Punishment and Public Authority

X Punishment and Public Authority MALCOLM THORBURN * Over the past thirty years or so, criminal law theory has lost sight of the important role of criminal justice in the business of government. Legal moralism, the school of criminal law theory that has dominated the debate over the past three decades, is an attempt to justify the state’s coercive criminal justice powers as mere tools for doing to moral wrongdoers what morality requires. Proceeding in this way allows moralists to avoid dealing with the intricacies of state institutions but it places enormous weight on one’s moral intuitions. For in order for the moralist position to succeed, we must first accept that moral wrongdoers actually deserve to suffer something like the hardship that criminal justice administers simply in virtue of the fact that they have committed moral wrongs. And this is not an intuition that many people share. As H.L.A. Hart famously put the point, it ‘seems to rest on a strange amalgam of ideas. … [It] is uncomfortably close to human sacrifice as an expression of religious worship’.1 T.M. Scanlon concludes that ‘the idea that when a person has done something that is morally wrong it is morally better that he or she should suffer some loss in consequence… [is] morally indefensible’.2 With the perceived failure of legal moralism to provide a plausible moral justification of the institutions of criminal justice, many scholars nowadays are turning back to an old idea: the thought that the moral justification of criminal justice turns on its role in a larger story about the morally important role of government. One way to tell this story is to focus on the role of government in setting standards of conduct and ensuring general compliance with them, all for the sake of the general welfare. Two hundred years * This chapter is a distant relative of the draft presented at the Uppsala workshop. I am grateful to Anthony Bottoms, Matt Dyson, Alon Harel, and Lucia Zedner for insightful comments on that earlier draft. Thanks also to my University of Toronto colleagues Vincent Chiao, David Dyzenhaus, Larissa Katz, Arthur Ripstein, and Hamish Stewart for extremely helpful comments and discussion on a later draft. I owe special thanks to Antje du Bois-Pedain for much helpful criticism, discussion and patience. 1 H L A Hart, Law, Liberty, and Morality (Palo Alto CA, Stanford University Press, 1962) 65. 2 T M Scanlon, What we Owe to Each Other (Cambridge MA, Harvard University Press, 1998) 274. 1 ago, Jeremy Bentham suggested that criminal justice should be understood as part of the state’s legitimate role in securing general compliance with morally valuable legal rules. ‘[A]ll punishment is mischief: all punishment in itself is evil’,3 Bentham wrote, but it could be morally redeemed if we could show that it serves to prevent even greater harm (by inducing compliance with morally valuable legal rules). On this account, crimes are just whatever forms of conduct the law prohibits and punishments are just whatever threats the state deems appropriate to deter such undesirable conduct. So long as the good we do indirectly (by inducing greater compliance with the law) is greater than the harm we do directly (by punishment and other coercive measures) and perhaps so long as those harms and benefits are fairly distributed among persons, it seems that the operations of criminal justice are morally justified. Like the legal moralist account of criminal justice, the instrumentalist account is very attractive in its simplicity. It seems plausible on first inspection that we might be able to redeem the terrible coerciveness of criminal justice institutions by showing them to be essential means to producing the great benefit of general compliance with morally valuable legal rules. But this vision of well-placed incentives seems to fit more easily with what are often called regulatory offences than it is with criminal punishment.4 For when we are required to pay a late fee at the public library or for overstaying at a parking meter or for being in possession of undersize lobsters,5 it might seem that there is nothing more to it than the fact that the state needs some way to induce compliance with its rules and these fines are effective tools for inducing compliance. But what is sometimes referred to as ‘true criminal law’—where crimes call for subjective mens rea,6 and where 3 ‘An Introduction to the Principles of Morals and Legislation’ in J Bentham A Fragment of Government with an Introduction to the Principles of Morals and Legislation edited by W Harrison (Oxford, Basil Blackwell, 1960) 281. 4 On the nature of regulatory offences (also called public welfare offences), see F B Sayre, ‘Public Welfare Offenses’ (1933) 33 Columbia Law Review 55; R M Perkins, ‘The Civil Offense’ (1952) 100 University of Pennsylvania Law Review 832. 5 The best-known Canadian case concerning regulatory offences is R v Pierce Fisheries [1971] SCR 5 in which the accused was found to be in possession of 26 undersized lobsters, contrary to the Lobster Fishery Regulations made pursuant to the Fisheries Act RSC 1952 C 119. 6 R v Sault Sainte Marie [1978] 2 SCR 1299 per Dickson J at 1309–1310: ‘Where the offence is criminal, the Crown must establish a mental element, namely that the accused who committed the prohibited act did so intentionally or recklessly, with knowledge of the facts constituting the offence or with wilful blindness toward them. Mere negligence is excluded from the concept of the mental element required for conviction’. 2 conviction involves formal state condemnation and a criminal record as well as the imposition of a penalty—seems to involve a good deal more than just this. In this chapter, I argue that the moral justification of ‘true criminal law’ is indeed wrapped up with the business of government, but it is concerned with something even more basic to the project of government than the enforcement of specific legal rules. The central focus of ‘true criminal law’ is the very authority of the state itself as the sole lawmaker for the jurisdiction. When the state makes laws for its people, it does not just set out some rules and then try to do what it can to ensure general compliance with them; it does so as the public authority—the entity that has the sole authority to determine the basic rules according to which people shall organize their shared lives together in the jurisdiction. So when the state makes a law, say, prohibiting assault, it purports to decide authoritatively that people shall not assault one another within the jurisdiction. When one of the state’s subjects intentionally assaults another in violation of the legal prohibition, he thereby arrogates to himself the mantle of law-maker, imposing his own favourite terms of interaction on the rest of us in place of the rules set down by the state. The offender does not merely fail to conform to the legal rule, he usurps the state’s role in setting the terms under which he may interact with others, thereby challenging the state’s claim to be the sole authority on the matter. In this way, the offender challenges the most basic promise of the rule of law: that we will never be subject to the arbitrary will of others, but only to the general laws that are the product of the legitimate public authority. If the offender’s claim were allowed to stand, it would show the state’s claim of sole authority on the matter to be a sham. It is for this reason that any state that claims the authority to make the law over its jurisdiction must claim those powers necessary to ensure that its authority is supreme within the jurisdiction. It is usually sufficient to induce compliance ex ante with deterrent threats and the use of pre-emptive force by the police. But where someone disregards deterrent threats and evades the pre-emptive force of the police, the state must have some other means of reasserting its sole law-making authority over the jurisdiction. The Jerome Hall is the best-known scholarly advocate for the centrality of subjective mens rea to ‘true criminal law’. See J Hall, General Principles of Criminal Law (Indianapolis, Bobbs-Merill, 1960) 146ff. 3 institution of criminal punishment is that means. For when the state threatens criminal offenders with a deterrent sanction, it claims not only the power to impose costs on those who would violate its laws. It also claims the last word over the normative position of the offender. The offender tried to take on decision-making authority over how he shall relate to others (by acting on his own view of the matter rather than according to the state’s law). If he were to evade punishment, then his decision would in fact be the last word on his normative relations with others. But where he is subject to criminal punishment, the state reasserts its authority over him (and everyone else in the jurisdiction). His act can then be seen as just another factual precondition to the operation of the state’s laws: those who violate the laws shall be subject to the punishments prescribed by law. The institution of punishment allows the state’s authority to expand to cover even those cases where it appears that individuals have been able to evade it. Seen in this