Agency, blame, and humanity in criminal law theory and practice
Writing in 1969 in a standard text on justifications of punishment, the philosopher Ted Honderich commented that ‘there no longer are defenders of the traditional retributive theory. … At any rate, there are no defenders writing in the usual places’ (1969: 148). In this, across the USA and UK (as well as elsewhere), theory reflected practice.1 The use of indeterminate ‘public protection’ sentences was common, and the Model Penal Code that had been adopted by the American Law Institute in 1962 gave as the general purposes of “the sentencing and treatment of offenders”:
(a) to prevent the commission of offenses;
(b) to promote the correction and rehabilitation of offenders;
(c) to safeguard offenders against excessive, disproportionate or arbitrary punishment;
(d) to give fair warning of the nature of the sentences that may be imposed on conviction of an offense;
(e) to differentiate among offenders with a view to a just individualization in their treatment … (American Law Institute 1962 (1985): 1.02(2)).
Over the following half century, things changed dramatically. Beginning with work in the mid 1970s – including John Kleinig’s Punishment and Desert (1973) and Andrew von Hirsch’s Doing Justice (1976) – by the 1984 revision of Honderich’s text book a postscript was needed dealing with the ‘new college industry [which] turns out theories of retribution’ (1984: 10). Again, theory reflected practice. Indeterminate sentences had given way to more rigid sentencing guidelines, and in many cases to mandatory minimum sentences. The transformation seemed complete when the 2007 revision to the Model Penal Code gave primacy to proportionality in relation to blameworthiness and explicitly relegated the other concerns of sentencing to a subservient role. The general purposes now being:
(i) to render sentences in all cases within a range of severity proportionate to the gravity of the offences, the harms done to crime victims, and the blameworthiness of offenders;
(ii) when reasonably feasible, to achieve offender rehabilitation, general deterrence, incapacitation of dangerous offenders, restoration of crime victims and communities, and the reintegration of offenders into the law-abiding community, provided these goals are pursued within the bounds of proportionality [as specified above] (American Law Institute 2007: 1.02(2), emphasis added).
As we will see, in penal practice this picture of a landscape transformed from a consequentialist/preventive to a retributive criminal justice system was never fully descriptively accurate. However, in penal theory it is possible to delineate the contours of a broadly retributivist understanding of punishment that remains in the ascendancy in the Anglo-American world.
Begin with the proposition that the rationale for the criminal law is to provide for the conviction and punishment of those who culpably do serious public wrongs, the distinctive technique being to declare those wrongs, to provide for the public censure of those who commit them … and to provide for punishment up to a proportionate maximum.
(Ashworth 2011: 129)
As Ashworth notes, the distinctiveness of the criminal law lies – or is often presumed to lie – in its censuring and punishing. Its legitimacy depends on its punishing only those who deserve censure to the degree proportional to the censure that is deserved. Moreover, in some leading retributive thinking – and in the minds of many in the general public – the justification of the system of criminal sanctions is to be found in its overlap with morality. We are justified in punishing those who have committed morally wrong acts (at least those that fall in the subset of morally wrong acts that are also serious public wrongs) to the degree proportional to their moral blameworthiness.
This commitment to a broadly retributivist account of punishment has generated wide agreement on two key themes: that the punishment of a particular individual expresses blame (censure) for the serious public wrong done and that both the blame and the sanction that follows must be proportional to that wrong and to the culpability of the offender where culpability ‘tracks’ (in some sense) the offender’s responsibility. These themes are in turn reflected in two growing literatures. One on what is properly criminal (what are – and, perhaps more important, what are not – the serious public wrongs that ought to be the concern of the criminal law) (Husak 2008; Duff, Farmer et al. 2010; Duff, Farmer et al. 2012). The other on what kinds of persons, with what kinds of capacities, are rightly held responsible, and what kinds of circumstance excuse or reduce culpability (where the literature is now vast).
The broad conclusions that flow from these literatures follow from the key themes: the criminal law ought to be reserved only for serious public wrongs and blame and sanctions reserved only for those offenders whose actions can be properly attributed to their voluntary choices (an agent who has ‘the general capacity to understand and to be guided by reasons … or the general capacity to understand the law’s commands and the consequences for violating them’ and who is not non-culpably irrational such that s/he ‘is unable rationally to comprehend the facts that bear on the morality of his action or is unable rationally to comprehend the applicable moral or legal code that provides the good reason not to breach’) (Morse 2013: 121).
This broadly retributive picture is of course not uncontroversial – and retributivist theories continue to come in a variety of guises2 – but it nevertheless captures something of the spirit of the age in both theory and practice. It also, for those who endorse it, provides the template against which contemporary policies are to be judged. That said, criminal law theory and practice do not map easily onto this retributivist picture for at least three reasons. First, the transformation of the criminal justice landscape was never as complete as some accounts of the rise of retributivism would have it. With respect to the key claims above, second, criminalization and the criminal law have continued to expand in ways that have increased the gap between retributive theory and the phenomenon which it is meant to theorise. Third, the notion of agency and responsibility – the culpable person presumed by retributive theory – has continued to prove problematic. It is worth saying a little about each of these as a way of setting up the debate and motivating the responses to that debate.
Although it is accurate to describe retributive thinking as having the ascendancy since (roughly) the mid 1970s, elements of the other rationales of punishment have not of course completely disappeared in that time. Concern for prevention, reform, rehabilitation, and so on, have continued to inform policy-making and penal practice. For example, §142 of the British Government’s Criminal Justice Act of 2003, lists five ‘purposes of sentencing’:
1. Any court dealing with an offender in respect of his offence must have regard to the following purposes of sentencing—
(a) the punishment of offenders,
(b) the reduction of crime (including its reduction by deterrence),
(c) the reform and rehabilitation of offenders,
(d) the protection of the public, and
(e) the making of reparation by offenders to persons affected by their offences.3
These multiple purposes are reflected in penal practice. Diversion programmes (drug courts and the like) and attempts at treatment and rehabilitation (in particular through Cognitive Behaviour Therapy) have continued and to these have been joined practices of restorative justice. Beyond these familiar rehabilitative measures, there has also been significant growth in preventive policies to control those deemed ‘dangerous’. This in part reflects the rhetorical division promoted by populist politicians and media between the law-abiding ‘us’ and the dangerous ‘them’ – rhetoric that was fuelled by the attacks of 9/11 on New York and 7/7 on London and more generally by the rise of global terrorism – and in part the growth of a ‘risk society’ in which the management of risk becomes the predominant concern (for a classic statement on the ‘risk society’, see Beck 1992; for application to criminal justice see Hudson 2003). These ‘dangerousness’ policies include statutes to increase dramatically the sentences for repeat offenders (most notoriously ‘three strikes’ rules); sex-offender registers and so-called ‘Megan’s Law’ statutes; sentencing guidelines that increase the penalty for those perceived as a future risk to others; and ‘sexual predator’ laws that allow the continued detention of those considered to present a continuing danger to the public on completion of their penal sentence.
As noted above, the retributive model has at its core the idea of censuring those who do moral and legal wrongs in proportion to the harm done and to their blameworthiness in performing those wrongs. However as Vincent Chiao notes, this looks increasingly out of touch with the reality of the law both ‘on the books’ and as practiced ‘on the ground’. As Chiao puts it, ‘if we look at actually existing institutions it becomes quite implausible to describe those institutions as exclusively, or even, for that matter, primarily, devoted to censorious punishing of prejusticial wrongs’ (Chiao: 8; these ideas are discussed further in Chiao Forthcoming, Chapter 2). The basis for Chiao’s assessment lies in a number of empirical observations.
