Theories of Criminal Law

Philosophical ‘theories of criminal law’ may be analytical or normative (§ 1). Once we have identified the salient features that distinguish criminal law from other kinds of law (§2), we ask whether and why we should maintain such an institution (§3). Instrumentalist answers to this question portray criminal law as an efficient technique that helps us achieve worthwhile ends; non-instrumentalist answers portray it as an intrinsically appropriate response to certain kinds of wrongful conduct (§4). By considering the question of how the criminal law should address citizens (§5), we can discern the truth in the non-instrumentalist perspective. The next question concerns the proper scope of the criminal law: what kinds of conduct should be criminalised? Several candidate principles of criminalisation are critically discussed (§6), including the Harm Principle, and the claim that the criminal law should be concerned with ‘public’, rather than merely ‘private’, wrongs.Further questions are raised, however (§7), by the increasingly important phenomenon of international criminal law.

1. Different Kinds of Theory

‘Theories of criminal law’ could just be general theories of law applied to the particular case of criminal law: proponents of legal positivism, of natural law, of economic analysis of law, of Critical Legal Studies and other schools of legal theory will expect to be able to say about the criminal law what they say about law in general (for examples of the last two approaches, see Posner 1985; Kelman 1981). Questions raised by theories of this kind will figure in what follows—for instance whether it is part of the essence of criminal law that it must satisfy, or make, certain kinds of moral demand; whether criminal law can be adequately understood in purely instrumental terms; whether we should take the criminal law's apparent pretensions to rationality and principle seriously, or should rather see it as an oppressive exercise of political or economic power, or as the site of conflicts which produce an irredeemably contradictory, unprincipled set of doctrines and norms (see Norrie 2001). Such questions are important, but we will not begin with them. We should, instead, begin by asking what is distinctive about criminal law. What marks it out from other kinds or aspects of law? What are its distinctive institutional structures, purposes, or content?

Philosophical theories of criminal law can be analytical, or normative (see Husak 1987: 20-26). Analytical theorists seek to explain the concept of criminal law, and related concepts such as—most obviously—that of crime (metaphysically more ambitious theorists might seek an account not merely of the concept of criminal law, but of its real, metaphysical nature; see Moore 1997: 18-30). They need not look for a strict, ahistorical definition—an account of the necessary and sufficient conditions given, and only given, which a human practice counts as a system of criminal law; we have no reason to think that any such definition will be available. But they can hope to identify and explain the central or salient features of systems of criminal law—features at least some of which will be exhibited by anything we can count as a system of criminal law; and to develop an account of a paradigm of criminal law, on the basis of which we can recognise as systems of criminal law other practices that resemble that paradigm sufficiently closely, even though they do not quite fit it.

Normative theorists seek an account not just of what criminal law is, but of what it ought to be (and whether it ought to be at all). Should we maintain a system of criminal law? If so, what goals should it serve, what values should inform it, what should its scope and structure be? Any such normative theory must presuppose some analytical account of that whose goals, values, scope and structure are being discussed. Whether analytical and normative theorising are related more closely than this will depend on what kind of analytical theory we develop: a legal positivist will insist that, here as elsewhere, the question of what law ought to be is quite separate from, and left open by answers to, the question of what law is; a Natural law theorist will argue that an adequate analysis of the concept or the metaphysical nature of criminal law will reveal the moral purposes or values that a practice must serve (or at least claim to serve) if it is to count as a system of criminal law at all (see Moore 1997: 23-35).

Philosophical theories of criminal law, whether analytical or normative, cannot subsist in isolation. They must have some regard to the empirical actualities of that which they theorise: to the histories of the different systems of criminal law, and to sociological inquiries into their actual operations. Some critical theorists believe that such historical or sociological inquiries will undercut the pretensions of philosophical theorising: that what needs analysing is not the superstructure or superficial self-presentation of the criminal law, on which philosophers tend to concentrate, but the social, political and economic realities lying beneath that surface; and that given the oppressive or conflictual nature of those realities, philosophical theories cannot amount to anything more than doomed attempts to rationalise what is inherently irrational or a-rational (see Kelman 1981; Norrie 2001; also Law and Ideology). The only adequate reply to these critiques of philosophical theorising is to show how such theorising can assist both an understanding of what criminal law is, and the discussion of what it ought to be, by taking seriously the concepts in terms of which it presents itself: that is the task on which we embark in what follows.

(Another way in which philosophical theories of criminal law cannot subsist in isolation is that they cannot be wholly separate from other branches of philosophy. They must draw, most obviously, on political philosophy, since they must depend on some conception of the proper aims of the state and of the proper relationship between a state and its citizens (see e.g. Pettit 1997, 2002; Duff 2001: chs 2-3; Dubber 2005; Dagger 2008). They must draw on moral philosophy, insofar as the criminal law properly aims to define types of moral wrong and to punish those who culpably commit them (see e.g. Moore 1997; Tadros 2005). They must draw on philosophy of action and on philosophy of mind, if they are to explicate ideas of wrongdoing and of fault that are appropriate to law's wrong-defining role (see e.g. Moore 1993; Green 2005; Gardner 2007).)
2. Aspects of Criminal Law

We can usefully begin by identifying some of the salient features of the systems of criminal law with which we are familiar (contemporary systems of municipal criminal law): features by which we can distinguish the criminal law both from non-legal phenomena and from other types of law. It would be unproductive to ask whether all these are strictly necessary features of criminal law, or whether we might still count a practice that lacked one or more of them as a system of criminal law; the most we can sensibly claim is that these are defining features of the paradigm of criminal law as we understand and experience it.

The criminal law deals with crimes: but what is a crime—and how does the criminal law deal with it?

Crimes, we might initially say, are kinds of conduct that are defined by the law as wrong. However, even this crude initial approximation must be qualified. First, we can say that crime always involves ‘conduct’ only if we stretch the meaning of that term so far as to empty it of substantial content (see Husak 1987: ch. 4): we can (whether justly or not) be held criminally liable not merely for what we do, or fail to do, but for what we are, perhaps even for what we think—for what we intend, for instance. But for the moment we can talk of ‘conduct’, since it captures the most familiar kinds of crime. Second, we must not, or must not yet, read ‘wrong’ here as ‘morally wrong’: it will be a further question whether the criminal law either must of its very nature, or should as a matter of normative theory, portray the conduct it criminalises as morally wrong; all we should say so far is that it portrays it as being in some way wrong or defective, something that those bound by the law should not do (this point is often expressed by saying that the criminal law ‘prohibits’ the conduct that it defines as criminal, but we will see in section 5 that this is misleading). That is, for instance, the defining difference between a law that defines a certain kind of conduct as a crime which is punishable by a fine, and one that subjects that conduct to a tax: both laws might be intended to reduce the incidence of the conduct, but the former, unlike the latter, does by defining and punishing it as wrong.