way, the institution of criminal punishment, is the ultimate vindication of the rule of law. This chapter proceeds in four parts. In part one, I review some of the better known versions of legal moralism and of the instrumentalist account of punishment as a deterrent sanction. I do not purport to provide anything like a conclusive rebuttal of their positions; my purpose is simply to contrast their positions on some of the basic issues of criminal law theory with the view I espouse in this chapter. In part two, I provide an account of what I call here the state’s ‘robust public authority’—the claim that it and it alone is in charge of making the law in the jurisdiction. I consider the place of criminal justice, and especially criminal punishment, in an account of the state’s robust public authority. In part three, I explore a number of arguments one might propose for why the state’s robust public authority should be so important to our moral life. In part four, I present some conclusions, qualifications and clarifications of the view set out in this chapter. I. Criminal Justice without Public Authority Punishment is morally problematic in a way that many other activities are not. It is not just that punishment involves the imposition of hardship on the punished person. Many 4 worthwhile activities impose hardship on someone. We generally take this (within limits) to be morally acceptable. What makes punishment so particularly difficult to justify is that the imposition of hardship in this context is not merely a side-effect of our pursuit of some good; it is in some sense the whole point of the exercise. If we could pursue most of our ends without imposing any costs on others, we would take this to be an unmitigated good. But if we were to try to impose punishment in a way that imposed no hardship on anyone, this would simply constitute a failure to punish.7 In recent years, criminal law theorists have pursued two quite different strategies to justify the practice of criminal punishment. The first strategy – legal moralism – aims to show that it is a basic fact about the structure of interpersonal morality that the punishment of wrongdoers is required, or at least permissible. The second strategy – instrumentalism – aims to show that the punishment of wrongdoers is acceptable as an essential part of a larger instrument for the pursuit of our collective welfare. Although these two justificatory strategies differ in many of their most fundamental aspects, they share one crucial feature: they both try to make sense of criminal justice without reference to the idea of the state’s public authority. The justification of punishment, they assume, can be understood simply as a function of the morality of inter-personal relationships. A. Legal Moralism In recent years, legal moralism of one sort or another has come to dominate criminal law theory in the English-speaking world. The dominant strain of moralism (in the United States, at least) is what we may call ‘moral retributivism’.8 Moral retributivists insist that it is simply a moral fact that wrongdoers deserve to suffer in virtue of their wrongdoing. Although they might disagree about whether the suffering of wrongdoers is a requirement 7 This distinction between direct and oblique intention has its roots in the Catholic doctrine of double effect. In the years since Philippa Foot’s celebrated article on the topic first appeared in 1967, this distinction between what we set out to do and what we merely foresee as a side-effect of our actions has taken centre-stage in moral theory. See ‘The Problem of Abortion and the Doctrine of Double Effect’, reprinted in P Foot, Virtues and Vices (Berkeley CA, University of California Press, 1978). 8 Alan Brudner uses this term to distinguish this school of thought from what he calls ‘legal retributivists’. (A Brudner, Punishment and Freedom (Oxford, Oxford University Press, 2009) 19ff.) The latter category includes Brudner’s own view as well as the position I espouse in this chapter. 5 of justice9 or whether it is even a good to be pursued,10 they agree that interpersonal morality endorses the suffering of moral wrongdoers simply in virtue of their wrongdoing. Now if this is true, then there is no great puzzle at all about the moral justification of criminal punishment: the state is simply doing what interpersonal morality endorses or even requires. The trouble here is to explain why it is that interpersonal morality could ever demand or even permit that individuals should suffer hard treatment simply because of their past wrongdoing. Michael Moore has tried to defend retributivism as a basic feature of interpersonal morality by means of a thought experiment. He invites us to imagine that someone (oneself or another person) has committed an unspeakably awful crime and then asks: ‘Should you or the other offender be punished, even though no other social good will thereby be achieved?’ He replies: ‘The retributivist’s “yes” runs deep for most people’.11 Although we might hesitate to punish others where no good will come of it, Moore suggests, most of us feel that offenders still ought to suffer punishment for their moral wrongs. Our responses to these sorts of thought-experiments, Moore asserts, are reliable guides to the structure of inter-personal morality. But, as Hart, Scanlon and many others have pointed out, these intuitions are not nearly so widely shared as Moore would like us to believe. Antony Duff’s communicative account of criminal punishment takes a different approach to the legal moralist intuition: he tries to turn it into a justified system for the communication of moral censure.12 Duff writes: ‘[W]hatever puzzles there might be about the general idea that crimes “deserve” punishment… there is surely nothing 9 Michael Moore endorses this view in M Moore, ‘The Moral Worth of Retribution’, in F D Schoeman (ed), Responsibility, Character, and the Emotions (Cambridge, Cambridge University Press, 1987) 179 at 182: ‘Retributivism… is truly a theory of justice such that, if it is true, we have an obligation to set up institutions so that retribution is achieved’. 10 Michael Moore argues in M Moore, ‘Justifying Retributivism’ (1993) 27 Israel Law Review 15 at 19 that, ‘what is distinctively retributivist is the view that the guilty receiving their just deserts is an intrinsic good’. 11 M Moore, Placing Blame (Oxford, Oxford University Press, 2010) 163. 12 Duff is building on Joel Feinberg’s claim that punishment is essentially expressive in ‘The Expressive Theory of Punishment’ in J Feinberg, Doing and Deserving: Essays in the Theory of Responsibility (Princeton, Princeton University Press, 1970). 6 puzzling about the idea that wrongdoing deserves censure’.13 So, Duff argues, there is nothing morally troubling about criminal punishment if we think of it as just a highly formalized and especially emphatic mechanism through which the state communicates the political community’s moral censure of offenders for their wrongdoing. The trouble is that Duff’s argument cannot explain why the state should be entitled to communicate its message of moral censure through the means of hard treatment.14 For there is something quite distinctive about criminal punishment that is not captured in Duff’s account, in at least two ways. First, we might all be entitled to criticise others for their moral failings as we see fit, but it would be a crime for any of us to take it upon ourselves to fine, to imprison or to impose any of the other standard forms of criminal punishment on another person. Second, it is acceptable for someone to receive moral criticism from any number of sources for a single wrong; but it would not be acceptable for him to be punished multiple times for the same wrong. The puzzle is not how one could have the standing to communicate moral criticism; it is how anyone could have the standing to punish—to deprive someone of his basic legal rights on the grounds that he deserves it in virtue of his criminal conduct. The problems with legal moralism do not end there. For not only does the legal moralist account seem unable to justify the punishment of moral wrongdoers, it also seems unable to provide a plausible account of what constitutes a criminal wrong. According to legal moralists, criminal offenders deserve moral censure and perhaps punishment because their conduct constitutes a freestanding moral wrong. Of course, they insist, what constitutes a moral wrong depends on the factual context. Most importantly, Duff points out, it is morally wrong to disregard the safety and convenience of others. This, he argues, renders it morally wrong to drive on the left-hand side of the road where all others drive on the right-hand side (because there are regulations in place requiring that one drive on the right-hand side).15 The question of whether something 13 R A Duff, ‘Punishment, Communication and Community’ in M Matravers, Punishment and Political Theory (Oxford, Hart Publishing, 1998) 48 at 50. 14 As Scanlon famously put the point: ‘Insofar as expression is our aim, we could just as well “say it with flowers” or, perhaps more appropriately, with weeds.’: T M Scanlon, ‘The Significance of Choice’ in S McMurrin (ed), The Tanner Lectures, Vol 7 (Salt Lake City, University of Utah Press, 1986) 149 at 214. 