‘On the books’, US Federal Law is estimated to contain over 3,000 distinct offences only a tiny minority of which could possibly be described as serious moral wrongs. Moreover, the figure of 3,000 does not include the thousands of regulatory offences created by governmental agencies of various kinds.
‘On the ground’, arrests for drug offences, property offences, public order offences, and misdemeanours far exceed those for the ‘core’ of serious assault, rape, homicide, and so on. In short, large parts of the criminal justice systems of both the USA and UK resemble more a system of rules for co-operative living than they do a system for censuring and sanctioning those who culpably commit serious moral wrongs.
Finally, there is an increasing gap between the retributive model of the offender – a responsible agent whose voluntary choices determine his culpability – and the target of criminal law. This gap is evident in three ways. First, in the growth of strict liability offences (offences in which at least one element is strict). On the retributive model such offences ought to be outliers or even, as H.L.A. Hart put it, compromises that if not justified by some important policy goal render strict liability ‘an odious and useless departure from proper principles of liability’ (2008: 34). Yet, far from being outliers – or thought ‘odious’ – strict liability is now a commonplace across the criminal law (for a discussion of its prevalence in English and Welsh criminal law see Ashworth and Blake 1996). Second, and as mentioned above, the criminal law and criminal ‘punishments’ are increasingly used not to censure people for their past wrongs, but to control and prevent people from performing future wrongs. This is not merely a matter of the extension of inchoate offences, but includes additional punishment for ‘dangerousness’, sex-offender registers, and the other policies described above. For some of those who fall under these regimes, the accompanying rhetoric is one that is, from the retributive point of view, deeply confused. The ‘sexual predator’, for example, is both a culpable agent who deserves punishment for his past act and a continuing menace who cannot help himself and who thus needs incapacitating in one way or another. Third, and finally, the default view of the person as an agent who acts voluntarily in response to reasons such that it is fair to blame him for his actions is coming under increasing pressure given the realities of crime and punishment and from various disciplines that offer explanations of human behaviour.
With respect to crime and punishment, the sad reality of criminal offending does not reflect the neat abstractions of choosing agents guided by reasons. Instead, the unemployed, the socially and economically disadvantaged, members of minority racial groups, those from communities in which crime is rife and commonplace whose options are few and whose position is often desperate, make up the bulk of those processed by an impersonal criminal justice system that is far removed from the model of a respectful dialogue through which to bring the offender to an awareness of the moral wrong he has done.
With respect to the pressures that arise from causal explanations of human behaviour, of course some psychiatrists and social scientists – and now neuroscientists – have long pressed their claims that human action is caused in ways that render judgements of praise and blame illegitimate so the point should not be pushed too far. Nevertheless, the basic legitimating story on which retributivism depends – that punishment is only of the blameworthy in proportion to their blameworthiness – is one that has been both subject to criticism from the outside (by, for example, neuroscientists; Greene and Cohen 2004) and to a degree accommodated from within (by, for example, developments in defences for ‘battered women’ and so-called ‘cultural defences’) in ways that threaten to render the default position unstable.
It is worth adding two features to the landscape; features that challenge retributive and other theorists alike (albeit in different ways). First, the aim of much of the retributive work done in the 1970s was to criticise, and to find an alternative to, indeterminate sentencing. The goal was to reduce the use of prison (as well as to try to ensure greater equality in the treatment of offenders of different classes and races). However, by promoting proportionality in the language of desert these theorists unwittingly played into the hands of populist politicians who colonised the language of retributivism to promote harsher penalties, mandatory minimum sentences, and other draconian policies (see Matravers 2011). The result, nearly half a century later, is a staggering increase in the prison populations of the USA and UK.4
Second, to repeat the point part made above, most crimes (other than those for which high levels of education or status are required) are carried out by young men who are disproportionately drawn from the ranks of the disadvantaged. Their lives will already have been blighted by limited opportunity, unemployment, and disadvantage, and there is a good chance that the same will apply to their victims. Crime and punishment do not exist in a vacuum sealed from the facts of socio-economic and racial injustice. However, it is also worth noting that most will stop offending – ‘desist’ – as they grow older. Being a ‘criminal’ may well be a social state from which many offenders never escape, but being someone who commits crime, for many individuals, is a transitory state that may have begun in their early teens and that they will grow out of by their mid to late twenties.
The gulf between the retributive theory of the criminal law and the actual state of the criminal law is something that ought to concern retributivist theorists. As Nicola Lacey reminds us, philosophers who are ‘engaged in theorizing social phenomena which have a “real” existence’ bear a special responsibility, insofar as their ‘philosophical account is … in some sense answerable …’ to the contours of that social phenomena (Lacey 2007: 138). Of course, Lacey does not mean that philosophical theories ought to bend uncritically to accommodate what is happening in the world. Many theories – not merely of the criminal law, but for example, of justice – have as an important function to provide ideals not only to which we should aspire, but that allow us to see how far short of ideal is the world we inhabit. What kind of challenge, then, is it that taken as a whole our systems of criminal justice so little resemble retributive theory?
One possibility is as mentioned: that retributive theory correctly articulates the ideal system and provides the basis on which to challenge existing non-retributive practices. This might be done wholesale or in part. Amongst ‘wholesale’ theorists Michael Moore stands out as insisting that ‘we are justified in punishing because and only because offenders deserve it’ (1997: 91, emphasis added). For Moore, then, although crime reduction might be a happy side effect of justified punishment, criminal law practice ought to be brought into line with strict retributivist principles.
For others, there is no need to deny that punishment can have other aims, or effects, than ‘doing justice’. For example, in different ways both Andrew von Hirsch and ‘Limiting Retributivists’ such as Norval Morris and Michael Tonry allow that penal hard treatment can aim at prevention but insist that this can only be done in ways consistent with the demands of justice. In the language of the Model Penal Code 2007, all other goals must be ‘pursued within the bounds of proportionality’ (Morris 1974; Morris and Tonry 1990; von Hirsch 1993).5 Finally, there are those such as Stephen Morse and Paul Robinson for whom the issue is one of the distinct purposes of criminal justice as against preventive justice. For Morse and Robinson, the aim of preventing future harm is an acceptable one, but it is distinct from the aim of punishing past wrongdoing (Morse 1999; Robinson 2001).
The differing responses of the ‘wholesale’ and ‘in part’ retributivists reflect to some degree their willingness to include the prevention of future crimes amongst the overall purposes of the criminal justice system. For Moore, criminal justice exists only to give offenders the punishment that they deserve. Similarly, for Duff, it exists only to censure offenders and to allow them the opportunity and vehicle for penance. Others, like von Hirsch and Morris and Tonry, accept that the overall purpose – in H. L. A. Hart’s famous phrase, ‘the general justifying aim’ (Hart 2008: 4) – of the system of criminal justice includes the prevention of future crimes, but insist that proportionality (understood strictly or within limiting retributivism) is always a side constraint when pursuing this goal. For these theorists, the task is to mediate the potentially different demands of prevention and proportionality and, looking at the landscape as it is, to distinguish between those initiatives that are consistent with their position and those that are not.6
What follows does not engage point by point with retributivist theory with a view to revising or rebutting it. Rather, what is offered is a sketch of an alternative vision of criminal law and punishment. It will be clear in what ways the sketch differs from retributivism (in short, what is offered below is robustly political and institutional rather than moral), but there is no decisive argument in favour of the account or against the retributive one. As this is a relatively unusual way for a philosophical argument to proceed it is worth saying something about what does (and does not) motivate it.
