just the victims’ rights but also victims’ duties, and thus the real topic of this chapter, must be understood. My criticism is that the statist view construes ‘public’ in quite the wrong way.

II.  Sharing Wrongs and Sharing Responsibilities To understand the nature of law we have to understand its role as partly constitutive of a political community and therefore as an object for identification, as playing an important role in a people’s sense of who they are.25 The state, on the account on which I rely in what follows, is ‘the institutional manifestation or mechanism of . . . a political community’.26 Citizens are responsible both to the state (insofar as they are responsible to one another in their roles as citizens) and for the state. Seen in this light, the criminal law and its processes are something for which citizens, qua citizens, are responsible not merely as subcontractors of responsibilities to officials, but as participants sharing responsibility for the criminal law with the officials. In this way the different role-based duties of citizens and officials are interconnected and give the criminal law its institutional form. In order to get clearer then about the victim’s role in the criminal law of a political community, let us return to a seemingly simple difference between civil and criminal cases. A civil case is listed as ‘Smith v. Jones’: a plaintiff, Smith, brings a case against Jones, a defendant, complaining that Jones has wrongfully harmed her; if the court finds for Smith, it upholds that complaint, thus validating Smith’s account of what happened. It is up to Smith whether to sue Jones or not, whether to settle without going to trial, and whether to enforce a judgment in her favour. A criminal case, by contrast, will be listed as ‘People v Jones’ or ‘State v Jones’ or ‘Commonwealth v Jones’: by a title, that is, that portrays the complainant not as an individual but as the polity. In a polity that aspires to be democratic, this means that the complainant is not just the individual victim (if there is one), not just a singular ‘I’, but the polity, a collective ‘we’. We, who include the victim, complain that Jones has wronged Smith in a way that properly concerns us as a violation of the values that go to define us as a political community. Upon proof of his guilt in 25  Joseph Raz, Between Authority and Interpretation (Oxford: Oxford University Press, 2009), 106. 26  Introduction to this volume, p. 20.

Victims of Crime

299

a fair trial (one that summons Jones to answer the charge, and gives him a fair hearing), that complaint becomes an authoritative condemnation of Jones’s conduct as a wrong that was in a certain sense public. A trial thus calls a defendant to answer not just to an alleged individual victim, though it does indeed do that, but also to the whole polity for the wrong that he allegedly committed; and it constitutes, in part, an expression, articulation, and application of what are purported to be the shared, ‘public’ values of the polity. It is also an expression of solidarity with the victim and with the alleged offender, for they are both fellow citizens. Such articulation and application involves, of course, the interpretation of those values in relation to the particular series of events that constituted the crime; in this way the individuals involved get to be recognized as individuals. Values are shared in that they are the values which are internal to the structure of the institutions and practices that go to form the political community and the nature of citizenship within that community. Such a characterization of these values as part of the structure of the polity is to a degree abstract: it says nothing at all about what those values might be, or how rich a range of values there might be, or what importance any particular value has in the structure. All of this will vary from one political community to another. There will be some limits on what values could form this structure, but these limits are given only by what can intelligibly count as a political community—they are conceptual limits. Moreover, in this sense of the ‘shared values’ of a political community, it does not follow that there is no scope for disagreement as to how those values are to be interpreted in their particular applications. Nevertheless, without some shared and roughly agreed conceptions there will not be a political community (or any other kind of community for that matter). From this point on, however, I shall suppose that we are concerned with a roughly liberal polity, with shared values of, for instance, liberty and equality. We can see that even here there is plenty of room for people to disagree about what those values amount to and how they are to be weighed against one another: the crucial point is that it is because these are the shared values in the sense I have specified that disagreement is possible at all. In a liberal polity the range of values constituting the public domain will be limited. They will not seek to govern all aspects of citizens’ lives; rather they specify the quite limited, normative terms of our cohabitation as citizens. Moreover, liberal polities may instantiate different interpretations of those values and there may be, from time to time, disagreement as to how best to express those values in the criminal law. Nevertheless, what is claimed for those values understood as the polity’s public values is that they both protect and bind us. They protect us from being victimized by criminal wrongs and bind us in requiring us to refrain from criminal wrongdoing. They protect us not only as actual or potential victims but also, crucially for my arguments later on, as actual or potential offenders. We can all be assured that we will be called to public account through a coercive and condemnatory criminal law, but only for the wrongs defined by the criminal law, as authoritatively interpreted by the courts. The criminal law constituted by these publicly defined wrongs binds us, furthermore, not only as actual or potential offenders but also as witnesses, as jurors, and as victims: those values demand our allegiance and

300

S. E. Marshall

respect not just in requiring that we do not directly violate them by committing crimes, but in requiring us to play our part in their application and enforcement. This is what is involved in seeing the criminal law, as citizens in a democracy should be able to see it, not as a law to which we are subjected by ‘the state’, but as our law—a law by which we bind ourselves and each other. Then, if it is indeed our law, we must be ready to play our part in the activities that make up this political institution: activities that include not just legislation but also the roles indicated above.

III.  Taking Part What is that part? The account of the relationship between citizens, between citizens and the state, and between citizens and the law, sketched above clearly has some affinities with the republican tradition of political thought which emphasizes active citizenship and the idea that citizenship is an office which brings with it public duties, an important one of which, for my purposes, will be the duty to speak out. Wenar argues that ‘[T]‌his picture of the citizen as the bearer of public duties has faded in common sensibility’.27 Still, even if he is right that such a conception of citizenship no longer resonates in ‘common sensibility’ as strongly as it once might have done, living on only, as he notes, in some contemporary republican theory,28 nevertheless I suggest that it still whispers to us through the institution of criminal law. Thus, citizens are summoned to jury service, not merely invited to participate, and are excused only under certain circumstances—that is to say, reasons need to be given, not simply an RSVP declining an invitation. Jurors, once they are engaged, then participate in proceedings which include a shared interpretative responsibility: they have to share in the interpretation of the law in applying it to the particular case. A jury might have to decide, for instance, whether a defendant’s disregard of a ‘substantial and unjustifiable risk’ involved a ‘gross deviation from the standard of conduct that a law-abiding person would observe’, in order to decide whether she acted recklessly;29 or whether the defendant’s appropriation of another’s property was ‘dishonest’, in order to decide whether he is guilty of theft.30 In making these determinations, the jury has to make normative judgments as a collective ‘we’ whose responsibility is to determine the meaning of the polity’s public values as they apply to the instant case. Furthermore, in line with the ‘duty-based desire’ account of roles sketched briefly above, the role of juror is partly structured by the desire to perform the role well, and with this comes the right to be enabled