Crimes differ from extra-legal wrongs in that they are defined as wrongs by the law: they are not, or not just, wrongs in terms of some extra-legal social standard of morality, prudence, or etiquette, but wrongs that are defined and recognised as such by the law. (This leaves open the question of whether the criminal law can create wrongs, or whether it rather gives formal recognition to wrongs whose wrongfulness is initially determined by extra-legal standards.) But not all legally defined wrongs are criminal wrongs.

First, some legal systems distinguish between ‘crimes’ properly speaking and other kinds of penalised conduct. So German law distinguishes ‘Strafrecht’ and ‘Straftaten’ (criminal law and crimes) from ‘Ordnungswidrigkeitenrecht’ and ‘Ordnungswidrigkeiten’ (regulations and violations; Weigend 1988); and the American Law Institute's Model Penal Code distinguishes ‘crimes’ from ‘violations’ (s. 1.04). Violations might include conduct that other legal systems count as criminal, although even in systems in which it counts formally as criminal, it is often seen as not ‘really’ criminal (thus German Ordnungswidrigkeitenrecht includes many traffic violations that English law defines as crimes, although many drivers would deny that they are ‘real’ crimes). They are distinguished from crimes by the procedures for dealing with them, the relative mildness of the sanctions they attract, and the absence of some of the other consequences that typically attach to conviction for a crime—such as a criminal record. I will not discuss this distinction further here, save to note that it can be justified (if it is justifiable) either on pragmatic grounds—mere violations are not dangerous enough to justify mobilising the expensive resources of the criminal justice system; or on principled grounds—they do not involve serious enough wrongs to merit the condemnation that a criminal conviction, as we will see, involves (for further, more critical discussion, see Weigend 1988; Steiker 1997; Duff et al 2007: ch 6.5.

Second, most legal systems distinguish criminal from civil wrongs: wrongs that ground a criminal prosecution, from those that ground a civil case for damages brought by the injured party. We can clarify the concept of crime by focusing on this distinction. The same conduct often constitutes both a criminal and a civil wrong, as is shown most dramatically when, after a failed prosecution or a decision not to prosecute, the victim or her family bring a civil case for damages against the alleged wrongdoer: but we can still usefully ask what the difference is between defining and treating conduct as a criminal wrong and defining and treating it as a civil wrong (see generally Murphy & Coleman 1990: ch. 3; Boston University Law Review 1996).

Civil wrongs are typically treated as ‘private’ matters in the sense that it is for the victim to investigate what happened, to identify the alleged wrongdoer, and to bring a case against him. The law provides the institutions (the courts, arbitration panels) through which that case can be brought; it lays down the norms by reference to which the case is decided; it specifies what remedies are available; it might also help successful plaintiffs to extract damages from unwilling defendants. But it is for the injured party to bring, or to decide not to bring, a case; to pursue, or to abandon, that case; to insist on extracting the damages the court awarded, or to forgo them. The case is described and understood as ‘P v D’: P sues D, and the case thus belongs to her. The criminal law, however, provides for the public investigation, prosecution and punishment of crimes: for a police force, tasked with investigating (as well as preventing) crime and detecting criminals; for a system of criminal courts, in which defendants are tried for the crimes that they are alleged to have committed (and whose workings are structured by a complex array of procedural rules and requirements); for a system of punishments that will be imposed by the courts, and administered by other institutions and officials. Now the police act in the name and with the authority not just of the victim, but of the whole polity; it is for the prosecuting authority, not for the victim, to decide whether, and on what charge, anyone will be prosecuted. If the victim does not want the case to go to court, the prosecutors will in fact often not proceed with it—because it would be hard to do so without the victim's willing co-operation, or out of concern for the victim's feelings; but cases can be prosecuted despite the victim's unwillingness (this can be an important issue for prosecutors dealing with domestic violence; see Dempsey 2007). When the case comes to court, it is described not as ‘P v D’, but as ‘State v D’, or ‘People v D’, or ‘Queen v D’: D is prosecuted not by an individual victim, but by the polity—or, in societies that have not yet shaken off the trappings of undemocratic monarchy, by its sovereign. (Some legal systems allow the possibility of private prosecutions; this is one of several ways in which the distinction between criminal and civil law is neither sharp nor watertight.)

The difference between the public character of criminal wrongs and the private character of civil wrongs is also evident in the outcomes of the two kinds of legal process. A civil case typically results in a finding either for the plaintiff, or for the defendant; if the plaintiff wins, the defendant may have to pay her damages, as compensation for the harm that she suffered, and for which she has sued. Criminal cases, by contrast, result in a conviction or an acquittal; if the defendant is convicted, he is liable to suffer a punishment. Criminal convictions express an explicit condemnation of the defendant: he has been proved guilty of doing wrong, and the verdict is focused on that wrong. A verdict for the plaintiff in a civil case will typically imply that the defendant acted wrongfully, but the focus of the case, and thus of the verdict, is more on who should pay for whatever harm was caused (see Ripstein 1999: chs. 2-4).