Although the sketch below is offered as a better fit with the realities of criminal justice presented above, it is not an attempt to justify or rationalise current criminal and penal policy. The extensions of criminal law and their disproportionate application to those already disadvantaged are rightly the subject of severe criticism.
That said, partly what motivates the account is a certain frustration with the existing state of the debate. Philosophers have offered sophisticated arguments in good faith for and against retributive and consequentialist accounts of punishment for years and in the meantime criminal law and criminal justice practices have continued to fail to adhere to either paradigm. Politicians and legislatures – often pushed by actual or perceived public opinion – extend the scope of the criminal law and add to penal severity in the name of ‘keeping us safe’, ‘protecting the law-abiding’, and ‘giving offenders what they deserve’, as it suits them (and their electorates).
Aligning retributive arguments with moral theory exacerbates this problem and has, to a large degree, left retributivist penal theorists shouting (sometimes entirely apt) denunciations of penal policy from the side lines (consequentialist theorists can at least engage with the ‘what works?’ agenda on the basis of empirical research). Perhaps this is in part a result of believing that the alternatives are mutually exclusive. In Stephen Morse’s terms, that the alternatives are ‘desert jurisprudence’ or ‘disease jurisprudence’ (or sometimes ‘the justice model’ or ‘the treatment model’) (Morse 2000, 2013) and that any move away from the justice model will result in a slide towards the treatment model. If this were right, then we have good reason to be careful. Being a criminal is not a disease, and it is important that the state treats its (sane adult) citizens as responsible agents. However, as the metaphor of ‘sliding’ suggests, there is conceptual space between ‘desert’ and ‘disease’. Finally, there is the possibility that political philosophy just is a suggestive discipline. That it progresses not by ‘proof’ and ‘refutation’, but by the suggestion that looking at a certain thing in a certain way might shed on it a different and potentially more interesting light. In any event, that is the hope in what follows.
The chapter began with a retributive account in which the criminal law is concerned with offences that involve a moral wrong done by a responsible agent in virtue of which that agent deserves blame in a sense that brings with it deserved sanctions. The demands of agency and responsibility are intimately tied to the idea of deserved sanctions. On this picture it is legitimate to use the power of the criminal law to punish only and because the responsibility of the offender underwrites her deserving the sanction. However, as we have seen, neither criminal practice nor theory fully accords with this picture. In practice, many things are criminalized that are not moral wrongs and the law fails to track responsibility. Moreover, non-desert based, non-proportional practices continue to flourish. These variations from the retributive norm can be deliberate and explicit, as in the case of strict liability with respect to responsibility and therapeutic jurisprudence in the management of offenders, but are often neither as in the case of ‘three-strikes’ laws and sexual predator statutes.
What follows is recognisably a ‘mixed theory’ in that it accords place to both the thought that the whole edifice of criminal justice, law, and punishment must have something to do with reducing future offending and the thought that ‘punishment’ must be ‘of an offender for an offence’ (Hart 2008).
The starting point for many mixed theories is with the values ‘in the mix’ (for example, with prevention and desert). This then structures what follows, which is to deploy philosophical resources to show how these values can both be realised in one scheme. What this threatens to miss is that it is the state that criminalises and punishes. As T.M. Scanlon puts it, ‘Governments owe it to their citizens to affirm their rights by condemning serious rights violations (whether or not this makes those violations less likely to occur). They also owe it to their citizens to discourage violations by condemnation and, insofar as this is necessary, by threatening wrongdoers with forms of hard treatment’ (Scanlon 2013: 103).
Thus, before we can ask about the justifiability of punishing this or that offender to this or that degree, we have to ask what justifies the existence of a set of state rules that have attached to them the threat of sanctions for those who culpably disobey those rules. The values to be affirmed need to be understood in the light of this question as do the particular features of the criminal law and state punishment.
In a liberal democratic state, political power – ‘coercive power backed by the use of sanctions’ – is ‘ultimately the power of the public, that is, the power of free and equal citizens as a collective body’ (Rawls 2005: 136). It is a power that we impose upon ourselves. A heuristic question can then be put, ‘under what circumstances and for what reasons would I, together with others, agree to a set of coercive rules which have attached to them the threat of sanctions for those who disobey?’
Heuristic devices of this kind – particularly those that use the language of ‘contract’ – can arouse suspicion, so it is worth being clear as to its purpose here. The point is simply to try to make the steps in the argument clear; to illustrate how reasoning in a certain way from a certain position gives rise to certain results (different reasons and/or different positions would of course result in different results).7
The circumstances in which the question of punishment arises cannot be entirely abstract. However, we can borrow from Rawls’s account of the circumstances of justice (itself borrowed from Hume). Amongst what Rawls calls the ‘objective circumstances’ are that ‘many individuals coexist together at the same time on a definite geographical territory’; the capacities of these individuals are roughly similar so that ‘no one among them can dominate the rest’; each is ‘vulnerable to attack, and all are subject to having their plans blocked by the united force of others’; and society is characterized by a moderate scarcity such that ‘natural and other resources are not so abundant that schemes of cooperation become superfluous, nor are conditions so harsh that fruitful ventures must inevitably break down’.
Amongst the ‘subjective circumstances of justice’ are that persons have their own ‘plans of life’. These plans are ‘in various ways complementary’ so that living together with others is both possible and advantageous, but nevertheless people have ‘different ends and purposes’, and they ‘make conflicting claims on the natural and social resources available’. Finally, ‘we suffer from various shortcomings of knowledge, thought, and judgment’. Our ‘knowledge is necessarily incomplete, [our] powers of reasoning, memory, and attention are always limited, and [our] judgment is likely to be distorted by anxiety, bias, and a preoccupation with [our] own affairs.’ As Rawls notes, ‘some of these defects spring from moral faults …’, but many are simply features of our imperfect natures (Rawls 1971: 126–27).
Given the circumstances of justice there is a need for co-ordination and regulation. To give a simple example, we all have an interest in getting around quickly and safely; we have roads and cars, so we need a rule for whether we drive on the right or the left as well as many (formal and informal) conventions governing how we indicate to each other our intentions, allow one another to pass, and so on. Our lives are regulated in a wide variety of ways – from rules of etiquette to the criminal law – and of the nature and scope of these we can always ask ‘under what circumstances and for what reasons would I, together with others, agree to them?’8
What circumstances could give rise to the need for a system of criminal law and punishment? First, like any regulative system, there is the ‘declarative’ need identified by Ashworth. The criminal law needs to identify what must not be done (sometimes what must be done). This is so that people can adjust their behaviour as needed so as to live together successfully. In this, the criminal law is no different from, for example, the conventions that regulate discussion in a seminar. If everyone were to speak simultaneously and as a result no-one was able to hear anything other than babble, discussion would not work. By ‘declaring’ that one person speaks at a time, participants know what to do and what not to do, and there is a chance of effective discussion.