27  Wenar, ‘The Nature of Claim-Rights’ (n. 14), 221. 28  cf. Phillip Pettit, Republicanism: Theory of Freedom of Government (Oxford: Oxford University Press, 1997); Richard Dagger, Civic Virtues (Oxford: Oxford University Press, 1997). 29  See Model Penal Code §2.02(2)(c). In English law the comparable question would be ‘whether the risk was one which a reasonable and prudent person might have taken’: see David Ormerod, Smith and Hogan’s Criminal Law (13th edn.; Oxford: Oxford University Press, 2011), 118. 30  Theft Act 1968, s. 1(1).

Victims of Crime

301

to perform the task and not be obstructed. Officials, including judges, then have responsibilities to the jury to aid it in its deliberations.31 The jury is then one example of the way in which citizens and officials share responsibility for the criminal process, but there are other roles for citizens—­ witnesses and victims—which are crucial in the response to criminal wrongdoing, and which though less formally structured are nonetheless structured by duty. So, while the actualization of the response to criminal wrongdoing is largely the responsibility of various kinds of official (police, prosecutors, the judiciary, penal officials), citizens also have a responsibility to assist in that response: to report crimes, to assist in their investigation, to give evidence in court. These responsibilities are, I suggest, grounded in a general duty to bear witness, to ‘speak out’. This latter way of putting it is related to a traditional ideal, alluded to above, of an active citizenry whose responsibility is to hold its government and state to account. However, citizens have a duty not just to hold their government to account but, as members of the political community, to hold themselves and one another to account: bearing witness to wrongdoing by reporting crime is one way in which they discharge this duty. Now there is ample room for disagreement both about how extensive and stringent such responsibilities are, and about what force they should have. At this point I take these duties to be civic rather than necessarily legal duties. I will return, briefly, to the question of legal duties later. So, just what responsibilities do citizens have, for instance, to report crimes or to assist in their investigation? One kind of answer might have to do with the prevention of crime: someone who has knowledge of a planned crime has a duty to report it in order to prevent the crime from actually taking place; someone who witnesses a crime in progress has a duty to report it in order that the crime be frustrated; someone who knows of a past crime has a duty to report it in order that the offender be prevented from committing crimes in the future. There are certainly arguments to be had about all three of these putative duties and the way in which they may, or may not, infringe liberty or undermine human dignity or undermine trust in such ways as to outweigh their preventive value.32 My argument, however, grounds the duty to bear witness, or to speak out—and thus the duty to report—not in prevention but in the requirement to address public wrongs. It is not limited, therefore, to cases where crime might be prevented. We must also bear in mind, of course, that such responsibilities could be unqualifiedly asserted only in polities whose criminal laws and procedures were wholly legitimate. To the 31  For a discussion of the duty to do jury service which is close to the spirit of my account here, see Sherman Clark, ‘The Courage of our Convictions’, Michigan Law Review, 97 (1998–9), 2381; Albert W. Dzur, Punishment, Participatory Democracy, and the Jury (Oxford: Oxford University Press, 2012). For an interestingly ambitious account of the interpretative role of the jury, see Robert Burns, A Theory of the Trial (Princeton: Princeton University Press, 1999). 32  In some jurisdictions, e.g. Israel and the United States, there is a legal duty, in some form, to report. In the United Kingdom there is a duty to report in special cases: e.g to report information about terrorist offences (Terrorism Act 2000, s. 38B). An extensive discussion of the duty to report can be found in Miriam Gur-Ayre, ‘A Failure to Prevent Crime—Should it be Criminal?’, Criminal Justice Ethics, 3 (2001), 4.

302

S. E. Marshall

extent that the criminal law is radically imperfect in its content, in its procedures, in the claim it has on the allegiance and obedience of all citizens, those citizens’ responsibilities become more complex and more qualified. What needs to be clear at this point is that the citizen’s duty to report crime, as one form of the general duty to bear witness to criminal wrongdoing, is not a duty to seek out or investigate crimes. Any claim to a duty to seek out crime, or even to a general habit of watchfulness, would be in tension with, and would more than likely undermine, the kind of civic trust on which a genuinely communal life for citizens depends.33 So, ‘neighbourhood watch’ schemes, often characterized as a commendable form of citizens’ engagement in their ‘local community’, would require a different form of justification: crime prevention, for example, and even then what gets justified should be something very limited indeed. Certainly such a justification seems unlikely to sanction a permission, let alone a duty, to investigate. So, to the extent that a liberal polity’s criminal law can properly claim to speak in the voice of its citizens, who share in the values it embodies, they must recognize that they have responsibilities to support and assist it that go beyond the responsibility not to commit what it defines as crimes. In meeting such responsibilities citizens will most certainly be required to exercise judgment: not all breaches of the law, even of the criminal law, are equally serious; just as officials exercise discretion in carrying out their duties, so must citizens in carrying out theirs. It will not make sense to suppose, and it does not follow from my account, that every time I spot another driver committing a driving offence, however trivial, I should report what I see to the police: being overzealous in the performance of duties is as much a (moral) failing as being indifferent to them. This is not to say that I should never report a driving offence, and there may be some offences the reporting of which is never discretionary, even where there is some risk to the witnesses themselves.

