Sentencing and Punishment

John Lea 2004

justiceYour first instincts when you are robbed at knifepoint in the street or you come home and find your house burgled and ransacked are not nice. You would like to do various things to the offenders which are best not repeated here. One of the reasons why a Criminal Justice System exists and why we are prohibited by the state from 'taking the law into our own hands' is that ideas of vengeance and personally inflicted retribution, as well as being a receipe for disorder and social breakdown, are seen as most likely disproportionate to the actual offence. By handing over our victimisation to the state and subscribing to a modern notion of criminal justice in which it is society as a whole and not (just) the victim who is violated by the crime, we give the state the responsibility of working out the appropriate punishments and from a perspective which is more than simply vengeance. Thus a key aspect of modern punishment is that the State (the criminal justice system) is the only agency enabled to respond legitimately by inflicting punishment. The Court takes over vengeance and turns it into something else which meets other requirements. What those requirements are and how they are combined in the form of individual sentences for convicted offenders is our topic in this lecture. We shall concentrate on the basic ideas behind punishment rather than a detailed study of the development of sentencing policy in England

At the present time the main justifications for punishment can be considered under four headings: retributivism, deterrence, rehabilitation and incapacitation


The essence of retributivism as a theory of punishment is the idea of 'just deserts' in which punishment is inflicted on the offender because it is 'deserved' by the crime which he or she has committed. The offender 'pays for' his or her crime by means of a punishment should 'fit' the crime. That is to say the punishment should not be excessive; it should be what is required by the seriousness of the crime. In the seventeenth and eighteenth centuries the idea of retributivism came to displace older ideas of revenge. The state takes over revenge and reserves to itself the right and the duty to inflict it. But at the same time revenge becomes humanised into a punishment which is not excessive but precisely calibrated in accordance with the seriousness of the crime. The offender, in suffering the punishment, makes reparation to society through suffering the punishment inflicted by the state. A secondary aspect of retributivism can thus be said to be a denunciatory function in which the state, by inflicting the punishment, acts on behalf of society to denounce, show disapproval of, or censure, the crime. The censure is more than simply the extraction of reparation. It might be argued, for example, that if punishment is simply 'paying for the crime' it is a retrospective licence for the offence. It's okay to commit the crime as long as you are prepared to pay the punishment ('do the crime, do the time'). Punishment as censure is saying: 'you should not have done this and the fact that you have been punished in no way means it was okay for you to have done it.' Nevertheless there is a sense that if you have suffered the punishment then you can become a full member of society again.

In the retributive approach punishment is therefore entirely concerned with the past behaviour of the offender, with the crime that he or she has committed. It is the duty of offenders, as beings possessing free will, to make reparations for their crime; not directly to the victims but indirectly to the state (the penal system). The state, unlike the victim, who may still be grieving and thinking about excessive revenge, is capable of inflicting the correct proportional punishment which correctly reflects the seriousness of the offence.

Not only is it the duty of the offender to make reparation to the state for his or her crime it is also, in the retributive view, the duty of the state to punish offenders. This is not due to any consequences of punishment such as the deterrence of further offending but because retribution confirms the free-will of the offender. This view, associated with the eighteenth century German philosophers Immanual Kant and Georg Hegel, sees the criminal as willing his own punishment in so far as he is a free agent. One would not, for example, punish a dog or a cat (except metaphorically) because they have no free-will. But a human being has the choice of not committing the crime but chooses to do so. It is his or her moral responsibility and thus the infliction of punishment is a recognition that the criminal act was chosen rather than the result of some automatic impulse over which the offender had no control.

The rate of exchange (the tariff)

So far so good. But the big problem with the retributive approach becomes clear when we come to decide what is the appropriate punishment for a particular crime. The most primitive punishments are those that repeat the crime. The offender is punished by suffering the same hurt as he or she inflicted on the victim. This doctrine of  'an eye for an eye, a tooth for a tooth' (Lex Talionis) is regarded as unjust and unworkable in most Western societies. It remains, if at all, only in the form of the death penalty for murder.

