Criminal Law and Philosophy

, Volume 2, Issue 1, pp 59–65 | Cite as

Liberalism and the Changing Character of the Criminal Law: Response to Ashworth and Zedner


Ashworth and Zedner outline seven changing aspects of the criminal law––namely, the greater use of diversion, fixed penalties, summary trials, hybrid processes like ASBOs, strict liability, incentives to plead guilty and preventive orders. They criticise these changes insofar as they diverge from what they call ‘a liberal conception of criminal justice’ which ‘emphasises both the purpose of the criminal law in providing for censure and punishment and the need to respect the autonomy and dignity of individuals in the criminal process’.1 According to this liberal model, criminal law is an institution that responds to responsible individuals, punishing them only for serious wrongs they performed as responsible individuals. Few of the seven changes listed escape criticism when held up to the requirements of that liberal model. The only changes compatible with this model are perhaps (1) the use of diversion for young people and for minor offences and (2) the use of fixed-penalty notices for minor offences. These two, insofar as they involve only minor penalties, are arguably appropriate liberal responses to situations where the criminal law should not have a role because the individuals concerned are not fully responsible or the harms involved are not serious wrongs.

After detailing and assessing these seven changes from their liberal perspective, the authors go on to explain some of the forces driving the changes. They offer three ways in which state functions have developed that, it is held, explain the non-liberal changes to the criminal law. First, the state has become a regulatory state that does not provide services, but instead regulates their provision and direction––and criminal law is here seen as a form of regulation that works by affecting agents’ decisions as self-interested rational actors. Secondly, the state has become a preventive state, that aims to manage the future, rather than simply respond to past wrongdoing. Thirdly, the state has become an authoritarian state, willing to be punitive and retaliatory, to deploy its powers with little restraint.

In the first section of this critical commentary, I shall ask whether there are further explanations for the seven changes that Ashworth and Zedner have outlined, and in particular whether some of the impetus for these changes comes from two aspects of the very liberal conception that the authors favour. My first worry is that, as outlined by Ashworth and Zedner, the liberal conception is insufficiently communitarian: this conception does not highlight the fact that the individual who is tried and perhaps found guilty is one of us, a member of the same community that is conducting the trial and imposing the law. In my view, an additional explanation for the seven changes Ashworth and Zedner have outlined is the move towards a ‘them-and-us’ view of criminal law, where the law is conceived as imposed on people who, by their criminal behaviour, have opted out of our community of law-abiding ordinary people. This kind of ‘them-and-us’ thinking will readily support additional authoritarian punitivism and a regulator’s sense that ‘they’ are just there to be controlled or managed by ‘us’.2

My fear is that this ‘them-and-us’ view is supported by a simple liberalism that focuses on punishing responsible individuals. A view that is closer to the moral truth will accept that whenever a crime is committed, we all bear some communal responsibility for it, because the crime has been committed by one of ‘us’ (the relevant ‘us’ here is at least all citizens, including defendants and convicts). This point is not the controversial causal claim that an individual’s criminal actions will usually be caused, in part, by his or her social circumstances, to which all members of the community contribute. That causal claim might be true, but my point is a conceptual rather than a causal one. The point is that any action of an individual within our community is always in some morally relevant respect also our action, as it is an action done by one of us. By belonging to the community one of whose members performed a criminal action, we cannot avoid this action being collectively ours. This point does not deny that individuals should bear primary responsibility for their actions, and that punishment should normally be targeted at the individual who performed the crime. But the point suggests that we should not be willing to adopt the managerial stance that views crime simply as a problem created by external others, that has been lumped on us to be solved. The willingness to adopt this stance can perhaps be traced to liberalism itself, in its more individualistic forms.

A closely related issue deserves further mention. Someone who takes too individualistic a view might end up thinking that concern for the well-being of the perpetrator and concern for the well-being of the victim and the wider community must necessarily be opposed. But a properly rounded communitarian moral outlook will reject the claim that there is necessarily a tension here. Instead, concerns for the well-being of perpetrator and of victim and community can generate identical requirements: requirements that the perpetrator be tried fairly, publicly, and if found guilty, be subjected to genuinely hard treatment in the form of punishment. In particular, concern for a perpetrator’s well-being will recognize the very great importance for that perpetrator, of punishment including full hard treatment. It is bad in some respect even for the murderer or assaulter herself, to be ‘let off’ punishment. This thesis is generated by a properly communitarian moral outlook, which recognises that an individual’s well-being depends significantly on the well-being of other members of her group, and thereby recognises that an individual’s well-being depends in part on the moral quality of her actions, and on whether the individual has been able appropriately to make amends for actions that wrong others.3 I suggest that an individualistic liberalism tends to overlook this thesis that proportionately hard punishment is good for the perpetrator of crime. For this reason, individualistic liberalism lends support to the authoritarian view that being concerned for the perpetrator, or seeing her as one of us, will necessarily result in implausibly lenient punishments. And it lends support to the managerialist view that if criminals are rational actors, then their good necessarily involves their escaping punishment.