IV.  Back to Victims On the account I am offering here, the witness’s and victim’s duties to report, to testify and give evidence in court, are grounded not simply in the need to prevent crime, but in the importance of addressing public wrongs and calling offenders to account. Indeed, in the case of the victims it is particularly inappropriate to rest such duties in prevention. It would seem odd to suggest that the duty of the rape victim to report and testify is simply grounded in the need to prevent further rapes—this would make the victim primarily of instrumental interest. The best way to focus the preventive aspect would be to talk in terms of her responsibility 33  This is not to say that there are never any, particular, circumstances in which ‘watchfulness’ is appropriate. The public notice warnings seen at large railway stations to ‘watch out for’ pickpockets are not unreasonable as guides to tourists—though such warnings do not impose duties on travellers. But, still, there may be some argument to be had about warnings to ‘watch out for’ and ‘report’ sightings of ‘unattended bags’ at e.g. airports.

Victims of Crime

303

to help protect her fellow citizens against this person. The duty to bear witness to wrongdoing, as I am characterizing it, is better seen as something which victims owe both to themselves and their fellow citizens as a matter of civic solidarity34 and civic dignity. We need then to notice that among the ‘fellow citizens’ to whom the duty is owed are the accused persons/offenders themselves—to call a wrongdoer to account is to acknowledge the wrongdoer as ‘one of us’ with equal standing, where that means being equally bound by the law as well as protected by it. How, then, does this reflect on the role that victims have in the criminal law process? To flesh the account out a little further, consider ‘Sarah’, with whom I began my discussion. Clearly, in falsely claiming that the rape did not take place, she at least failed in her civic duty not to impede the officials—police and prosecutors— in the performance of their duties, but more than that she failed in her duty to bear witness to wrongdoing. She had a duty not just not to make false claims that she had not been raped, but also a duty to report the rapes in the first place and not to withdraw the claim. Notice too that, in so far as she owes these duties to her fellow citizens, she also owes the duty to her husband—not because he is her husband but because he is a fellow citizen. In failing to report his wrongdoing she failed to take him seriously as a responsible citizen whose actions need to be addressed both by him and the community of fellow citizens. This is not yet to say that we should give these civic duties the force of law, or criminalize failures to discharge them. A fully developed account of these matters would need to be part of a larger discussion about which of our civic duties should be legal duties, and whether the victim’s duty to report is more stringent than that of other witnesses. Still, since my argument has stressed the role-based, active, and participatory nature of these duties, there seems to be at least a prima facie case for arguing that victim’s duties should be legal duties. In this respect their role will be similar to that of jurors and other witnesses, whose roles involve legal duties with which we are already familiar. It might, at this point, be argued that we could at least start by distinguishing between failures to engage in the process of responding to crime—failures to report, withdrawals of claims that amount to a withdrawal from the process—and engagements that aim to impede or pervert that process. But this distinction is not quite as straightforward as it at first appears, for at least some failures to engage with the process will be ways of impeding it. So, where there is a legal duty to bear witness, e.g. the legal duty to appear as a witness if one is summoned, then failure to turn up to court will impede the trial process and undermine it, although failing to turn up looks like one form of just not engaging with the process. It might also look as if, where a victim is not required as a witness, the duty to bear witness will nonetheless require the victim to make a Victim Impact 34  ‘Solidarity’ is a complex notion which requires more unpacking than I have space for here but an intriguing discussion of it has been started in Alan Norrie, ‘The Man Who Will Turn the World Upside Down: Critical Realism, Freedom and Political Theory’, presented at the International Association for Critical Realism Annual Conference, Rhodes University, South Africa, July 2012.

304

S. E. Marshall

Statement,35 which would turn what is currently a right that the victim may choose not to exercise into an enforceable duty. Now there is much more to be said (but not here) about Victim Impact Statements. There are questions, in particular, about the kind of content they should have, and the extent to which they should reach beyond the immediate impact (the intrinsic character) of the crime itself to include its further, consequential effects on its direct victim or on others. There are questions about the role such statements should play in the criminal process:36 should they, for instance, be a part of the sentencing process, whether as information on which the sentence may draw, or in the form of comments on the appropriate sentence?37 However, all that my argument here implies is that victims have a civic duty to bear witness, within the criminal process, to the wrong that was done to them: the precise form, content, and role of that bearing witness are matters for further discussion. But whether we are talking only of civic duties or also legal duties, what I say might seem unduly or harshly demanding of victims. (This might be the real burden of the protests from ‘Sarah’ and her supporters reported at the beginning of this chapter.) Now, it must certainly be recognized that victims can be, though not inevitably, under considerable pressure not to report or not to testify, and that victims, when they do fail in their duty, may have a reasonable excuse which can provide grounds for a plea in mitigation, or even a defence. I am less inclined to think that victims would have a justification for not reporting a case that ought to be reported (bearing in mind what I said about the exercise of judgment by citizens here), even in the face of strong pressure. Holding firm against such pressures may indeed take courage. Nevertheless, it is not unreasonable to expect citizens to be brave in the face of adversity. Moreover, as I argued earlier, even in the face of strong pressure, victims also have rights of an enabling kind. And it is precisely in the context of pressures not to report that we see the importance of these rights. If victims have the kinds of role duty I am ascribing to them, then they also have the right, first of all, not to be obstructed in the performance of that duty; but further they have the right to positive assistance from their fellow citizens and, most importantly, from officials. Apart from some very general role-derived rights—to be taken seriously when they report crimes, to be protected from intimidation, for example—the precise form which such assistance should take will depend upon the nature of the criminal justice processes: the differences between adversarial and inquisitorial systems, for instance, will impose different obligations on officials as to the treatment of witnesses and victims in the trial process. Thus a full account of these rights will need to show how and in what ways the roles of officials—police, prosecutors, court officials, judges—articulate with the citizens’ roles in the whole

35  Or a ‘Victim Personal Statement’ in England: see Crown Prosecution Service, ‘Victim Personal Statements’, . 36 See A. J. Ashworth, ‘Victim Impact Statements and Sentencing’, Criminal Law Review (1993), 498. 37  As in some American jurisdictions: see E. Erez, ‘Victim Participation in Sentencing: And the Debate Goes on . . .’, International Review of Victimology, 3 (1994), 17.