As we shall see, the most important form of punishment in modern societies is the deprivation of liberty or resources which may take the form of imprisonment or other restrictions on freedom. These restrictions are measured in terms of time or money. The various equivalences of so many years, or months imprisonment for various types of offences is known as the 'tariff'.  The question is therefore how to work out how much time a particular crime deserves. For more spectacular crimes many people  think they know the answer: the appropriate tariff for murder is life imprisonment. But how many years does a rape or a financial fraud deserve. Or, if we are talking about fines (deprivation of resources) then how much should a speeding motorist or a petty shoplifter be fined? More than the value of the goods stolen?

Other problems become clear when we take into account the fact that we live in a society of profound inequality and diversity. Such differences concern both the impact of sentences and public consensus about their appropriateness.

The impact of sentences

But the inequalities in society produce differential motives to commit crime. How can someone be censured for crime when society has driven them to it?  Poverty may drives large numbers of young single parents into shop lifting in order to feed their children. Surely we should be more concerned with the poverty that has driven them to shoplifting than with sentencing their offences?  As Alan Norrie (1993) puts it: "We hold the individual responsible for what he has done to society, when in reality, his crime is a product of what society has done to him." (page 208) I remember many years ago talking to a group of Magistrates in the North of England. Most of them were concerned with youth offending. One of them said to me: "I wish I could sentence the social conditions that drive these young lads into crime." 

Looking at the matter from the standpoint of a particular sentence it is clear the the same fine for a rich or poor offender will take away a different percentage of their resources. Fining a large company for polluting the environment, even if the fine runs into millions of pounds, may well be 'peanuts' and can easily be passed onto the customer. Or in terms of age and background: it is one thing to give a hardened professional bank robber in his 50s another 'stretch' but the same number of years custody for a young offender of say 21 may have a completely different effect on their life. The same sentence may in fact be quite different punishments.

Some reference to social conditions and the particular background of offenders can be taken into consideration by the Court in mitigation as can the differential impact of sentences

Consensus and Knowledge about crime

These problems apart, to the extent that retributive punishment is denunciation and censure on behalf of society then some sort of consensus about what are appropriate punishments must be presupposed. But this is a big assumption. Attitudes certainly differ between societies. Thus if you vandalise a motor vehicle in England you are likely to get a fine or a short community sentence. The same offence in Singapore will 'deserve' six strokes of the cane, four months in prison and a fine of 1000. But even within society consensus may be a problem, particularly in a modern society characterised by cultural pluralism and diversity. Can women and men, different ethnic groups, people of different sexual identities, incomes and lifestyles any longer agree on what a particular crime 'deserves'? Do women and men have the same perception of the seriousness of sexual crimes? Does everyone agree that racial motivation may make an assault more serious than say, simply financial gain? Have we become too diverse to agree or is there a fundamental consensus about the seriousness of most forms of criminality irrespective of these diversities. Underlying the philosophical justification for retributive punishment we find a sociological issue of what the social consensus of values actually is.

There is also the problem that in modern complex societies there are a growing number of crimes in the criminal code which society as a whole has little view of and doesn't know what to think about in terms of punishments. For example what serverity of punishment does someone convicted of technical infringement of financial regulations 'deserve'? 

public opinion about sentencing

How much do the public know about what the courts do? 
Home Office survey of public attitudes derived from the 
1996 British Crime Survey

2002 Sentencing Advisory Panel research panel on public
attitudes to sentencing for domestic burglary and date rape

An interesting research paper (published in 2002 on public attitudes to sentencing, punishment and other aspects of the criminal justice system

The conclusion is that the clarity of retributivism as a doctrine of punishment depends for its workability not only on moral arguments about 'desert' but also sociological assumptions about the degree of agreement in society about harm and how it should be dealt with. Thus there is plenty of room here for conflicts, among judges, between judges, politicians and different sections of the public concerning the appropriate 'just deserts'

Retributivism however remains an important doctrine of punishment despite its problems. It is most important probably in a negative role as a check against excessive punishment. It is a due process oriented approach to punishment. We are now going to move on to look at other theories of punishment. Deterrence theory says that the goal of punishment is to deter the criminal behaviour in question by the offender or other potential offenders. That is a quite different view to the focus on how much someone deserves. The importance of retributivism is that even if, for example we decided that a life sentence for petty theft might act as a deterrent, it would be unjust because it would be in excess of what such an offence deserved.