So far, I have focused on liberalism’s individualism as a source of some of the seven changes in criminal law listed by Ashworth and Zedner. A second aspect of liberalism that has arguably driven some of these changes is a contemporary liberal assumption that the enforcement of justice should be state-centred and algorithmic. In his recent book on ethics, David Wiggins accuses Rawls of working with this conception of justice. 4 The idea of the enforcement of justice as state-centred and algorithmic encompasses the twin assumptions that securing justice is overwhelmingly a matter of setting up appropriate state institutions, and that the role of these institutions will be to guide individuals’ decisions, leaving very little space for independent judgement. For example, on this view the most just society would leave very little space for teachers to make independent judgements about how to punish their students; instead, teachers would simply apply a set of state-imposed punishment guidelines that would outline what they had to do in almost any particular situation. Similarly, on this view the most just society would leave very little space for individual citizens to make independent judgements about how to censure anti-social behaviour. Instead, this view invests great confidence in the algorithmic application, by the state, of general principles of justice that have specific implications for almost any relevant situation.

This conception of justice is, I suspect, a further driving force behind some of the changes outlined by Ashworth and Zedner. For instance, the use of Anti-Social Behaviour Orders (ASBOs hereunder), and the extension of strict liability, can be seen in part as attempts to extend state control into areas that were previously left to non-state actors such as individual citizens, teachers, family members. Some such as Wiggins would argue that this over-confidence in the use of the state for enforcing justice is itself a product of liberalism, which can be seen in some aspects of Rawls’s focus on state institutions, and in his focus on a limited number of principles with fairly specific implications.5 It is unclear that Wiggins is correct to see Rawls as the source of this problem, though the problem arguably stems from misconceptions of liberalism that fixate on first identifying and then applying liberal principles. Whether or not Rawls is the culprit, there is certainly a problem here: a decline in trust for the judgement of ordinary citizens and of professionals, and an over-confidence in the state’s ability to ‘micro-manage’ individuals’ behaviour in order to achieve justice. This is based on a failure to recognise that the achievement of justice cannot be algorithmic, but must inevitably give a central role to the independent judgement of ordinary citizens and professionals. We should add this problem to the sources of the seven changes outlined by Ashworth and Zedner.6

I have proposed two additional sources that supplement the three sources Ashworth and Zedner find for recent changes to the criminal law. I have suggested that these two additional sources are, at core, misconceptions (excessively individualistic and algorithmic) of the very liberal conception of criminal law that the authors favour. Liberalism need not be misconceived in these ways: the key liberal commitments to individual liberty, rights and political participation are compatible with recognition of the core role that the community’s good plays in defining the individual’s well-being, and with recognition of the centrality of non-algorithmic individual judgements to the achievement of justice.7 And it is notable that acknowledging the importance of these misconceptions of liberalism would not significantly alter Ashworth and Zedner’s main argument. For a more communitarian, less algorithmic liberal model of criminal law would still yield roughly the same criticisms that the authors offer, concerning the seven changes to the criminal law.

In the second half of this critical commentary, I think further about what should follow from Ashworth and Zedner’s liberal criticisms of the seven changes to the criminal law. Suppose Ashworth and Zedner are correct: then we have fairly strong reasons to restrict the use of diversion to only minor offences or offences committed by young people; and to restrict the use of fixed penalties to only minor offences; and possibly similarly to restrict the shift to summary courts; and to abolish the use of hybrid processes like ASBOs; and greatly to restrict the use of strict liability; and greatly to limit the incentives for guilty pleas; and greatly to limit the use of preventive orders. Suppose we were to reform criminal procedures in this way. The result would be something like the ‘purist’ position that Ashworth and Zedner outline: ‘the criminal law concerns itself with a core of seriously wrong conduct, and [...] full procedural protections [...] are applicable in such cases’.8 In particular, no person would be liable to imprisonment or commensurately hard treatment without having been convicted with full procedural protections. Suppose we decide to adopt this purist position. What, we must ask, should we then do or care about all the crimes or quasi-crimes that were once subject to fixed penalties, or prohibited by ASBOs? What should we do or care about the behaviour that would have been subject to a preventive order? And what should we do or care about cases that would have generated guilty pleas, or would have been convicted under strict liability? In a nutshell, how should the purist cope with the behaviour that cannot be criminalised under the strict liberal model?