15 R A Duff, Answering for Crime (Oxford, Hart Publishing, 2007) 92. 7 constitutes a crime thus depends on our moral evaluation of the conduct in light of all the facts (among which are the relevant legal regulations in place within the jurisdiction). This argument allows the moralists to expand the scope of moral wrongs considerably, but it still fails to track anything like the actual scope of criminal wrongs as found in the positive law of most jurisdictions. Section 126 of the Criminal Code of Canada makes clear this basic structure of ‘true crimes’ in the following words: ‘Every one who, without lawful excuse, contravenes an Act of Parliament by wilfully doing anything that it forbids or by wilfully omitting to do anything that it requires to be done is… guilty of an indictable offence’.16 That is, we suffer criminal condemnation and punishment for contravening the demands of the positive law with mens rea, whether or not the law in question identifies a form of conduct that is morally wrongful. Legal moralists, who insist on turning the criminal law into a proxy for the enforcement of interpersonal morality, are simply unable to account for this basic feature of the law. B. Instrumentalism and Distributive Justice For those who find legal moralism to be a non-starter, there is a second strategy available for consideration. HLA Hart’s account is the standard-bearer for those who have tried to turn criminal punishment into a generic instrument of state policy.17 Hart takes it as given that, as Jeremy Bentham puts it, ‘all punishment is mischief: all punishment in itself is evil’.18 So if it is to be justified at all, this cannot be as something valuable in itself, but only as a means to some end beyond itself. Hart suggests that we should think of punishment in the same way as the harmful side-effects of our collective pursuit of valuable policy goals. For example, if we are able to dispose of waste in a way that serves the public good, individuals should be willing to accept some costs (in smells, noise, even some risks to their personal safety) that are necessary parts of the system that is required 16 Criminal Code of Canada s 126(1). Section 127 of the Criminal Code is also relevant to this point. This makes it a crime to ‘disobey a lawful order made by a court of justice or by a person or body of persons authorized by any Act to make or give the order’. 17 I discuss Hart’s account in greater detail in M Thorburn, ‘The Radical Orthodoxy of Hart’s Punishment and Responsibility’ in M D Dubber (ed), Foundational Texts in Modern Criminal Law (Oxford, Oxford University Press, 2014). 18 Bentham, ‘Principles of Morals and Legislation’, above n 3 281. 8 to secure that public good.19 Indeed, one could even build into the system distributional constraints to ensure that no particular individual would have to bear more than his fair share of the burden of our collective pursuit of a valuable end. The same, Hart argues, should be said of criminal punishment. The system of punishment, like the system of waste disposal, is a policy instrument designed to secure some collective good (in this case, the reduction of anti-social behaviour). Since the imposition of deterrent sanctions is a necessary part of the system, we should think of the costs it imposes on offenders as just the unavoidable side-effects of our collective pursuit of a valuable aim. So long as the benefits of the system outweigh its costs, Hart argues, and so long as the costs are distributed fairly,20 we should think of the system of hard treatment as a morally justified ‘form of social hygiene’.21 A number of recent accounts have sought to render the basic Hartian insight more palatable to contemporary tastes by replacing his choice principle of distribution with some other principle (such as one or other form of egalitarianism: luck, opportunity or outcome; Pettit-style republicanism; etc.) or by expanding the scope of the theory’s concern to include not only the allocation of punishment but also the allocation of security and other goods.22 Notwithstanding these differences, however, the basic structure of such accounts is the same as Hart’s: we are concerned with the minimization of harm and with the fair distribution of the benefits and burdens within a community. Old-fashioned talk of punishment as deserved in virtue of the offender’s wrongdoing has no place in such instrumentalist accounts. The trouble is that these instrumentalist arguments succeed in justifying something that looks more like regulatory offences rather than ‘true criminal law’. For on the instrumentalist account, there is nothing more to criminal punishment than the 19 I take this example from T M Scanlon in his discussion of the logic of what he calls ‘substantive responsibility’ in Scanlon, What we Owe, above n 2 at 256ff. 20 Like most contemporary instrumentalists, Hart insists that there are side-constraints on the pursuit of our aims. He chose a principle of fair choice, but one might choose any number of other principles depending on what one takes to be the relevant conception of distributive justice in this context. 21 H L A Hart, Punishment and Responsibility: Essays in the Philosophy of Law, 2nd edn, edited by J Gardner (Oxford, Oxford University Press, 2008) 6. 22 The instrumentalist style includes a great many self-described ‘political’ theories of criminal punishment including the explicitly policy-oriented approach set out in V Chiao, ‘What is the Criminal Law For?’ (2016) 35 Law and Philosophy 137. 9 imposition of an effective deterrent sanction: for the instrumentalist, the positive rationale for punishing offenders (what Hart calls the ‘General Justifying Aim’ of the practice) is that this will deter undesirable conduct. But something much weightier seems to be at work in the punishment of ‘true crimes’. Common law courts presume a subjective mens rea element for all ‘true crimes’—something that (as Hart himself points out)23 is beside the point when it comes to the fair distribution of deterrent sanctions. And when they impose criminal punishment (rather than mere regulatory penalties), they claim that this is deserved punishment for the offender’s wrongdoing. On the instrumentalist account, criminal offenders may be seen as ordinary citizens shouldering their part of the collective burden in society’s pursuit of some valuable collective end (as we might think of ourselves as we go down to the library to pay our overdue fines). But when it comes to the punishment of ‘true crimes’, it seems that, as John Gardner put the point, ‘punishment must, by its nature, be imposed by reason of actual or supposed wrongdoing’.24 As I shall argue below, the nature of the wrong involved in ‘true crimes’ is of a very particular kind—a wrong against the state’s claim of legitimate law-making authority—but it is a wrong all the same. Any account of ‘true crimes’ that does away with this aspect of criminal law fails to account for one of its most basic features. Recognising the weaknesses of both Hart’s instrumentalist strategy and Duff’s communicative account, Andrew Ashworth and Andrew von Hirsch have suggested that we should combine the two models into a sort of hybrid with the benefits of both.25 That is, they side with Duff that the central role of criminal punishment must be to communicate moral censure for wrongdoing, but they also recognise that punishment necessarily involves hard treatment. So, they argue, criminal punishment should be thought of as a hybrid of two distinct elements: the communication of censure for moral wrongdoing coupled with hard treatment imposed as a deterrent sanction. In this way, von Hirsch and Ashworth suggest, we can provide a neat justification for both aspects of 23 Hart, Punishment and Responsibility, above n 21 at 113ff. J Gardner, ‘Introduction’ in Hart, Punishment and Responsibility, above n 21 at xxv-xxvi (emphasis added). 25 A Ashworth and A von Hirsch, Proportionate Sentencing: Exploring the Principles (Oxford, Oxford University Press, 2005) at 21ff. T M Scanlon makes a similar sort of suggestion in Scanlon, What we Owe, above n 2 at 267. 