The Theories of Punishment

Sentencing amounts to the use of state coercion against a person for committing an offence. The sanction may take the form of some deprivation, some restriction, or a positive obligation.

Deprivations and obligations are fairly widespread in social contexts — duties to pay taxes, to complete various forms, and so on. What distinguishes a sentence is the added element of condemnation, labelling or censure of the offender. In view of the direct personal and indirect social effects this can have, it calls for justification.

That is the framing you should open with. And when comparing the theories, three things should be identified in each: its behavioural premise — what it assumes about how people act; its political premise — what it assumes about the standing of the individual against the community; and its empirical claims and practical influence.

Five theories are conventionally listed: deterrent, retributive, preventive, reformative and expiatory. The aim of all of them is to transform law-breakers into law-abiders, and the kind of punishment imposed depends on, or is influenced by, the kind of society one lives in.

• Deterrent. To deter means to abstain from doing an act. Severe punishments are inflicted so that the offender abstains from crime in future and so that other members of society learn what the consequences of crime can be.
• Retributive. Based on the principle of an eye for an eye, a tooth for a tooth; to retribute means to give in turn. The object is to make the criminal realise the suffering of the pain by subjecting him to the same kind of pain he inflicted on the victim. The theory aims at taking revenge rather than at social welfare and transformation, and has not been supported by criminologists, penologists and sociologists, who feel it brutal and barbaric.
• Preventive. This too aims to prevent crime rather than avenge it; the idea is to keep the offender away from society, and the criminal is punished with death, life imprisonment and the like.
• Reformative. The most humane of all, aiming to reform the offender by individual treatment; the idea is that no one is a born criminal, that criminals are also human, and that if trained and educated they can be transformed into law-abiding citizens.
• Expiatory. That if the offender expiates or repents and realises his mistake, he must be forgiven.

The modern arrangement is more useful for an examination, because it forces you to state premises rather than sentiments.

Desert, or retributive theories

Retributive theories have a long history including the writings of Kant and Hegel. In their modern guise as the desert approach they came to prominence in the 1970s, propelled to some extent by the alleged excesses and failure of rehabilitative ideas.

Punishment is justified as the natural or appropriate response to crime — a fundamental intuitive claim — and its quantum should be proportionate to the degree of wrongdoing. The justification of the institution of punishment also incorporates the consequentialist element of needing to deter crime, since without the institution anarchy might well ensue.

• Behavioural premise: individuals are responsible and predominantly rational decision-makers.
• Political premise: all individuals are entitled to equal respect and dignity; an offender does not forfeit all rights on conviction, and has a right not to be punished disproportionately to the crime committed.

Proportionality is the key concept. Cardinal proportionality requires that the overall level of the penalty scale should not be out of proportion to the gravity of the conduct: five years' imprisonment for shoplifting clearly breaches that principle, but beyond such extreme cases there is much room for debate, and social conventions and cultural traditions tend to determine what levels of sanction are thought appropriate in a particular national or historical context. Ordinal proportionality concerns the ranking of the relative seriousness of different offences; in practice much depends on how conduct is evaluated, especially by sentencers, and on social assumptions comparing traditional or real crime such as street crime with new types of offence such as city fraud and pollution. In theory it requires the creation of a scale of values to assess the gravity of each type of offence, into which culpability, aggravation and mitigation must then be assimilated.

The Indian formulation of the same idea is worth memorising: the principle of proportion between crime and punishment is a principle of just desert that serves as the foundation of every criminal sentence that is justifiable, and as a principle of criminal justice it is hardly less familiar or less important than the principle that only the guilty ought to be punished. Indeed the requirement that punishment not be disproportionately great, a corollary of just desert, is dictated by the same principle that does not allow punishment of the innocent, for any punishment in excess of what is deserved for the criminal conduct is punishment without guilt.

Deterrence theories

Deterrence theories regard the prevention of further offences through a deterrent strategy as the rationale of punishing: as an exercise of state power, sentencing can be justified only by its consequences.

Individual deterrence sees the deterrence of further offences by the particular offender as the measure of punishment. A first offender may require little or no punishment; a recidivist might be thought to require an escalation of penalties; the seriousness of the offence becomes less important than the prevention of repetition. There is little modern literature on it, though traces are detectable in modern sentencing practice — and some critics of desert theory claim that in taking limited account of prior record it incorporates a covert deterrent element.

General deterrence calculates the penalty on the basis of what might be expected to deter others from committing a similar offence. Major utilitarian writers such as Bentham, and economic theorists such as Posner, develop the notion of setting penalties at levels sufficient to outweigh the likely benefit of offending.