Victims of Crime

305

criminal justice process. If these victim rights are properly recognized within the system, then the failure to fulfil those duties is one for which even a victim can be held accountable. At this point it might strike one that the stress I have placed on the importance of keeping in mind the fact that the accused/offenders, as well as their victims, are citizens, invites the question of how the duty to bear witness to criminal wrongdoing applies to offenders, with respect to their own crimes. Do offenders have a duty to report their own crimes? Do they have a duty to hand themselves in to the police? Do they have a duty to plead ‘guilty’, if indeed they are guilty and know they are, when required to enter a plea? I cannot do proper justice to the important question of the accused/offender’s responsibilities here, but a simple beginning of my answer would be that only in a properly just system could there be such a duty, and even there we might be wary of making it a legally enforceable duty, since the risks of undue pressure on those who are not guilty will always be too great.38

V.  Finally—Victims beyond the Criminal Law Throughout this discussion of victims of crime and their duties I have stressed particular, related conceptions of public wrongs and the roles embodied in citizenship. It is important to be clear, though, that the associative relationship of citizenship and its role-derived, associative duties comprise a limited aspect of our lives—there is more to social, communal, life than citizenship. I suggest further that much of that social life is also structured by associative relationships and roles, sometimes of a closer kind than that of citizenship—friends, family, colleagues—but also professional and formal. The theory of associative duties had its most recent starting point in Ronald Dworkin’s Law’s Empire and has been further articulated by a number of political theorists since. John Horton, whose argument on political obligation seems to me to be persuasive, glosses the general idea in this way: It is this ordinary idea of obligation, arising from social practices rather than voluntary choices or from our common humanity, which the conception of associative duties or obligations seeks to capture. They are the obligations of family, collegiality and political community. Such obligations are not owed to everyone: they are obligations owed to other members of a particular group or association to which we belong. But unlike special obligations that are created through voluntary choices or decisions, such as those that arise from promises or a decision to join a club, associative obligations cannot be explained in terms of individual voluntary acts or decisions. It is this combination of not being owed to everyone and not arising from voluntary choice that makes associative obligations different and distinctive; and also, it should be noted, what makes them for many philosophers especially controversial.39 38  For an earlier discussion of e.g. the right to silence see Antony Duff, Lindsay Farmer, Sandra Marshall, and Victor Tadros, The Trial on Trial, vol. iii (Oxford: Hart, 2007), 203–13. 39  John Horton, Political Obligation (2nd edn; London: Palgrave, 2010), 148.

306

S. E. Marshall

Most of us will not become victims of crime—not even minor crimes, let alone very serious ones. However, it is hard not to think that nearly all of us will be subject to some, mostly minor, wrongdoing by others, and also be wrongdoers ourselves. The scope then of ‘victims’ duties’ could be wider than just the victims of crime, for there are all sorts of associative relationships in the context of which individuals do wrong: all sorts of moral wrongs which are not the public wrongs that fall within the scope of the criminal law. Do those who are wronged, the victims of such wrongdoing, have a duty to the wrongdoer, analogous to the duties which I have ascribed to the victims of crime? This might look like a rather recherché piece of philosophical speculation until one reflects more carefully on ‘the components of moral experience’.40 Letters to advice columns in newspapers and magazines quite often address just these questions: whether the writer should tell a friend of some wrongdoing by another; whether the writer should confront the friend they think has wronged them, or just walk away. How much do they really owe that friend, for friendship’s sake, to address the wrongs that stand between them? Luckily for the reader, further discussion of these questions is well beyond the scope of this chapter.

40  ‘Ordinary moral opinion continues to see associative duties as central components of moral experience’: Samuel Scheffler, Boundaries and Allegiances: Problems of Justice and Responsibility in Liberal Thought (Oxford: Oxford University Press, 2001), 64.

Index Acts of Parliament (UK) 61 Acts of the Scottish Parliament  57, 61 actus non facit reum nisi mens sit rea 8 actus reus  8, 97, 133, 156 administrative offences  9, 33, 77, 116 and charges/penalties  34, 112, 116, 119, 130 admission-cost regulation  134–5, 140–42 advanced marginality  271, 274–6, 279, 281, 285–6, 288 Alexander, Larry  6, 45, 50 Allen, Sir Carleton 105 alternative vote referendum  64, 70 American Gooseberry Mildew Order 1920: 113 Animal Welfare Act 1981: 111 Animal Welfare Act 2006:  112, 125 anti-social behaviour  4, 39, 41, 75, 115, 254 anti-social behaviour orders (ASBO)  4, 41, 75, 115 anti-terrorism legislation 86 Aquinas, St Thomas 170 Archbold 57 Aristotle  205, 234 ASBO (see Anti-Social Behaviour Orders) Ashworth, Andrew 57 assisted suicide  12, 13 autonomy  45, 47, 49, 124, 126 personal  124, 126, 127, 130 sexual 97 Baker, Dennis 210 behaviour-influencing taxation 120 behaviour modification 114 Bentham, Jeremy 102 Berlin, Isaiah 185 bigamy  209, 216–8 Black, Julia 114 Blackman, Justice 208–11 Blackstone, W.  23, 96, 102, 169, 255–7, 259 bloodfeud 92  Bottoms, Anthony  ix, 22 Bourdieu, Pierre  14, 271, 274, 278, 282, 285 Bourdieusian  vii, 270 bureaucratic field  274, 278, 283–5 cultural capital  283, 286 economic capital  283, 286 physical force  270, 283 (and see Harm Principle in relation to) symbolic capital  283 Bowers v. Hardwick 478 US 186 (1986): 208, 209 Braithwaite, John  6, 133 Brownlee, Kimberly 294