24 10 criminal punishment. Attractive though this hybrid might at first appear, however, it still does not provide the right sort of justification for criminal punishment. For the point, of course, is not simply that the hard treatment of criminal punishment can accompany the communication of moral censure, but that it is itself imposed by reason of wrongdoing. We are still no further ahead. C. The Inconveniences of the State of Nature There is something missing from both the legal moralist and the instrumentalist strategies for the moral justification of criminal punishment. And yet each approach seems to have something important to recommend it. The moralist strategy, although it fails to provide a compelling moral justification for punishment, works with a plausible understanding of what punishment is: the very nature of punishment requires that it be imposed as a matter of justice in response to the offender’s wrongdoing. The instrumentalist strategy, although it fails to justify the imposition of hard treatment as punishment for wrongdoing, makes clear that deterrence must play a central role in the justification of criminal punishment, for it is only by connecting punishment to this positive objective that we are able to show that it is capable of performing a morally valuable task. The trouble is that neither account has the tools necessary to draw together these two aspects of criminal punishment into a single, plausible account. In the next two sections of this chapter, I argue that the element that is missing from both the moralist and the instrumentalist strategies—the element that can connect the role of punishment as a deterrent to its place as the deserved response to wrongdoing within a legal order—is a robust account of the authority of the state over the punished wrongdoer. II. Public Authority and the Form of Punishment The criminal law theory debate of the past century or so has made one thing abundantly clear: despite enormous ingenuity and creativity, no one has come up with a compelling account of how we can square our basic assumptions about the structure of morality with the right of one person to punish another for his wrongdoing. It is not difficult to show 11 (with Duff) that we have good reason to communicate moral censure of others for their moral wrongdoing. And we might even be able to show (with Hart) that we can set up incentives for others to deter them from violating legal rules. But it is quite another thing to be able to show that we are entitled to impose hard treatment on criminal offenders as a deserved punishment for their wrongdoing. This, it still seems, is ‘morally indefensible’.26 In this section of the chapter, I argue that the many proposals that have littered criminal law theory over the past century have failed because they have tried to make sense of it as a practice that exists between individuals taken simply as natural persons, none answerable to any other. And this is something that simply cannot be done. For, as Thomas Aquinas pointed out many centuries ago, ‘this coercive power is vested in the whole people or in some public personage, to whom it belongs to inflict penalties’.27 That is, we cannot make sense of the institution of criminal punishment simply as a feature of the relations between individuals as moral equals; it is an institution that has its place only within a relationship of robust public authority. Now, if that is true, then the moral justification of punishment will have to proceed rather more indirectly than was contemplated by either of the strategies we have considered so far. That is, to justify the institution of criminal punishment, we must first make clear how it fits into the relationship of robust public authority; once this step is in place, we can then consider how such a relationship of robust public authority could be justified given our conception of ourselves as free and equal moral persons. In this part of the chapter, I sketch out a conception of robust public authority, indicating how criminal law and punishment are essential parts of it. I leave until the next section the question of how we might justify the existence of a public authority of the sort sketched out here. In one way, then, we might think of this section as the part where we engage in punishment theory properly speaking. In the next section of this chapter (where I consider various ways in which one might justify this conception of public authority) we depart the field of punishment theory properly speaking and enter into political theory 26 Scanlon, What we Owe, above n 2 at 274. St Thomas Aquinas Summa Theologica I-II Q. 90 a. 3 ad 2 in Introduction to St. Thomas Aquinas edited by A Pegis (New York, Modern Library, 1948) 614. 27 12 more broadly. By distinguishing these two moments of the argument, I mean not only to show how political philosophy and criminal law theory are distinct from one another, but also to show how they are importantly related, as well. A. Parental Authority Criminal punishment is in many ways a unique institution: unique in its harshness, unique in the demandingness of its procedural requirements,28 and unique in the role it plays in defining a society’s values.29 Nevertheless, it may be helpful as a way of understanding the institution of criminal punishment and its place within a larger system of public authority to consider a somewhat less procedurally complex and less harshly punitive institution: parenthood.30 The relationship between parents and their minor children is apposite because it is one of the few relationships of robust authority other than the one between state and subject—exhibiting the sort of deep moral inequality between the parties required for the justified application of punishment. Although it is generally agreed that children are bearers of the same basic rights (to life, liberty, security and the rest) as all other human beings,31 they are nevertheless incapable of exercising meaningful choices on their own behalf in a wide array of areas. The moral relationship between parents and children is one that is designed to square the status of children as beings of the same moral worth as adults with the fact of their inability to make meaningful decisions for themselves over certain matters. Parents have a robust form of authority over their minor children. By this I mean that it is up to them—and not up to their children—to decide how their children may behave in a great many respects (which of course diminish in number as a child’s 28 Although many of these procedural protections are now on the wane. See: A Ashworth, ‘Is the Criminal Law a Lost Cause?’ (2000) 116 Law Quarterly Review 225. 29 Winston Churchill (Hansard 20 July 1910 col 1354): ‘The mood and temper of the public in regard to the treatment of crime and criminals is one of the most unfailing tests of civilisation of any country’. 30 I have drawn this comparison before. See: M Thorburn, ‘Justifications, Powers, and Authority’ (2008) 117 Yale Law Journal 1070. 31 But see M Moore, Law and Psychiatry: Rethinking the Relationship (Cambridge, Cambridge University Press, 1984) 91: ‘[B]uilt into the concept of a right is an image of its holder as one who chooses. The idea of a choice, in turn, implicates rights holders as beings who act for reasons in exercising the choice of whether to enforce their rights. … [R]ights are meaningfully ascribed only to those who reason practically and are thus capable of choice’. 13 capacity grows). This is a very familiar claim but it is also a very unusual one for it appears to run counter to our understanding of all persons as free and equal. How could one person be in charge of the most basic decisions relating to another person’s life: where he shall live, what he shall eat, how he shall be educated, and so on? Although parents are not free to decide these matters in any way they would like, of course (for they must always make such decisions ‘in the best interests of the child’),32 it remains the case that it is up to the parents (and not up to the children themselves) to decide all these matters. We shall return to the question of whether such a relationship of robust authority between two human beings is ever morally justifiable in the next part of this chapter. For the time being, we can simply note that we quite easily recognise this sort of relationship in the parental context. We can now move on to explore its implications for the justification of punishment. The story so far about the rights of parents to make decisions for their minor children is quite familiar. What is slightly less obvious is what is required for parents actually to maintain their decision-making authority over their children. For children might be the objects of parents’ decision-making authority, but they are also human beings with wills of their own, and this adds a very important complication into the picture. For, as any parent knows all too well, it is a very real possibility that a child will not act as his parent has decided that he should. What, then, is required for parents to maintain decision-making authority over their (sometimes recalcitrant) children? There are three related mechanisms that are required. The first is the one we have already considered: parents may engage with their children as agents who are capable at least on some level of recognising and responding to reasons. In doing so, they set down directives for their children to follow: ‘don’t eat those sweets’, ‘don’t cross the street on a red light’, ‘go to bed’, etc. In at least some cases, the children might simply comply with these authoritative directives from their parents. And when they do, the parents will have thereby actualised their decision-making authority over their children. 32 This has long been part of our understanding of the parent’s relationship to his children. Young v Young [1993] 4 SCR 3 at 38 per L’Heureux-Dubé J: ‘It has long been recognised that the custodial parent has a duty to ensure, protect and promote the best interests of the child’. 