Duff's Communicative Theory of Punishment

(1)

Department of Law Fall Term 2019

Master’s Thesis in Legal Theory 30 Credits

Duff’s communicative theory of punishment

Author: Kristian Enblom Supervisor: Joel Samuelsson

(2)

(3)

Table of Contents

1 Introduction... 1

1.1 Background... 1

1.2 The place/concern of the debate ... 2

1.3 The traditional justification of punishment ... 3

1.4 Non-consequentialist theories ... 5

1.4.1 Crime and unfair advantage ... 5

1.4.2 The annulment theory ... 6

2 Duff’s communicative theory ... 8

2.1 Liberal-communitarianism ... 8

2.2 Punishment as communication ... 14

2.2.1 Punishment as deserved censure ... 14

2.2.2 Repentance, reform and reconciliation ... 18

2.3 An example: combining probation and community service ... 21

2.4 Imprisonment ... 23

2.5 Liberal-Communitarianism and the communicative theory of punishment ... 25

2.5.1 Communitarian virtues ... 28

3 Discussion ... 29

3.1 Introduction... 29

3.2 Duff’s view of the offender ... 32

3.3 Punishment as recognition ... 35

Sources ... 41

(4)

1

1 Introduction

1.1 Background

How do we justify punishment? In almost all societies there is the institution of punishment, which is brought about as a consequence for the commitment of certain acts. It mostly takes the form of imprisonment, probation or monetary fines, but it always involves some element of unpleasantness or even suffering for the offender.

Typically, punishment is not such that it simply restores the state of affairs to that before the crime either, as might be the case when someone replaces a thing they accidentally broke. This is simply not possible for most crimes, for example, an assault cannot be undone no matter how much the offender pays the victim. Furthermore, punishment is often thought of as something “extra”, added on top of any material restoration owed to the victim. If that is the case, what makes punishment justified?

Normally we consider the infliction of suffering to be wrong, and it is not obvious how the fact of the offender’s previous wrong makes punishment right. It is this question that so-called theories of punishment try to answer.

The question of what justifies punishment is not a legal one, it is not enough that the institution of punishment has all the required legal support, the question is what the whole system rests on - our reasons for it and whether these reasons can be

considered a satisfactory justification for something that would otherwise be

considered unjust. For a system of punishment gives rise to great costs, both human and economical. It is not enough that the institution exists and has existed for all of history. (A similar argument might have been given in support of slavery, for example.) Our answer, if we find that punishment can be justified at all, should also guide the shaping of our justice system, from the relevant factors in sentencing to the form and harshness of punishment itself. For example, if we take rehabilitation to be the primary goal of punishment, we should not impose punishments that are counter- productive to this aim, perhaps by being too long or too harsh. But on a more punitive view where the goal of punishment is to give the offender “what he deserves”, harsh and long prison sentences might be considered appropriate.

(5)

2

The question of what justifies punishment has seen many different answers through history. Within the Western philosophical tradition, they are typically divided into two broad categories, consequentialist theories and non-consequentialist theories.

Consequentialist theories try to justify punishment by the good consequences it is thought to bring about, like the deterrence of crime, its reformative or rehabilitative effects and the incapacitation of criminals who might otherwise offend if left

undisturbed. These aspects are all part of a utilitarian justification of punishment. Non- consequentialist theories differ more among themselves, but they typically tie the justification of punishment to the wrongdoing itself and not the consequences which punishment may bring about. For example, that punishment serves as a sort of debt repayment from the offender to society, or that some kinds of wrongdoing simply deserves to be punished.

In this paper I will consider the recent and influential theory by Antony Duff called the communicative theory of punishment, presented in “Punishment, Communication, and Community”¹. I will present my interpretation of Duff’s theory and evaluate it as best I can. I will argue that within Duff’s framework there is room for another kind of

justification of punishment that Duff does not consider, one that is victim-focused unlike Duff’s account which is focused on the offender. The purpose of this paper is to give a critical examination of the theory, and if possible suggest an alternative ground for the justification of punishment that hasn’t been previously considered. Before I get to Duff’s theory I will give some more background surrounding the debate around punishment, as to give a better idea of where Duff’s theory is located in relation to other theories, and what questions it tries to answer.

1.2 The place/concern of the debate

While punishment might be administered by parents, teachers or umpires, the debate concerns state punishment within the law, like fines or imprisonment. What’s more, we are interested in the justification for the institution of punishment, as separate from a particular act of punishment. A specific act of punishment may be unjustified even when the institution is.

1 A. Duff, Punishment, Communication, and Community, Oxford: Oxford University Press, 2001

(6)

3

Since the justification of punishment concerns state punishment, the question is nested within a political theory of the proper role of the state. Some justifications won’t be compatible with some other conceptions of society. For example, a

justification of punishment on the ground that it inspires fear and obedience among citizens might be compatible with a monarchy or a dictatorship, but not a liberal democratic society. Some of the strongest criticism of consequentialist theories build on the view that it treats citizens merely as objects to be coerced or manipulated into obedience.² For the purpose of this paper I will assume a liberal political theory (as does Duff), as large parts of the debate hinges on this perspective. While assuming a certain political perspective limits the applicability of any theory that follows, it is also necessary. It is not possible to answer what justifies state punishment independently of an idea of the state.