Buckland, Frank 125 Buckland, William 125 burden of proof  9, 43, 64, 123, 124, 147, 226, 231 Burger, W.E. 210 Canada  156, 158, 217, 258 Canadian Charter of Rights and Freedoms  156, 217 Supreme Court of  156 Carr, Cecil 108 Castel, Robert 271 Castells, Manuel 271 Cato’s letters 136–7 Census Act 1920: 113 ‘centaur state’  273, 286 Chalmers, James  7, 10 Christie, Nils 296 Cicero (see also Republican Theory)  236 citizens’ responsibilities (putative duties) to 302 report crime  301–5 assist investigation  310 prevent crime/wrongdoing  18, 34, 37, 297, 301 (see also Harm, prevention of) citizenship  138, 164–7, 170, 179, 206, 274, 280, 299, 300 judicial citizenship  285 civic duties/responsibilities  26, 171, 301–4 civic enterprise  53, 166, 167 Civil Investigation of Fraud (CIF) 32 civil penalties  77, 110, 117, 119 Clegg, Nick 58 Cloward, R.A. 278 coalition  69–71, 76 coalition government  61, 65 Code for Crown Prosecutors (England) 246, 259 Common law  2, 55, 57, 107, 110, 111, 113, Community  85–6, 92, 135, 141–9, 161, 166, 176, 180–1, 198–9, 238, 240 civilized  47, 82–3 [as] local group  251 moral  162, 165 political  15, 20, 21, 22, 25, 42, 153, 293, 297–9, 301 [as] State  255–8, Companies Act 2006: 127 condemnation authoritative  34, 145, 299 concept of  118–9 public  34, 86, 142, 153, 166 moral  100, 193, 209 communal 142

308

Index

Contaminants in Food (England) Regulations 2010: 74 contractarianism  24, 151–4, 161, 164–6, 167, 170, 175, 180 and criminal law theory  24, 151–2, 155, 161, 164–5 control orders  4, 41 conviction  56, 141–2, 144, 146–7, 226, 246, 252–3, 261, 291–2 as condemnation  118–9 for ‘real’ criminal offence  9, 11–3, 28 for regulatory offences  32–3 penalty on  70–2 stigma of  122–4 in strict liability offences  156, 159, 161 corporate criminal liability  121, 123, 130 courts  2, 8, 11, 54, 64, 105, 128, 129, 170, 188, 208, 243, 273, 299 administrators of law  13, 14 burden on  110 Canadian 217 Civil 256 Court of Appeal  79 criminal  55, 115, 117–8 function of  139 and legal services  71 Magistrates 57 Crime and Courts Act 2013: 130 crimes against the State  160, 169–70 bribery of public officials  170 contempt of court  170 counterfeiting  160, 170 subversion of justice  170 tax evasion  32, 160, 170 treason  96, 160, 169, 170 criminalizable  21–3, 29, 49, 50, 53, 153, 155–6, 161, 173 criminalization Bourdieusian perspective on  270–290 criminological approaches to  86 criminological writings on  99, 100 crisis of   48, 56, 152, 181–2 deterrence  80, 83–4, 121, 134, 178, 228–9, 261–2 enforcement  17, 19, 22, 44, 74, 85 formal  54, 55, 56 group ‘trouble-making’  115 individual ‘trouble-making’  115 legal-institutional approach to  244, 267–9 non-institutional approach to  18 normative theory of  1–3, 5–6, 8, 10, 14, 17, 25, 36, 42–4, 51, 81, 94–9, 233, 252 outcome or practice  14, 56, 114, 174 over-criminalization  48, 84, 112, 118, 152, 181–2, 189, 203, 221, 230 by legislative over activity  6–11 of particular groups  86 of polygamous marriage  216, 222, 224–5, 228–9

in republican theory  16, 24, 132–3, 136–7, 139–40, 142, 147–8, 300 rethinking theoretical approaches to  80, 99, 250 sanction-based regulation of acts  133–4 substantive  45, 55, 56, 130 criminal justice process, role of ‘other citizens’ 294 bystanders 294 offenders 294 victim 294 witnesses 294 criminal law bearing witness to  301, 312 as censure  27, 29, 33, 40, 42, 85–7, 118–9, 161, 173, 233, 249 as cultural practice  88 distinctiveness of  98 institution of  89, 94 last resort (ultima ratio)  27, 55 Marxist analysis of  88 Prevention of harm/wrongdoing  32, 37–41, 45–7, 80–1, 167, 177, 192, 204–8 Principles of limiting 83 organizing  83, 98 protection of private interests  84, 86 purposes of  6, 35, 38, 47, 80–2, 92–3 social function or meaning of  88 transnational  15, 16, 274 ultimate guardian of trust  259 Criminal offences creation  55, 58, 59, 69, 70, 73, 76, 77, 109, 120 environmental 112 geographical extent of  64, 66 inchoate  39, 53 international 16 municipal criminal  16 new (between 1997–2006)  7, 39 omission-based  62, 63, 64, 66, 95, 101, 121, 122, 124, 127, 130, 204, 233 paradigm  39, 100, 122, 135 pre-inchoate  124, 126 regulatory  4, 8, 9, 10, 105, 106, 107, 108, 111–3, 117 special capacity of  64, 73, 74, 75, 78 strict liability  8, 57, 105, 106, 121 criminal sanctions  34, 78, 112, 117, 140, 178–9, 185–6, 188 absence of  140, 185–6 Crofting Reform (Scotland) Act 2010: 74 Crown Prosecution Service (CPS) (England and Wales)  10, 246 Dangerous Drugs Act 1920: 113 Dangerous Wild Animals Act 1976: 125 Davis, Adrienne.  224–5, 229 ‘deadly triangle’  277, 289