14 But children will not always obey their parents: they might choose to disobey or to ignore their parents, they might not hear or (particularly with very young children) they might simply not understand what the directive requires. In order for parents to be able to actualise their decision-making authority over their children in these more complex situations, they must act toward their children in another way, by using preemptive coercive force over them. In order for parents actually to have the power to decide how their children shall live, it must be possible for them to take sweets out of their child’s hand, to carry their recalcitrant child to bed, to restrain their child from crossing the road on a red light, etc. When asked why it is that the parent uses preemptive coercive force against her child, one might be tempted to say that the use of force is justified merely as a means to induce compliance with the parent’s directive. But this would not be quite right. Of course, the point of pre-emptive force is to bring about a certain result, but the relationship between the use of coercion and the directive it is meant to enforce is not just one of a means to a distinct end. This is so for the same reason that we also deny that the parent’s communicating of the rule to her child is merely a means to inducing compliance with her decisions.33 It would be more accurate to say that both the communication of rules and their pre-emptive enforcement are partly constitutive of the parent’s robust authority over her child: if she could not act in these ways, we could not properly say that she had such authority at all. In order for parents actually to maintain a position of decision-making authority over their children, however, they must be capable of acting toward them in yet a third way. For there will be occasions when children will not only disobey their parents, they will also evade their parents’ pre-emptive coercion. So, if parents had no other mechanisms through which to vindicate their authority than the issuing of directives and the use of pre-emptive force to ensure compliance, their authority over their children would quite regularly fail. When parents threaten to punish their children for disobedience of their directives, this opens up a way for them to vindicate their authority over their children, even when those children have in fact disobeyed them. This third way 33 Joseph Raz suggests that the point of the rule of law virtues of clarity, prospectivity, etc is just this—as a means to ensuring greater compliance with the law’s directives. See: ‘The Rule of Law and its Virtue’ in J Raz, The Authority of Law: Essays on Law and Morality (Oxford, Oxford University Press, 1979). 15 of relating to one’s children is even more complex in its structure than the first two, however. Indeed, it seems rather mysterious how threatening and carrying out punishment for disobedience could be aspects of the parental decision-making power at all. Surely, one might object, if a parent decides that her child shall do one thing and in fact he does another, this shows that she does not, in fact, have decision-making authority over the child. Tempting though that conclusion might be, it is mistaken. For this would be to confuse the question of whether parents can actually ensure compliance with each and every directive with the question of whether they remain in the position of robust authority over their children at all times. It is important to recall what we mean by the claim of robust authority: it is the claim that parents (and not their children) are in charge of their children’s lives (with respect to certain matters). Now, if we believe that it is morally significant that parents and children should stand in this sort of ‘in charge of’ relationship of robust authority, then we will think that there is good reason to try to vindicate it—to ensure that this form of relationship always governs relations between the parties—even in the face of conduct that is inconsistent with its terms. It is for this reason that we will look to remedies whose aim it is to restore the parent’s robust authority in the face of conduct that attempts to challenge it. Remedies aimed primarily at restoring a certain form of relationship between the parties are most clearly in evidence in private law.34 If I own a new car and you steal it from me, then you must give it back to me. This is necessary to vindicate my sole authority to decide whether or not you may have it in the face of your conduct that attempted to undermine that authority. If you scratch my new car, you must pay me the amount of money necessary to restore it to the status quo ante. This is necessary to vindicate my sole authority to decide what happens to my car; anything you do to it35 must either be undone or endorsed by me so that only my decisions affect what happens to my car. And so on. 34 At least, on one reading of private law. For the most thorough recent treatment of this way of thinking, see A Ripstein, Private Wrongs (Cambridge MA, Harvard University Press, 2016). 35 In this case, what I do to something is not the same as the consequences of my actions on a thing. Where I use my own body or property legitimately, this might have certain (negative) consequences for others, but they are not things that I do to others in the technical sense I am invoking here. A full account of this distinction would involve a discussion of the terms of equal freedom to use one’s body and property as one would like—a topic which is well beyond the scope of the present chapter. 16 If we mean to vindicate the parent’s claim of robust authority over her child even in the face of conduct that is inconsistent with that claim, things are somewhat more complex, for the nature of the challenge is somewhat deeper. If we are only concerned with possession or a loss that is inconsistent with claims of right (as in many private law cases), then the remedy simply requires us to adjust factual possession or loss to accord with the allocation of rights. But when we are concerned with a direct attack on a claim of robust authority itself (as in the case of parental authority), it is not quite as obvious how we might bring the situation back into alignment with rightful relations. The solution lies in an extension of the parent’s authority into the future—into a situation that the child’s disobedience cannot disturb. When the parent threatens punishment for disobedience, she extends her claim of authority in just this way. Without the threat of punishment, the parent has nothing more to say in response to the child’s disobedience. It appears that it is the child, rather than the parent, who determines how things shall stand between them. Once we include the threat of punishment, however, the parent has a further outlet for her claim of robust authority over her child. For now the child’s disobedience is set in a very different light. Once the parent renders judgment and administers punishment, the child’s wrongdoing can be seen as merely a factual precondition to the effectuation of the parent’s ultimate authority. Although the child’s disobedience remains regrettable—indeed, we should properly see it as a genuine moral wrong against the parent’s authority—it is no longer a genuine threat to the parent’s claim of robust authority over the child. B. Interlude: Two Kinds of Authority It is important to distinguish the robust sort of authority parents hold over their children from the sort of authority set out by Joseph Raz and that has become the common currency of much contemporary legal philosophy. For Raz seems to deny the core of the idea of robust authority I am concerned with here: that sometimes it is not up to each of us to decide how we should act. On Raz’s account, it must always be up to each person to decide how he should act—to suggest otherwise would seem to be almost an absurdity. So the account of authority he proposes must turn on some other idea. That other idea is a sort of instrumental rationality. According to Raz, ‘the normal way to establish that a 17 person has authority over another person involves showing that the alleged subject is likely better to comply with reasons which apply to him… if he accepts the directives of the alleged authority as authoritatively binding and tries to follow [its directives], rather than by trying to follow the reasons which apply to him directly’.36 That is, to say that A is an authority for B is not to insist that it is up to A to decide how B should act. Rather, it is simply that B has good instrumental reasons to do as A says—for it is most likely to guide B to do as he already has reason to do. As a result of this difference, a claim of robust authority of the sort I am concerned with here and a claim of Razian practical authority have a number of significantly different implications. As Raz himself notes, the Normal Justification Thesis, ‘invites a piecemeal approach to the question of the authority of governments, which yields the conclusion that the extent of governmental authority varies from individual to individual, and is more limited than the authority most governments claim for themselves in the case of most people’.37 This follows from the fact that authority is a matter of the instrumental value to the subject of following the putative authority’s directives. Where this is of instrumental value in complying with reasons that apply to us, we have reason to follow the putative authority’s directives. But where they are not, there is no broader reason for us to defer to them whatsoever. By contrast, on the account of robust authority I am concerned with here, the concern is not with the instrumental value to the subject of following any particular directive from the authority. It is, rather, the justification of the relationship of robust authority as such. Once we have established that parents have robust authority over their children, the conclusion of that argument is that certain decisions about their children’s lives are simply up to them. Now, of course, it remains to be seen whether it is possible to justify such a relationship of robust authority—but this is a task for part three of this chapter. 