In a formal setting like a university seminar, the tutor may attach sanctions to the rules of discussion. For example, that if a participant persistently interrupts, then he will be barred from speaking for 15 minutes. The circumstances that would motivate such a move are of course disputed by retributivists and their opponents (the debate does not disappear – and nor should we expect it to – when we use a heuristic device to shed light on it). A consequentialist will argue that what explains the attaching of a sanction is the risk of interruptions and the need to deter those who are, or would be, motivated to interrupt. Should there be no such risk, there would be no reason to sanction. By contrast, the focus of the retributivist is on the need to sanction the person who interrupts. That person has done something wrong, and there is a need to mark that wrong by sanctioning him.
That said, if there were no great need for seminar discussions – or for conversation generally (perhaps because human beings developed the ability to communicate silently by reading each other’s minds) – we surely would not need to bring the rules of discussion into existence merely so that we could sanction interrupters when necessary. The need to sanction, where it exists, is a consequence of having systems of regulation, which are themselves a consequence of the structures of social interaction, not a reason to bring such systems into existence.
To pinpoint the circumstances that could give rise to the need for a system of criminal law and punishment requires that we identify some characteristics of a criminal, rather than any other, system of regulation. Begin with those on the table: the criminal law is distinctive because of the way in which it sanctions in the quantitative sense that its sanctions are severe and in the qualitative sense that they express a particular form of public censure.
The ‘circumstances of criminal justice’ borrow from the ‘circumstances of justice’. Given that we have reason to live together, we need systems of regulation. Some of those systems will govern activities that are vitally important because, for example, they include aspects of our lives that are essential to our individual plans of life (such as our bodily or sexual integrity) or because they include interactions that are essential to the functioning of the system as a whole (such as rules of property ownership).9 Thus the criminal law declares that citizens ought not to infringe on others’ sexual integrity without consent and that they ought not to steal from one another (and so on). In designating these things as criminal wrongs it adds weight to the declaration.
The grounds on which to add sanctions to the regulations of the criminal law are that there is a risk that its injunctions will not be obeyed. This risk arises in a number of ways. There are those who will disobey in any case. There are those who will only disobey if there is no sanction (elsewhere (Matravers 2016), I have called this group ‘potential direct offenders’). In addition, there are those whose obedience is conditional on their being assured that others are obeying and that they are not the victims of free-riding (‘potential assurance offenders’). Finally, it is worth recalling Rawls’s subjective circumstances of justice. We are all (to greater and lesser degrees) subject to various ‘shortcomings of knowledge, thought, and judgment’ and our reasoning ‘is likely to be distorted by anxiety, bias, and a preoccupation with [our] own affairs.’ Any one of us, then, might disobey the reasonable demands of the criminal law through momentary bad judgement.
Attaching sanctions to the rules mitigates these risks in different ways. For example, there are those who absent sanctions would conceal their incomes so as to avoid paying tax (potential direct offenders). There are those who would not do this if they believed others were complying, but who would do so if they came to believe that tax avoidance was widespread and that they were the victims of others’ free riding. Finally, there are those who might be generally obedient, but whose reasoning is prone to letting them down (who think, for example, that paying tax is an obligation with which one ought to comply, but who then think that it will not ‘really matter’ if in their particular case they do not declare some bit of unexpected income).
If these risks did not exist, and (enough) citizens would obey the criminal law in the absence of any sanctions, then there would be no reason to attach sanctions to the declarations of the criminal law. If the risks do exist and the threat of sanctions is effective, then there is reason to attach sanctions, but of course were the threat of these entirely successful, then they would never need to be used.
At this point, the wholesale retributivist demurs. For him, the rules and their associated sanctions are needed in order to punish those who do (serious public) moral wrongs. That is no part of this picture. The performance of moral wrongs may rightly elicit moral criticism, but criminal regulation and sanctions represent a discrete form of regulation whose rationale lies in the need for regulation if we are to live together successfully.10
However, that still leaves blame. So far, what has been offered is a purely instrumental account of sanctions – as needed to render the system of criminal regulation effective – with no mention of blame. If we return to our heuristic device, under what circumstances and for what reasons would we incorporate blame as an essential element in the system of criminal regulation to which we subject ourselves individually and as a public body?
Consider again the example of a seminar discussion. Imagine that one person (call him Bill) consistently interrupts and generally shows disregard for the conventions that regulate such discussions. Assuming that Bill is an adult and does not suffer from some kind of disorder that renders him unable to control his behaviour or unable to recognise normal social cues, it is likely that this will trigger in the other participants certain ‘reactive attitudes’ such as resentment (Strawson 1962). In these circumstances, the other participants might criticise Bill and challenge him for his behaviour and, at the end of the seminar, they might not extend to him the invitation to join the group in continuing the discussion in the pub. In doing these things, the group may have no intention of making Bill suffer (although of course it is possible that he may feel uncomfortable at being criticized). Rather, they wish to mark their disapproval of his conduct, call him to account for it, and perhaps give him the opportunity to explain or to apologise.11
This example suggests two reasons for the inclusion of blame in the system of criminal regulation. First, there is the possibility that ‘strong feelings produced by … crimes … would remain unassuaged’ and that ‘revenge would be the only source of the indispensable comfort now provided by criminal justice’ (Gross 2012: 2). This argument – that the criminal law is needed to avoid a society plagued by private justice and vendettas – has a long history (dating back to Athena’s taming of the Furies), and to a large degree its importance depends on the empirical matter of what might in fact satisfy our need to blame and to punish (such as it is).
Second, and far more important, is the claim that by engaging with Bill and subjecting him to criticism the group treats him as a responsible agent. They would not, for example, behave the same way had they been interrupted by a crying infant. Of course, this does not mean that the group always has decisive reason to voice their criticism or to exclude Bill from the pub (perhaps it is someone’s birthday and there is a desire not to spoil the atmosphere), but there is a prima facie reason for so-doing, and it would certainly be inappropriate for the group to treat him as if he were a child rather than a responsible adult.
For Strawson, the reactive attitudes and the conduct that goes along with them constitute what it is to treat one another as responsible adults. In a further step, he argues that the withdrawal of good will that accompanies the reactive attitudes can involve ‘modification of the general demand that another should, if possible, be spared suffering’ and a ‘preparedness to acquiesce in that infliction of suffering on the offender which is an essential part of punishment’ (Strawson 1962: 90), but nothing underwrites that claim (cf. Scanlon 2013: 108). The idea that you are appropriately the subject of criticism cannot ground a separate claim that it is appropriate that you be made to suffer (other than the suffering felt as a result of being criticised).
We need not commit to this Strawsonian picture of responsibility to find a place for blame in the account. If there is reason to regulate some aspects of our lives together through the criminal law, that regulation will plausibly need to be backed up by the threat of sanctions. In addition, given the existence of the system of regulation those who culpably violate the rules can be appropriately criticised for so-doing. Of course, insofar as the regulations of the criminal law govern particularly important matters for us as individuals and as members of a society, we should expect their content to overlap extensively with moral regulations, but it is as a criminal not a moral wrongdoer that the offender is called to account.
This leaves of course the question of how to integrate sanctions and censure into a coherent account. Although the appropriateness of being liable to criticism cannot underwrite or ground liability to suffering, it might be the case that the way in which serious criticism is expressed is through the imposition of suffering. If so, the deterrence and censure rationales could contingently come together. Attaching sanctions to criminal rules both serves the consequentialist function of reducing violations of those rules and simultaneously provides the vehicle for expressing society’s criticisms. This neat solution has been pioneered by Andrew von Hirsch (1993) with the important difference that the offender is censured for his moral failings as revealed in his offending. This incorporation of morality puts pressure on the account as the degree of proportional moral censure can often differ from the degree of sanction needed to contribute effectively to crime prevention. By making both aspects of criminal regulation, this pressure can be relieved.