A theory of punishment also rests upon a conception of crime. While a theory of punishment tells us why we are justified in punishing certain wrongdoings, it does not tell us what those wrongdoings are. For that we need an overall moral theory,

combined with a theory of the proper ends of the state. For example, while most in the Western world would agree that cheating on your partner is wrong, they would also agree that it is outside the proper interest of the state and that it has no business policing such matters. Not so with murder, which is both considered wrong and within the interest of the state, i.e. a proper crime. This background theory is largely separate from the one of punishment however, and different moral theories ought for the most part be interchangeable with respect to a theory of punishment.

1.3 The traditional justification of punishment

Punishment is, and has been, thought of as justified primarily on utilitarian grounds – Not just by philosophers but politicians and law-makers as well.³ The positive effects of punishment are thought to be the rehabilitation of the offender, the deterrent effect of punishment on crime and the incapacitation of the offender. As long as the net benefits of these effects outweigh the suffering caused by punishment it is justified on

2 A. Duff and D. Garland, A Reader on Punishment, Oxford: Oxford Univ. Press, 1994, p. 10-12.

3 A. Duff and D. Garland, A Reader on Punishment, Oxford: Oxford Univ. Press, 1994, p. 8-12, 16-20.

(7)

4

utilitarian grounds. One of the biggest objections to the utilitarian theory is that it would justify the imprisonment of innocent people. A famous example runs as

follows⁴; A man in a town consisting of two segregated social groups commits a crime towards a member of the other group. Because of the tension between these groups violent riots are likely to break out unless the criminal is apprehended, however, he is nowhere to be found. In this situation it would, on utilitarian grounds, be justified for the police to frame and punish an innocent man for the crime as a means of

preventing the riots from happening. Variations of this example exist, and they all purport to show that utilitarianism would justify the punishment of people who are not guilty (in certain cases). The argument is that this is obviously wrong, hence the utilitarian theory can’t be right. One reply to this problem is that it lies in the very definition of “punishment” that it can only be dealt to a guilty person. An innocent person cannot properly be said to be “punished” for something he or she did not commit.⁵ This does not solve the problem however, even if we define “punishment” as something that can only be administered to a guilty person the objection to the

utilitarian theory still stands. It would still justify the administration of punishment-like treatment of innocent persons, only we would have to call it something else. Since the problem is a moral one and not semantic, nothing is solved by restricting the definition of “punishment” to only be applicable to guilty people.

Another reply to the objection that utilitarianism supports the punishment of innocent people is to question the validity, and relevance, of examples like the one above.

Utilitarians can argue that, in the real world, punishing innocent people produce worse results than only punishing guilty people. These unlikely scenarios are irrelevant to the justification of punishment in our actual society, since they virtually never occur.⁶ But such a reply seems to miss the point of positing these far-fetched thought

experiments. By imagining these scenarios, we can better judge whether our supposed principles hold in our current situation as well. The question at hand is, on what

grounds do we justify punishment? For the utilitarian the only moral principle is to

4 From H. J. McCloskey, A non‐utilitarian approach to punishment, Inquiry, 8:1-4, 1965, p. 249-263.

5 See A. Quinton, “On Punishment”, Analysis, 14(6), 1964, p. 133-142.

6 See C. L. Ten, Crime, Guilt and Punishment: A Philosophical Introduction, Oxford: Clarendon, 1987, p.

18-32 for a discussion on the relevance of unlikely thought-experiments.

(8)

5

produce the best outcomes. Even if it is almost always negative to punish innocent people such a principle is secondary, or derivative, from the first principle. There will always be possible situations where the punishment of someone innocent produces positive effects (even if they are an exception), and in these situations utilitarians would have to think that it is right. If in some situations utilitarianism recommends us to punish innocent people, and we judge this to be unacceptable, utilitarianism can’t be right. It seems as if a notion of guilt is fundamental as to who may be justifiably punished, and that is a distinctly non-consequentialist principle.

1.4 Non-consequentialist theories

Non-consequentialist theories, also called retributive theories, are more numerous and varied than the consequentialist ones. But they all see punishment as an inherently appropriate response to criminal wrongdoing. Often they involve the notion of “just desert”, that is, that punishment is deserved by the criminal simply because of his or her wrongdoing. The difficulty lies in explaining that connection. Here I will briefly present two retributive theories, for a more thorough overview and critique see Ted Honderich’s book “Punishment, the supposed justifications revisited”.⁷

1.4.1 Crime and unfair advantage

Several writers have proposed a theory of punishment as a means of removing an unfair advantage gained by the criminal.⁸ The law confers benefits to all members of society by protecting certain areas of life. These benefits are only possible because law-abiding citizens choose to abstain from invading these areas. By breaking the law the criminal neglects to do his or her part in upholding this social contract, while continuing to enjoy the benefits that come from it. Thus, the criminal gains an unfair advantage over the law-abiding citizen, one that comes at their expense. Punishment, it is suggested, removes this unfair advantage by imposing a burden on the offender thus canceling out the advantage gained by the offender.

7 T. Honderich, Punishment: The Supposed Justifications Revisited, rev. ed., London: Pluto Press, 2006

8 See J. Murphy, “Marxism and Retribution”, Philosophy & Public Affairs, vol. 2, no. 3, 1973, p. 217–243.

and H. Morris, “Persons and Punishment”, The Monist, vol. 52, no. 4, 1968, p. 475–501.