One possibility is that this behaviour is unresponsive to the ASBOs, fixed penalties and other measures that liberals would exclude, so taking the liberal purist line would not alter this behaviour because the preventive and hybrid measures etc. had no effect on the incidence of this behaviour. This might be supported by considering the high proportion of ASBOs that are breached.9 But even if this is true, and so the changes to criminal law do not affect the behaviour targeted, we still need to ask what, if anything, we should care or do about this behaviour. And obviously this question is more pressing if ASBOs and fixed penalties etc. actually help reduce such behaviour, but are incompatible with our favoured liberal model.

We can distinguish three routes that the liberal purist might take:

1. Endure the behaviour.

Some types of unpleasant behaviour cannot be justifiably restricted by the state at all, either through civil or criminal law or other state procedures. For example, consider burping in public, or teasing others. Perhaps some of the behaviour at which ASBOs, fixed penalties and preventive orders are targeted comes into this category. For example, ‘anti-social behaviour’ includes abusive language which one might think need not be restricted.10 But clearly not all at which the changes in criminal law are targeted will come into this category.

2. Improve and adjust criminal procedures within the liberal model.

Much of the behaviour at which ASBOs and fixed penalties etc. are targeted is already, and should be, criminal. So why not ensure the resources are there to prevent and punish this behaviour through traditional criminal procedures? This could involve a combination of measures, including additional funding for criminal procedures, and at the same time money-saving measures (e.g. perhaps more legal functions should be performed voluntarily by ‘lay’ participants).

In addition, we could refine our understanding of criminal procedures, adapting these where the liberal model permits it. Ashworth and Zedner suggest at least two areas where this might be possible: the definition of criminal responsibility and the understanding of the law of evidence. Regarding the second area, Ashworth and Zedner consider admitting intercept evidence in court.11 The liberal view on this issue will depend closely on the liberal view of the moral justification for secret service activities; this seems precisely one of the cases which illustrate the authors’ claim that ‘a normative conception of the criminal law cannot be distinguished from a theory of good government’.12

Regarding the first area, Ashworth and Zedner ask whether ‘it is the criminal law’s traditional indicators of blame that are inadequate (intention, knowledge, recklessness), and [whether] on some issues it is morally and socially appropriate to convict and punish without proving one of those indicators because the factual situation was such that the person ought to have known what to do and what not to do?’.13 This latter view might seem to motivate extensions of strict liability of the type that Ashworth and Zedner go on to criticise; but they suggest that exceptionally a strong justification could succeed in supporting extensions of strict liability.14 Certainly liberal views of morality allow that sometimes an agent can be held morally responsible for an act (in the sense of being morally required to apologise or compensate for that act), even if the agent did not know all the relevant facts, the act was not reckless and the wrongdoing was not intentional: ‘if Oliver is caught in a wholly unforeseeable and unprecedentedly large traffic jam, and if he is consequently unable to keep his promise to meet up with Susan at an appointed time, he is morally obligated to apologize to her’.15 We might see this extension of moral responsibility to include actions in which the agent gets non-recklessly ‘caught up’ as akin to my earlier suggestion that a given individual will share some moral responsibility for the actions of members of her group, simply due to their being members of her group. In both cases, the individual bears moral responsibility for actions that fail to satisfy ‘the criminal law’s traditional indicators of blame’. But we should be extremely wary of translating this into the realm of criminal responsibility. People should obviously not be held criminally responsible for everything for which they are morally responsible (among the many good liberal reasons for this are the public aspect of criminal proceedings and the severe punishments that these proceedings can impose). Thus my point is the restricted one that some careful, limited and principled extension of criminal responsibility beyond ‘the criminal law’s traditional indicators’ might be compatible with a liberal moral outlook, and if we are lucky it might pick up some of the most serious wrongs16 that are currently inappropriately tackled by ASBOs or preventive orders etc. but would not be tackled by ‘the criminal law’s traditional indicators of blame’. But I must confess that I am unsure about this; the liberal model should draw the indicators of criminal responsibility much more tightly than those of moral responsibility.

3. In addition to enduring some behaviour, and to improving and adjusting criminal procedures within the liberal model, the further liberal purist response to the issues that have prompted the seven changes outlined by Ashworth and Zedner is to try other non-criminal measures.

This is obvious in outline, but I do not think we have much understanding of the specifics of what is required here. Ashworth and Zedner mention civil law and administrative processes as possible channels for dealing with behaviour that does not merit significant punishment.17 In addition, factors such as education and community involvement seem desperately important to tackling the behaviour for which the criminal law has been extended in the seven ways outlined by Ashworth and Zedner. Especially important, I suggest, is the need for a shift towards more communitarian ways of thinking, a shift necessary in civil service language and cost-benefit analyses, as much as in public discussions. And equally necessary is further work on the ethics of non-state sanctions, and the extent to which non-state actors should be trusted, or morally required, to engage in such sanctions. When and how ought I to chastise other people’s children? When and how ought teachers to do this? When and how ought I to respond to my neighbours’ anti-social behaviour? More generally, what is the appropriate role of non-state actors in securing justice? As cited before, Ashworth and Zedner suggest that ‘a normative conception of the criminal law cannot be distinguished from a theory of good government’; I would add that a full understanding of the appropriate role of the criminal law also requires a complementary normative understanding of non-criminal and non-governmental sanctions. There is much philosophical work to be done here. If Ashworth and Zedner’s critique of the seven changes to criminal law is to be taken seriously, then it is incumbent on us to go on by addressing the question of what if anything to do instead of these changes––and this will require us to step outside considerations of both criminal and civil law, and consider when and how non-state sanctions are justified or required within a broadly liberal moral outlook.18


  1. 1.