It is worth bringing together the various strands of the argument. The question is ‘under what circumstances and for what reasons would I, together with others, agree to a set of coercive rules which have attached to them the threat of sanctions for those who disobey?’ It is important that this question is understood from an individual’s perspective. That is, in favouring sanctions for reasons of reducing future offending I am not arguing that the ‘general good’ or ‘total utility’ is to be maximised, but rather that each of us has reason to agree to rules that have attached to them the threat of sanctions in circumstances where temptation, the lack of assurance, or mistakes of reason and judgement will give rise to levels of disobedience such as to render social life unstable. In addition, we have reason to attach to this system an idea of blame or censure where appropriate as a means of marking the responsibility of the agent (and possibly of assuaging the reactive attitudes of others).
In one sense that completes the sketch by providing an alternative vision of the grounds of criminal regulation to that given by the retributivist with which the chapter began. However, and although there is no possibility of filling out the account in any great detail, this section pushes the argument a little further in part by returning to the landscape of the criminal law described earlier.
The heuristic method remains the same, but for this section more detail is needed (put another way, having given a general account of the circumstances that give rise to the need for, and the rationale of, criminal regulation, the question arises as to its shape and for this it is not enough merely to know the circumstances of (criminal) justice).
In reflecting on the general shape of criminal regulation, (at least) three additional factors matter.12 First, the criminal law and criminal sanctions are the most powerful weapons in the state’s arsenal (at least with respect to domestic policy). Even at their best they impose significant burdens on the punished. At their worst (as is often the case), criminal sanctions are blunt tools whose impact is not limited to the offender. They often are, as Hyman Gross puts it (2012: 5), ‘a terrible source of human misery’ that can and often do disrupt and ruin the lives of those punished and of their families and dependents.
Second, precisely because of the seriousness of criminal sanctions, and the element of blame that accompanies them, the agreement to have a set of coercive rules of criminal law will be conditional on a set of protections such as due process, the presumption of innocence, a high burden of proof that falls on the accuser and not the accused, and so on. This in turn means that the investigative process itself will often be burdensome and invasive. Together with the costs of sanctions, it will also be expensive. In short, in reflecting on the existence of the criminal law and asking under what circumstances and for what reasons one would endorse it, the case needs to be compelling. Criminal justice is hugely burdensome and expensive for the guilty and innocent alike.
Finally, criminal liability can befall anyone. We are all susceptible to temptation and all capable of mistakes of judgement. If one is lucky, one’s lapses of judgement will lead to nothing (as when one drives over the speed limit, but arrives home without incident) or to no more than a social faux pas and slight embarrassment. In others (as when speeding and hitting a child who has stepped into the road from behind a parked car) it may lead to criminal prosecution and punishment (the effects of) which might last a lifetime.
Given the above, there is reason to endorse a criminal law of a limited range.13 There are many forms of regulation of social behaviour. To fail to behave in accordance with their demands may make one liable to criticism and to certain social costs. Some, like the rules governing corporate finance, are very important and the penalties involved may dwarf those routinely imposed in the criminal law. Nevertheless, there are some individual interests and elements of co-operative living that are sufficiently central to our lives and to living together to make it sensible to give these the imprimatur of ‘crime’ and to threaten as sanctions not merely financial or other penalties, but both censure and potentially severe sanctions such as imprisonment. Precisely what these are will depend on the social meanings of the society in which they are to be applied, but given human nature (and biology) they are likely to include serious attacks on, and endangerments to, the person. Given current socio-economic arrangements, they will also include some aspects of property rights, and so on.14
Given this, the account here joins forces with the retributivist in criticising the current range of offences. Much that is currently criminal – for example, in England the failure to buy a television licence – needs regulating, but does not need to be included in the criminal law. It would be enough to penalise those who fail to buy a television licence (perhaps by increasing its cost by a small percentage for each month during which the person had watched television without a licence) to provide the incentive needed to potential direct offenders and the assurance needed by potential assurance offenders. In the end, the criminal law might be required as a backstop, but then the offence would not be the failure to buy a licence, but the failure (after repeated opportunities) to obey the law.
Under ‘scope’ is included those aspects of criminal liability that are not covered by the cases above. In particular, inchoate offences and criminal complicity (Ashworth and Horder 2009: 6). In many cases, the argument here will again support the retributive critique. Consider the offence (in England and Wales) of ‘publishing a statement that that is likely to be understood as glorifying acts of terrorism, intending to encourage others or reckless as to whether others are encouraged to commit or prepare for such acts’ (Terrorism Act 2006: s1(2); Ashworth and Horder 2009: 41). This has been rightly criticised as it does not require that any encouragement is actually given, any harm (or preparation for harm) done, or that one intends to encourage others (recklessness is sufficient). Moreover, given the general law of attempts, the creation of this offence extends the scope of criminal liability still further by creating an associated attempt offence.
What has gone wrong is not just the extension of the scope of criminal liability, but the thinking behind the offence that implicitly assumes a ‘good guys/bad guys’ or ‘them and us’ view of the criminal law. That is, the assumption is that the enforcement agencies and those who make decisions over prosecution will distinguish between, for example, an academic writing on ‘the war on terror’ and the dangerous terrorist doing the same. The former can be assured that she has nothing to fear while the latter is swept up by the legislation. This is incompatible with the reasoning captured by the heuristic device. Each citizen must endorse (or be able reasonably to endorse) the proposal to institute criminal regulation of a certain shape (and each must be cognisant of the risk of themselves falling foul of criminal regulations).
What of other offences in which the harm is remote or need not have resulted for their being complete? In at least some of these cases, focussing on the fallibility of agents’ knowledge and reasoning opens up a set of reasons that risks being obscured by the retributivist. Consider offences of possession of firearms, knives, or other weapons. A standard and important justification of these is that they ‘promote an environment the enjoyment of which is not fraught with the risk of harm, unless running that risk of harm is integral to an activity with important value’ (Ashworth and Horder 2009: 333). That is right, but what it misses is that an agent considering whether to endorse such rules has an additional reason to do so grounded in recognition that we might all fall foul of temptation and mistakes of judgement. Put crudely, if I am on a night out and I am not carrying a knife, I significantly reduce the risks of a momentary lapse of judgement ruining my life.
This line of reasoning empowers the state to enact legislation designed in part to save me from myself. This raises fears of paternalism, but paternalism imposed by a public power understood as ‘the power of free and equal citizens as a collective body’ need not be objectionable (for example, we might endorse taxation to provide health services and pensions because we know that left to our own devices we tend to discount the future in irrational ways and so are unlikely to make adequate provision for our elderly selves. Similarly, we might endorse ‘cooling off’ periods on certain kinds of financial commitments given that we know our first judgements are not always our best ones).
This argument is not meant as a defence of the recent expansion of the scope of criminal liability and, of course, it is not by itself enough to distinguish the justified from the unjustified elements of that expansion. Rather, it stresses a way of thinking about the criminal law that risks being forgotten in the standard analysis. The criminal law not only regulates all of us, but it threatens all of us, and given our everyday fallibility we have good reason to protect ourselves from falling into its clutches.