(9)

6

While this theory provides a non-consequentialist reason for punishment it does not explain why punishment should be in the form of fines or imprisonment – all it requires is that some sort of disadvantage be imposed on the offender. Another potential problem lies in the way the theory characterizes the punishment-deserving element in breaking the law, an objection Murphy himself (being an initial advocate of the unfair-advantage theory) finds convincing.⁹ While it might seem correct for tax- evasion for example, the wrong in an assault does not lie in the offender taking an unfair advantage over all non-assaulting citizens but in the assault itself. The theory establishes a connection between wrongdoing and just punishment, but it does not seem to give a complete picture of why some wrongs deserve punishment (even if it might give a partial answer).

1.4.2 The annulment theory

Another interesting theory comes from B. Cooper, who provides an interpretation of Hegel’s justification of punishment.¹⁰ According to Hegel punishment “annuls crime”.

On the face of it this seems wrong, punishment does not annul crime in the sense that it makes the crime undone – nothing could undo a murder for example. But Cooper argues that we should understand the annulment claim in a different way. According to Hegel a criminal act constitutes an implicit denial of the right which the criminal code in question establishes, for example the right to one’s life. To leave the crime unpunished would be a tacit admittance that this denial was in fact correct, while punishment asserts that the right exists – it annuls the denial of the right.

This theory rests on the view that legal rights are so-called performatees, that is, their existence rests on the actions of people. For example, the existence of a social

convention is made by the fact that people follow said convention. If a substantial percentage of people didn’t follow it the convention could not be said to exist in the first place. Similarly, Cooper argues that a “right” cannot be said to exist in any real way unless those who break them are apprehended and punished. That they exist “on

9 J. Murphy, “Legal Moralism and Retribution Revisited”, Criminal Law and Philosophy, vol. 1, no. 1, 2007, p. 14.

10 D. Cooper, “Hegel’s Theory of Punishment”, in Z. A. Pelcynski (ed), Hegel’s Political Philosophy:

Problems and Perspectives, Cambridge: Cambridge University Press, 1971

(10)

7

paper” is not enough, they must be forcibly upheld. The theory is somewhat limited in scope since many laws don’t take the form of conferring rights to some individual or entity (for example laws against speeding). But one could drop the talk of “rights” to broaden the applicability of the theory further. Instead, one could argue that legal rules in general need some form of sanction to exist, for otherwise they would be mere guidelines that people could break at their own discretion. A rule without

sanction is like having no rule at all. The result of the theory is to tie the justification of punishment to the justification of the legal rule. If the rule is justified then so is

punishment, for the rule could not exist without it.

One objection to this theory is that it does not follow from it that punishment as we know it is justified, only that some sort of negative consequence is imposed on the criminal. Why, for example, could it not take the form of a public denunciation made by the courts? Cooper replies by making some concession to the utilitarian theory, arguing that such responses to crime would not be as effective in deterring criminals.¹¹ It seems the theory needs to make some sort of reference to effectiveness, or

alternatively, to the severity of the sanction. For an ineffective or light sanction does little more to establish a rule than no sanction at all.

Another critique would be that Cooper avoids answering the actual question. Even if he succeeds in tying the justification of punishment to the justification of legal rules, we are left wondering about the justification of legal rules. On the face of it we might think it’s obvious that a rule against murder is justified. But mustn’t we ask the same sort of questions as before? That is, on what grounds are legal rules (or rights) justified now that they necessarily involve punishment? Are they justified even if they do not produce any positive effects and involve the infliction of suffering on offenders, on what grounds? In a way we are back at square one.

Both the theory of unfair advantage and Coopers theory that punishment establishes rights contains elements that can be found in Duff’s communicative theory. From the theory of unfair advantage the idea that punishment can serve as a sort of debt repayment by the offender, and from Cooper the importance of the actions of the

11 D. Cooper, “Hegel’s Theory of Punishment”, p. 166-167.

(11)

8

larger community in responding to crime. Regarding its classification, Duff’s theory is located in the non-consequentialist camp, even though it ties the justification of punishment to some future goals (and not solely on the past wrong).

2 Duff’s communicative theory

2.1 Liberal-communitarianism

Duff begins his account with some political theory, something he thinks is lacking in many other theories of punishment. The relevance of a political theory lies in the constraints it sets on the kind of justification we can give for punishment. As we’ve seen, some justifications are not compatible with a modern liberal democracy, the political system which Duff assumes for his theory. To properly understand Duff and where he is coming from, we need some understanding of the political theory as well.

Duff calls himself a liberal-communitarian, which he thinks of as a sort of hybrid between liberalism and communitarianism.¹² Liberalism, like any political theory, sets out to legitimize the state and to define its proper ends, and the rights and obligations people have between themselves and between the state. It aims to answer, for

example, why we as citizens of a society are obliged to follow its laws even though we have never explicitly agreed to them. It also aims to answer if, why and to what extent the state can use force against its citizens, or lay claim to their time and energy (for example through political or military service). Liberal theorizing often begins with the idea of the social contract, which is used to ground the legitimacy of the theory. We begin by imagining individuals in a sort of “pre-social” state, as individuals who don’t live in a society or associate with any group. Realizing the benefits that would come from co-operating, they would decide that banding together was in their best interest.