If retributivism as a justification for punishment oriented to the past behaviour of the offender (i.e. the crime that has been committed) deterrence theory is oriented to the future behaviour of the offender. (i.e.what punishment is necessary to stop them doing it again.) Whereas retributive theory focuses on the gravity of the offence: what punishment does this crime justly deserve, deterrence theory focuses on the nature of the offender: what punishment will stop this person from doing it again. To put it another way, retribution is concerned with paying the price for the crime; deterrence is concerned with influencing the future behaviour of the criminal. Deterrence is oriented to crime control rather than due process. For this reason deterrence theory is sometimes referred to as Utilitarianism as it is an application of the Utilitarian philosophy expounded by the eighteenth century English philosopher Jeremy Bentham who argued that actions (including punishments) were to be justified by reference to whether they increased or diminished the sum total of human happiness. Deterrence is also sometimes referred to as consequentialism for the obvious reason that it justifies punishments by referring to their consequences in terms of reducing offending

There are severe practical problems with too much emphasis on deterrence. If the problems facing retributivism are philosophical and moral problems (how much deprivation of liberty does a particular crime deserve? Is it fair to punish those driven to crime by poverty?); the problems facing deterrence theory tend to be psychological and sociological. 

The idea that a particular sentence will deter someone from committing the same crime again may make assumptions about their personality makeup. Deterrence theory often seems to assume people are like the economist's model of the rational calculator and that they will calculate that if the gain from the crime is less than the pain of the punishment then they will not repeat the criminal behaviour. However:

  • They may calculate rationally but with more focus on the chance of getting caught than the severity of the sentence
  • The risk of being caught may be precisely why crime is an exiting and challenging activity!

Here the sociologist may point out that 

  • in particular subcultures (e.g. professional criminals) having a prison sentence might enhance one's status rather than act as a deterrent to future activity.

  • For others the deterrent effect may exist but be cancelled out by the stigmatization and labelling which a prison or other sentence produces. Someone with a criminal record will find it harder to get a job. This will lead to poverty and social exclusion and a return to a life of crime. (not to mention that a prison sentence may be itself "an expensive way of making bad people worse."). 

In fact it may be argued that deterrence works best on the offender who neither comes from, nor returns to, a criminal subculture. The middle class motorist is most likely to conform to the model of the rational calculator. A prison sentence and a driving ban will concentrate the mind wonderfully. The social disgrace of a prison sentence, and loss of income consequent on a driving ban may be acutely felt. But for others the great paradox of punishment may be that the people who end up in front of the courts awaiting sentence are disproportionately those from the poorest sections of the working class who have little to lose from punishment and therefore are least likely to be deterred by a prison sentence or other punishment.

individual and general deterrence

We have spoken about deterrence thus far as if it was oriented simply to the future behaviour of the individual who had committed the crime. But deterrence can also be aimed at the future behaviour of the population at large. A judge may say for example when sentencing an individual: "there has been much too much of this sort of thing lately so I'm going to make an example of you." The notion of an 'exemplary sentence' would not be easily accepted by retributivists. They may argue that it is immoral to punish one person with the aim of discouraging others. For example when he was leader of the Conservative Party, William Hague proposed automatic life sentences for drug dealers who are twice convicted of supplying hard drugs to children. This would be accepted by retributivists only if they were convinced that individuals who engaged in such activity deserved such punishment. Deterrence theorists might say ""this is just the sort of tough sentence we need to discourage this sort of activity." They may be thinking as much about 'sending a message' to wider population of drug dealers as deterring the actual individual being sentences.