The conditions of criminal liability refer to the fault requirements and ultimately to the nature of agency in the criminal law. As we have seen, this is an area that has come under pressure from three directions. First, the expansion of strict liability, which requires no fault. Second, the expansion of preventive legislation designed not to punish for past offences, but to control ‘dangerous’ persons from committing offences in the future. Third, the greater sophistication of explanatory accounts of human behaviour that threaten the idea of the responsible agent whose acts can be unproblematically assigned to his voluntary choices.
In one sense the account given here – which stresses, on the one hand, the political and institutional nature of the criminal law and, on the other, the fallibility of persons’ knowledge and judgement – has little to differentiate it from the retributive account when delineating the conditions of criminal liability. Given the seriousness of criminal sanctions, we have reason to be wary of strict liability. One ought always to have what H. L. A. Hart calls the ‘fair opportunity’ to avoid criminal liability (I have tried to develop an account of preventive justice along these lines in Matravers 2013).
However, that does not necessarily render strict liability ‘odious’ or unjustified. Rule of strict liability might sometimes be justified as ways of putting one on one’s guard that one is entering risky territory. For example, discussing strict liability in relation to an adult engaging in sexual activity with a child under a specified age, Antony Duff comments ‘if part of the rationale for such stipulations is that citizens should not trust themselves to make such substantive judgements as whether their intended sexual partner is mature enough, the child’s age becomes a guiding reason to which the would-be sexual partner should pay careful attention’ (Duff 2007: 259).
In relation to dangerousness: as noted above legislation driven by dangerousness has greatly expanded the reach of the criminal law. Moreover, much of this has been done under the guise of criminal law and punishment. Nothing in the above argument adds or detracts from the retributive critique of this.
There are people who present a continuing significant danger to others. If they can be identified with an acceptable degree of certainty, we have reason to incapacitate them (using the least intrusive measures possible). However, this is not punishment. The grounds and legitimacy of the system of punishment are entirely different, and both are undermined by the pretence that the situation is otherwise (Robinson 2001).
More generally, the account defended here has as one aim that we should remember that criminal law and punishment are expensive, invasive, and often have terrible consequences not only for those who are subject to it, but also for their families and communities. It ought to be a last resort in enabling us to live together given the circumstances of (criminal) justice. The concerns of ‘dangerousness’ detached from the idea of ‘the dangerous person’ – that is, understood as concerns to reduce future serious violations of rights – are genuine, but they are best and most effectively addressed by regulative systems other than the criminal law. In short, and to state the obvious, if our concern is with reducing future harms we have reason to invest in situational crime control and even before that in education and social justice.
It is with respect to blame and agency – to the responsible agent – that the retributive account and the one articulated here come apart in interesting ways. As we have seen, the retributive model stresses the moral culpability of the agent. The offender has performed a moral wrong for which he deserves censure in proportion to his culpability. The pressures that result from making criminal blame a species of moral blame appear in a number of places and underwrite the concerns of those who worry that the legitimacy of the criminal law – which depends on punishing only those who are blameworthy – is threatened by explanatory accounts of human behaviour. Consider two examples that push in slightly different directions.
First, impossible attempts. In a much used example, a committed believer in the power of voodoo pushes pins through the heart of a wax doll of his victim intending to kill the victim and confident that the method chosen will achieve this end. Such a person is surely morally culpable, but there is something odd about deploying the criminal law in a case where the harm is not merely remote, but impossible (for a clear account of this position, see Robinson 2001).
The institutional account, which stresses the regulative purposes of the criminal law in enabling us to live together, suggests an answer in which the moral culpability of the agent is orthogonal to the issue of whether to criminalise. Although the agent clearly embodies homicidal intent, there is no reason to criminalise actions that could not under any circumstances cause harm (for a sophisticated discussion of impossible, and other, attempts see Duff 1996).
Second, consider a case of what I will call ‘no wonder’ criminal activity.
Imagine an eighteen-year-old male gang member who was brought up in a disorganized, broken family living in a dirty, dangerous, disorganized, deprived community. Assume that the gang member is of average or below-average intelligence and does not have much education, but he is not cognitively disabled. Perhaps he is even functionally illiterate. From his pre-teen days, family, school, and church life had little emotional hold on him, but the gang in his neighborhood recruited him. The gang offered him the sense of identity, belonging, structure, meaning, and self-worth that his family and community failed to provide. Starting at an age when he was not a fully responsible moral agent – say, as early as ten or eleven – the gang encouraged him with its emotional leverage and perhaps threats to engage in various forms of antisocial conduct. He complied, and by age eighteen, he is a hard guy whose allegiance is firmly to the gang. Now, the gang asks him to execute a rival gang member. The gang no longer needs to threaten him or in any other way to manipulate him. He is committed to the gang and its projects, and he carries out the request, perhaps even proudly.
(Morse 2000: 148)
As Morse notes, following Watson, our reaction to cases such as this need not be that ‘it had to be’ or ‘it was determined to be’, but rather ‘no wonder’ (cf. Watson 2004: 242). Given the background as described, no doubt some people would have resisted the gang’s demand to execute a rival, but some will not and ‘no wonder’.
Such cases can be accommodated within the retributive picture of the criminal law. Unless gang membership has genuinely undermined the agent’s capacity to appreciate right and wrong, the agent is criminally liable. He is not insane, and he does not face an intolerable hard choice. The issue, though, is not whether this or that agent is properly held responsible within the criminal law, but whether the criminal law (as understood by the retributivist) meets its own legitimacy condition. Is it fair to blame this agent for his conduct and to ignore the story that accounts for how he came to be the person he now is?
To a degree, the retributivist tries to have it both ways. Yes, the agent is responsible and so can be found guilty of murder. But, the sentencing scheme available to the judge ought to allow that the degree of his culpability is recognised in the sentence he receives. Given his background, we might think he is less culpable and so deserving of a lesser sentence (although quite why these kinds of consideration can make such a significant difference to the penalty whilst being completely excluded from the earlier question of criminal liability is left unanswered).15
In what way will the account sketched here differ from this? Clearly, there are good reasons to criminalise murder and good reasons to treat the gang member as an agent (not as a child or a dangerous animal). The difference lies in the nature of the blame communicated to the offender. The example above is a particularly extreme one, but the pressures of life it describes are all too common amongst many of those who regularly fall within the purview of the criminal justice system. Ours is a complicated world scarred by distribute injustice. Against that background, the language and practices of moral blame fit very uneasily. Moreover, they are unnecessary. Our task is to respond to a criminal wrong in a way that characteristically marks the harm done to the victim and accords respect to the offender as an agent. That can be done in ways that resist moral blame and yet do not reduce to the treatment model.
One possibility would be to do without blame entirely. But, as noted above, blame can operate as an important marker of responsibility. When a free and equal citizen co-participant in the project of living together acts in ways that are justifiably criminalised, the public institution of the criminal law holds him to account not only to affirm the rights that he has violated, but as a responsible citizen who is answerable to himself and others. The issue is not whether to blame or not, but how and for what blame is conveyed.
As noted above in the discussion of rude Bill, our immediate reactions to wrongdoing are often hostile, and we feel that they are deserved by the offender in virtue of his wrongdoing. This is what Nicola Lacey and Hanna Pickard characterise as ‘affective blame’ (Lacey and Pickard 2013: 3) and Scanlon as ‘blame as sanction’ (Scanlon 2008: Chapter 4).
