The ORE of Politics will be available for subscription in late September. Speak to your Oxford representative or contact us to find out more.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 18 August 2017

Punishment and Political Philosophy

Summary and Keywords

The punishment of criminal offenders constitutes a topic that has for many years received comprehensive attention, both in narrower academic circles and in broader public debate. This is not surprising. State-mandated infliction of death, suffering, or deprivation of freedom on citizens should from the outset be met with hesitation, and constitutes a practice which clearly calls for more profound considerations. Though the theoretical discussion of punishment has dealt with many conceptual and ethical issues, from an overall point of view, it is dominated by two questions.

The first question, as indicated, concerns the justification of legal punishment. Why and under what conditions is it justified for the state to impose punishment on perpetrators? The traditional answers have been split between the utilitarian approach, according to which punishment can be justified in terms of its future desirable consequences, mainly crime prevention, and the retrospectively oriented retributivist approach, which justifies punishment in terms of just deserts. In the modern discussion, the picture has become more diverse. Consequentialist and retributivist justifications have been developed in many different versions and several attempts have been made to combine forward- and backward-looking considerations into coherent schemes. Moreover, genuinely new accounts of penal theories have also been presented.

The second question concerns the issue of how different crimes should be punitively responded to. Though this question is obviously theoretically closely related to the first, it is also clear that the question of how individual offenders should be punished for their respective misdeeds prompts a plethora of more detailed challenges such as: What should determine the gravity of a crime? How should one determine the severity of a punishment? Are there types of punishment that should never be used in the criminal justice system (e.g., capital or corporal punishment)? Much of the contemporary discussion within penal theory is devoted to the task of providing principled solutions to these detailed challenges.

Keywords: consequentialism, crime, distribution, penal theory, proportionality, punishment, retributivism, severity, utilitarianism

Punishment: Normative and Conceptual Challenges

The punishment of criminal offenders constitutes a topic that has received comprehensive attention, both in narrower academic circles and in broader political debate. While the impetus behind the attention that is directed to punishment in political discourse and by decision makers is not always clear-cut or transparent and may well be multifarious, the point of departure of the academic philosophical approach to punishment is more simple, namely, that punishment involves a way of acting that is prima facie morally problematic.

Depriving people of their property by taking their money, limiting their freedom in various ways—most radically by locking them up under heinous conditions—inflicting bodily suffering, or putting people to death all constitute examples par excellence of behavior that would usually be regarded as morally wrong. Thus, since state-mandated punishment may consist precisely in these kinds of coercive and violent practices imposed on citizens who have broken the law, it is natural to ask how (if at all) such practices can be morally justified. It is this basic question of justification that has traditionally constituted the turning point of the philosophical approach to punishment. However, prior to this normative question often lie conceptual considerations.

When we are considering the justifiability of punishment it is reasonable to ask what we are talking about in the first place. What is a legal punishment? The simplest response to this question would be to point to paradigmatic examples of punishment, such as the state’s imposition of fines, incarceration, bodily suffering, or death (see, e.g., Montague, 1995). However, it is also agreed by many penal theorists that this type of ostensive characterization of punishment is not fully satisfactory. (For instance, questions sometimes arise as to whether a particular type of sanction can serve the function of a genuine punishment.) For this reason, and more generally, in order to be able to pursue the question of justification in a satisfactory manner, it has often been held to be crucial that one is able to more precisely identify the properties characterizing legal punishment. However, as is the case with regard to many other important societal concepts (e.g., torture, discrimination, terrorism), the quest for a definition has prompted several intricate questions.

For instance: Must a punishment involve pain, suffering, or at least be burdensome to the person upon whom it is imposed (or, what if a fine is imposed on a very rich offender; if a community service is experienced as harmless; or—perhaps more extremely—if an offender finds incarceration preferable to her other available alternatives? [Adler, 1991; Primoratz, 1990; Snook, 1983])? Must a punishment be deliberatively inflicted (or should there be room for regarding some natural consequences that might befall an offender as a result of her crime as part of the punishment? [Husak, 1990])? Does a punishment presuppose a transgression of a legally prohibited act (or does this conceptually block the possibility of punishing innocents? [Primoratz, 1990; Scheid, 1990; Sverdlik, 1988])? Must a punishment express disapproval or condemnation (or does this beg the justification question by conceptually favoring some theories of punishment to other theories? [Boonin, 2008; Duff & Garland, 1994; Feinberg, 1970; Husak, 1990; von Hirsch, 1993])? Does a punishment presuppose authorization of the party who is executing it (or are there circumstances under which harm inflicted on offenders by private persons does not amount to mere vigilantism but could properly be regarded as legal punishment? [Primoratz, 1990; Reiff, 2005])?

As these questions indicate, the task of defining legal punishment is far from simple. A proposal for a minimal definition that may overcome some of the challenges raised by these and related questions would be to hold that legal punishment involves: The imposition of something that is intended to be harmful and reprobative, on a supposed offender for a supposed crime, by someone who possesses the authority to do so (Boonin, 2008; Duff, 2013). However, it is fair to underline that there is no consensus either when it comes to the content of the definition or, more generally, with regard to the usefulness or possible futility of the endeavor of developing such a definition in the first place.

The purpose in the following is to address two overall questions. The first section will provide an overview of some of the central theories that give answers to the question of how and under what circumstances legal punishment is justified. The standard distinction between a forward-looking consequentialist and a retrospectively oriented retributivist approach provides a framework for the outline of the theoretical field which, as will be made clear, has become quite diverse. The second section addresses the challenge of penal distribution, that is, the questions of how different crimes should be punished and, not least, how severely. Though these questions are typically closely related to the more basic justificatory question, it is also clear that challenges of punishment distribution give rise to numerous more detailed theoretical problems that have not necessarily been answered merely by subscribing to a particular overall theory of punishment.