Finally, the punishments imposed for crimes differ from the damages that are awarded as a result of a civil suit—and not just in the fact that whereas the successful plaintiff can forego the damages she is awarded, it is not for the victim of a crime to decide whether the sentence imposed by the court should be carried out. Often punishments take a different material form from civil damages, as when an offender is imprisoned or put on probation. Even when their material form does not differ, however (as when a convicted offender is fined £1,000, and the defendant who loses a civil case is ordered to pay £1,000 in damages), their meanings differ. First, even if the severity of criminal punishments is to some degree determined by the extent of the harm caused (itself a controversial matter), it typically also depends on the nature and degree of the offender's culpability for that harm: someone who kills or injures recklessly can expect to be punished more severely than someone who causes death or injury by a negligent act or omission. Civil damages, however, are proportioned to the harm actually caused; some kind of culpability, such as negligence, might be a threshold requirement, in that the plaintiff must prove that the defendant was at least negligent in relation to the harm he caused; but the damages are not proportioned to the degree of the defendant's culpability, since their purpose is simply to provide compensation for the harm caused. Second, punishments are intended to be painful or burdensome, whereas civil damages are not (indeed, this is standardly cited as a defining feature of punishment). If I am ordered to pay £1,000 compensation for damage that I negligently caused to your property, making that payment might be burdensome for me, if I am not well off, or no burden at all, if I am rich (or have suitable insurance): but the damages serve their purpose in either case. If, however, I am fined as punishment for a crime, that fine is intended to be burdensome: that is why many sentencing authorities aim to proportion fines to the offender's means, to ensure that both rich and poor offenders are burdened fairly and proportionately; and that is what is wrong with the idea that one might take out insurance to cover the cost of fines.

These distinctions between criminal and civil outcomes are often blurred in practice. For instance, English criminal courts can include a ‘compensation order’ in the sentence that they impose, thus bringing a dimension of civil law into the criminal process (see Powers of Criminal Courts (Sentencing) Act 2000 s.130); some civil systems allow for the award of ‘punitive damages’, which are intended to burden the defendant. More generally, civil plaintiffs might be seeking an apology rather than damages that could compensate for the material harm they have suffered—this is often true in libel cases, for instance, or in cases in which bereaved parents sue the hospital or corporation that negligently caused their child's death: in such cases, any damages that are awarded might best be rationalised as a way of giving material form and force to the apology; but if they are to serve that role, they must be burdensome (see Duff 2001: 94-96). Nonetheless, we can still draw a useful analytical distinction between two paradigms: a civil paradigm, which focuses on harm that has been caused and on the question of where the costs of that harm should fall; and a criminal paradigm, which focuses on a wrong that has been done and on the question of who—if anyone—should be condemned and punished for that wrong. The civil paradigm is a matter of private law in the sense that it aims to provide compensation and satisfaction for the aggrieved plaintiff, if she chooses to pursue the case; the criminal paradigm is, by contrast, a matter of public law in the sense that the case is brought and the punishment is imposed in the name and on behalf of the whole polity rather than any individual victim.

We now have a sketch of the criminal law as a distinctive kind of human institution. This then raises three further questions. First, should we have such an institution at all? Second, if we should, what goals or purposes should it serve? Third, what should its content be, or how should that content be determined: what kinds of conduct should be criminal, and how should we go about deciding that issue?
3. Should We Abolish the Criminal Law?

Although the criminal law is a pervasive, and might seem to be an inescapable, feature of the developed societies in which we live, there are those who argue that, precisely in virtue of the paradigmatic features identified in the previous section, it is an institution that we should seek to abolish: this is a central strand of the ‘abolitionist’ movement which, whilst often focusing most directly on the abolition of criminal punishment, also incorporates a critique of criminal law (see Legal Punishment ss. 2, 7). Abolitionist critics focus on three aspects of criminal law which, they argue, make it an utterly unsuitable institution for the kinds of social life and the kinds of relationship that we should seek (see Christie 1977; Hulsman 1986; Bianchi 1994).

First, the criminal law purports to declare and enforce authoritative standards of value, in particular of moral value: it claims the authority to tell us how we should live, and to enforce its demands on us if we disagree or disobey. But this, critics argue, amounts to an illegitimate attempt to impose a moral consensus—inevitably the consensus of those with political power—on societies which are rather characterised by radical moral disagreement; it denies to those who do not share that consensus the freedom to think and live as they see fit.

Second, the criminal law ‘steals conflicts’ from those to whom they properly belong. Of course citizens often find themselves in conflict with one another; their relationships are often impaired by various ‘troubles’. Such conflicts and troubles must be resolved; any harms that have been done must be repaired. But that is a task for those most directly involved—for the ‘victim’ and the ‘offender’ (though we should be cautious about such notions), with the help of their local community. The criminal law, however, in defining such conflicts or troubles as criminal wrongs to be dealt with by a public criminal process, steals them: it transfers them to the professionalized context of a criminal justice system in which neither victims nor offenders are allowed really to participate; it thus denies those to whom the conflict belongs the chance to work it out for themselves.

Third, the criminal law deals in punishment—in ‘pain delivery’—when what is needed is instead a process that will repair whatever harm was caused, reconcile the people involved in the conflict, and thus restore the relationships that the conflict damaged. Criminal punishment cannot contribute to those appropriate ends: it reflects a primitive, backward-looking concern with retributive justice, whereas we should rather be seeking a forward-looking restorative or reparative justice.

I will not discuss the third objection here, since it belongs with the discussion of criminal punishment (see Legal Punishment ss. 2, 7): but we will see that the familiar consequentialist and retributivist models for the justification of punishment have analogues in accounts of the proper aims of criminal law more generally. The following sections will constitute an answer to the first two objections. Abolitionists are right to highlight these two features of a system of criminal law: it claims the authority to declare certain public norms of conduct (norms that must, as we will see, claim a moral foundation), and to insist on respect for those norms even from those who do not share them; and it makes breaches of those norms its business, and so the business of the whole polity in whose name and on whose behalf the law claims to speak and to act, rather than leaving them as the business purely of those who are most immediately involved. The question is whether and how we can justify maintaining institutional practices of this kind.

The first objection is, as it stands, unimpressive, and sometimes expresses an incoherent moral relativism which makes the moral demand that we should not make moral demands of others (see B. Williams 1976: 34-9). It does reflect two general issues that face any attempt to justify systems of political authority and law: the question of how far a polity depends for its legitimacy on a normative consensus, at least a Rawlsian overlapping consensus, amongst its members, and how far law and polity are possible in contexts of radical disagreement; and the question of whether and how a polity can claim legitimate authority over those who reject its central values and its normative claims. We cannot pursue these questions here, although we may note that they are as urgent for abolitionists as they are for advocates of the criminal law, since their favoured practices and institutions depend, just as a system of criminal law does, on the legitimacy and authority of the polity that sustains them. However, this first objection does also raise a question that is more specific to the criminal law, and that must be answered by those who would defend the criminal law: what kinds of norm, with what kind of claimed authority, does or should the criminal law declare—and should we maintain an institution that seeks to declare and support such norms?