• Behavioural premise: responsible and predominantly rational, calculating individuals.
• Political premise: the greatest good of the greatest number is the supreme value and the individual counts only for one — so it may be justifiable to punish one person severely in order to deter others effectively.

The evidence is the weak point. Satisfactory empirical evidence of the effect of deterrent sentencing on individual behaviour is difficult to obtain, because the conditions must be such that non-offending can safely be ascribed to the deterrent effect of the legal penalty rather than to any of the myriad other influences on conduct — the perceived risk of detection, the opinions of significant others, and so on. Few research findings meet that criterion, and those that do support general deterrent sentencing in only a few types of situation.

Rehabilitative sentencing

Sentencing aimed at the reformation of the offender's character has a lengthy history, evident in the early days of probation and of Borstal institutions. The rationale is to prevent further offending by the individual through rehabilitation, which may involve individual case-work, therapy, counselling or intervention in the family.

Still a leading rationale in many European countries, it reached its zenith in the United States in the 1960s and then declined spectacularly in the 1970s. Research by Martinson was widely represented as demonstrating that treatment programmes usually failed, swamping the more qualified judgment of an English survey by Brody and a subsequent clarification or retraction by Martinson himself. On effectiveness, the true position is probably, as with deterrence, that certain rehabilitative programmes are likely to work for some types of offender in some circumstances; and a humanitarian desire to provide help for those with obvious behavioural problems has ensured that various treatment programmes continue to be developed, with some authors calling for the rehabilitation of rehabilitative theory.

• Behavioural premise: some or many criminal offences are to a significant extent determined by social pressures, psychological difficulties or other problems impinging on individuals. The links with positivist criminology are strong.
• Political premise: offenders are seen as unable to cope and in need of help from experts, and therefore perhaps as less than fully responsible individuals.

The approach indicates that sentences should be tailored to the needs of the particular offender, and places no limits on the extent of the intervention; it emphasises diagnosis and treatment by trained professionals. In practical terms the pre-sentence report, formerly the social inquiry report, is an essential element in its pursuit.

Incapacitative sentencing

The incapacitative approach identifies offenders or groups likely to do such serious harm in the future that special protective measures should be taken against them. Programmes of selective incapacitation have focused on groups of recidivists; in England the emphasis has been on identifying individuals predicted to be likely to commit serious harm, increasingly through the discretionary sentence of life imprisonment and through public protection sentences for violent and sexual offenders considered likely to do serious harm.

• Behavioural premise: none. It looks chiefly to the protection of potential victims and applies whether the offender is a rational calculator or is driven by pressures.
• Political premise: often presented as utilitarian, justifying incapacitation by reference to the greater aggregate benefit — but even a rationale emphasising the continuing rights of a person who has offended must deal with the possibility that those rights conflict with the rights of a potential future victim.

The fatal difficulty is prediction. The Floud report found that predictions of dangerousness tended to be wrong more often than not, and that fallibility has been repeatedly confirmed. It calls into question the justification for any lengthening of sentences on grounds of public protection; and yet political pressure to have some form of incapacitative sentence available has been felt in most countries. If that is the reality of penal politics, there is a strong case for procedural safeguards to ensure that predictive judgments are open to thorough challenge.

Restorative and reparative theories

These are not theories of punishment. Their argument is that sentences should move away from punishment of the offender towards restitution and reparation, aimed at restoring the harm done and calculated accordingly. They are therefore victim-centred, though some versions encompass reparation to the community for the effects of crime. They envisage less resort to custody, with onerous community-based sanctions requiring offenders to work in order to compensate victims, and also contemplate support and counselling for offenders to reintegrate them into the community.

• Behavioural premise: similar to rehabilitation.
• Political premise: compensation for victims should be recognised as more important than notions of just punishment on behalf of the state.

Legal systems based on a restorative rationale are rare; but the increasing tendency to insert victim-oriented measures such as compensation orders into sentencing systems structured to impose punishment provides a fine example of Garland's observation that institutions are the scenes of particular conflicts as well as being means to a variety of ends, so that each particular institution combines a number of often incompatible objectives and organises the relations of often antagonistic interest groups.

Restitution as a fifth theory gained significant ground in the twentieth century. It interprets the debt to society the criminal incurs in a more mercantile and perhaps more humane sense: in addition to suffering society's retribution, the criminal's debt may also be paid through valuable service to the community and to the individuals he harmed. The methods that make this possible are more intensive probation, jail time served at weekends, and house arrest. And compensation fits the traditional purposes surprisingly well: uncompensated labour can be a very unpleasant experience for people accustomed to thinking only of themselves; the burdensome inconvenience of house arrest, weekends in jail, community service and handing away hard-earned paycheques can serve as adequate deterrence and can satisfy the requirements of retribution; strict probation is usually enough to restrain a majority of offenders; and the potential for rehabilitation is considerably stronger for criminals given the opportunity to experience what it is to work and give back to the people they have injured.