The Justification of Punishment

A Forward-Looking Justification

A standard approach to the justification challenge is to hold that legal punishment is justified in terms of its beneficial effects. More precisely, the consequentialist approach to punishment implies that punishing people who have broken the law is justified insofar as this constitutes a practice that best promotes the good (see, e.g., Bagaric, 2001; Boonin, 2008; Duff & Garland, 1994; Walker, 1990). This approach to punishment has often been presented as an instantiation of a general consequentialist position in ethics, according to which the right act is the one that produces the best consequences compared to any available alternative. But it is clear that one can subscribe to a consequentialist approach to punishment without being committed to consequentialism in general. However, the consequentialist approach presupposes that one is able to provide an account of what should be regarded as morally desirable consequences (i.e., what is intrinsically good). Though, as we shall return to shortly, different answers have been given to this question, the traditional view has been the utilitarian approach, according to which the good is specified in terms of well-being. Thus, given this approach, when, more precisely, would the punishment of offenders be justified?

From a utilitarian point of view, the first thing that should be noted is that the imposition of a punishment on an offender, when viewed in isolation, is morally wrong (Bentham, 1838). A punishment may well significantly reduce the well-being not only of the person upon whom it is inflicted but also of parties such as family and other relatives. Therefore, the question is how punishment can produce beneficial effects that are sufficient to outweigh this initial drawback. This can happen in several ways. The standard answer has been to point to the crime preventive effects caused by punishment. One way to achieve crime prevention is by influencing the behavior of the actual offender, so as to get her to abstain from committing future misdeeds. This is known as particular prevention. There are different ways in which this can be done (Bagaric, 2001; Primoratz, 1990). A first possibility is that a punishment may disable an offender from engaging in new criminal activity. The typical example is incarceration, which may incapacitate the offender from committing new crimes (though, of course, some crimes can still be carried out even if one is placed behind bars). A second possibility is the use of punishment as an instrument to eliminate an offender’s propensity to engage in future criminal activity. A reformative effect is achieved when a punishment causes a change in inclinations, motives, habits, or character, so that the offender will desist from committing a new crime even if she could do so without being discovered. Finally, a third possibility is that a punishment may deter an offender from repeating her misdeed. While an unpunished crime may induce the offender to commit new crimes, the frightening experience of being punished may induce her to desist.

However, even though there are several ways in which punishment can have a preventive effect on those who have already broken the law, the main focus of the utilitarian justification of punishment has traditionally been on the possible effects which punishment may have on parties other than the individual offender, that is, on general prevention (Bagaric, 2001; Primoratz, 1990) A general preventive effect can be achieved in different ways, but the main focus has usually been on deterrence. The evil of punishment may deter potential criminals, that is, people other than the offender herself, from breaking the law. Alternative ways in which punishment may have a general crime preventive effect could be by affirming and consolidating social norms, thereby contributing to the internalization of these norms in the motivational systems of citizens; or by preventing the distrust that would follow in the absence of appropriate punitive reactions and which may reduce a citizen’s willingness to comply with the law; or simply by preventing people from taking the law into their own hands.

There are also other consequences that should be taken into account. In so far as the feeling of trust, of being secure, or perhaps the satisfaction of vindictive feelings have an impact on people’s well-being, these reactions should also figure in the utilitarian calculus. However, there is no doubt that the focus has typically been placed on the crime preventive effects of punishment.

As this overview clearly indicates, the utilitarian justification of punishment may (unsurprisingly) be quite complex. What generates this complexity, of course, is that the justification is contingent on purely empirical facts. The utilitarian can hold that legal punishment is justified if it causes the outlined effects, but the truth of this implication hinges on facts that have to be established through the studies of criminologists and other social scientists. Furthermore, it should be kept in mind that the maximizing structure of utilitarianism implies that the justification of punishment cannot be provided without considering the extent to which nonpunitive measures may serve the function as instruments to crime prevention. Even if it is an empirical fact that legal punishment has a particular and general crime preventive effect of a magnitude that clearly outweighs the harm caused by imposing punishment on offenders, this is not sufficient to justify the use of punishment. We also have to know that the desired effect cannot be achieved in other and less harmful ways. If a particular crime preventive effect can be reached by assisting, educating, training, or treating offenders, and, correspondingly, if a general crime preventive effect can be reached through the implementation of social and educational initiatives, that is, in short, if the desired effects can be reached by applying nonpunitive methods, then ceteris paribus these methods are what the utilitarian approach would prescribe. Moreover, as this indicates, it does not make sense to consider the utilitarian approach to punishment independently of the question as to how other societal institutions may contribute to the promotion of well-being. The criminal justice system is costly and thus requires resources which could be used on other societal institutions and initiatives that promote well-being. Thus, even though, as mentioned, the subscription to the utilitarian view of punishment does not commit one to utilitarianism in all other aspects of ethical decision-making, it is hardly possible to provide an adequate justification of punishment without engaging in considerations of how limited resources should be distributed between different societal institutions. This, as mentioned, means that the utilitarian guidelines for punishment become quite complex; however, it is also fair to say that the utilitarian approach ties the question of justification to the debate of priorities of (limited) societal resources, which has often been totally ignored by rival penal theoretical positions (Avio, 1993; Ryberg, 2013).

While the utilitarian approach to legal punishment had earlier played a dominant role in the penal theoretical discussion, its influence has significantly decreased in the modern academic literature on punishment (Duff & Garland, 1994; Tonry, 2012; von Hirsch, 1998). The most familiar moral objections have been to hold that the utilitarian approach has counterintuitive implications by prescribing, under particular circumstances, punishments which seem manifestly unjust (including punishment of innocents or excessively harsh punishment of the guilty) (Boonin, 2008; Duff, 1986; McCloskey, 1965; Primoratz, 1990; Ten, 1987). Responses to these objections have varied: holding that the utilitarian view does not have these implications; that even though the theory does in principle have such counterintuitive implications, this is not sufficient to refute the theory; or that the alleged unjust implications, on closer scrutiny, are not counterintuitive after all (Bagaric, 2001; Braithwaite & Pettit, 1998; Smart & Williams, 1996).

A Retrospectively Oriented Justification

The consequentialist approach to punishment, as mentioned, dominated penal philosophy in the first and central years of the last century. A retributivist approach to punishment was often regarded as an inhumane or even barbarous position, a polite term for revenge, or was presented as “the only moral theory … which has been definitely destroyed by criticism” (Mabbott, 1969, p. 39). In this light, it is not surprising that the revival and dominance of retributivism in modern penal theory has often been described as nothing but a striking theoretical reorientation (Davis, 1992, 2009; Tonry, 2012; von Hirsch, 1998).