Index decriminalization  11–3, 55, 60, 224–5 Deferred Prosecution Agreement  32, 130 Department for the Environment and Rural Affairs (DEFRA) 115–6 Department for Trade and Industry 116 deterrence  80, 83–4, 121, 134, 178, 228–9, 261–2 Devlin, Lord Patrick  199, 265 Dicey, A.V. 107 Dignitas 12 Dignity  18, 45, 46, 47, 49, 50, 124, 130, 301, 303 Dimock, Susan  10, 24 Director of Public Prosecutions (DPP)  11–13, 31 Discretion  10, 32, 107, 117, 142, 144–8, 175, 181, 203, 230, 302 domination  22, 136, 138–9, 142, 145–6, 148, 279 non-domination  22–4, 133, 136–42, 144–6, 148, 279 Douglas, Justice 207 Duff, Antony  77, 86, 119, 166, 172, 174, 176, 198–200, 233, 249, 254–6, 258, 260, 265, 266, 268, 292 Dunn, John 238 Durkheim, Émile  88, 271, 278, 286 Duties civic  26, 172, 301, 303, 304 legal  301, 304 public 300 Dworkin, Ronald 305 Law’s Empire 305 economic crimes 168 Employment of Women, Young Persons and Children Act 1920:  113, 114 Engels, F. 278 ethnicity  vii, 148, 270, 272–6, 282, 285, 288, 290 ethnoracial classifications (US)  272 ethno-national classifications (EU)  272 European Convention on Human Rights (ECHR) 34 European Court of Human Rights 33–4 European Directives  67–9, 78 European Legislation see legislation Equine Identification (Wales) Regulations 2009: 74 Excise Goods Regulations 1992: 79 Factories Acts 1833: 102 fair labelling 60 fair notice  65, 77–8, 230 Farmer, Lindsay  169, 250 Fassin, D. and Fassin E. 272 De la question sociale à la question sociale? (2006) 272 fault  101, 165 accompanying wrongdoing  101 characterization of an offence  10

309

elements such as intention and recklessness  122–3, 129 required for criminal liability  35, 116 Feeding Stuffs (Establishments and Intermediaries) Regulations 1999: 62 Feinberg, J.  44, 46–7, 50, 82–3, 159, 193, 198–200 ‘free-floating’ obligations  198 ‘free-floating’ evils  46, 198 non-grievance morality  198 The Moral Limits of the Criminal Law 46 felonies  96, 169 Ferguson, P.R.  57, 69 Ferzan, Kimberly  6, 45, 50 Financial Conduct Authority 118 fine 71 nominative scale  72 standard scale  72 unlimited 72 Finkelstein, Claire 177–8 Firearms Act 1920: 113 Foot, Philippa 266–7 Fortes, Meyer  239, 242 Foucault, M.  245, 278 free citizen ideal (see also Republican Theory) 139 freedom  184, 186, 205, 238, 265, 267 loss of  179 [as] non-domination  135–7, 140–3, 148 restriction of  154 Republican conception of  24 [of] speech  177 Fuller, L. 105 Gaius Institutes 245 Game theory 173 Gardner, John 293–4 Garland, David  88, 278 Gauthier, D.  152–3, 162, 164, 168 Geremek, Bronislaw 273 ghetto  274–8, 280, 284, 285 African-American 276 Anti-ghettos  275–6, 285 Communal ghetto 275 Ghettoization 288 Hyperghetto  275–6, 285, 288 Giddens, Anthony 239 Goffman, E. 285 Goggins, Paul 10 Goldsmith, Oliver 107 Hale, M. 102 Halsbury’s Statutes of England and Wales  57–8, 69 harm  40, 45, 48, 60–3, 88, 99, 106, 114, 122–6, 144–5, 167, 211 based theories  6, 80–5 compensation for  297 conceptual vocabulary of  22–3 conduct 4

310 harm (cont.): connection to, direct/indirect  251–2, 254 criteria for  249 emotional 254 financial 254 future 33–4 harmful wrongs  83, 195–8 harmless wrongs  159 see also Feinberg, J. [to] individual interest  80 innocently caused  160–1 inter-generational 181 ‘non-trivial harm or evil’  38–9, 43, 156 [to] others  78, 190, 193, 198, 258 past 32 polygamy as  219–24, 226–8 prevention of  46–7, 49 80, 81, 84, 85, 154–9 reduction  82, 84, 125, 261 regulatory 128–9 restorative approach to  29, 30–2 risk of  171–2, 174 [as] serious moral wrongs  204–5 social harm  27, 86, 87 specification of  83 Harm Principle  18–9, 22–3, 46–9, 50, 82, 156, 159–60, 189–93, 210, 250, 266 Harrington, J. (see also Republican Theory) 146 Hart, H.L.A. on punishment  252, 265 general justifying aim  252 how much should we punish?  252 who should we punish?  252 Harvey, David  275, 279 Hawkins, K. 55 Hinde, Robert 238 Hirsch, Andrew von  47, 233, 249–54, 261, 265 Hobbes, Thomas  95, 138, 152 Holmes, Oliver, W. 3 Home Secretary 67 honour  14, 273, 278, 282 dishonour  270–1, 274, 287–9 Horder, Jeremy  7, 9, 10 (101-131) Horton, John 305 Huhne, Chris 67 Husak, D.  43, 47, 60, 157–9, 171, 17–5, 198–200 (213-231) hybrid offences / crimes  168–9, 170, 172, 174–7, 179 implicit endangerments  174 explicit endangerments  174 hyperincarceration  271, 285, 286 ‘implied special category’ 74 imprisonable offences  71–2, 77 imprisonment  33, 42, 71–3, 130, 242, 247 Independent Health Care (Wales) Regulations 2011: 70 information-gathering 114 Innes, Martin 240