But to create a community they would need to agree to some set of rules regarding their rights and obligations towards each other, and how the community should be structured. The social contract is meant to represent the agreement which these individuals would reach. The social contract is not thought of as an actual contract

12 A. Duff, Punishment, Communication, and Community, p. 35.

(12)

9

which citizens enter into upon birth, but a hypothetical contract (although some would claim it is implicitly agreed to). People would consent to it under the conditions

imagined, and this hypothetical agreement grounds the legitimacy and claim of the liberal state over its citizens.

So what would these people decide on? Duff identifies a couple of values and principles which are central to any liberal account.

“Central to any version of liberalism is an insistence on the moral standing and rights of individual agents: as agents who are, or have the potential to be, autonomous—

determining their own actions in the light of their own self determined ends and conceptions of the good; as agents who should be left free to determine those ends, goods, and actions—who should be allowed and perhaps helped to develop and to exercise that capacity for autonomy; as agents who should be allowed an extensive realm of privacy, of freedom from the uninvited interference of other people or the state, in which to exercise that freedom and autonomy; and as agents who should not simply be sacrificed for some larger social good—whose individual rights "trump" the consequentialist claims of social goods.”¹³

These central values, autonomy, freedom and privacy (and the principle of the individual right over the social good) all set some limits on the justification a state could give to punishment, and the interest it could take in its citizens lives. To punish criminals as a way of deterring crime is to use them “as means” for the social good of crime prevention. Neither should the state seek to “reform” criminals as that is to treat them as mere objects to be remolded, not as autonomous agents who are free to make up their own mind about what is right or wrong. This does not mean that all attempts by the state or the community to influence an offender’s conduct is illegitimate, as we will see later. Rather, this objection targets the kind of forceful reform which is often implied in consequentialist theories, where criminals are something to be reformed rather than being given the opportunity for reform.

Another objection to justifying punishment by reform stems from the right to privacy, and this objection will also be relevant against Duff’s own theory. In a liberal society

(13)

10

there will be some areas of life that count as private, areas which the state should not concern itself with. Such areas are, for example, people’s sexual orientation, religious or political views or how they live their lives (as long as they don’t harm or threaten other citizens rights). But it also covers people’s moral character. The state should not inquire into people’s moral virtues, or work to change them. It is an area where the state has no legitimate authority or interest. That this area is private rests not just on the contention that it should be private, but the stronger point that the state lacks the appropriate standing to make someone answer for their supposed moral shortcoming.

As an analogy, suppose someone comes up to me and chides me for not donating enough to charity. This person might make a compelling argument for why I should donate to charity, even that it would be immoral for me not to do so. Still, I have no obligation to explain or justify my behavior to this person. It is simply not their business, and I don’t answer to them in this regard. So too with the state. Someone can be called to answer for such actions that harm or threaten to harm other people, but not for being a bad person. The problem with reform, from a liberal perspective, is that it does involve such an interest in people’s moral character. It aims to “make its citizens morally better”.¹⁴ This objection is also one that is relevant for Duff’s theory of punishment, and something I will return to in section 2.5.

Much can be said about the idea of the social contract, and the values that liberals claim such a contract would contain – Whether these really are the values that “pre- social” individuals would agree to, and what exactly they mean. They are by no means clear or uncontroversial, and we can find plenty of exceptions to them in our societies.

It is also important to keep in mind that they are ideals, not descriptions or our current societies, and that they are part of liberal political theory which is not the only political theory out there. Nonetheless Duff subscribes to these values and takes these

objections seriously. But he is also what is called a “communitarian”, a political theory which pose some other perspectives on the state and its limits and obligations.

Communitarianism gives rise to some further concerns or objections where liberalism says nothing, and pose some other duties for the community.

14 A. Duff, Punishment, Communication, and Community, p. 36-37.

(14)

11

It is hard to compare liberalism with communitarianism, but a good starting point is perhaps the difference in their grounding of the community and political obligation.

Whereas liberalism grounds membership in a community through the hypothetical consent of its members, communitarianism sees membership more as a given.

Membership in a community arises simply by being born in to it, along with the rights and obligations that come with. As an analogy we could think of the family. Quite plausibly, being a member in a family involves some obligations towards one’s family members, which arise simply by being born in to it (or by having children oneself). For example, it seems we are obliged to provide food and shelter for our children, along with any other material needs, and to sacrifice some of our own time and energy in their favor. It would be wrong to think that these family obligations arise through something like a “family contract”, something that I could distance myself from and be bound to only insofar as it is something which I would agree to. We are not free to give up these obligations in this way, it seems we are bound by them in virtue of the special kind of relationship alone. According to the communitarian the same holds to some extent for the larger community we are born in to. The rights and obligations which a community confers to its members, and the community itself, is a given between its members and not based on something like a contract between them.

Why does this difference between liberalism and communitarianism matter?

According to Duff the difference in grounding of the political theory leads to a difference in the kind of questions liberals and communitarians ask regarding our normative theorizing, and a difference in which areas or aspects of society these theories focus on.¹⁵

The liberal perspective starts with individuals and asks what they, as individuals, should agree upon. Liberals have to start at this point since, on this view, a society is only legitimate insofar as it begins with a sort of contractual consent of otherwise free individuals. The communitarian however, thinks that we are already part of the

community that we happen to be born in to. The normative question we should ask is not how we as individuals should enter into community, but how we as a community

(15)

12

should live. This leads to a difference in the sort of appeals and considerations communitarians make, where community enters as a point of consideration. Most importantly for Duff, considerations surrounding social inclusion and exclusion: A community should (has a duty to) treat all its members as full members of the community – it should strive to be inclusive and to avoid being exclusive.