    Ashworth and Zedner (2007), subheading ‘Introduction’.

  2. 2.

    This way of thinking will, of course, have impacts in other areas of social policy (e.g. social work, housing, taxation) as well as criminal law.

  3. 3.

    For a picture consonant with this view, see Duff (2001).

  4. 4.

    Wiggins (2006), chapter 10. Kant has also been accused of an excessively algorithmic view of justice and of morality in general; for rebuttals of this charge see, e.g., O’Neill (1989), chapter 8.

  5. 5.

    See Rawls’s restriction of the scope of his principles of justice to ‘the basic structure’, and his famous exposition of just two or (depending on how one counts them) three principles of justice that should be applied to this ‘basic structure’ (Rawls 1971), at, e.g., pp. 7–11, 60–65.

  6. 6.

    For a related discussion, see O’Neill (2002).

  7. 7.

    See, for example, Duff’s liberal communitarianism or Wiggins’s liberal neo-Aristotelianism (Duff 2001; Wiggins 2006).

  8. 8.

    Ashworth and Zedner (2007), subheading ‘Liberal values and the criminal law’.

  9. 9.

    Ashworth and Zedner (2007), subheading ‘Greater use of hybrid civil-criminal processes’.

  10. 10.

    The UK media cites many examples of ASBOs imposed for behaviour which, arguably, should not be restricted by the state at all. See, for example, the discussion at, which mentions ASBOs imposed for ‘criminal damage and throwing eggs and snowballs’. While the former (criminal damage) is independently criminal and could be punished without an ASBO (see §2 below), snowballs should normally be endured by liberals, or perhaps in certain cases tackled by non-state sanctions (see §3 below)––unless the snowballing is so severe that it qualifies as (e.g.) an assault, in which case it will again already be covered by the criminal law.

  11. 11.

    Ashworth and Zedner (2007), subheading ‘Liberal values and the criminal law’, last alinea.

  12. 12.

    Ashworth and Zedner (2007), subheading ‘The authoritarian state’, last alinea.

  13. 13.

    Ashworth and Zedner (2007), subheading ‘Greater use of strict liability’.

  14. 14.

    Ashworth and Zedner (2007), subheading ‘Greater use of strict liability’.

  15. 15.

    Kramer (2005), p. 331. The debates about ‘moral luck’ are obviously relevant here (see, e.g., Williams 1981, chapter 2).

  16. 16.

    Less serious wrongs should not be criminalised; for suggestions on how to handle these, see §3 below.

  17. 17.

    Ashworth and Zedner (2007), subheading ‘Liberal values and the criminal law’.

  18. 18.

    Many thanks to participants at the British Academy’s symposium, Why Criminal Law? (13 January 2007). Special thanks are owed to Andrew Ashworth and Lucia Zedner for their excellent paper, and to Antony Duff and especially to Claire Grant, for organising the symposium and for very useful comments.


  1. Ashworth, A. & Zedner, L. (2007). Defending the criminal law: Reflections on the changing character of crime, procedure, and sanctions. Criminal Law and Philosophy, doi: 10.1007/s11572-007-9033-2.
  2. Duff, R. A. (2001). Punishment, communication and community. Oxford: Oxford University Press.Google Scholar
  3. Kramer, M. H. (2005). Moral rights and the limits of the ought-implies-can principle: why impeccable precautions are no excuse. Inquiry, 48(4), 307–355.Google Scholar
  4. O’Neill, O. (1989). Constructions of reason: Explorations of Kant’s practical philosophy. Cambridge: Cambridge University Press.Google Scholar
  5. O’Neill, O. (2002). A question of trust. Cambridge: Cambridge University Press.Google Scholar
  6. Rawls, J. (1971). A theory of justice. Oxford: Oxford University Press.Google Scholar
  7. Wiggins, D. (2006). Ethics: Twelve lectures on the philosophy of morality. Penguin.Google Scholar
  8. Williams, B. (1981). Moral luck. Cambridge: Cambridge University Press.Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2007

Authors and Affiliations

  1. 1.Department of PhilosophyUniversity of StirlingStirlingScotland

Personalised recommendations