Index institutional normative order  22, 89, 91, 235–6, 266 ‘institutional fact’  90, 91 intellectual property law 169 Jareborg, Nils  232, 244, 246, 254 Johnston, David 266 Jurisdiction  16, 27, 55, 60, 77, 96, 97, 109, 156, 158, 230, 245 jury, jurors  299, 300 JUSTICE, Breaking the Rules 1980:  57, 64 Kantian autonomy  187, 202, 204, 206 Kelsen, H. 104 Kennedy, Justice  209, 210–1 Kenny, C.S. 103–4 King’s Peace 96 Lacey, Nicola  14, 54, 102, 104, 265 Law Commission  57, 58, 60, 68, 109, 112, 128 ‘law-jobs’  237, 241, 243 Lawrence v. Texas 539 US 558 (2003): 208, 209, 228 Lee, Ambrose  256, 258 legal moralism  6, 18, 45, 83, 85, 155, 191, 199, 215, 248 legislation against moral-wrongdoing  196, 197 anti-terrorism 86 coercive  208, 209, 211 costs of  44 criminal  43, 45, 107, 111, 114, 124, 184, 188–195, 198–200, 222, 248, 265, 268–9 European  65, 67 general  6–11, 14, 15, 17, 29, 61–3, 86, 87, 104, 115, 122, 302 long-term effects of  262 primary  58, 72, 73, 111 secondary  58, 59, 66, 79, 108, 111, 113, 116 subordinate 73 legitimacy  130, 153, 209, 214, 228, 253, 263–4, 279 [of] criminalization  15, 16, 17 [of] offences  168 political 24 [of] state coercion  118 legislative drafting  62, 63, 64, 65 legislative instrument  55, 59, 62, 68 Leverick, Fiona  7, 10 Liberty,  19 184, 185, 187 Llewellyn, Karl  241, 242, 243, 247, 258, 260 Lochner v. New York 198 US 43 (1905) 206, 207, 208 Locke, John  94, 178, 246, 247 Lockean proviso  152, 162 MacCormick, Neil  89–94, 96, 100, 232, 234, 236, 244, 261, 266, 267, 268 Machiavelli, N. 135 see also Republican Theory  136, 139

Index ‘Madisonian Compromise’ 184 Magna Carta 105 Maitland, F.W.  102, 108, 122, 127 mala in se offences against the state  168, 169 general  8, 25, 29, 151, 152, 154, 160, 165 hybrid offences related to regulatory state  168, 169, 171–2, 174–77, 179 justification for  165–67 mala prohibita  3, 8, 9, 48, 151, 111, 143, 151–5, 161, 165, 219, 236 against the state  168, 169 hybrid offences related to regulatory state  168, 169, 171–2, 174–77, 179 inchoate offences  53 involving market behaviour  10, 168 justifying of  154, 164, 167–8, 180, 181 [as] offences against public order  143 Markets and Fairs Act 1847: 102 Marketing of Fresh Horticultural Produce Regulations 2009: 75 Marine and Coastal Access Act 2009: 75 Marshall, Sandra E.  85, 166, 198, 199, 200, 233, 255, 256, 258, 265, 268 Marx, Karl  88, 109, 178 Marxist analysis of criminal law (see also criminal law)  88 Matravers, Matt 163 mens rea  8, 9, 52, 64, 73, 121, 147, 156 mercy  137, 148, 149 Merchant Shipping and Fishing Vessels (Health and Safety at Work) Regulations 1997: 63 Mill, John Stuart  19, 44, 47, 82, 97, 159, 189, 190, 191, 205, 208, 209, 265 (see also Harm Principle) Millian autonomy  187, 190, 192, 194, 201, 204, 206 Milton, J. (see also Republican Theory)  136 Ministry of Agriculture, Fisheries and Food (MAFF)  115, 116 Ministry of Justice  58, 59, 66, 67, 109 Model Penal Code (US)  80, 83, 266 Monetary Administrative Penalties (MAP) 116 Moore, M.S.  6, 45, 82 moral obligations  196, 197 agent-relative  166, 196 consequentialist duty e.g. duty to rescue 196 categorical 196 deontological 196 moral realism 151 moral reasons 195 aretaic 195 deontic 195 More, Sir Thomas 182 Morrill Act 1962 (US) 217 Morris, N. 58 murder  23, 25, 39, 60, 135, 152, 169, 199, 204, 213, 255, 256

311

National Assembly for Wales Referendum (Assembly Act Provisions) (Referendum Question, Date of Referendum Etc.) Order 2010: 70 Ndiaye, Pap 272 La Condition noir (2008)  272 neo-liberalism  274, 276, 279, 282, 289 New Labour  7, 61, 65, 68–72, 74–6 Newsam, Sir Frank  108, 109 North, Sir Peter QC  263, 264 Nuclear Explosions (Prohibitions and Inspections) Act 1998: 67 Official Secrets Act 1920: 113 Office of Fair Trading 128 opportunity-cost regulation  134, 135, 140, 141, 142 Ordnungswidrigkeiten (regulatory offences)  4, 33, 77, 144 Page, Edward 68 Parliamentary Voting System and Constituencies Act 2011:  64, 70 parsimony, principle of  63, 146 participation  26, 296, 298 paternalism  82, 126, 190, 192, 271 liberal paternalism  285 Peckham, Justice  208, 210 penal populism 7 penal segmentation 285 penality maximum  59, 64, 71–3 as an operator of  negative sociodicy 287 penalization  14, 124, 130, 270, 274, 277–82, 286–9 perjury  144, 160, 170, 292 Permanent Under-Secretary of State for the Home Department 108 personal autonomy  124, 126, 127, 130 perverting the course of justice  291, 295 Pettit, P.  6, 24, (132-150) Piven, F.F. 278 police  17, 25, 35, 71, 122, 133, 247, 260, 273, 296, 301–5 cost of  188 duties  240, 243 [as] enforcement officials  3, 10, 11,180, 262, 264 [in the] European Union  279 function of  37, 38 [in] Latin America  281 [as a] public institution  284, 286 police powers  55, 86, 87, 96, 147, 208 political philosophy  5, 15, 17, 24, 222, 260 Polkinghorne, John 237 Polybius (see also Republican Theory)  136 polygamy and children 230 decriminalization  214, 215 external constraints on  227–231