There is nothing wrong with affective blame, and there are occasions on which it is the only appropriate response. If one reads, and sees pictures, of Nazi extermination camps it is morally fitting to feel ‘hatred, anger, resentment, indignation, disgust, disapproval, contempt and scorn’ (Lacey and Pickard 2013: 3).
In some cases that will also be the appropriate moral reaction to more ordinary criminality, but other than in exceptional cases it is not the appropriate legal reaction. The state’s role is to mark the violation of rights and to hold the violator to account as a responsible agent for his legal failing. It can do this ‘through the simple judgement’ of blameworthiness, which Lacey and Pickard (2013: 19) characterise as ‘detached blame’ (and Scanlon calls ‘blame as evaluation’).
Separating punishment from the hostile elements of affective blame does not of course resolve the question of whether it can ever be fair to punish (given that we are all products of causes that stretch back to points before we were born). Rather, it speaks, on the one hand, to the appropriate role of political institutions and, on the other, to the awareness we should all have that ‘what separates us’ from those who find themselves subject to the criminal law is ‘not just, as we would like to think, that we behave better and choose more wisely, but also our luck in being the kind of people who respond in these ways’ (Scanlon 1998: 294).
Finally, in adopting a more neutral stance with respect to blame, the opportunity arises to use the attribution of responsibility as a rehabilitative tool. Most offenders wish to escape the treadmill of disadvantage, offending, and punishment – ‘to turn their lives around’16 – and it is in this spirit rather than in a retributive one that punishment should be imposed. In practice, this will require greater use of meaningful community penalties, but also where custodial sentences are necessary, they should allow offenders to learn to read and write (where they cannot do so), to gain skills that might enable them to get jobs on release, and to help them with drug and alcohol addictions, and so on (cf. the report of the then Chief Inspector of Prisons for England and Wales, which was imaginatively titled ‘Doing Time or Using Time’, Great Britain 1993).
This is not to embrace ‘an objectifying stance of treatment’ (Kelly 2013) or a dystopian vision of a state that uses criminal justice to ‘fix’ people. It is rather to endorse a humane system of punishment fashioned in the light of the twin demands of living together and our own fragile natures.
This chapter began with the observation that the dominant model of criminal law theory was not descriptively accurate and has argued that it is also not normatively adequate. Using the heuristic tool of asking under what circumstances and for what reasons each of us has reason to endorse a system of criminal law and punishment – given that we understand ourselves as jointly constituting a society whose public power is simply the power of its members – it has offered an alternative vision of the nature and place of the criminal law in modern pluralistic societies. The vision, like so many visions, is short on detail, but the intention of the chapter was not to provide a blueprint. Rather, its purpose is suggestive. Once we think of the criminal law as one amongst a number of regulative systems that allow us to live together, and detach it from moral judgement, it becomes easier to map the theory onto practice and, in particular, to map the deep imperfections of any system of criminal law onto the deep imperfections in ourselves and our societies. Throughout, then, the focus has been on the criminal law as a necessary political institution and on imperfect citizens whose flaws in both knowledge and judgement might at any time render them liable to its reach.17
1 Throughout the chapter, reference is made to the theory and practice of the criminal law. There is of course no such thing as the criminal law, but only the criminal laws of different jurisdictions (and even within jurisdictions what is criminal might be difficult to determine). The focus in this chapter is on what Hyman Gross calls ‘a composite of convenience that might be called the Anglo-American legal system’ (Gross 2012: vii).
2 Famously, John Cottingham distinguished nine varieties of retributivism in his 1979 paper, which was enough to make him doubt the usefulness of the term (Cottingham 1979).
3 The Sentencing Guidelines Commission subsequently addressed this seeming jumble of purposes by giving priority to a proportionality principle found in §143, which requires that ‘in considering the seriousness of any offence, the court must consider the offender’s culpability in committing the offence and any harm which the offence caused, was intended to cause or might foreseeably have caused’ (Ashworth 2010: 78).
4 Although the political systems in the USA and UK are very different, Nicola Lacey argues that the electoral systems, which promote an adversarial approach in which political parties can become locked in an ‘arms race’ of escalating policy on being ‘tough on crime’, makes them particularly susceptible to this issue (Lacey 2008).
5 There are significant differences between von Hirsch’s ‘proportionality model’ and Morris and Tonry’s ‘limiting retributivism’. Von Hirsch demands a fairly strict proportionality scale that leaves relatively little room for discretion. Limiting retributivists allow for a wider range of ‘not unjust’ sentences within which the sentencer can choose based on other values. From the perspective of the argument here what matters is that both are clear that should the demands of proportionality conflict with other demands, proportionality must triumph (they disagree, of course, about how often such conflicts will arise).
6 For example, habitual offender statutes that significantly increase the sentence for recidivist offenders on the grounds that a prior record is an indicator of future dangerousness are clearly illegitimate. This contrasts with a legitimate policy in which a first offender receives a modest discount that is reduced by each subsequent offence until it runs out (see von Hirsch and Ashworth 2005: 148–55).
7 It is not a criticism of this kind of contract account – although it might be of an overambitious contract theorist – that the outcomes are determined by the choice situation (that is the point) (see Rawls 1971: 41–2, 138).
8 For complicated reasons, I think this question is often better asked in the negative form of ‘under what circumstances and for what reasons have I decisive reason to reject such regulation’, but that is not relevant here (for discussion see Matravers 2000: Chapter 7).
9 Note, these are contingent claims. It is possible to imagine a society in which sexual integrity was not particularly valued by its members, and there are societies in which property ownership is relatively unimportant.
10 I am aware that this is not an argument against the wholesale retributivist position. As noted above, the method here is only to present an alternative vision of criminal law and punishment, not to try to rebut retributivism. That said, in this instance the retributive position strikes me as implausible. We are surely not required to bring into existence ‘the institution of property’ so as ‘to punish thieves’ (to borrow an example from Rawls 1971: 313).
11 In an important contribution to penal theory, Antony Duff has built an account of the criminal trial and punishment as centrally involving censure of this kind. The offender has violated the law and is called to account. Assuming he is fit to answer, he may deny that he did it, or offer an excuse or justification, or he may plead guilty. If he is guilty, he is appropriately censured. As noted above, the difference between Duff’s account and that being sketched here is that for Duff the criminal law and criminal sanctions exist in order to provide the opportunity to censure, and the vehicle for penance for, those who commit serious public moral wrongs (which are, for Duff, the substance of the criminal law).
12 Of course, were one to try to reason to particular features of a given criminal code, far more factors would need to be included.
13 The language of ‘range’, like that of ‘scope’ and ‘conditions’ below is taken from Ashworth and Horder’s Principles of Criminal Law (2009: 5–7, 17).
14 In other times and places, they might have included, for example, attacks on, or endangerment to, personal honour (a matter we would now consider one for private law). That is not to say that ‘anything goes’ and these are entirely matters for social meanings and conventions. Our reasoning must be coherent in light of the best evidence and reasons available.
15 In some jurisdictions the illegitimate confusion of dangerousness and punishment will render him more ‘culpable’ in that his gang membership and enculturation will be aggravating factors (based on predictions of likely future dangerousness).