The central retributivist view is that punishment of offenders is justified in terms of “just deserts.” In other words, the (positive) retributivist would hold that the guilty deserve a punishment; that is, more precisely, that the desert of an offender constitutes a necessary and sufficient condition for justified punishment. The temporal orientation for the retributivist view thereby is the exact opposite of that of the consequentialist justification. Retributivism is offense-centered in regarding punishment as justified in terms of the misdeed that has taken place, independently of the consequences the punishment may have in the future. The central idea of regarding punishment as a deserved response to an offense, however, prompts several questions. First and foremost: How can the idea that offenders should be punished because they deserve this kind of reaction be theoretically sustained? And to what precisely does this idea amount, that is, what is it the offender deserves?

The first question has in the modern retributivist area been answered in many different ways (Boonin, 2008; Cottingham, 1979; Duff, 1986; Ten, 1987; Walen, 2014; Walker, 1999). For instance, according to one influential answer, the idea of penal desert should be analyzed in terms of fairness in the relation between law-abiding members of the community and those who break the law. In this view, the law can be regarded as a kind of cooperative venture which involves costs and benefits of all parties involved. What happens, then, when someone breaks the law is that this person gets the “benefits of the system,” that is, the benefit of the fact that law-abiding members of the community have restrained their actions, without bearing a corresponding cost of keeping the law herself. The offender gets an unfair advantage in comparison to those who keep the law and, therefore, deserves a punishment as a way of restoring the initial equilibrium of benefits and burdens. This fairness-based idea of retributivism has been developed in many different versions (Dagger, 1993; Davis, 1992, 2009; Morris, 1968; Sadurski, 1985; Sher, 1987). A very different account of the basic idea of deserved punishment has been to suggest that the giving of their just deserts to offenders is a way of recognizing them as rational and autonomous individuals who are responsible for their behavior (von Hirsch, 1993; von Hirsch & Ashworth, 2005). Somewhat differently, it has also been suggested that the most coherent account of our judgments in individual cases (involving punishments both of offenders and of innocents) is given by a theory of deserved punishment (Moore, 1997). Several other ways of sustaining a deserved-based approach to punishment have also been developed.

With regard to the question as to what it is an offender deserves, different answers have also been presented. The traditional view has been that an offender deserves to suffer or at least undergo some sort of burdensome treatment (Davis, 1972; Kleinig, 1973). However, another answer that has gained significant influence in the modern retributivist era is to hold that what the offender basically deserves is appropriate censure or condemnation (Bennett, 2008; Duff, 1986, 2001, 2013; Markel, 2011; von Hirsch, 1993; von Hirsch & Ashworth, 2005; Wringe, 2015). The point of departure of expressivist or communicative versions of retributivism has been that a punishment can serve the function of expressing or conveying censure or condemnation and that this feature is precisely what distinguishes a punishment from other types of burdens that might be imposed on a person (e.g., a fine in this way is different from a tax, and a prison term different from other types of detention or quarantine). The question that confronts theories perceiving punishment mainly as an instrument for the deliverance of a condemnatory message, however, is why, when it is indisputable that condemnatory messages can be conveyed in other ways, should the state resort to punishment at all? At this point, communicative-minded retributivists have been split between different answers.

One answer to the “why hard treatment?” question has been to hold that hard treatment is necessary to show the offender that the conveyed message is “really meant” (Baldwin, 1999). Another answer has been that hard treatment is justified as an instrument to increase the crime preventive efficacy of the law (Narayan, 1993; von Hirsch, 1993). Yet another approach has more ambitiously been that hard-treatment communication can function as an instrument to education and reform, and, in the end, may enable offenders to reconcile themselves with their victims and the community at large (Duff, 1986, 2001).

As this brief overview indicates, the basic idea of retributive justice, namely that punishment is justified as a deserved response to criminal offending, has been developed in many different directions. This is also reflected in the critical discussion which has often consisted more of the presentation of objections against particular versions of retributivism (frequently presented by retributivists adhering to other versions of retributivism) than in challenges to the basic idea of justifying legal punishment in terms of desert. However, it is clear that, insofar as the retributivist believes that normative theories of punishment can be rejected on the ground of possible counterintuitive implications, the question naturally arises as to whether the retributivist herself faces these kinds of implications. For instance, an absolutist interpretation of retributivism—according to which even the slightest deviation from what justice requires would be unacceptable, even if this could prevent terrible outcomes—may well be regarded as highly counterintuitive. Attempts to avoid such implications by rejecting the absolutist interpretation might face the challenge of explaining in a nonarbitrary way when highly undesirable consequences have sufficient weight to justify deviations from justice. As yet, this question has been addressed only marginally by recent retributivists (Ryberg, 2010).

Other Theories

The philosophy of punishment has often been presented as being marked by the division between the utilitarian and the (positive) retributivist approaches. From a historical point of view, this way of outlining the theoretical field is understandable (reflecting, for instance, the traditional opposing thoughts of Bentham and Kant). However, it is also clear that the modern theoretical approach to punishment constitutes a much more diverse field. Here are a few examples of alternatives.

One possible alternative is to maintain the basic forward-looking consequentialist approach to punishment, while at the same time rejecting the monist utilitarian theory of the good. This can be done in several ways. It might be held that a more plausible theory is either pluralist, by insisting that there are several intrinsic values, or offers an alternative to the focus on well-being. For instance, one account that has been defended specifically (though not exclusively) in relation to punishment has been to hold that what matters is the promotion of dominion (Braithwaite & Pettit, 1998). Another suggestion has been to incorporate the idea of desert into the consequentialist theory of the good. This can be done in several ways; for instance, by holding that the value of well-being depends on the degree to which this well-being is deserved or undeserved by the person upon whom it falls (Feldman, 1997; Ryberg, 2011). Such a more complex value theory might be held to capture some retributivist intuitions by implying (in contrast to the utilitarian view) that suffering imposed upon someone who deserves to suffer is intrinsically valuable. However, even though, as this indicates, the consequentialist view can be worked out in many different ways—some of which have been designed specifically to deal with penal theoretical challenges—it is also fair to hold that if one is persuaded by traditional criticism that the utilitarian theory of punishment opens up unacceptably unjust instances of punishment, then one would hardly accept alternative consequentialist theories. For any consequentialist theory it follows that if the punishment of an innocent maximizes what is intrinsically valuable then this is what the theory prescribes. In other words, since the standard objections have been directed against the consequentialist theory of the right (and not against a specific theory of the good), these objections concerning the punishment of innocents and excessive punishments can still be raised against other consequentialist approaches to punishment (Boonin, 2008).