36 37 J Raz, The Morality of Freedom (Oxford, Oxford University Press, 1986) 53. Ibid at 80. 18 C. The King’s Peace and the Authority of the State It is perhaps easy to concede that parents have robust authority over their minor children in at least some aspects of their lives: young children are simply not capable of making sensible decisions about many aspects of their lives, so it is incumbent upon their parents to make those decisions for them. Those decisions are properly up to the parents to make and not to the children. But it is not nearly so easy to see how competent adults could be subject to a similar sort of robust authority from the state of which they are subjects. We do not conceive of competent adults as incapable of exercising meaningful choice on their own behalf. How, then, can we reconcile our self-conception as free moral agents with the thought that it is the state (and not us) who has the standing to make certain decisions about how we should live our lives? In this context, the very thin picture of authority painted by Raz and his followers might seem more plausible. On the Razian picture, all that we mean when we say that the state has authority over us is that we have good reason simply to follow the state’s directives rather than trying to figure out how to act from first principles. This is not because these decisions are the state’s (and not ours) to make; rather, it is simply that as we make our own decisions about our lives, we will tend to make better decisions if we acted as though these decisions were up to the state to make. On this picture, we do not suggest that decision-making authority has passed from individuals to the state over any particular matter; it is just that we will exercise our own decision-making authority better if we acted as though this were the case. In this section of the chapter, I do not present a substantive argument for why we should recognise the robust public authority of the state. I take up that question in part three. For now, my focus is on what criminal law and punishment would look like when seen in the context of such a commitment. I take the idea of the king’s peace as an organizing idea as the clearest example of how we might think of the state’s powers to make laws, to prevent wrongdoing with the use of pre-emptive force, and to punish those who violate its laws as parts of a single idea: the state’s robust public authority over its subjects. The Roman Empire claimed not merely the power to extract taxes over its subject lands and peoples; it claimed—and simultaneously promised—to impose its 19 peace, the pax romana over those lands and peoples.38 Similarly, in England, from the time of Henry II, the king claimed and simultaneously promised to impose his peace: ‘the king’s peace’.39 The possessive form is significant. That is, the king did not merely promise to provide peace for his subjects; he promised to provide his peace—that is, peace under his terms. By imposing his peace on his subjects, the king would tell them what they could and could not do—but they could also be sure of the terms under which they would live with one another. That is, to be under the king’s peace was to be simultaneously under the authority of the king and under his protection—all according to terms set by the king. We see this idea of the king’s peace at work in a variety of ways in the early English criminal law. Acts committed in war that would otherwise be crimes were deemed not to be crimes because they were not committed under the king’s peace. That is, the reason why killing in war was not subject to criminal punishment was simply because it was not committed in violation of the king’s laws—for the king’s jurisdiction did not extend to acts committed in war.40 Similarly, the status of ‘outlaw’ was for many years a recognised status in English law: to be an outlaw was to be someone who was not under the king’s peace.41 Although it is always morally wrong to assault or to rob or to murder a person, such offences would not be punishable if committed against an outlaw. The reason, once again, was bound up with the idea of the king’s peace: if the victim was not subject to the king’s peace, then the king did not claim jurisdiction over any offences committed against that person. 38 T B Lambert and D Rollason (eds), Peace and Protection in the Middle Ages (Toronto, Pontifical Institute for Medieval Studies, 2009). 39 According to G O Sayles, The Medieval Foundations of England (London, Methuen, 1948) 170–2, ‘the king’s peace’ had earlier been used to refer to a special privilege granted to those favoured by the king. It was under Henry II that this took on a broader meaning of a form of protection for all. 40 J Fitzjames Stephen, A History of the Criminal Law of England, Vol 2 (Cambridge, Cambridge University Press, 2014) at 62: ‘The older definitions of murder expressly say that it is the killing of a person ‘within the King’s peace,’ but an open enemy is not within the King’s peace’. 41 Henry de Bracton, On the Laws and Customs of England, Vol 2, edited by G E Woodbine, translated with revisions and notes by S E Thorne (Cambridge, Cambridge University Press, 1968) 357: After people are declared outlaws, ‘they bear the wolf’s head and in consequence perish without judicial inquiry; they carry their judgment with them and they deservedly perish without law who have refused to live according to the law’. 20 On this understanding of sovereign authority, we treat the power to legislate, the power to use pre-emptive force to ensure compliance and the power to threaten and carry out punishment of those who disregard the king’s directives as three incidents of a single normative position of robust public authority akin to the position of parents over their children.42 As we saw in the case of parents and their children, there are three powers implicit in the position of robust authority. First, the king shall communicate his decisions to his subjects so that they may conform their own conduct to those directives. Second, the king must also have the power to use pre-emptive force to ensure his subjects’ compliance with his directives where necessary. Thus, the king’s officials must have the power to use coercive force to prevent individuals from committing criminal offences.43 And third, the king must have the power to threaten and to carry out punishment for disobedience of his directives. D. Wrongs and Crimes I return to these early roots of the criminal law in England because I believe this illustrates an important aspect of criminal punishment that has been lost in contemporary criminal law theory. This is that the idea of criminal punishment is tied up with a larger notion of the king’s (or, in modern times, the state’s) exclusive authority over a certain subject matter. Take theft. Where one person takes the property of another, this might be a matter of moral significance—within a reasonably just system of property, there are good reasons of interpersonal morality not to take the property of another without her consent. But the fact that theft is morally wrong is (as we have seen at some length in section one, above) not sufficient grounds for anyone to punish the thief. The crucial question for criminal law theory is not what makes theft morally wrong, but what makes 42 As Martin Loughlin (citing F W Maitland) points out, Parliament ‘came into existence as an emanation of royal power’. That is, it is quite plausible to see the legislative power as part of the king’s public authority. See M Loughlin, ‘The State, the Crown and the Law’, in M Sunkin and S Payne (eds), The Nature of the Crown (Oxford, Oxford University Press, 1999) 33 at 47. 43 As Sir Frederick Pollock points out in F Pollock, ‘The King’s Peace in the Middle Ages’ (1899) 13 Harvard Law Review 177 at 186, the king in medieval England did not in fact make good on this promise in any serious way. Instead, the king simply assumed the power to supervise the enforcement of his laws by local authorities. It was not until the nineteenth century that the state in the common law world attempted to control crime directly through public policing. 