16 I have borrowed this phrase from Anthony Bottoms, who used it in a discussion of ‘desistance’. Although a common phrase, it seems to me to capture something very important about both the majority of those who are swept by the criminal justice system and how we ought to respond to them.
17 I am grateful to the editors of this volume for inviting me to participate in it and to Jonathan Jacobs in particular for comments on an earlier draft. As will be clear from the text, I am deeply indebted to Andrew Ashworth, Antony Duff and Nicola Lacey from whom I have learned an enormous amount (although they will think ‘clearly not enough’). I am also grateful to the Independent Social Research Foundation for a Mid-Career Fellowship award, which allowed time for the thinking underpinning this piece.
American Law Institute, The. (1962 [1985]) The Model Penal Code: Official Draft and Explanatory Notes, Philadelphia, PA: The American Law Institute.
American Law Institute, The. (2007) Model Penal Code: Sentencing, Tentative Draft No. 1, Philadelphia, PA: The American Law Institute.
Ashworth, A. (2010) Sentencing and Criminal Justice, Cambridge: Cambridge University Press.
Ashworth, A. (2011) ‘Attempts’, in J. Deigh and D. Dolinko (eds.) The Oxford Handbook of Philosophy of Criminal Law, Oxford: Oxford University Press, pp. 125–46.
Ashworth, A. and Blake, M. (1996) ‘The presumption of innocence in English criminal law’, Criminal Law Review, pp. 306–17.
Ashworth, A. and Horder, J. (2009) Principles of Criminal Law, Oxford: Oxford University Press.
Beck, U. (1992) Risk Society: Towards a New Modernity. London: Sage.
Chiao, V. (2014) What is the Criminal Law For? http://www.law.utoronto.ca/utfl_file/count/documents/chiao/WPS%202014-05.pdf.
Chiao, V. (Forthcoming). Criminal Law in the Age of the Administrative State, New York, NY: Oxford University Press.
Cottingham, J. (1979) ‘Varieties of retribution’, Philosophical Quarterly, 29: 238–46.
Duff, R.A. (1996) Criminal Attempts, Oxford: Oxford University Press.
Duff, R.A. (2007) Answering for Crime: Responsibility and Liability in the Criminal Law, Oxford: Hart Publishing.
Duff, R.A., Farmer, L., Marshall, S., Renzo, M. and Tadros, V. (2010) The Boundaries of the Criminal Law, Oxford: Oxford University Press.
Duff, R.A., Farmer, L., Marshall, S., Renzo, M. and Tadros, V. (2012) The Structures of the Criminal Law, Oxford: Oxford University Press.
Great Britain, H.M.P.S. (1993) HMCIP Review of Regimes: Doing Time or Using Time? The Prison Service Response, H.M. Prison Service.
Greene, J. and Cohen, J. (2004) ‘For the law, neuroscience changes nothing and everything’, Philosophical Transactions of the Royal Society London B: Biological Sciences: 1775–85.
Gross, H. (2012) Crime and Punishment: a Concise Moral Critique, Oxford: Oxford University Press.
Hart, H.L.A. (2008) Punishment and Responsibility: Essays in the Philosophy of Law, Oxford: Oxford University Press.
Honderich, T. (1969) Punishment: The Supposed Justifications, London: Hutchinson.
Honderich, T. (1984) Punishment: The Supposed Justifications, Harmondsworth: Penguin.
Hudson, B. (2003) Justice in the Risk Society: Challenging and Re-affirming Justice in Late Modernity, London: Sage.
Husak, D. (2008) Overcriminalization: The Limits of the Criminal Law, Oxford: Oxford University Press.
Kelly, E. (2013) ‘Beyond blame (reply to Fried)’. Retrieved 28/07/2015 from http://www.bostonreview.net/forum/beyond-blame/blame-corrupts-criminal-justice-system.
Kleinig, J. (1973) Punishment and Desert, The Hague: Martinus Nijhoff.
Lacey, N. (2007) Interview. Legal Philosophy: 5 Questions. M. Nielsen, Automatic Press: 125–41.
Lacey, N. (2008) The Prisoners’ Dilemma: Political Economy and Punishment in Contemporary Democracies, Cambridge: Cambridge University Press.
Lacey, N. and Pickard, H. (2013) ‘From the consulting room to the court room? Taking the clinical model of responsibility without blame into the legal realm’, Oxford Journal of Legal Studies, (1): 1–29.
Matravers, M. (2000) Justice and Punishment: The Rationale of Coercion, Oxford: Oxford University Press.
Matravers, M. (2011) ‘Is twenty-first century punishment post-desert?’, in M. Tonry (ed.) Retributivism Has a Past: Has it a Future?, New York, NY: Oxford University Press.
Matravers, M. (2013) ‘On preventive justice’, in A. Ashworth, L. Zedner and P. Tomlin (eds.) Prevention and the Limits of the Criminal Law: Principles and Policies, Oxford: Oxford University Press.
Matravers, M. (2016) ‘Punishment, suffering and justice’, in S. Farrall, B. Goldson, I. Loader and A. Dockley (eds.) Justice and Penal Policy, New York, NY: Routledge.
Moore, M. (1997) Placing Blame: A General Theory of the Criminal Law, Oxford: Oxford University Press.
Morris, N. (1974) The Future of Imprisonment, Chicago, IL: University of Chicago Press.
Morris, N. and Tonry, M. (1990) Between Prison and Probation: Intermediate Punishments in a Rational Sentencing System, New York, NY: Oxford University Press.
Morse, S.J. (1999) ‘Neither desert nor disease’, Legal Theory, 5: 265–309.
Morse, S.J. (2000) ‘Deprivation and desert’, in W.C. Heffernan and J. Kleinig (eds.) From Social Justice to Criminal Justice: Poverty and the Administration of Criminal Law, New York, NY, Oxford: Oxford University Press, pp. 114–60.
Morse, S.J. (2013) ‘Compatibilist criminal law’, in T. Nadelhoffer (ed.) The Future of Punishment, Oxford: Oxford University Press, pp. 107–31.
Rawls, J. (1971) A Theory of Justice, Cambridge, MA: Harvard University Press.
Rawls, J. (2005) Political Liberalism: Expanded Edition, New York, NY: Columbia University Press.
Robinson, P. (2001) ‘Punishing dangerousness: Cloaking preventive detention as criminal justice’, Harvard Law Review, 114(5): 1428–56.
Scanlon, T.M. (1998) What We Owe to Each Other, Cambridge, MA: Harvard University Press.
Scanlon, T.M. (2008) Moral Dimensions: Permissibility, Meaning, Blame, Cambridge, MA: Harvard University Press.
Scanlon, T.M. (2013) ‘Giving desert its due’, Philosophical Explorations, 2: 101–16.
Strawson, P. (1962) ‘Freedom and resentment’, Proceedings of the British Academy, 48: 1–25.
von Hirsch, A. (1976) Doing Justice: The Choice of Punishments: Report of the Committee for the Study of Incarceration, New York, NY: Hill and Wang.
von Hirsch, A. (1993) Censure and Sanctions, Oxford: Oxford University Press.
von Hirsch, A. and Ashworth, A. (2005) Proportionate Sentencing: Exploring the Principles, Oxford, New York:, NY Oxford University Press.
Watson, G. (2004) ‘Responsibility and the limits of evil: reflections on a Strawsonian theme’, in G. Watson (ed.) Agency and Answerability: Selected Essays, Oxford: Oxford University Press, pp. 219–59.