The second objection focuses our attention on the distinction between civil and criminal law sketched in the previous section. We might agree with the abolitionists that our existing criminal procedures do not allow either victims or offenders the actively participatory roles that they should be able and encouraged to play, but the basic question is still this: should we maintain a system of law that defines and responds to a category of ‘public’ wrongs—wrongs that concern not only the particular victim and offender, but the whole polity; wrongs which are ‘our’ business collectively as a polity, and which must therefore be investigated and dealt with by a public process—which inevitably involves taking them out of the hands of those most immediately affected by them?
4. Instrumental and Moralistic Conceptions of Criminal Law

We can begin to tackle these two questions by distinguishing two radically different ways of conceptualising criminal law. We might decide, in the end, that a plausible account will have to draw on both kinds of conception; but we can usefully begin by contrasting simple, pure versions of each.

One conception is instrumental. The criminal law is a technique or instrument that can be used to serve various possible ends. We are justified in maintaining a system of criminal law if it is an efficient technique for achieving worthwhile ends; its structure and content should then be determined by asking how it can serve those ends most efficiently.

What worthwhile ends could a system of criminal law serve? We cannot simply say that it should prevent or reduce crime, since without the criminal law there would be no crimes—no conduct would count as criminal. However, a number of plausible goals could be posited, reflecting a range of views both about human goods and about the proper roles and functions of the state. The American Model Penal Code, for instance, declares that:

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 [s. 1.01(1)].

We begin with a set of individual and public interests that merit protection, given their role in human welfare: they can be protected by various methods, including various state activities; a system of criminal law makes its distinctive contribution to their protection by forbidding and thus preventing conduct that threatens substantial harm to them. German criminal law theory posits a similar starting point: a set of individual and collective Rechtsgüter (a Rechtsgut is a good which the law properly recognises as being necessary for social peace or for individual well-being, and as therefore meriting legal protection) which the criminal law protects against conduct that seriously threatens them (see Roxin 2006: 8-47; for critical discussion of the utility of the idea of Rechtsgüter see Wohlers et al 2003). As we will see in § 6, it is not yet clear whether or how individual as distinct from public or collective interests should figure in an account of the protective aims of the criminal law, and some accounts certainly emphasise the collective dimension. Thus on Walker's ‘pragmatic’ account, the criminal law should aim to further the “smooth functioning of society and the preservation of order” (Walker 1980: 18, quoting Devlin 1965: 5)—collective or shared goods which provide essential preconditions for individual flourishing.

Two aspects of such instrumentalist accounts are worth noting here. First, they typically limit the criminal law's concern to serious harms to the specified kinds of interest, that cannot be otherwise prevented: thus the Model Penal Code refers to “substantial harm”, and German theorists argue that criminal law should be used only as a last resort against seriously harmful conduct (see Roxin 2006: 45-7; also, more generally, Husak 2004; Jareborg 2005). This kind of limitation can itself be rationalised in instrumental terms. The criminal law is a blunt and oppressive technique, which impinges seriously on the interests of those who are subjected to its coercive attention: not just those who are convicted and punished, but also those who are caught up in police investigations, or who are tried and acquitted, A consequentialist calculus of costs and benefits is therefore unlikely to favour its use unless it is the only feasible method of preventing quite serious harm.

But, second, the Model Penal Code also limits the criminal law's concern to conduct that “unjustifiably and inexcusably inflicts or threatens substantial harm” (see lso Feinberg 1984: 31-6); and most criminal codes include similar limitations. The ‘unjustifiably’ limit might still be justified instrumentally; we should not want to prevent conduct that justifiably causes harm. Some theorists argue that we can also justify the ‘inexcusably’ limit in instrumental terms (e.g. Braithwaite & Pettit 1990): that the criminal law's goals are not efficiently served by criminalising faultless or excusable conduct. Others, however, ground this limit in a non-instrumental side-constraint on the aim of harm-prevention: a purely instrumentalist theory cannot justify criminalising only culpable conduct; we must instead appeal to a non-instrumentalist demand of justice, that those who lack fault should not be liable to criminal punishment (see Hart 1968: 17-24, 28-53).

What emerges here is a familiar difference between two types of instrumentalist theory (see Braithwaite & Pettit 1990: 26-36). A pure instrumentalist seeks to explain every aspect of a justified system of criminal law in consequentialist terms; in designing a system, we need only ask which doctrines, practices and rules will efficiently serve the goals we have posited. A side-constrained instrumentalist, by contrast, argues that our pursuit of those goals is also constrained by non-consequentialist values—for instance by requirements of justice—which might preclude some practices—for instance the criminalisation of faultless conduct—even if those practices would efficiently serve the system's goals.

For any instrumentalist theory, whether pure or side-constrained, it is an open empirical question whether we should maintain a system of criminal law at all: we should do so only if this is an efficient means to whatever goals the theory posits. For a pure instrumentalist, the proper structure and contents of a system of criminal law also depend on an empirical inquiry into how those goals can be most efficiently served, whilst side-constrained instrumentalists must also attend to whatever non-consequentialist constraints bear on these issues. We cannot pursue the debates between these two kinds of account, but should note one set of questions that must figure in them, about the relation between criminal and moral wrongs.

Some instrumentalists hold that we should criminalise only conduct that is in some way immoral, and should punish only agents who are morally culpable for such conduct: thus, for instance, Braithwaite and Pettit “assume … that only persons who are morally culpable for a prescribed [sic] encroachment upon the dominion of others should be convicted” (1990: 99), whilst the Model Penal Code declares another purpose of the criminal law to be “to safeguard conduct that is without fault from condemnation as criminal” (s. 1.02(1)(c)). Others seem less sure about this. Walker, for instance, sees reason to criminalise conduct that provokes social disorder, even if we would, speaking morally, blame that disorder on the intolerance of others rather than on the moral wrongness of the conduct (1980: 21). As for moral culpability, Hart argues that we should explain excuse doctrines not as aiming to exempt the morally faultless from criminal liability, but as aiming to protect individual freedom by subjecting to liability only those who had a fair opportunity to avoid it (1968: 17-24).