21 it properly the object of criminal punishment by the state. And this is a question of robust authority. If the king claims the exclusive right to decide what property belongs to whom within the jurisdiction, then anyone who takes it upon himself to upset that arrangement by stealing the property of another has violated the king’s authority over the allocation of property. And in order to vindicate the king’s claim of robust authority over the matter, punishment—making good on the king’s threat—is required. On this account, then, the relationship that is of crucial significance to the criminal law is not the one between victim (where there is a victim) and perpetrator; rather, it is the one that obtains between state and citizen. Returning once more to the parental analogy, it is the parent who is entitled to punish her child for wronging another child because punishment is an aspect of the parent’s robust authority over her child. The wrong done to the other child, from the point of view of punishment, is of significance insofar as it constitutes an attack on the parent’s claim of robust authority. The victim, from the point of view of punishment, is neither here nor there. A child may disobey his parent by wronging a specific victim (say, by hitting his sister) but he might just as easily do so by committing a victimless wrong (such as eating sweets when forbidden to do so). Similarly, the fact that the offender has committed a moral wrong against a particular victim is not in itself significant to criminal law and punishment. It is significant only insofar as the state has prohibited that form of conduct and attached the threat of punishment to anyone who chooses to violate that prohibition. III. Justifying Robust Public Authority It is a commonplace in much of the history of political thought that private persons cannot legitimately punish one another for wrongdoing, for this is something that only a robust public authority can do to its subject.44 We can now see why. As we saw in the 44 Eg, Aquinas, above n 27: ‘this coercive power is vested in the whole people or in some public personage, to whom it belongs to inflict penalties’; Thomas Hobbes, Leviathan edited by J C A Gaskin (Oxford, Oxford University Press, 1996) 205: ‘A punishment is an evil inflicted by a public authority, on him that hath done, or omitted that which is judged by the same authority to be a transgression of the law; to the end that the will of men may thereby the better be disposed to obedience’ [emphasis added]; Immanuel Kant, Metaphysics of Morals Ak 6:331 in I Kant, Practical Philosophy translated and edited by M J Gregor 22 first part of this chapter, criminal law theorists have been unable to come up with a good moral justification for criminal punishment without recourse to the idea of robust public authority. But as we saw in the second part of this chapter, the idea of punishment can be understood as one aspect (along with the ideas of setting norms of conduct and of enforcing them with preemptive coercion) of a larger idea of robust authority. The question that remains, then, is whether there is a plausible moral justification for such a relationship of robust authority between the state and its subjects. If such an argument can be found, it will provide what has hitherto proven elusive in the philosophy of punishment: an argument for why it is morally justifiable for the state to punish criminal offenders. In what follows, I do not pretend to provide anything like an exhaustive treatment of these issues. My aim is simply to show that there are a number of efforts in this regard in the history of political philosophy that offer very promising beginnings for a theory of justified criminal punishment. A. Parenthood Again As we saw in part two, parents claim robust authority over their children in a great many areas of life: where they may live, how they shall be educated, how they shall eat, and how they shall behave more generally. These questions concerning the most basic aspects of children’s lives are up to their parents to decide, and not up to the children themselves. In many cases, of course, it seems that almost any account of authority would recognise that this is as it should be. Even on Raz’s instrumentalist account of authority, children ought generally to defer to their parents’ directives rather than trying to conform to the reasons that apply to them on their own. But on Raz’s account, this is just because they would be more likely actually to conform to those underlying reasons if they deferred to their parents than if they tried to do so on their own. (Cambridge, Cambridge University Press, 1996) 472: ‘The right to punish is the right a ruler has against a subject to inflict pain on him because of his having committed a crime’; John Locke, ‘A Letter Concerning Toleration’ in J Locke, A Letter Concerning Toleration and Other Writings edited by M Goldie (Indianapolis, Liberty Fund, 2010) 14: ‘Every man is entitled to admonish, exhort, and convince another of error, and lead him by reasoning to accept his own opinions. But it is the magistrate’s province to give orders by decree and compel with the sword’. 23 The account of robust authority, however, provides considerably more assistance to us as we try to make sense of state authority in matters of criminal law. In the case of parents and children, the account of robust authority does not merely insist that children generally have good instrumental reasons to defer to their parents’ directives. Rather, it insists that many decisions are simply not up to the children to make in the first place. In the case of minor children, this is because they are simply not equipped to make such decisions—they are incapable of exercising the kind of judgment that is required for us to be able to say that they are exercising their basic rights. Whether it is because their decisions are irrational, unintelligible, too radically uninformed or for some other reason,45 we recognise that minor children are simply not in a position to make responsible decisions over certain matters. So although they are, in principle, holders of the same rights as competent adults, they are not capable of exercising meaningful choice about the exercise of those rights. So the problem that gives rise to robust parental authority is this: children are bearers of certain rights but they are incapable of exercising responsible agency in the enjoyment of those rights. The role of parents, then, is not to provide useful guidance to the perplexed child, allowing him to make wiser decisions than he otherwise would. It is to take over the role of decision-maker altogether in certain matters.46 Minor children are not merely in need of guidance as they exercise rational agency in certain matters; they are in need of someone to take over their decision-making authority altogether. But in order for parental decision-making over matters concerning their children’s lives to be consistent with the children’s claim of autonomy, we insist that parents must exercise that decision-making authority in the best interests of their child. That is, we must think of parents (and they must think of themselves) as making decisions not as dispassionate experts but as defenders of the child’s interests, acting in the child’s name. The rationale for robust authority in the case of parents, then, is not an instrumentalist one—assisting children in the exercise of their agency. It is, rather, an 45 Moore, Law and Psychiatry, above n 31 explores several of the grounds for denying that a particular agent is capable of responsible agency. 46 I do not mean to ignore the educative role of parents. But this is an obligation parents bear in addition to their obligation to make decisions for their children. 24 argument about the necessary conditions for the possibility of children enjoying rights in any way at all. If parents were not available to exercise decision-making authority over their children, their claim of individual rights would be meaningless. Parental authority, therefore, need not be justified as an infringement on children’s rights, but as a necessary condition for their very existence. This leaves only the question of why it should be parents rather than any other competent adult who should make such decisions. The most plausible answer is that parents must take on this responsibility insofar as they are the cause of the child’s predicament in the first place. It is through the parents’ choice that the child was brought into the world in a situation where he is unable to exercise decision-making authority over his rights, so it the parents who must bear the duty to make decisions on behalf of their children and in their best interests. B. State Instrumentalism In earlier times, the authority of the king over his subjects was often derived from the authority of God over all his children, and often explained in terms of the authority of a father over his children.47 But that sort of argument has not had any real purchase for many years. The beginning of any plausible moral justification of the state’s authority to decide how we may live and interact with others today must be a recognition of the status of all persons as free and equal in our moral standing. Whether it is Raz’s instrumentalist account of authority or the account of robust authority contemplated here, any plausible modern account of authority must take the claim of all persons to equal freedom as a starting point.48 The division among accounts, then, will turn on differences among them in their understanding of precisely what a claim of individual freedom entails, how claims of individual freedom and equality might relate to one another, etc. 47 The best-known example of such an argument is R Filmer Patriarcha edited by J Somerville (Cambridge, Cambridge University Press, 1991, to which Locke’s Two Treatises are said to be a response. Another interesting example is the treatise penned by James I of England (James VI of Scotland) arguing that the normative basis of his right to rule lay in the Christian doctrine of the apostolic succession. See ‘The True Law of Free Monarchies’ in The Political Works of James I edited by C H McIlwain (Cambridge MA, Harvard University Press, 1918). 