Social theories

There has been a resurgence of writings emphasising the social and political context of sentencing — Garland's analysis of the theoretical underpinnings of historical trends in punishment, and Hudson's argument for a shift towards a more supportive social policy as the principal response to crime.

Hart's distinction is between the general justifying aim of punishment, in his view utilitarian or deterrent, and the principles of distribution of punishment, in his view retribution or desert. Lacey challenges the dichotomy, arguing that both issues raise questions of individual autonomy and of collective welfare, and that rather than denying the conflict we should address it and strive to ensure that neither value is sacrificed entirely at either stage; in developing this she explores the political values involved in state punishment and argues for a clearer view of the social function of punishing.

Braithwaite and Pettit's republicanism has at its heart the concept of dominion — liberty not in the sense of simple freedom from constraint by others, but in the form of a status of guaranteed protection from certain kinds of interference. From this they propose that punishments should increase the dominion of victims with the least loss of dominion to the offenders punished. Since dominion emphasises reassuring citizens about the prospect of liberty, it might require preventive sentences based on deterrence or incapacitation; they refer to unspecified limits on severity but not to lower limits, and hold that the censuring function of the criminal justice system can and should so far as possible be fulfilled by means other than punishment. There is thus no recognition of an individual's right not to be punished more than is proportionate: all depends on what will advance overall dominion, which might happen to be more or less than the deserved punishment in any individual case.

In law, there is a hierarchy. Under the Criminal Justice Act 1991 desert is the primary rationale, except in the relatively rare cases where the conditions for imposing an incapacitative sentence are met; deterrence may not be used to justify a disproportionately severe sentence; rehabilitative considerations become important when choosing among community orders of similar severity, and justify probation orders and supervision after early release; compensation orders appear to have priority over fines and not over custody. That framework is hardly conflict-free, but it establishes a greater degree of certainty than the previous cafeteria-style system in which sentencers usually had wide choices of approach.

In practice, it is less clear-cut. Punishment is a social institution, so discussion ought to take account of its relation to other social institutions and of the relationship of sentencing to other aspects of the criminal justice system; and the form and content of punishment in a given society are likely to reflect a historical mixture of political and social events and influences.

On effectiveness, doubts attach to individual deterrence, general deterrence, rehabilitation and incapacitation alike. The point that wins marks is that what is at stake is marginal preventive effect: a sentencing system based on desert is likely to deter and incapacitate to a certain degree anyway, so proponents of deterrent or incapacitative theories must justify the search for extra increments of prevention by reference to evidence of likely success and to the measures needed to achieve it — and it is there that concrete evidence is wanting.

On rights, even satisfactory evidence of efficacy would leave the question of the rights of victims and offenders. The victim's right to receive compensation from the offender is surely undoubted, but in what circumstances should it give way to the offender's interest in not being utterly impoverished for months or years to come? There is controversy about whether victims should be able to express an opinion on sentence to the court, but this overlooks the point that sentencing is a state function which should be determined by public policy rather than private preference. As to offenders: deterrence theory seems to regard individual offenders as mere units in the overall calculation; incapacitative theory may recognise a general right not to be punished more than is proportionate but overrides it in situations that may be more or less well defined; and rehabilitative theory has often failed to recognise any such general right, having been invoked in support of indeterminate sentences. It may therefore be seen as a strength of desert theory that it limits punishment to what is proportionate and proposes criteria for determining proportionality.

On punitiveness, it is sometimes argued that desert theory leads to harsh penalties, but there is no such connection: California has harsh penalties whereas Finland and Sweden have relatively low levels of punitiveness, and all three regard desert as the primary rationale. Much depends on the political climate and the attitudes of key actors, and reductions in punitiveness may be better achieved through specific principles of restraint in the use of custody, or even prison capacity constraints, than through the promotion of a general rationale.

1. Open with the definition and the need for justification, in the terms given above. It takes two sentences and it frames everything after it. 2. Give each theory with its two premises, not merely its slogan. A candidate who states the political premise of general deterrence — that it may be justifiable to punish one person severely to deter others — has already shown the examiner what is wrong with it. 3. Do not treat restorative theories as a sixth theory of punishment. Say in terms that they are not theories of punishment at all. 4. Use the marginal-effect argument when asked whether deterrence works; it is more sophisticated than saying the evidence is mixed. 5. Close on the Indian trend. The successive amendments to the sentencing provisions show Parliament taking note of contemporary criminological thought and the movement from deterrence to reformation; and reformation and rehabilitation of offenders, not mere deterrence, are now among the foremost objects of the administration of criminal justice in this country. </content>