Thus an instrumentalist approach to the justification of criminal law seems to leave it as something of an open question whether the law should criminalise only immoral conduct, or should subject only morally culpable agents to criminal liability. At the other extreme of the spectrum of theories of criminal law, by contrast, we find accounts that make immorality and moral culpability central to the proper concerns of the criminal law.

Criminal law, Stephen notoriously argued, “is in the nature of a persecution of the grosser forms of vice”; conduct is properly criminalised

not only because [it is] dangerous to society, and so ought to be prevented, but also for the sake of gratifying the feeling of hatred—call it revenge, resentment or what you will—which the contemplation of such conduct excites in healthily constituted minds (1873/ 1967: 152).

One could read Stephen as offering what is still an instrumentalist account of criminal law; it is important to satisfy that “feeling of hatred and the desire of vengeance … a regular public and legal manner” (loc. cit.), because otherwise they will find more violent, uncontrolled and socially harmful expres​sion(compare Gardner 1998: 31-32). But he clearly also believed that such feelings and desires were intrinsically appropriate responses to the grosser forms of vice, which deserved to be thus satisfied; and we find a contemporary version of this kind of view in Moore's claim that criminal law should be understood as a functional kind, whose function is to achieve retributive justice by punishing “all and only those who are morally culpable in the doing of some morally wrongful action” (Moore 1997: 35). This is, as it stands, a wholly non-instrumentalist, intrinsicalist account of the proper purpose of the criminal law: it has no purpose beyond itself, beyond the punishment of culpable agents for their immoral conduct; it does not even, apparently, aim to reduce the incidence of such conduct.

Moore offers what looks at first like an extreme and simple version of ‘Legal Moralism’, the view that “all and only moral wrongs should be criminally prohibited” (1997: 662), but in fact the implications of his account are less dramatic than this might suggest. The immorality of a given kind of conduct creates a presumption in favour of criminalising it—it ‘should be criminally prohibited’. However, that presumption can be defeated by other considerations to do with the impact of criminalisation; in particular, a proper regard for individual liberty will dissuade us from actually criminalising much wrongful conduct (see Moore 1997: ch. 18).

There are, of course, other types of Legal Moralism than Moore's. Any version of Legal Moralism claims that the immorality of a given kind of conduct is significantly relevant to the question of whether it should be criminalised. We can then distinguish positive from negative versions. Positive Legal Moralists hold that immorality is a good reason for criminalisation—not necessarily that it creates a presumption in favour of criminalisation, but that it provides a reason that should carry some weight in our deliberations (see Feinberg 1984: 27; 1988: 324). Negative Legal Moralists hold instead that immorality constitutes only a necessary condition for criminalisation: we must not criminalise conduct unless it is immoral, but its immorality does not give us any positive reason to criminalise it. Negative Legal Moralism, like negative retributivism (see Dolinko 1991: 539-43), acts as a side-constraint on our pursuit of the goals that provide our positive reasons for maintaining a system of criminal law, whereas a positive Legal Moralism helps to set those goals. We should note too that a positive Legal Moralist as defined here need not be a negative Legal Moralist: one can believe that immorality provides a good reason for criminalisation whilst also believing that there are other reasons, including reasons for criminalising conduct that is not immoral. (We will also look later at the argument that whilst immorality as such provides no reason for criminalisation, immorality of the right kind does provide a good reason.) Furthermore, even positive Legal Moralists need not think, with Moore, that the reason for criminalising immoral conduct is precisely and only to secure its retributive punishment: she could instead believe, as Feinberg's Legal Moralist does, that we should criminalise it in order to prevent it, and therefore only if criminalising it would be likely to reduce its incidence (see Feinberg 1988: 324).

It might seem that negative Legal Moralism is straightforwardly undermined by the fact that many of the offences defined by a modern criminal law constitute mala prohibita rather than mala in se. Mala in se, as normally understood, are crimes consisting in conduct that is wrong independently of the criminal law—that would have been wrong even had there been no criminal law. Mala prohibita, on the other hand, consist in conduct that is not wrongful independently of the law that prohibits it: if they are wrong, their wrongfulness depends essentially on their illegality. The distinction between mala in se and mala prohibita is neither clear nor uncontroversial, but does point to something important: malicious killing, for instance, is wrong, something that we all have very good reason not to do, independently of law of murder; by contrast, there is nothing wrong with driving from north to south down a narrow street in the absence of a regulation making it a south to north one-way street. However, Legal Moralists can easily justify a category of mala prohibita. If the legislature is justified in creating the kinds of regulation that mala prohibita involve (such as traffic regulations), to serve some aspect of the common good, breaches of such regulations might be moral wrongs (though their weongfulness will need to be shown, not just assumed); that wrongfulness provides, for the Legal Moralist, a reason to criminalise such breaches—to define them not merely as morally neutral breaches of a regulation, but as criminal (see further Green 1997; Duff 2007: chs 4.4, 7.3; and, for criticism, Husak 2005).

Were we faced by a stark choice between an instrumentalist view of the criminal law and a view like Moore's, we might think that some form of instrumentalism has to be right. Even if we think that, once we have a system of criminal law, we must justify criminal punishment in retributivist terms (which is itself controversial), it seems much less plausible to think that we should create and maintain the whole edifice of criminal law simply in order to condemn and punish immoral actions; surely at least a central part of the purpose of a modern system of criminal law must be to protect citizens against various kinds of harm, by preventing kinds of conduct that cause such harm. We must also ask whether positive Legal Moralists are right to believe that every kind of immoral conduct is, in principle, the law's business—that even if in the end the balance of reasons argues against criminalising some kind of immoral conduct, its immorality provided a good reason in favour of its criminalisation. Suppose that I betray a friend by frivolously revealing a secret that she had entrusted to me and that I knew mattered a lot to her: I have done her a grievous moral wrong, and might indeed have fatally damaged our friendship; but are we really to say that such conduct should (ceteris paribus) be criminal, or that its immorality gives us good reason to criminalise it?