What do these terms mean? Duff lists for common areas of focus for communitarians where the community can succeed or fail in being inclusive. These are political, material, normative and linguistic.¹⁶

Political inclusion or exclusion refers to the degree which people can take part in the community’s decision-making process. It can include such things as voting rights, or if the community’s political posts and positions are open to all. In a democratic society all citizens should have at least the possibility to make their voices heard and to influence the rules or laws that are passed. A community excludes its members politically insofar as they are excluded from the decision-making process, either on purpose or effectively.

Duff defines material inclusion as “the extent to which people can share in the material resources and benefits available within the community”.¹⁷ It refers not only to physical goods but things like education, health care and employment. There is a problem here in defining what these goods being shared entails, whether they should be provided to all or if it is enough that everyone has the opportunity to obtain them. Regardless, people are excluded from these material goods if they have no fair chance of obtaining them.

Normative inclusion refers to the extent which we treat people as if the community’s values apply to them, and as if the community’s values were their own. It means to extend to everyone the rights and respect which flows from those values. But it also means to treat everyone as if the obligations which come from those values are binding on them.

(16)

13

The last area of community is linguistic, and it serves as a prerequisite for political and normative community. There is linguistic inclusion in the sense of there being a

common language between all the members in the community. To the extent that some people don’t speak this language, they are excluded not just linguistically but often politically as well. Members are also excluded from the normative community if they cannot speak or understand the language where the normative discourse takes place. It also seems like Duff understand “sharing a language” in a more figurative sense, for he gives an example of linguistic exclusion in the criminal trial as it is today.

For most offenders “their trial is an alien process that they cannot understand and in which they can play no real part and insofar as, given the content of their punishment and the context and manner of its imposition, the language in which they can

understand it as addressing them is not that of legitimate authority but that of brute power”.¹⁸

While the political theory which Duff presents is communitarian, it is nonetheless liberal-communitarian. It involves a liberal answer to the communitarian question of how we as a community should live. This entails an emphasis on such values as privacy, autonomy and freedom, and of individual rights which are not to be sacrificed for some greater good. But the added communitarian perspective presents some additional challenges for an account of just punishment. A big problem according to Duff lies in the way punishment excludes people from the community. Some theories of punishment are exclusionary by their nature, for example, the justification of punishment by its deterrent effect. Such theories portray offenders as people outside the normative community, as people who need additional reasons to refrain from crime apart from its being intrinsically wrong according to the community’s values.

(Though some would argue that this is the way the state should address potential lawbreakers.)¹⁹ The way punishment is used today is also often exclusionary. It is not uncommon that those serving prison sentences lose their right to vote, for example in the UK or the US.²⁰ Punishment is also often stigmatizing, those who have been

19 R.J. Lipkin, “Punishment, Penance and Respect for Autonomy”, Social Theory and Practice, vol. 14, no.

1, 1988, p. 87-104.

(17)

14

sentenced find it harder to find employment or housing, excluding them materially as well.

A central question for Duff is, “can a system of punishment still treat those who are punished as full members of the political community?”²¹ That is, can punishment be inclusive, and not something which serves to exclude those who are subjected to it from society? Much of Duff’s account of punishment will be devoted to showing that it can. In the section that follows I present Duff’s account, followed by a look at how it deals with the requirements presented in this chapter.

2.2 Punishment as communication

Duff’s basic contention is that punishment should be understood, and justified, as communicating to the offender the censure his crime deserves, with the aim of

“repentance, reform and reconciliation”.²² This account, I think, contains two parts.

First, punishment as deserved censure, and second, the goals of repentance reform and reconciliation which are to be achieved through that censure. I will start by

elaborating on these two ideas, and then see how Duff relates them to the proper role of the state, the liberal concerns previously discussed, and the inclusionary demands which communitarians place on a system of punishment.

2.2.1 Punishment as deserved censure

One of the problems with retributive theories of punishment was how to explain the connection between what an offender deserves and punishment. Simply saying that wrongdoers deserve to suffer is not enough. Neither do all possible explanations seem plausible even if they do establish a link between wrongdoing and deserved

punishment. An example of such a theory was the idea that punishment removes an unfair advantage held by the offender (see section 1.4.1).

Duff thinks that the best way to make sense of the desert claim is by thinking of punishment as a way of communicating censure. To censure someone is to express disapproval, criticism or otherwise fault them for the conduct. If there is one thing

(18)

15

wrongdoers deserve it is to be criticized for their actions. Duff thinks, this point is not mysterious like some other theories’ connection between wrongdoing and

punishment.²³ Part of the reason why wrongdoers deserve censure has to do with meaning what we say: If we declare certain conduct to be wrong we are committed to abstaining from such conduct ourselves. If we nonetheless engage in it our declaration comes off as insincere or dishonest. But, we are also committed to criticizing others who engage in it. Not doing so would make our declaration insincere in a similar way.

For example, say that we as a community declare assault to be wrong. If someone assaults someone else we should censure the assailant for his behavior. Not doing so would signal either that we don’t really take the assault to be wrong, or that the victim lacks the worth or standing of other members of society (and that therefore the assault was not wrong). To take a wrong seriously involves at the very least some negative reaction towards the offender. Duff thinks that this perspective can be applied to the criminal law as well, as a kind of collective declaration of wrongs. To be serious about these declarations of wrongful actions necessitates some censure by society towards those who commit them.²⁴

It is in this part of Duff’s theory that we find the similarity with Cooper (section 1.4.2).