inconclusive evidence

The problem here is that the justification for general deterrence, like individual deterrence, relies on there actually being consequences in terms of reducing offending. Of course we never know how much crime is actually deterred for the simple reason that it doesn't take place! One of the most well known examples in the UK of exemplary sentencing was in 1973 a street robber in Birmingham was given a sentence of 20 years. There had been a 'moral panic' about street robbery (mugging) at the time and the judge decided to 'send a message' to anyone else thinking of engaging in this form of anti-social activity. The question is: who actually got the message? There was no noticeable change in the crime rate following the sentence. Researchers concluded that the young people most likely to engage in street theft were also least likely to know about the sentence that had been handed down.

So there are some sever problems both with retributivism and deterrence theory. For retributivism the problem is: how can you punish (or in particular what is the tariff ) for people who don't have free will, who are driven by circumstances to commit crime? For deterrence the problem is: the people you most want to deter and least likely to be deterred because of the factors mentioned above. Towards the end of the nineteenth century these doctrines were coming in for considerable amount of criticism. Two further justifications for punishment moved onto the agenda: rehabilitation and incapacitation

a couple of interesting essays on punishment 
that I found on the web

On punishment  
Can punishment be justified?

Rehabilitation and Incapacitation

Both rehabilitation and incapacitation start from the notion that it is the role of the state to effectively reduce crime, that is to say with the consequences of punishment. They are, in a way, attempts to move beyond deterrence theory. The model of the rational calculator which lay behind deterrence theory was coming to be regarded as increasingly unsophisticated as both psychology and social policy made headway during the latter part of the nineteenth century. 


As the state became increasingly concerned with issues such as the health of the population, the notion of intervening to change people's living conditions became acceptable. After all, how could we have a strong economy and army without a healthy workforce? New ideas in punishment were part of this trend. The object of punishment was seen as not simply deterring the offender but attempting to change the offender's outlook on life such that he or she would not want to offend again. This could be done in some cases by psychological or psychiatric intervention: therapy and education to change the offender's personality, to remove aggression etc. In other cases what was seen as important was education and skills training: to equip the offender with some minimal life skills so that, on coming out of prison, they would not immediately sink back into the lowest strata of society and start committing crime again. How to enable the individual to live a good and useful life. 

An issue which arises with rehabilitation strategies is of course the length of the sentence. The demands of 'just deserts' might be one thing and the requirements of effective rehabilitation and re-education programmes quite another. There is a danger of the 'indeterminate sentence' in which the offender, quite irrespective of the requirements of just deserts, is only released upon successful completion of the rehabilitation programme. The complaint that offenders, particularly young people, were finding themselves in never-ending treatment programmes, was one of the driving forces in the 'back to justice' movement during the 1980s. 


The opposite side of the coin is incapacitation. Whereas rehabilitation looks for the offender who can be changed and re-educated, incapacitation is concerned with the fact that some offenders are 'dangerous' and need to be removed from public circulation for a time for the sake of protecting the public. Of course, it might be argued that all offenders are by virtue of their offending, dangerous and thus a prison sentence can be viewed as a form of incapacitation. "Prison works because while people are in prison they are not a danger to the public."  But in recent years the view that one of the key tasks of punishment is to keep dangerous people out of circulation has gained ground.

An orientation to rehabilitation suggests a fairly cohesive society in which the offender is seen as someone who has done wrong but still a citizen nevertheless. The task is to reform the offender and enable him or her to function again as a (law abiding) citizen. By contrast an orientation to incapacitation suggests a much more fragmented society in which the majority of the population see themselves as surrounded by various dangerous individuals and groups against whom they need to be protected.  The 'protection of the public' becomes a major concern rather than the rehabilitation of the offender.