48 To be explicit: the class of justifications I am concerned with are broadly liberal ones. In this, however, I do not diverge from the main currents of criminal law theory today. 25 Arguments for the state’s robust public authority turn on a claim of necessity of some kind. In one version, the necessity is practical. John Locke is usually understood to argue that a state is necessary in order to allow people to escape the ‘inconveniences’ of the state of nature.49 That is, given certain predictable facts about human beings, conditions of life will be miserable unless we are able to live together under a central authority with the jurisdiction to set laws of general application and to settle disputes under that law, to enforce laws with preemptive coercive force and to punish ex post facto those who choose to disobey the law. This is because when people take it upon themselves to interpret the demands of morality, they will tend to disagree; when people take it upon themselves to prevent violations of morality’s demands, they will tend to overreact; and so on. On this account, the price we pay in a loss of autonomy is well worth paying for the benefit we receive in securing the stability, predictability and evenhandedness of state action in all these areas. An instrumentalist argument for robust public authority of this sort, based on the costs and benefits of granting the state robust public authority over us, is liable to a number of trenchant critiques. Most importantly, it is unclear whether the cost-benefit justification of state authority can ever succeed in accomplishing its aim. For if, as seems clear from part one of this chapter, individuals do not have the authority to impose laws on one another and punish those who violate such laws (or to do many of the other things that states usually do: collecting taxes, adjudicating disputes, waging war and much else), then they could not grant such a power to the state even if they wanted to. If the state is nothing more than a tool we have created together to deal with a number of inconveniences in the state of nature, it could not have any powers beyond those we each possess as individuals. This critique is especially important today. For we should keep in mind that none of the accounts we considered in part one of this chapter were concerned with private punishment of wrongdoers. Both the direct strategy of Moore, Duff, et al and the indirect 49 J Locke, Second Treatise of Government in Two Treatises of Government edited by P Laslett (Cambridge, Cambridge University Press, 1960) 326-7: ‘For where-ever any two men are, who have no standing rule, and common Judge to Appeal to on Earth... they are still in the state of Nature, and under all the inconveniences of it’. 26 strategy of Hart and his many contemporary followers were self-consciously set out as accounts of state punishment. The trouble that all these accounts shared was that they could not explain how the state could have a power to set norms of conduct for its subjects and to enforce them with coercive force in a way that private individuals did not already have. So if we were to follow an instrumentalist account of state authority of the sort Locke proposes, we should not be able to provide a plausible basis for the sort of robust public authority that is required to solve the problem of justified punishment. C. Equal Freedom The instrumentalist intuition—that life without a stable state to secure civil peace would be unbearable—is a powerful one, but it is incapable of providing the sort of normative justification for state authority that our account of criminal punishment requires. For even if we are willing to make a trade-off of individual liberty for security, we will find that the deal does not provide what we need—as the state that results from this account cannot possess any powers beyond those that each of us already possessed ex ante. We need an account of robust state authority in order to explain how it could do what no individual has the right to do: to impose laws of general application and to enforce them with preemptive force and ex post facto punishment. No account based on a transfer of individual powers to the state can do that. Instead, we may look to an argument for the moral necessity of robust state authority on quite different grounds. In much of the natural law tradition, the argument for robust state authority is based on the recognition that the state is necessary precisely because we require it to do things that no private person is morally entitled to do. On one reading of that account, a public authority is morally necessary in order to make it possible for persons to live together under the rule of law. The central problem with private punishment and private lawmaking is that it presumes that one person has the authority to make and to enforce a set of norms over another person. So if one person assumes that role over another, this undermines the equal status of each. But if we try to do without anyone setting and enforcing norms of conduct, then everyone is vulnerable to the unilateral norm-setting of others at all times. Where there are no coercively-enforced 27 norms in place, the stronger (or the more cunning or the more reckless) will be able to impose the terms of interaction of his choosing on others, without any answer. The role of punishment in a regime of robust authority, as we have seen, is to reassert the continued authority of the legitimate authority in the face of an attempt by wrongdoers to usurp that role. Punishment is not designed to undo the harm caused by the wrong (this, if anything, is to be accomplished by damage awards in private law); its purpose is to carry through on the authoritative ex ante threat of punishment, showing that that exercise of authority to be determinative of the relations between the authority and the offender. IV. Conclusions, Qualifications and Clarifications The two main strategies for justifying the workings of criminal justice just as a matter of interpersonal morality—either as matter of giving moral wrongdoers their due or as a scheme for encouraging people to conform to morally valuable legal rules—have generally failed to convince. My central aim in this chapter has been to revive a way of thinking about criminal justice that once dominated the debate. From Aquinas through to Hobbes, Rousseau, Kant and many others, it was generally assumed that criminal justice was not something we could justify in terms of interpersonal morality; it was something that only a robust public authority could undertake—and the form of moral justification appropriate to this enterprise was one that reflected the moral powers, obligations and limits of such an entity. These writers recognised that there are many things that individuals are morally justified in doing to one another—communicating censure, offering incentives and much else—but these activities are importantly distinct from the practices of criminal justice. Although we might be tempted to do away with criminal justice as traditionally understood and replace it with something that can fit into our favourite moral accounts, we might benefit from broadening our moral horizons to recognise other forms of moral justification that might be better suited to making sense of criminal justice as we know it. The strategy outlined here is only the beginning of a larger research program. For once we have identified the connection between robust public authority and the 28 operations of criminal justice, there are still at least two major tasks ahead. The first is to sketch out an account of the legitimacy conditions for a robust public authority: what sorts of matter are properly within the public authority’s jurisdiction and what modes of deliberation ought to constrain the public authority as it addresses these matters? This first task is a matter of constitutional theory, broadly understood.50 The second is to contribute to the substantive political debate about how the public authority ought to exercise its legitimate powers in the operation of criminal justice institutions today. This task is a matter of substantive moral theorising: an attempt to provide our own favoured answer to the moral questions relevant to the institutions of robust public authority. 50 To be more specific: it is not a matter of how to interpret a particular constitution document. Rather, it is a matter of working out what the very idea of a constitution means and what limits it might impose on robust public authority. 29