Both theories place an importance on the response of the community towards crime.

For Cooper, the response (punishment) establishes the “right” which the crime implicitly denies. For Duff the response makes up the censure which offenders deserve. I understand Duff as making a sort of appeal to intuition, or common sense, when he argues that wrongdoers deserve censure, an intuition he tries to strengthen by looking at the act of declaring something as wrong and meaning it. He thinks, surely we can agree that someone who has done something which we take to be wrong deserves to be censured for it. Thinking that those who do wrong should be censured seems to be a part of thinking it wrong in the first place. At the same time, Duff presents no purely logical or factual reasons why we should censure those who do wrong – and there probably aren’t any either. If the contention that wrongdoers deserve censure rests on intuition, what makes it a better foundation for a theory of

(19)

16

punishment than the simple intuition that wrongdoers deserve to suffer? This is an intuition which many people have, yet most writers on punishment find it lacking.

Duff’s answer seems to be: more intuition. He writes “Second, it is not puzzling—in the way that critics of retributivism find the idea of penal desert puzzling—that offenders deserve censure for their criminal wrongdoings.”²⁵

Is this a weakness in Duff’s theory? I think, it doesn’t need to be. A lot of our moral though seemingly rests on intuition about what is proper, without the possibility to ground it in anything more solid. What makes Duff’s intuition better might simply be that it appears more sensible, and less questionable, to most people.

One reason why it might be easier to accept Duff’s initial idea is that it is not yet very strong. All he has argued for so far is that wrongdoers deserve censure. But censure is not the same as punishment. One kind of censure is simple verbal criticism, or a reprimand. Duff’s next step is to show why the deserved censure should take the form of punishment (or what he sometimes calls “penal hard treatment”.

As a first step, punishment can function as a form of censure: “Given an appropriate set of conventions and a shared understanding of those conventions, a term of

imprisonment or compulsory community service, or a fine, can communicate to those on whom it is imposed (and to others) an authoritative censure or condemnation of the crime for which it is imposed.”²⁶ But why should it be the preferred method of censure, as opposed to something else?

One answer is briefly discussed by Duff, and then discarded.²⁷ It is the idea that only punishment could serve as the appropriate censure for crime. On this line of thought, only punishment is severe enough, in that only the censure it communicates is severe enough. This would justify punishment solely as deserved censure, on the ground that no other censure would communicate the level of censure which is deserved. Duff has two problems with this proposal. First, that we must explain why only punishment could adequately communicate the censure which offenders deserve, an idea which

(20)

17

Duff is skeptical about. Second, why it should be the state’s duty or task to administer it.

I think this is a part of Duff’s theory which is a bit unclear. While he discards the idea that punishment is to be justified solely as deserved censure, he seemingly endorses this view in other parts of his book, or at least comes close to it. For example, he writes:

“a principle of proportionality between the seriousness of the crime and the severity of the punishment is intrinsic to my account. If punishment is to communicate the

censure that the offender deserves for the crime, it must communicate the appropriate seriousness or degree of censure. It must neither exaggerate nor depreciate the seriousness of the crime. But the seriousness or degree of censure is expressed by the severity of the punishment. A harsher sentence portrays the crime as more serious, a lighter sentence portrays it as less serious”²⁸

At another point he writes “We must determine not just that an offender deserves censure but how severe that censure should be: the more serious the crime, the more severe the deserved censure. That censure is communicated by penal hard treatment, and severity is a dimension of penal hard treatment as it is of censure. Thus the

severity of the penal hard treatment will communicate the severity of the censure: the more severe the hard treatment, the more severe the censure it communicates.”²⁹ While Duff does not write that only punishment could communicate the appropriate censure, he does draw a connection between the severity of punishment and the level of censure – harsher punishment equals stronger censure. Once we allow that such a material dimension plays into the level of censure it is hard to see how anything but punishment could reach this level. It seems very unlikely that, for example, a symbolic or mere verbal censure could ever match the censure communicated by a long prison sentence. If wrongdoers deserve censure, and the severity of that deserved censure can only be reached by punishment, it seems that punishment could be justified solely as deserved censure.

(21)

18

Either way, this is not how Duff wants to justify punishment as the preferred method of censure. Duff argues that punishment should be the preferred way of

communicating deserved censure because it serves the further aims of repentance, reform and reconciliation. These aims are not a separate justification of punishment, on top of it constituting deserved censure. Rather, censure has these aims intrinsically, and punishment best fulfills it.³⁰ (To put it in everyday terms, censure and the aims of repentance, reform and reconciliation are two sides of the same coin.) Thus

“Punishment should be understood, justified, and administered as a mode of moral communication with offenders that seeks to persuade them to repent their crimes, to reform themselves, and to reconcile themselves through punishment with those they have wronged.”³¹

2.2.2 Repentance, reform and reconciliation Repentance

Repentance is the act of regret, remorse or self-reproach for something one has previously done. It is to disown an act, wishing one had not done it. Duff argues that intrinsic in censuring others lie the aim of inducing repentance. “When we censure others for their wrongdoing, our intention or hope is that they will accept that censure as justified. But to accept censure as justified is to recognize and accept that I did wrong, the wrong for which I am censured; and an authentic recognition that I did wrong must bring with it repentance of that wrong.”³² Why is punishment the

appropriate way to induce repentance, instead of other forms of censure? Duff gives a couple of reasons. First, we often take the matter or remorse too lightly, we might say that we did wrong or acknowledge it superficially. But, at least for serious wrongs, this is not enough. Proper repentance requires that we focus on our wrong for a longer period of time, and with the right intensity. Punishment can best facilitate this, partly because of the time requirements involved, but also because it can bring home to the offender the seriousness of his offense – since the strength of censure follows the

(22)

19

strength of the punishment, the punishment communicates to the offender the seriousness of his wrong.