Of course it is possible to argue that this is a false opposition: if offenders are properly rehabilitated then they are no longer going to be a threat to the public. But it is a matter of political emphasis and orientation. Governments in most industrial countries see the protection of the majority of the population against risk and fear of crime as the key task and incapacitation does not necessarily mean rehabilitation. Locking people up in prison for a long  time or releasing them only on condition that they constantly report their whereabouts and activities to the police is nothing to do with rehabilitation. The assumption is that the offender is simply 'neutralised' whereas rehabilitation suggests that the offender is retrained and requipped so that he or she can once again play their full part in society.

Some of the problems of incapacitation are similar to those of rehabilitation. Both strategies face the issue of the indeterminate sentence. how long should the sentence be? If someone is dangerous then shouldn't they be permanently locked up? We could end up in the absurd position that all offenders considered dangerous should have mandatory life sentences. Obviously, a plausible incapacitation regime has to focus on individuals for which it can be known in advance that there is a high risk of re-offending. Certain types of mentally ill offenders may fall into this categories. But even here there needs to be some prospect that, as a result of various medical or therapeutic intervention such individuals could cease to be dangerous and therefore released back into society.

In recent years much debate in the UK has surrounded the issue of paedophiles. It has been regarded that convicted paedophiles will, while not necessarily facing permanent imprisonment, should necessarily face some regime of permanent surveillance to counter the risk of re-offending. Thus the Crime and Disorder Act 1998 enables police forces to apply for sex offender orders prohibiting a convicted paedophile offender from certain activities and locations. Further, Multi-Agency Public Protection Panels, made up of police and probation and other agencies such as social services and housing manage the risk posed by violent and sexual offenders released from custody. Paedophiles who have committed an offence since 1997, or were in custody at that time, are required to register their name and address with the police for a specified period of time under the Sex Offenders Act 1997. 

The Criminal Justice Act 2003 introduced much wider measures oriented towards incapacitation. This was part of a comprehensive change in many aspects of sentencing, the role of Parole and the Probation Service. In particular the Act introduced a new category of indeterminate public protection (IPP) sentences. According to the act, judges must impose an IPP - life in all but name - on any person convicted of any one of 153 separate violent and sexual offences, if they believe, in the words of the Act, that there is "a significant risk to members of the public of serious harm from the commission of further specified offences".

This has been a very controversial measure. Critics argue that the prisons are filling up with individuals on de-facto life sentences who may in fact have committed quite minor crimes. What is supposed to happen is that offenders serving these IPP sentences undertake 'rehabilitation' training courses which, if successfully completed, will render them no longer a danger to the public. In August 2007 the Court of Appeal ruled on the case of two prisoners who claimed they had been denied access to such courses (due to lack of resources) and were therefore facing extended terms in prison (read the story in the Guardian here) The government appealed against the judgement but was unsuccessful.

read this Briefing Paper from the Probation Service (or National Offender Managment Service, as it is now called) on the various new types of sentences introduced in the Criminal Justice Act 2003. 

read this article in the New Statesman magazine (March 2007) by journalist David Rose for some criticisms of Indeterminate Public Protection Sentences

another critical briefing on IPP sentences, from the Prison Reform Trust


Determining the Sentence

So, finally, who determines the sentence? There are in fact several agencies with an influence on sentencing 

The court

Obviously the judge at the end of a trial at which the accused is convicted is the primary sentencer.  The actual determination of the sentence by will involve a consideration of all of the perspectives outlined above. These may conflict and point in different directions. For this reason there may be variation in sentencing between what on the face of it are similar cases. The judge will have to make an individual decision in a particular case.

Punishment administrators

Prison governors, probation and parole officers, prison medical staff may all have an influence on the length of sentence actually served. A prison contains a potentially volatile collection of people who are there against their will. The governor needs various mechanisms of control apart from the presence of prison officers and guards. The ability of the prison regime (the governor, parole board etc) to vary sentence length as a reward for good behaviour or a punishment for bad behaviour is regarded as an important control mechanism. Those concerned with rehabilitation may well influence the actual release date in terms of their assessment of the degree of rehabilitation achieved. 