A very different set of theories has been developed sharing the basic view that one need not accept the seemingly irreconcilable choice between retributivism and consequentialism. If one believes that considerations of justice must have a role to play in a proper theory of punishment, but also that the prevention of future crimes or other like desirable social consequences should be pursued, then it is tempting to believe that a satisfactory theory of punishment is some sort of middle way, that is, a theory which attempts to combine the insights of both approaches. This has been done in different ways. One approach has been to contend that consequentialism and retributivism provide the right answer to different questions; that is, that one should distinguish between the justification of the punishment system—which is answered by a consequentialist view—and the questions of who and how offenders should be punished; a question that should be answered by following retributivist prescriptions (Clark, 1997; Hart, 1968). Another possibility is to hold a negative retributivist view, that punishment is justified only if it is deserved, combined with consequentialist considerations (together implying that it is never acceptable to punish offenders more than they deserve, and that the degree to which they should be punished below the upper level of desert should be determined on the ground of the possible desirable effects [Dolinko, 1991; Duff, 2001]). Or, slightly differently, that desert considerations allow for a range of deserved punishments for a particular crime, within which consequentialist considerations could be applied as an instrument for fine-tuning appropriate levels of punishment (Morris & Tonry, 1990). Yet other possibilities have also been suggested (Berman, 2008). However, while the impetus behind theories that seek to provide a compromise between justice and consequence-based considerations is easy to understand, the question as to whether the resulting theories have actually succeeded in avoiding the challenges faced by retributivism and consequentialism in their pure forms, while maintaining the insights of both, is much more controversial (Duff, 2013; Kaufman, 2008).

Another type of approach has been to reject standard consequentialist and retributivist theories (or possible combinations of these) by holding that legal punishment finds its justification in an entirely different set of considerations. For instance, it has been suggested that a person who commits an offense can be regarded as consenting to a loss of immunity from punishment. Thus, on this view, it is the idea of consent that makes it morally permissible for the state to treat an offender in ways that would otherwise be impermissible (Nino, 1983). Other theorists have suggested that punishment can be justified on the ground of considerations of self-defense (Ellis, 2003) or that punishment is justified as a type of compulsory moral education (Hampton, 1984).

What several of these theories have in common is that, even though they do not hold that offenders are currently being punitively treated as they should be by the criminal justice system, they nevertheless subscribe to the view that legal punishment can be justified. However, it should be underlined that not all penal theorists share this view. A certain strand in the theory of punishment has been that, rather than seeking to reform the criminal justice system in accordance with what constitutes the most plausible justification, what we should do is to abolish the practice of punishment (Bianchi, 1994). The abolitionist outlook has taken many different forms. For instance, some advocates have suggested that one should seek to eliminate the concept of crime in favor of “troubles” or other concepts that do not invite punishment as the obvious reaction (Christie, 1977; Hulsman, 1986). Another approach has been to appeal to an account of justice that does not at all regard punishment as the proper reaction to crime. On such a view, a crime is regarded as an offence by one individual against the rights of another. The just reaction, therefore, consists in making the offender restitute the victim for the loss she has suffered. In this sense, a restitutionist approach to crime basically concerns justice in the relation between the offender and the victim. The state’s role consists merely in ensuring that proper restitution is exacted and not in the imposition of punishment (Barnett, 1977; Boonin, 2008; Ellin, 2000; Ryberg, 2012b). Other arguments in favor of the abolitionist outlook have also been presented (Zimmerman, 2011), but it is clear that the abolitionist outlook constitutes a minority view in penal theory.

What this outline of theories of punishment indicates is that, even though it still makes sense to place different theories in certain overall rough categories (such as whether they provide forward- or backward-looking justifications), it is not the case that the philosophy of punishment has converged on two or three rival theories. It seems more apposite to characterize current considerations of the justification of punishment as constituting an increasingly diverse theoretical field.

The Distribution of Punishment

Suppose that the question concerning the justification of punishment has been solved. Does this imply that the normative challenge raised by legal punishment is thereby settled? The answer is in the negative. In so far as punishment—as most theorists believe—can be justified, this conclusion immediately leads to the kind of question which criminal courts are dealing with and which, in striking contrast to the justification question, has usually dominated the broader political discourse of punishment: namely, how should punishments be distributed? More precisely: How severely should offenders be punished for their misdeeds? And which types of punishment should the state impose on different offenders? Let us consider these questions in turn.

How Severe a Punishment?

How severely should a pickpocket, a burglar, or a murderer be punished for their respective crimes? The obvious answer is that this depends upon how punishment is justified, that is, on what constitutes the purpose of imposing punishment on offenders. However, not all the theories advanced in response to the justification question have also been engaged with the more detailed challenges of penal distribution, by far. Let us here stick to the consequentialist and retributivist answers.

For the consequentialist, the answer to the question of penal distribution is provided directly from the basic theory of the right; that is, the justified levels of punishment (or justified sentences in individual cases) are those maximizing the good. As mentioned, this principle has been thought to justify more general prescriptions, such as, for instance, that penal levels should be kept to the minimum necessary to reach a crime-preventive effect. Moreover, different procedures to reach the appropriate penal levels have been suggested, such as the use of a strategy of decrement, according to which existing penal levels should be gradually reduced until a level is reached at which the crime rates increases unacceptably (Braithwaite & Pettit, 1998). However, it is clear that, given the theory of the good (e.g., a theory of well-being), the challenge of how precisely different crimes should be punished is, at the end of the day, an empirical question. Perhaps due to the combination of the facts that the determination of penal levels is contingent on research conducted by other scientists (e.g., criminologists and economists) and, not least, that current penal theory has been dominated by retributivist thinking, it is fair to say that the theoretical attention that has been directed to the question of punishment distribution has been limited.