Why should we strive for repentance? Duff’s answer is that repentance is the proper response towards one’s own wrongdoing. Genuine remorse is also painful, thus repentance constitutes the suffering which offenders deserve (a notion other

retributive theories struggled to explain). Repentance is also a precondition for reform, the second aim of censure.

Reform

Proper repentance must also invoke a commitment to self-reform. In regretting a previous act we also aim to not do it again. Insofar as punishment is the proper way of inducing repentance, it also best serves the aim of motivating offenders to self-

reform.³³

How can Duff list “reform” as one of the justifying aims of punishment when this was one of the more criticized justifications in consequentialist theories? Not only is there little empirical support for the notion that punishment helps reform offenders, it was also criticized for treating offenders as objects to be remodeled – a view Duff himself argues is incompatible with a liberal society.

However, there is a big difference between the sort of reform Duff argues for, and the consequentialist aim of reform. For Duff, what partly justifies punishment as the appropriate method of censure is its supposed ability to motivate offenders to self- reform. What’s important is that this commitment comes from within, as something the offender himself sees to be necessary. For the consequentialist, reform is an end to be achieved by the most efficient means available. Whether or not it comes about voluntarily or through some coercive method is unimportant. This raises the question whether a commitment to self-reform induced through punishment really is voluntary.

This is part of the larger question of how Duff’s theory relates to the proper role of the state, and the limits put on the methods the state might utilize to seek its goals. (See

(23)

20

section 2.5 for the discussion on Duff’s theory and how it relates to the proper role of the state.)

One reply to the objection that punishment is ineffective as a means of inducing reform will run along similar lines; what justifies punishment is not that it is an effective method of reforming offenders, but that it aims at motivating offenders to self-reform (through the repentance of their crimes). This is true whether or not it actually succeeds in doing so. Of course, this reply seems somewhat disingenuous. If punishment motivated offenders to self-reform we would expect it to be somewhat effective, if it is not effective it seems it cannot be true that it motivates offenders to self-reform either. If this is the case, does it matter that punishment nonetheless aims at inducing self-reform? The same question can be asked of the aim of repentance, can this aim function as a justification of punishment even if it doesn’t succeed? Duff’s answer is no, but on a general level. These aims can still justify punishment in specific cases, even if they are likely to fail. He writes: “Punishment is justified as […]

persuading offenders to repent their crimes. Such a justification cannot be

independent of the extent to which that attempt is likely to succeed. First, however, success as thus conceived is not all that matters. We owe it to victims and to offenders to make the attempt to secure repentance, reform, and reconciliation. But that

attempt is worth making even if it is often likely to fail, since in making it we show that we do take crime seriously as a public wrong and address the offender as someone who is not beyond redemption”³⁴

What justifies punishment then, is not only that it constitutes deserved censure which achieves the goals of repentance, reform and reconciliation, but that the state owes it to its citizens to try and secure these goods even when it is likely to fail. It owes them this because in trying, the state shows that it takes crimes seriously and treats the offender as someone who could still redeem himself.

Reconciliation

The third aim of punishment is reconciliation. In cases of wrongdoing some apology by the wrongdoer is necessary if the parties are to reconcile. But, Duff argues, in serious

(24)

21

cases of wrongdoing a “mere verbal apology is not enough”.³⁵ The apology needs some more forceful, material expression, and the undertaking of a burdensome punishment can serve as that expression. Some kinds of punishments are well suited for this task.

For example, community service orders that involve some reparative work directly related to the crime (say, a vandal having to restore what he or she broke). By making up for the damage done the punishment can be seen as a sort of apology the offender makes in undertaking it. Duff argues, any punishment can be undertaken as an apology if, through it, the offender expresses his remorse of what he has done.

Duff meets two possible objections to the idea that punishment can serve to express an apology by the offender for what he or she has done. First, an apology is something we have to make for ourselves if it is to be sincere. But punishments are simply

imposed on an offender regardless of his thoughts or feelings about his crime, how could it serve as an apology then? Duff’s answer is that we should view punishment as a sort of formal apology, where what matters is that the offender partakes in the apologetic “ritual”, not his sincerity.³⁶

This answer ties together with the second objection, that to require an offender to apologize for his offence is inconsistent with liberal values. To require an offender to apologize for his wrongdoing is to force him to portray his own actions as wrong or morally blameworthy. But such a judgment must be left up to the offender if we are to respect him as an autonomous agent. Duff’s reply is that, since the apology takes the form of a formalized ritual, we are not requiring the offender to mean it, only to participate.³⁷ (Duff is not wholly convinced by this answer, but thinks that we can deal with any remaining doubt by giving offenders the opportunity to clarify that they are not undertaking their punishment as an apology.)