Politics: The Criminal Justice Act 2003

The political system may influence the sentence in two ways. Firstly in the general sense that governments may legislate regarding the length of sentences and, for example, make certain sentences mandatory upon conviction. In the English system Judges dislike mandatory sentences because they reduce the autonomy of the Judge to take into account all the relevant factors, including pleas of mitigation and to individualise the sentence for a particular case. For multiple murders where there is a "high degree of premeditation" there is a new tariff of 'life means life' with no chance of remission. This has been set by government under the Criminal Justice Act 2003 

In a narrower sense in the UK a government politician, until recently the Home Secretary may, in certain circumstances, vary the tariff for long term prisoners. The basis for this was that the Judge who made the original sentence would only be aware of the state of the offender at the time of sentence. Over the years if, for example, the offender had obviously reformed and become rehabilitated then there may be a case for early release. This should be decided by the Home Secretary. But the Home Secretary also had the right to lay down the minimum portion of a sentence that must be served. However, the UK Human Rights Act 1998, which incorporates European Union human rights law, has (by decision of the House of Lords in 2002) removed this power from the Home Secretary on the grounds that it is incompatible with human rights to give a political official the right to vary a judicial sentence. Such an arrangement is now considered a violation of  the separation of powers between the judicial and political spheres central to the rule of law. Making the judgement in the House of Lords Lord Bingham said the power exercised by the Home Secretary to decide how long sentences should be, was "incompatible" with article six of the European Convention on Human Rights - the right of a convicted person to have a sentence imposed by an independent and impartial tribunal.

Meanwhile the Criminal Justice Act 2003 as well as the new IPPs mentioned above, established a Sentencing Guidelines Council to produce a set of sentencing guidelines for all courts.This is follows a recommendation from the Halliday Report, that emphasised the importance of sentencing guidelines, and their impact on the sentences passed and on the wider understanding of why particular sentences are selected for particular offences. The Sentencing Guidelines Council will be chaired by the Lord Chief Justice and will comprise 12 members drawn from Judiciary and Magistrates but also including lay members, appointed by the Home Secretary, with experience in one or more of the following fields: policing, prosecution, criminal defence, the interests of victims, sentencing policy and the administration of sentences. Judicial members will be appointed by the Lord Chancellor, after consultation with the Home Secretary and the Lord Chief Justice. Non-judicial members will be appointed by the Home Secretary, after consultation with the Lord Chancellor and Lord Chief Justice. The Act also for the first time makes explicit the purposes of sentencing and puts them in statute. The purposes will be: public protection, punishment, crime reduction and reparation. The sentencing framework needs to address more directly than at present the purposes of crime reduction (through reform and rehabilitation and deterrence) and reparation alongside the continuing key and prominent aims of public protection and punishment.

The new key principles of sentencing will guide the court when determining the seriousness of the offence and the severity of the resulting sentence. The most important of these, is that previous convictions (where they are recent and relevant) should be treated as an aggravating factor when determining the sentence severity.

The government recently (2005) announced further measures derived from the 2003 Criminal Justice Act:  From December 2004 judges will be expected to reveal a defendant's prior convictions for the same offence they are charged with in the case before the court. In cases of theft and child sex offences Juries can be informed of any related past crime or conduct. An example could be that someone accused of child rape could have previous convictions for downloading child pornography from the internet revealed to the jury. According to a government If this proves successful, these special measures will be extended to cover other types of offences, such as murder. Civil libertarians and lawyers criticise these measures on the grounds that they are likely to lead to unfair convictions--where someone is convicted on the basis of their previous convictions whereas it should be purely the evidence in the current case before the court that establishes guilt beyond reasonable doubt. The response of the government is that the criminal justice system needs to be 'rebalanced' in favour of the victim and making sure the guilty are convicted. (read article in The Guardian for further details)


Norrie, A. (1993) Crime, Reason and History: A Critical Introduction to Criminal Law. London: Weidenfeld and Nicholson.