312

Index

polygamy and (cont.): gay marriage  228 general  193, 213, 216, 217 internal restraints  219–226 polygamist families  230 polygyny  217, 228 Poor Law Amendment Act 1834: 102 precariat post-industrial  271, 276 urban  274, 276, 277, 287, 289 preventive justice  4, 37, 38, 39, 41 anti-social behavior orders (ASBO)  4, 41, 42, 75, 115 control orders  4, 41 Price, R.  136, 137 Priestley, J. 136 Prisonfare  275, 276, 285, 286 private vengeance  92, 94, 96 prosecution,  55, 56, 86, 87, 147, 175, 242, 246, 295–7 criminal  4, 10–14, 30, 32, 33 fear of  76 formal  28, 29 of rape  292–3 time bar on  63 prosecutors  3, 11, 17, 55, 180, 243, 259, 260, 296, 301, 303, 304 [as] criminalizers of conduct  27 discretion of  107, 142 judgment of  230 ‘public interest test’ for  245, 246 public interest  11, 12, 13, 80, 83, 85, 86, 246, 259 public wrongs (crimes as)  86, 93, 98 punishment  56, 82–85, 88, 92, 94, 98, 99, 192 capital 145 corporal 145 punitive approach (to wrongdoing by companies) 123 punitive containment  271, 279, 281, 286 ‘quasi-criminal’  9, 77, 258 racialization  271, 274, 279, 281, 282, 283, 287–8 rape (see also sexual property, sexual autonomy)  85, 97, 291 false allegations of  293 Rawls, John  20, 132, 150, 176, 178, 186, 266 Ready Money Football Betting Act 1920: 113  Rechtsgüter 44 reciprocity principle of  233, 240, 266–8 reflective equilibrium  132, 149 Regulatory Enforcement and Sanctions Act 2008:  112, 128 Rehnquist, W. 211 Reid, C.T.  117, 118

Republican theory Cicero 136 free citizen ideal  138, 139 goal-centered approach to penalties/ punishment 147 Livy 136 Polybius 136 theory of government and law  24, 132, 133, 139, 140, 300 subjective 138 Restorative Justice  29, 30, 150, 296–7 retribution  80, 82–85 retributive theory  155, 191, 194 moral wrong principle  192 rights  142, 205, 211, 220–2, 227, 247, 250, 291, 295 accused, of  117 basic/fundamental 176 citizens, of  170 civil  255, 275 definition of  137 [and] duties  154 human 260 individual  84, 184 legal 84 liberty, to  194–8, 206–8 movement or association, of  138 pre-existing 84 property 84 private  84, 98 self-defensive 167 to be protected  83, 90 victims, of  296, 298, 304–5 risk-based nature (of regulatory criminal law) 124 risk-based offending 125 Ristroph, A. 151 Roads Act 1920: 113 Road Safety Act 1967:  233, 261, 262 Robinson, Paul  180, 181 Roe v. Wade 410 US 113 (1973) 208 RSPCA 109 R v. Courtie [1984] AC 463 (CA) 108: 112 R v G [2003] UKHL 50: 55 R v. Chambers [2008] EWCA Crim. 2467: 79 R v. Oakes (1986) 1 S.C.R.  105 217 Ryan, Joan 258 Scalia, Justice  207, 209, 210 Scottish parliament  57, 61, 66, 69 Scottish Statutory Instrument  61, 73 Seeds Act 1920: 113 semi-autonomous sentencing commission 146 sexual autonomy 97 Sexual Offences Act 2003: 10 sexual property 97 Shawcross, Sir Hartley 245–6 Sherman, Lawrence 240 Sidney, Algernon  136, 137 Sigman, M. 224

Index Simester, A.  47, 233, 249, 250, 252, 253, 254, 261, 265 ‘social capital’  251, 283 concept of dominion  251 social order 233 the concept of  233 [or] ‘civil peace’  238 regularity 239 rule 239 social peace  22, 232–3 sovereignty  18, 23, 45, 47, 49, 50, 53, 271 social solidarity  88, 234, 267, 271, 278 special capacity  64, 73–5, 78 specification of harms see Harm standard-setting  114, 125 state coercion  10, 118, 184, 186, 194, 195, 197, 202, 204, 210 legitimacy of  118 Statute law database (UK)  78, 79 Steadman, C. 102 Stephen, Sir James Fitzjames  97, 102, 209 Strassberg, M 221 Straw, J. 67 stop and search powers  86, 87 Stuntz, W. 60 symbolic interactionism 240 concept of ‘control signals’  240 ‘signal crimes’  240 ‘signal disorders’  240 symbolic power (Bourdieu) 274 Tadros, Victor  53, 100, 232 Tasioulas, John 252 terrorism  16, 39, 41, 69, 71, 104, 280 Thorburn, Malcolm  22, 23, 293–4 Tobacco Products Regulations 2001: 79 torts  4, 35, 139, 154, 199, 200, 256 Toulson LJ 79 trust  10, 172, 238, 246, 259, 261, 267, 268, 301, 302 between citizens  234, 235 loss of  161, 162, 165, 167 mutual 92 social 22 ultima ratio see criminal law as last resort Ulväng, N.  94, 232, 244, 246–50, 252–4, 261, 264, 265

313

United Nations Act 1946: 73 Urban Marginality  270, 273, 277, 278, 285, 290 Victim impact statement 303–4 victims of crime 291 wrongdoing or misfortune  293 victims’ rights 296 Volenti non fit injuria 156 Wacquant, Loïc  14, 270 Water Environment (Controlled Activities) (Scotland) Regulations 2010: 73 Weale, A. 163 Weber, Max  272, 276, 287 Welfare of Animals During Transport Order 1992: 111 Welsh Assembly  66, 70 White, R.M. 55 Wiggins, David  266, 267 Williams, Glanville.  98, 103, 104, 111 Wilson, Andrew 241 Winfield, P. Sir 104 witnesses  180, 188, 294,295, 299, 301 bearing witness  302–5 non-participant 203 Work Act orders 59 Workfare  275, 276, 277, 286, 289 Wrong, Dennis 238 wrongs  8, 17, 31, 40, 47, 48, 50, 83–5, 89, 93, 97, 99, 152, 226, 250 criminalizable  21, 23, 49, 160 ‘harmless’ 159 moral  21, 22, 23, 45, 98, 100, 191, 192, 194, 196–205, 210–11, 306 private  215, 216 public  13, 23–8, 30, 36, 53, 77, 86, 165–7, 170, 173, 233, 249, 255–60, 265–6, 268, 291–2, 301–6 ‘serious’  112, 113, 119 ‘shared’ 289 wrongdoing  80, 82, 86, 92, 93, 101 wrongful conduct  82, 83, 93 wrongful harms principle  83, 194 limiting principle  83 Zedner, Lucia 58