From a retributivist point of view, the picture is different. Much attention has been given to the question of what a desert-based approach to punishment implies with regard to penal distribution. The standard view for retributivists has been to subscribe to the principle of proportionality (for a discussion of a consequentialist justification of proportionality, see Bagaric, 2001). Even though the more detailed justifications of the view that there should be proportionality between the seriousness of a crime and the severity of the punishment have varied with the different versions of retributivism, the overall idea of a punishment that fits the crime seems to follow directly from the idea of desert: An offender deserves a punishment that reflects the gravity of the crime committed. More precisely, this idea of proportionality has usually been interpreted as comprising two different aspects of the relation between crime and punishment.

The first aspect of proportionality consists in the requirement that a punishment should comport with the severity of punishment for other crimes. For instance, if burglary is a less serious crime than rape, then ceteris paribus the burglar should be punished more leniently than the rapist. And, correspondingly, two crimes that are regarded as equally serious should be responded to with equally severe punishments. This purely comparative requirement is known as “ordinal” proportionality (Ryberg, 2004; von Hirsch, 1993; von Hirsch & Ashworth, 2005). The second aspect of proportionality concerns the way a punishment nonrelatively comports to a specific crime. For instance, a sentencing system which imposes several years of imprisonment for a minor theft might well be regarded as imposing grossly disproportionate punishments irrespective of what constitutes the punishment levels for other crimes. This idea of a nonrelative relation between crime and punishment is usually referred to as “cardinal” proportionality (Ryberg and Petersen, 2014; von Hirsch, 1993; von Hirsch & Ashworth, 2005).

In order for the principle of proportionality to provide specific guidance with regard to how different crimes should be punished, it is crucial, first, that one is able to specify what is meant by one crime being more serious than another. It must be possible to tell whether rape is a more serious crime than theft or whether burglary is more serious than reckless driving. Second, the principle presupposes that it is possible to determine whether one punishment, say a fine, is more or less serious than another punishment, such as a prison term of a certain length. Both types of comparative judgments have invited more thorough theoretical considerations.

When it comes to the ranking of crimes in seriousness, a standard answer in the modern retributivist tradition has been that crime gravity should be determined on the ground of harm and culpability (von Hirsch, 1993; von Hirsch & Ashworth, 2005). More precisely, the view is that the relative seriousness of a crime increases with the degree of harmfulness. Everything else being equal, a rape is a more serious crime than theft because it causes more harm to the victim. However, one of the things that complicates the harm dimension of crime and, therefore, has called for further clarification is that there are several crimes which do not, at least not in any straightforward way, involve harmful conduct (e.g., conduct that only risks or attempts harm) (Ryberg, 2004; Ryberg & Petersen, 2014; von Hirsch & Jareborg, 1991). With regard to the culpability dimension of crime gravity, this is usually held to include considerations of mens rea (i.e., the mental state of the offender, such as intention, knowledge, recklessness, or negligence), as well as considerations of criminal responsibility (e.g., whether an offender acted under constraints from defects of knowledge or will). The fact that the computation of crime gravity is in this way a polydimensional enterprise naturally prompts the theoretically challenging question as to how the different dimensions should be weighed against each other in order to finally determine the seriousness of a particular crime (Ryberg, 2004).

The ranking of punishments in severity is usually regarded as a less challenging task. Different types of punishments typically vary in severity across a single (or a few) dimension(s); for instance, the severity of a prison term varies with its duration, while the severity of a fine varies with the amount of money that is exacted. However, motivated by the wish to be able to make comparisons between different types of punishment, the task has been to specify a more basic denominator of severity. The obvious answer has been that punishment severity should be determined, and thus can be compared, on the ground of the suffering, deprivation, or inconvenience which a punishment inflicts on the punished. One of the things that complicates this suggestion, however, is that offenders may be affected very differently in terms of suffering or inconvenience by objectively identical punishments. For instance, a fine of a certain amount may be much more burdensome to the poor than to the rich offender. And some people are much more vulnerable than others in coping with life under prison conditions. These observations have led to discussion of whether the severity of punishment should be determined on the ground of an objectivist measure, that does not give room for the fact that individuals may be affected very differently by objectively the same punitive treatment or whether differences in individual sensitivity to punishment should be reflected in the determination of severity, with the possible implication that two offenders who have committed the same crime may end up receiving objectively very different punishments (Kolber, 2009; Ryberg, 2004, 2010; von Hirsch, 1993).

The ranking of crimes in gravity, and of punishments in severity, provides a precondition for the satisfaction of ordinal proportionality constraints. If an offender has committed a crime that is ranked higher than another crime on the gravity scale, then she should be met with a punishment that is comparatively more severe. However, this in itself does not say anything about how severely different crimes should be punished. In order to provide guidance with regard to this crucial question one will have to consider how the two scales, once constructed, should be connected. At this point retributivist theorizing has offered different answers.

A traditional answer has been that crimes and punishments should be linked by letting the punishment directly mirror the character of the crime that has been committed. This view (referred to as lex talionis) is expressed in the Biblical formulation “an eye for an eye, a tooth for a tooth.” It has been interpreted both in a literal sense, according to which the crime should be reversed against the offender herself, and in a less strict sense, prescribing equivalence between the harm of the crime and the harm inflicted by the punishment (Hegel, 1967; Kant, 1996; Kleinig, 1973). However, in the modern retributivist tradition, proportionality has not usually been interpreted as a “like for like” principle. But then how should one determine the proportionate punishment for a certain crime? One possibility is to hold that the scales of crimes and punishments should be linked by the use of “anchor points”; for instance, that the least severe punishment is linked to the least serious crime and the most severe punishment to the most serious crime, with the other crimes falling in between (Kleinig, 1973; Ryberg, 2004; Scheid, 1997). This proposal obviously presupposes that it is possible to specify what is meant by the most severe punishment and the most serious crime. Other more complex conjectures combining considerations of desert, prevention, and parsimony in punishing have also been suggested (von Hirsch, 1993). What this illustrates is that even among theorists subscribing to the principle of proportionality, there is currently now consensus on how the proportionate punishment for different crimes should be determined.