We will return to Legal Moralism (in § 6 below), but should not here that there are also some serious questions for instrumentalists. A purely instrumentalist account faces the same questions, the same moral worries, as does any purely consequentialist theory of moral, social or political action: put crudely, the general worry is that any such theory will fail to do justice to individuals and their rights, since it will too easily sanction unjustly sacrificing individuals to the greater social goods that it posits as the justifying aim of our actions. Side-constrained instrumentalists avoid that kind of objection, since the side-constraints that they recognise are precisely intended to rule out such injustice, such infringements of individual rights: but there is a serious question about their conception of criminal law—whether we should see it simply as a technique whose positive justification lies solely in its beneficial effects.

For instrumentalists, whether pure or side-constrained, it is a contingent, empirical issue whether the criminal law is an appropriate institution: it is appropriate if and because it does, as a matter of contingent fact, make an efficient contribution to whatever ends we posited for the state. Now we can agree that this instrumental dimension is crucial to the justification of a system of criminal law: we must surely believe, for instance, that a system which on balance did more harm than good could not be justified. However (leaving aside the question of what is to count as ‘harm’ or ‘good’ for the moment), it does not follow from this that instrumental efficiency is the only positive justifying reason for maintaining a system of criminal law: we could still also believe that such an institution can be justified only if it can be shown to be an intrinsically appropriate way of dealing with and responding to the kinds of conduct that fall within its proper ambit.

The upshot of this section is that we still lack any clear criteria or principles by appeal to which we can try to determine which kinds of conduct should be criminal. However, though this might be frustrating, we should at least by now be clearer about what kinds of claim we must be able to make about kinds of conduct that we want to show are apt for criminalisation. First, we must be able to show that and how they involve wrongdoing: for as we saw in s. 2, the criminal law focuses on wrongs that should be condemned, rather than just on harms that need to be repaired or compensated; and as we saw in s. 5, the criminal law must speak to us of wrongs that we should not commit. Second, we must be able to claim that the wrong is of such a kind that it should concern us all as citizens—we should not leave it to the individual victim to pursue, or not to pursue, a civil case against the wrongdoer. I have not suggested determinate criteria by which we can identify such wrongs, nor do I think that any determinate criteria can be provided; although theorists might yearn to find a single principle, or a single set of principles, by reference to which we could determine the (in principle) proper scope of the criminal law, such yearnings are doomed to be frustrated (for two recent examples, see Dan-Cohen 2002, Ripstein 2006; for critical discussion see Duff 2007: 138-9; Husak 2007). We can, however, identify the main kinds of consideration that should be relevant. Is the wrong one that injures ‘the public’ rather than any individual victim? Is it one that flouts or implicitly denies the core values by which we define ourselves as a polity, and which supposedly underpin our civic relationships? Is it one from which we should be able to expect the protection of our fellow citizens (which is to ask whether it is a wrong from which we should be able to expect to be categorically safe as we go about our normal lives, rather than a kind of wrong that we can be expected to risk on condition that we can seek compensation if we suffer it)? Answers to these questions will be contestable, and will properly emerge only from a collaborative attempt to understand what joins us as citizens and what we owe to each other as citizens—an attempt which will lead to different results in different political communities: but we have made progress if we have at least identified more clearly the questions that we must ask.
7. International Criminal Law

We have focused so far, explicitly or implicitly, on the domestic criminal law of nation states—a criminal law that typically claims jurisdiction only over crimes committed within the state's territory, by and against citizens of or visitors to that state (for some of the complications here, see Hirst 2003; Duff 2007: ch. 2.2). However, recent years have seen also seen significant developments in international criminal law, culminating in the creation of the International Criminal Court in 2002 (see Rome Statute of the International Criminal Court 1998; Cassese 2003: 340-405). Behind a range of particular questions about the proper scope of international criminal law (over what kinds of crime should the ICC have jurisdiction), about the relationship of the ICC to domestic courts and systems of criminal justice, and about the appropriate procedures for international criminal trials, lies a deeper question about the moral authority or legitimacy of any such court: by what moral right does it calim jurisdiction over this range of wrongs and over these wrongdoers? This question becomes particularly acute in relation to crimes whose impact is intra- rather than inter-national. Crimes committed by one state against another, or against another's population (war crimes, the crime of aggression; Rome Statute arts. 5, , are clearly better dealt with by an international court: but why should such a court have jurisdiction over ‘widespread or systematic attack[s]’ committed by a state's officials against its own citizens (Rome Statute art. 7, defining ‘crimes against humanity’)? What could give such a body the moral authority to hold those who commit such wrongs to account—and just what kinds of wrong can it claim as its business?

One answer to these questions appeals to the impersonal demand of retributive justice that those who commit such wrongs should not escape punishment: the ICC acts in the name of justice. Another answer is that the ICC acts in the name and on behalf of the more local polity within and against which the crimes were committed, when the domestic courts will not or cannot act effectively. Another answer takes seriously the idea of ‘crimes against humanity’, and portrays the ICC as acting in the name and on behalf of ‘humanity’: those who commit such wrongs must answer not merely to their particular polities, but to humanity itself, since their crimes ‘deeply shock the conscience of humanity’, and are ‘of concern to the international community as a whole’ (Rome Statute, Preamble). Each of these answers is problematic: a central task for theorists of criminal law is to work towards a clearer understanding of the questions to which such answers are offered (see generally Altman & Wellman, 2004; Luban, 2004; May 2005; Besson & Tasioulas 2008).
8. The Internal Structure of the Criminal Law

Once we have articulated an account of the proper aims and limits of the criminal law, we can tackle various issues about its internal structure—about the general principles and conditions of criminal liability (the so-called ‘general part’ of the criminal law), and about its definitions of specific offences (the ‘special part’). I cannot pursue these issues here, but our approach to them must clearly be based on our account of the proper purposes of the criminal law. Thus if the criminal law should aim to define, to condemn, and to call perpetrators to public account for wrongs whose character and implications are such that they properly count as ‘public’, its definitions of crimes and its principles of liability must be apt to identify such wrongs and the conditions under which agents can be justly condemned for them. In discussing such issues as whether the law should contain offences of ‘strict liability’, which can be committed by those who are not even negligent as to the harm they cause or risk; or what kind of ‘fault elements’ should be required for criminal liability, either in general or for particular crimes; or whether criminal liability should depend solely on the ‘subjective’ character of an action (on what the agent intended to do or believed herself to be doing), or also on its ‘objective’ character (its actual connection with and impact on the world); or whether and how the law should distinguish offences from defences, either substantively or procedurally; or what kinds of excuse or justification the law should recognise, and how they should be defined: we must first ask which doctrines and principles would capture the relevant kinds of wrongdoing and identify the culpable agents of such wrongs. This can only be the first stage of the discussion, since we will need to go on to ask whether those doctrines and principles can satisfy the wide range of other normative and practical constraints that must bear on a system of criminal law and justice, or how they could be adapted so as to satisfy those constraints: but it is where we must start. (For introductions to these issues see Fletcher 1978; Robinson 1997; Tadros 2005; Ashworth 2006; Dressler 2006; Simester & Sullivan 2007; Duff 2007; Gardner 2007.)

Theorists of criminal law must also attend, however, more seriously than they have often attended, to the criminal process that leads from (alleged) crime to punishment, and in particular to the criminal trial. It is through the criminal trial that criminal responsibility and liability are formally assigned, and the norms and doctrines of the substantive criminal law are articulated and applied. We therefore need an account of the proper aims and values of the criminal trial, and of the larger criminal process of which it is part—an account that can then underpin a more adequately grounded critique of our existing criminal processes. Should we, for instance, see the trial as an attempt to establish the truth (but what truth?), albeit an attempt that is constrained by a range of independent principles and rules that aim to protect us against the potentially oppressive and intrusive power of the state? Or should we see it as a process through which alleged wrongdoers are called to answer the charges that thay face and to answer for their crimes if their guilt is proved? What kinds of criminal process are appropriate to a liberal democracy that aims to treat all its members as responsible citizens? (See generally Burns 1999; Ashworth & Redmayne 2005; Duff et al 2007.)
Bibliography

* Altman, A., & Wellman, C.H. (2004), “A Defense of International Criminal Law”, 115 Ethics 35.
* Ashworth, A. J. (2006), Principles of Criminal Law (5th ed.), Oxford: Oxford University Press.
* Ashworth, A. J., & Redmayne, M. (2005), The Criminal Process, Oxford: Oxford University Press.
* Becker, L. (1974), “Criminal Attempts and the Theory of the Law of Crimes,” Philosophy and Public Affairs 3: 262-94.
* Besson, S., & Tasioulas, J. (eds.) (2008), Philosophy of International Law, Oxford: Oxford University Press.
* Bianchi, H. (1994), Justice as Sanctuary: Toward a New System of Crime Control, Bloomington: Indiana University Press.
* Blackstone, W. (1765-9), Commentaries on the Laws of England (4 vols), Oxford: Clarendon Press; available at http://www.yale.edu/lawweb/avalon/bl...e/blacksto.htm.
* Boston University Law Review (1996), “Symposium: The Intersection of Tort and Criminal Law,” 76: 1-373.
* Braithwaite, J. & Pettit, P. (1990), Not Just Deserts, Oxford: Oxford University Press.
* Burns, R. P. (1999), A Theory of the Trial, Princeton, NJ: Princeton University Press.
* Cassese, A (2003), International Criminal Law, Oxford: Oxford University Press.
* Christie, N. (1977), “Conflicts as Property,” British Journal of Criminology 17: 1-15.
* Cotterrell, R. (1995), Law's Community, Oxford: Oxford University Press.
* Dagger, R. (1993), “Playing Fair with Punishment,” Ethics 103: 473.
* Dagger, R. (2008), “Republican Punishment, Consequentialist or Retributive?,” in C. Laborde & J. Maynor (eds.), Republicanism and Political Theory, Indianapolis: Wiley.
* Dan-Cohen, M. (1984), “Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law,” Harvard Law Review 97: 625-77.
* –––. (2002), “ Defending Dignity”, in M. Dan-Cohen, Harmful Thoughts: Essays on Law, Self and Morality, Princeton, NJ: Princeton University Press, 150.
* Dempsey, M. M. (2007), “Toward a Feminist State: What Does ‘Effective’ Prosecution of Domestic Violence Mean?” Modern Law Review 70: 908-35.
* Devlin, P. (1965), The Enforcement of Morals, Oxford: Oxford University Press.
* Dimock, S. (1997), “Retributivism and Trust,” Law and Philosophy 16: 37-62.
* Dolinko, D. (1991), “Some Thoughts about Retributivism,” Ethics 101: 537-59.
* Dressler, J. (2006), Understanding Criminal Law (4th ed.), New York: Lexis
* Dubber, M. D. (2001), “Policing Possession: The War on Crime and the End of Criminal Law”, Journal of Criminal Law and Criminology 91: 829.
* –––. (2005), The Police Power: Patriarchy and the Foundations of American Government, New York: Columbia University Press.
* Duff, R. A. (1986), Trials and Punishments, Cambridge: Cambridge University Press.
* –––. (2001), Punishment, Communication, and Community, New York: Oxford University Press.
* –––. (2007), Answering for Crime: Responsibility and Liability in the Criminal Law, Oxford: Hart Publishing.
* Duff, R. A., Farmer, L., Marshall, S. E., and Tadros, V. (2007), The Trial on Trial (3): Towards a Normative Theory of the Criminal Trial, Oxford: Hart Publishing.
* Dworkin, G. (ed.) (1994), Morality, Harm and the Law, Boulder, Colorado: Westview Press.
* Feinberg, J. (1984), Harm to Others, New York: Oxford University Press.
* –––. (1985), Offense to Others, New York: Oxford University Press.
* –––. (1986), Harm to Self, New York: Oxford University Press.
* –––. (1988), Harmless Wrongdoing, New York: Oxford University Press.
* Fletcher, G. (1978), Rethinking Criminal Law, Boston: Little, Brown.
* Gardner, J. (1998), “Crime: In Proportion and in Perspective,” in A. J. Ashworth & M. Wasik (eds.), Fundamentals of Sentencing Theory, Oxford: Oxford University Press, 31-52.
* –––. (2007), Offences and Defences, Oxford: Oxford University Press.
* Glazebrook, P. R. (1978), “Situational Liability,” in Glazebrook (ed.), Reshaping the Criminal Law, London: Stevens, 108-19.
* Green, S. P. (1997), “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: 1533-1615.
* –––. (2005), Lying, Cheating, and Stealing: A Moral Theory of White-Collar Crime, Oxford: Oxford University Press.
* Harcourt, B.E. (1999), “The Collapse of the Harm Principle”, Journal of Criminal Law and Criminology 90: 109.
* Hart, H. L. A. (1968), Punishment and Responsibility, Oxford: Oxford University Press.
* –––. (1994), The Concept of Law (2nd ed.), Oxford: Oxford University Press.
* Hirst, M. (2003), Jurisdiction and the Ambit of the Criminal Law, Oxford: Oxford University Press.
* Holtug, N. (2002), “The Harm Principle”, Ethical Theory and Moral Practice 5: 357.
* Hughes, G. (1958), “Criminal Omissions,” Yale Law Journal 67: 590-637.
* Hulsman, L. (1986), “Critical Criminology and the Concept of Crime,” Contemporary Crises 10: 63-80.
* Husak, D. (1987), Philosophy of Criminal Law, Totowa, N. J.: Rowman & Littlefield.
* –––. (1995), “The Nature and Justifiability of Nonconsummate Offenses,” Arizona Law Review 37: 151-83.
* –––. 1998, “Does Criminal Liability Require an Act?” in R. A. Duff (ed.), Philosophy and the Criminal Law, Cambridge: Cambridge University Press: 60-100.
* –––. (2004), “The Criminal Law as a Last Resort”, Oxford Journal of Legal Studies 24: 207.
* –––. (2005), “Malum Prohibitum and Retributivism”, in R. A. Duff and S. P. Green (eds.) Defining Crimes: Essays on the Special Part of the Criminal Law, Oxford: Oxford University Press, 65.
* –––. (2007), Overcriminalization, Oxford: Oxford University Press.
* Jareborg, N. (2005), “Criminalization as Last Resort (Ultima Ratio)”, Ohio State Journal of Criminal Law 2: 521.
* Kelman, M. (1981), “Interpretive Construction in the Substantive Criminal Law,” 33 Stanford Law Review: 591-673.
* Kleinig, J. (1978), “Crime and the Concept of Harm,” American Philosophical Quarterly 27: 32-42.
* Lacey, N., Wells, C., & Quick. O. (2003) Reconstructing Criminal Law (3rd ed.). London: LexisNexis.
* Lamond, G. (2007), “What is a Crime?”, Oxford Journal of Legal Studies 27: 609.
* Luban, D (2004), “A Theory of Crimes Against Humanity”, 29 Yale Journal of International Law 124.
* Marshall, S. E., & Duff, R. A. (1998), “Criminalization and Sharing Wrongs,” Canadian Journal of Law & Jurisprudence 11: 7-22.
* May, L. (2005), Crimes Against Humanity, New York: Cambridge University Press.
* Mill, J. S. (1859), On Liberty, London: Parker.
* Moore, M. S. (1993), Act and Crime, Oxford: Oxford University Press.
* –––. (1997), Placing Blame: A Theory of Criminal Law, Oxford: Oxford University Press.
* Murphy, J. G., & Coleman, J. (1990), The Philosophy of Law (2nd ed.), Boulder, Colorado: Westview Press.
* Norrie, A. W. (2001), Crime, Reason and History (2nd ed.), London: Butterworths.
* Pettit, P. (1997), “Republican Theory and Criminal Punishment”, Utilitas 9: 59-79.
* –––. (2002), “Is Criminal Justice Politically Feasible?”, Buffalo Criminal Law Review 5: 427.
* Posner, R. A. (1985), “An Economic Theory of the Criminal Law,” 85 Columbia Law Review: 1193-1231.
* Raz, J. (1986), The Morality of Freedom, Oxford: Oxford University Press.
* –––. (1987), “Autonomy, Toleration, and the Harm Principle”, in R. Gavison (ed.), Issues in Contemporary Legal Philosophy, Oxford: Oxford University Press, 313.
* Rescher, N. (1972), Welfare: The Social Issue in Philosophical Perspective, Pittsburgh: University of Pittsburgh Press.
* Ripstein, A. (1999), Equality, Responsibility and the Law, Cambridge: Cambridge University Press.
* –––. (2006), “Beyond the Harm Principle”, Philosophy and Public Affairs 34: 215.
* Robinson, P. H. (1997), Structure and Function in Criminal Law, Oxford: Oxford University Press.
* Roxin, C. (2006), Strafrecht Allgemeiner Teil (4th ed.), vol. 1, Munich: CH Beck.
* Schonsheck, J. (1994), On Criminalization: An Essay in the Philosophy of the Criminal Law, Dordrecht: Kluwer.
* Simester, A. P., & Sullivan, G. R. (2007), Criminal Law: Theory and Doctrine (3rd ed.), Oxford: Hart Publishing.
* Simester, A. P., & von Hirsch, A. (2006), Incivilities: Regulating Offensive Behaviour, Oxford: Hart Publishing.
* Steiker, C. (1997), “Punishment and Procedure: Punishment Theory and the Criminal-Civil Procedural Divide”, Georgetown Law Journal 85: 775.
* Stephen, J. F. (1873/1967), Liberty, Equality, Fraternity (ed. J. White), Cambridge: Cambridge University Press.
* Tadros, V. (2005), Criminal Responsibility, Oxford: Oxford University Press.
* von Hirsch, A. (1996), “Extending the Harm Principle: ‘Remote’ Harms and Fair Imputation,” in A. P. Simester & A. T. H. Smith (eds.), Harm and Culpability, Oxford: Oxford University Press, 259-76.
* von Hirsch, A., Garland, D., & Wakefield, A. (eds.) (2000), Ethical and Social Perspectives on Situational Crime Prevention, Oxford: Hart Publishing.
* Walker, N. (1980), Punishment, Danger and Stigma, Oxford: Blackwell.
* Weigend, T (1988), “The Legal and Practical Problems Posed by the Difference between Criminal Law and Administrative Penal Law”, Revue Internationale de Droit Pénal 59: 67.
* Williams, B. (1976), Morality, Cambridge: Cambridge University Press.
* Wohlers, W., von Hirsch, A., and Hefendehl, R. (eds.) (2003), Die Rechtsgutstheorie, Baden-Baden: Nomos Verlagsgesellschaft.
* Wolfenden, J. (1957), Report of the Committee on Homo***ual Offences and Prostitution (The Wolfenden Report) (Cmnd 247), London: HMSO.