2.3 An example: combining probation and community service

The idea of punishment communicating deserved censure, with the aim of repentance, reform and reconciliation, can perhaps be made clearer with an example. Duff takes

“combination orders” consisting of probation and community service as one of the

(25)

22

more fitting forms of communicative punishment. While on probation an offender is subjected to some level of surveillance. He has to meet with a probation officer who’s job it is to make sure that the offender stops his criminal conduct and follows other possible conditions in the probation order. This surveillance communicates to the offender that he cannot be trusted in the eyes of the community. Other more specific requirements can be part of a probation order. As an example, Duff takes a football hooligan who is barred from attending matches as a part of his sentence.³⁸ Another example is someone who gets into fights while drunk, and who therefor is forbidden from going to places where alcohol is served. These sorts of restrictions can be tailored to the specific crime. Duff argues, they should be specific. There should be a clear connection between the restriction and the crime, if it is to communicate to the offender just what he has done wrong and in what way the community cannot trust him.

Another possible component of a probation order is a requirement to attend some program aimed at correcting the harmful conduct. Duff mentions the CHANGE

program in Scotland, a program for men who have been convicted of domestic abuse.

Such programs, Duff argues, should not be viewed as mere treatment or therapy but as punishment.³⁹ They aim at making these offenders understand what they have done, and how it could happen. Such programs are necessarily unpleasant, they try to confront the offender with what he has done, and to induce a change in his behavior.

The community service order can likewise be tailored so that there’s a clear

connection between the kind of work the offender has to undertake and his offence.

As an example, Duff takes a vandal who has to repair the damage he has done (or other similar damage caused by others).

The examples above helps illustrate what Duff has in mind when he writes that punishment should be communicative. Duff is not out to protect the status quo by arguing that our current penal practices can be understood (and justified) in a communicative light. In fact, he is very critical of how punishment is used in todays’

societies, and the form it takes (although his focus lie on the U.K. and the U.S.). More

(26)

23

specifically, he is critical of the way imprisonment is used both in its extent and its execution. Good communicative punishment establishes a connection between the punishment and the crime – the football hooligan is barred from attending more matches, the vandal has to repair the damage he has caused. This connection helps to communicate what the offender is being punished for, and also in explaining the notion that this censure is deserved. We might have a hard time explaining why an offender simply deserves to suffer, but not why a football hooligan deserves to be barred from more matches (even if he suffers when he is thus barred).

The connection between punishment and offence also make the goals of repentance, reform and reconciliation more plausible. Duff argued for punishment over other ways of communicating censure because punishment could serve these aims more

effectively. The best way to induce repentance was to focus the offenders’ attention to his crime. It seems plausible that a communicative punishment could do this. For example, the offender who has to attend a program for domestic abuse is likely reminded of what he has done. So too with someone who is barred from entering places where he has caused trouble, or someone who has to spend time repairing what he has damaged. Confronting the offender with the damage he has done is also likely to help motivate him to self-reform, insofar as it helps him understand the harm he has caused. As for reconciliation, if the punishment is to serve as an apology by the offender it can better do so by tying together with the crime. An apology carries greater force if it is in some way reparative, either literally (as in fixing something that was broken) or figuratively (as perhaps the attendance of some program).

2.4 Imprisonment

So far I think Duff has managed to offer an attractive picture of what justifies punishment. But he has done this, in part, by focusing on some specific kinds of punishment (and specific kinds of crimes). The kind of punishment that Duff has covered so far are typically thought of as lighter than, for example, prison sentences.

They also seem well-suited for those crimes that we think of as being less serious. But what of crimes like murder, rape or serous assault? Or those offenders we take to pose an ongoing danger to others? It is far from clear that we could employ probation or community service orders in response to these crimes, or that it would be appropriate.

(27)

24

They are typically punished by imprisonment, a mode of punishment which Duff hasn’t covered so far. Prison, on the other hand, does not make for good communicative punishment. Even if the prisoner is likely to know what he ended up in prison for, this mode of punishment does not serve as a reminder of the crime the same way

probation or community service can do. It does not “confront” the criminal with the damage he has done. It also offers little room for the offender to be an active participant in his punishment. Imprisonment does not require that the offender do something, it can simply be inflicted on him. This makes it harder to see the prison sentence as a form of apology by the offender, since he does not undertake to do something that could be interpreted as such. A prison sentence can of course include similar programs and work as probation and community service, but this is separate from the imprisonment itself – they can be appropriate communicative punishments, but they don’t make the imprisonment appropriate as a communicative punishment.

Despite this, Duff does find a place for imprisonment, but as mentioned before he thinks it should be reserved to the most serious crimes. Duff argues that, while imprisonment lacks some communicative virtues, they are not completely lacking.

Imprisonment can communicate to the offender that his conduct has rendered his further participation in the community (even supervised participation) impossible.

“The message of imprisonment is that the offender has not just damaged or

threatened, but has broken, the normative bonds of the community.”⁴⁰ As for the aims of repentance, reform and reconciliation, Duff reminds us that any punishment can serve these aims if it is done with them in mind (even if there are better modes of punishment for achieving these aims). Thus “Imprisonment […] must itself, even while excluding the offender from normal community, be a way of reconciling him with the community. By this I mean, not just that prisoners must be offered help during their prison term to equip them for life on their release, but that their imprisonment itself, as imprisonment, must be understood and administered as a way of reconciling them with those they have wronged.”⁴¹