What Type of Punishment?

Many discussions concerning penal distribution use fines and imprisonment as standard examples of punishment. And studies have shown that imprisonment usually comes to mind when people in general think of punishment. However, it is also clear that punishment systems usually offer wide-ranging punishment options, including intermediate punishments such as home detention, community service, day fines, electronic monitoring, and so on. It is also possible to imagine many other punishment options. The normative question then is which types of punishment the state should use in response to crime.

A simple answer is of course provided by the consequentialist approach; according to this, one should impose that type of punishment on offenders which provides the optimal balance between the suffering of the one who is punished and the beneficial effects produced by punishment. Interestingly, this idea has also been defended within a retributivist framework by holding that, while the amount of punishment imposed on an offender should be determined on the ground of the proportionality principle, the method of punishment should be determined on consequentialist grounds. Following this view, the retributivist could allow for substitutability between different types of punishment as long as the “penal bite” of a punishment is preserved (von Hirsch, Wasik, & Green, 1989). However, when it comes to the methods of punishment, the main focus has been on which types of punishment should be included at all in the set of acceptable punishment options.

This question has, on the one hand, been driven by the fact that new ways of dealing with criminal conduct have been proposed which do not fit into the standard conceptions of punishment. An example that has received considerable attention is the use of restorative justice procedures, which include processes in which victims and offenders (and other stakeholders) play an active part in deciding how the harm that has been caused by the crime should be repaired. Such processes may involve so-called victim-offender mediation, family group conferencing, and restorative circles. The discussion of such methods has been concerned with not only whether they should function as supplements to standard punishment, but also whether they should be regarded as preferable alternatives to punishment or, under certain conditions, could even function as a type of punishment (von Hirsch, Roberts, Bottoms, Roach, & Schiff, 2003). Another example that has directed attention to the scope of legal punishment is the use of neurotechnological treatment programs that could involve ways of changing the motives and inclinations of criminals, for instance, by increasing impulse control. A dispute has recently commenced on whether it is acceptable to instigate an offender by offering reductions in punishments in favor of participation in such treatment programs. It has even been suggested that such treatment schemes could, when properly imposed, function as a type of punishment (Bomann-Larsen, 2011; Ryberg, 2012a).

The main part of the discussion of the methods of legal punishment has not been concerned with which methods should be included in the set of punishment options, but has rather focused on which methods ought to be excluded. Given the widespread use of imprisonment in many countries, it is not surprising that some attention has been directed to the question of the ethical permissibility of this type of punishment (Lippke, 2007). However, the main criticism has typically not consisted in contending that incarceration per se should be regarded as morally unacceptable, but rather in holding that the conditions of prison life are often unacceptably heinous. Another important aspect of the discussion of the methods of punishment has focused on the use of bodily punishment (or different kinds of torture) and, in particular, on the permissibility of the use of capital punishment. Both consequentialist and retributivist arguments have been presented in the comprehensive discussion of the death penalty (Bedau & Cassell, 2004; Hodgkinson & Schabas, 2004). A particular perspective has in this connection been that there are certainly types of punishment that should never be imposed on criminals because they are cruel, inhumane, or degrading (Duff, 2005; Kleinig, 1998; Murphy, 1989). From a theoretical point of view, the challenge has consisted in specifying what is meant by these epithets and in explaining why punishment falling into these categories, even in cases where they are experienced as less harmful than acceptable types of punishment (such as long prison terms), should nonetheless never be inflicted on those who have broken the law.

Penal Theory and Penal Practice

The philosophical reflection of punishment has undergone a significant development. From having in certain periods been focused mainly or even exclusively on the basic question concerning the underlying theoretical justification of punishment, the philosophy of punishment is currently more properly characterized as a broader penal theoretical field addressing a plethora of detailed issues, as well as challenges confronting real-life sentencing practice (e.g., how should one punitively deal with recidivists or offenders who have committed a series of crimes before being apprehended?). But it should be underlined that penal theory is very much not a field that simply delivers theoretical support for the existing penal practice. On the contrary, many penal theorists subscribe to the views that there are offenders who are currently being punished for crimes that should not have been criminalized in the first place (Husak, 2008), that the existing penal levels are clearly excessive (Murphy, 1979; von Hirsch, 1993), or that basic societal preconditions for holding citizens answerable for their misdeed are not met in real life (Duff, 2001; Murphy, 1973). Thus, despite a wide theoretical disagreement between different approaches to punishment, it is usually very much agreed that the existing penal practice is not living up to what is theoretically desirable; in fact, penal practice is often seen as radically imperfect. Theoretically, this fact prompts the question as to what different theoretical approaches to punishment imply under nonideal conditions. As yet, this question has received only limited attention.

Further Reading

Brooks, T. (2012). Punishment. New York: Routledge.Find this resource:

Corlett, J. A. (2008). Responsibility and punishment. Dordrecht: Springer.Find this resource:

Davis, M. (1996). Justice in the shadow of death. New York: Rowman & Littlefield.Find this resource:

Ellis, A. (2012). The philosophy of punishment. Exeter: Imprint Academic.Find this resource:

Frase, R. (2013). Just sentencing: Principles and procedures for a workable system. New York: Oxford University Press.Find this resource:

Gross, H. (2015). Crime and punishment: A concise moral critique. Oxford: Oxford University Press.Find this resource:

Heffernan, W. C., & Kleinig, J. (Eds.). (2000). From social justice to criminal justice. New York: Oxford University Press.Find this resource:

Honderich, T. (2006). Punishment: The supposed justifications revisited. London: Pluto.Find this resource:

Lacey, N. (1994). State punishment. London: Routledge.Find this resource:

Lippke, R. (2007). Rethinking imprisonment. Oxford: Oxford University Press.Find this resource:

Matravers, M. (2000). Justice and punishment: The rationale of coercion. Oxford: Oxford University Press.Find this resource:

Murphy, J. G. (2014). Punishment and the moral emotions: Essays in law, morality, and religion. New York: Oxford University Press.Find this resource:

Ryberg, J., & Roberts, J. V. (Eds.). (2014). Popular punishment: On the normative significance of public opinion. New York: Oxford University Press.Find this resource:

Simmons, A. J., Cohen, M., Cohen, J., & Beitz, C. R. (Eds.). (1995). Punishment. Princeton, NJ: Princeton University Press.Find this resource:

Tandros, V. (2011). The ends of harm: The moral foundations of criminal law. Oxford: Oxford University Press.Find this resource:

Tonry, M. (Ed.). (2007). Why punish? How much? A reader on punishment. New York: Oxford University Press.Find this resource:

von Hirsch, A., Ashworth, A., & Roberts, J. V. (Eds.). (2009). Principled sentencing. Oxford: Hart.Find this resource:

White, M. D. (Ed.). (2011). Retributivism: Essays on theory and policy. New York: Oxford University Press.Find this resource:


Adler, M. D. (1991). The urgings of conscience: A theory of punishment. Philadelphia: Temple University Press.Find this resource:

Avio, K. L. (1993). Economic, retributive, and contractarian conceptions of punishment. Law and Philosophy, 12, 249–286.Find this resource:

Bagaric, M. (2001). Punishment and sentencing. London: Cavendish.Find this resource:

Baldwin, T. (1999). Punishment, communication, and resentment. In M. Matravers (Ed.), Punishment and political theory (pp. 124–132). Oxford: Hart.Find this resource:

Barnett, R. E. (1977). Restitution: A new paradigm of criminal justice. Ethics, 87, 279–301.Find this resource:

Bedau, H. A., & Cassell, P. (Eds.). 2004. Debating the death penalty. New York: Oxford University Press.Find this resource:

Bennett, C. (2008). The apology ritual. Cambridge, U.K.: Cambridge University Press.Find this resource:

Bentham, J. (1838). Principles of penal law. In J. Bowring (Ed.), The works of Jeremy Bentham (pp. 365–580). Edinburgh: William Tait.Find this resource:

Berman, M. N. (2008). Punishment and justification. Ethics, 118, 258–290.Find this resource:

Bianchi, H. (1994). Justice and sanctuary: Toward a new system of crime control. Bloomington: Indiana University Press.Find this resource:

Bomann-Larsen, L. (2011). Voluntary rehabilitation. Neuroethics, 6, 65–77.Find this resource:

Boonin, D. (2008). The problem of punishment. New York: Cambridge University Press.Find this resource:

Braithwaite, J., & Pettit, P. (1998). Not just deserts. Oxford: Oxford University Press.Find this resource:

Christie, N. (1977). Conflicts as property. British Journal of Criminology, 17, 1–15.Find this resource:

Clark, M. (1997). The sanctions of the criminal law. Proceedings of the Aristotelian Society, 97, 25–39.Find this resource:

Cottingham, J. (1979). Varieties of retribution. Philosophical Quarterly, 29, 238–246.Find this resource:

Dagger, R. (1993). Playing fair with punishment. Ethics, 103, 473–488.Find this resource:

Davis, L. H. (1972). They deserve to suffer. Analysis, 32, 136–140.Find this resource:

Davis, M. (1992). To make the punishment fit the crime. Boulder, CO: Westview.Find this resource:

Davis, M. (2009). Punishment theory’s golden half century: A survey of developments from (about) 1957 to 2007. Journal of Ethics, 13, 73–100.Find this resource:

Dolinko, D. (1991). Some thoughts about retributivism. Ethics, 101, 537–559.Find this resource:

Duff, R. A. (1986). Trials and punishment. Cambridge, U.K.: Cambridge University Press.Find this resource:

Duff, R. A. (2001). Punishment, communication, and community. New York: Oxford University Press.Find this resource:

Duff, R. A. (2005). Punishment, dignity, and degradation. Oxford Journal of Legal Studies, 25, 141–155.Find this resource:

Duff, R. A. (2013). Legal punishment. In E. N. Zalta (Ed.), Stanford Encyclopedia of Philosophy. (Summer 2013 Edition).Find this resource:

Duff, R. A., & Garland, D. (1994). Introduction: Thinking about punishment. In R. A. Duff & D. Garland (Eds.), A reader on punishment (pp. 1–43). Oxford: Oxford University Press.Find this resource:

Ellin, J. (2000). Restitutionism defended. Journal of Value Inquiry, 34, 299–317.Find this resource:

Ellis, A. (2003). A deterrence theory of punishment. Philosophical Quarterly, 53, 337–351.Find this resource:

Feinberg, J. (1970). The expressive function of punishment. In J. Feinberg (Ed.), Doing and deserving (pp. 95–118). Princeton, NJ: Princeton University Press.Find this resource:

Feldman, F. (1997). Utilitarianism, hedonism, and desert. Cambridge, U.K.: Cambridge University Press.Find this resource:

Hampton, J. (1984). The moral education theory of punishment. Philosophy and Public Affairs, 13, 208–238.Find this resource:

Hart, H. L. A. (1968). Punishment and responsibility. Oxford: Oxford University Press.Find this resource:

Hegel, G. W. F. (1967). Philosophy of right. New York: Oxford University Press.Find this resource:

Hodgkinson, P., & Schabas, W. A. (2004). Capital punishment. Cambridge, U.K.: Cambridge University Press.Find this resource:

Hulsman, L. (1986). Critical criminology and the concept of crime. Contemporary Crises, 10, 63–80.Find this resource:

Husak, D. (1990). Already punished enough. Philosophical Topics, 18, 79–99.Find this resource:

Husak, D. (2008). Overcriminalization: The limits of the criminal law. New York: Oxford University Press.Find this resource:

Kant, I. (1996). The metaphysics of morals. Cambridge, U.K.: Cambridge University Press.Find this resource:

Kaufman, W. (2008). The rise and fall of the mixed theory of punishment. International Journal of Applied Philosophy, 22, 37–57.Find this resource:

Kleinig, J. (1973). Punishment and desert. The Hague: Martinus Nijhoff.Find this resource:

Kleinig, J. (1998). The hardness of hard treatment. In A. Ashworth & M. Wasik (Eds.), Fundamentals of sentencing theory (pp. 273–298). Oxford: Clarendon.Find this resource:

Kolber, A. (2009). The subjective experience of punishment. Columbia Law Review, 109, 182–236.Find this resource:

Lippke, R. (2007). Rethinking imprisonment. Oxford: Oxford University Press.Find this resource:

Mabbott, J. D. (1969). Punishment. In H. B. Acton (Ed.), The philosophy of punishment (pp. 39–54). London: St. Martin’s.Find this resource:

Markel, D. (2011). What might retributive justice be? In M. D. White (Ed.), Retributivism (pp. 49–72). New York: Oxford University Press.Find this resource:

McCloskey, H. J. (1965). A non-utilitarian approach to punishment. Inquiry, 8(1–4), 249–263.Find this resource:

Montague, P. (1995). Punishment as societal defence. Lanham, MD: Rowman & Littlefield.Find this resource:

Moore, M. S. (1997). Placing blame. Oxford: Clarendon.Find this resource:

Morris, H. (1968). Persons and punishment. The Monist, 52, 475–501.Find this resource:

Morris, N., & Tonry, M. (1990). Between prison and probation. New York: Oxford University Press.Find this resource:

Murphy, J. G. (1973). Marxism and Retribution. Philosophy and Public Affairs, 2(3), 217–243.Find this resource:

Murphy, J. G. (1979). Retribution, justice, and therapy. Dordrecht: Kluwer Academic.Find this resource:

Murphy, J. G. (1989). Cruel and unusual punishment. In M. A. Stewart (Ed.), Law, morality, and rights (pp. 373–404). Dordrecht: Springer.Find this resource:

Narayan, U. (1993). Appropriate responses and preventive benefits: Justifying censure and hard treatment in legal punishment. Oxford Journal of Legal Studies, 13, 166–182.Find this resource:

Nino, C. S. (1983). A consensual theory of punishment. Philosophy and Public Affairs, 12, 289–306.Find this resource:

Primoratz, I. (1990). Justifying legal punishment. London: Humanities Press International.Find this resource:

Reiff, M. (2005). Punishment, compensation, and law: A theory of enforceability. Cambridge, U.K.: Cambridge University Press.Find this resource:

Ryberg, J. (2004). The ethics of proportionate punishment. New York: Kluwer Academic.Find this resource:

Ryberg, J. (2010). Mass atrocities, retributivism, and the threshold challenge. Res Publica, 16, 169–180.Find this resource:

Ryberg, J. (2011). Punishment and desert-adjusted utilitarianism. In M. Tonry (Ed.), Retributivism Has a Past: Has it a Future? (pp. 86–100). New York: Oxford University Press.Find this resource:

Ryberg, J. (2012a). Punishment, pharmacological treatment and early release. International Journal of Applied Philosophy, 26, 231–244.Find this resource:

Ryberg, J. (2012b). Restitutionism: A self-defeating theory of criminal justice. Social Theory and Practice, 38, 279–301.Find this resource:

Ryberg, J. (2013). Retributivism and resources. Utilitas, 25, 66–79.Find this resource:

Ryberg, J., & Petersen, T. S. (2014). Proportionality. In G. Bruinsma & D. Weisburd (Eds.), Encyclopaedia of criminology and criminal justice. Dordrecht: Springer.Find this resource:

Sadurski, W. (1985). Giving desert its due. Dordrecht: Reidel.Find this resource:

Scheid, D. E. (1990). Davis and the unfair-advantage theory of punishment: A critique. Philosophical Topics, 18, 143–170.Find this resource:

Scheid, D. E. (1997). Constructing a theory of punishment, desert, and the distribution of punishments. Canadian Journal of Law and Jurisprudence, 10, 441–506.Find this resource:

Sher, G. (1987). Desert. Princeton, NJ: Princeton University Press.Find this resource:

Smart, J. J. C., & Williams, B. (1996). Utilitarianism: For and against. Cambridge, U.K.: Cambridge University Press.Find this resource:

Snook, I. A. (1983). Scheid on punishment. Canadian Journal of Philosophy, 13(1), 129–132.Find this resource:

Sverdlik, S. (1988). Punishment. Law and Philosophy, 7, 179–201.Find this resource:

Ten, C. L. (1987). Crime, guilt, and punishment. Oxford: Clarendon.Find this resource:

Tonry, M. (2012). Retributivism has a past: Has is a future? New York: Oxford University Press.Find this resource:

von Hirsch, A. (1990). Proportionality in the philosophy of punishment: From why punish to how much. Criminal Law Forum, 1, 259–290.Find this resource:

von Hirsch, A. (1993). Censure and sanctions. Oxford: Clarendon.Find this resource:

von Hirsch, A. (1998). Penal theories. In M. Tonry (Ed.), The handbook of crime and punishment. (pp. 659–682). New York: Oxford University Press.Find this resource:

von Hirsch, A., & Ashworth, A. (2005). Proportionate sentencing. Oxford: Oxford University Press.Find this resource:

von Hirsch, A., & Jareborg, N. (1991). Gauging criminal harm: A living-standard analysis. Oxford Journal of Legal Studies, 11, 1–38.Find this resource:

von Hirsch, A., Roberts, J., Bottoms, A. E., Roach, K., & Schiff, M. (Eds.). (2003). Restorative justice and criminal justice: Competing or reconcilable paradigms? Oxford: Hart.Find this resource:

von Hirsch, A., Wasik, M., & Green, J. A. (1989). Punishments in the community and the principles of desert. Rutgers Law Journal, 20, 595–616.Find this resource:

Walen, A. (2014). Retributive justice. In E. N. Zalta (Ed.), Stanford Encyclopedia of Philosophy (Summer 2014 Edition).Find this resource:

Walker, N. (1990). Why punish? Oxford: Oxford University Press.Find this resource:

Walker, N. (1999). Even more varieties of retribution. Philosophy, 74, 595–605.Find this resource:

Wringe, B. (2015). An expressive theory of punishment. Basingstoke: Palgrave Macmillan.Find this resource:

Zimmerman, M. (2011). The immorality of punishment. Peterborough: Broadview.Find this resource: