Can restorative justice reduce the burden on the criminal justice system?

Can restorative justice reduce the burden on the criminal justice system?

Martin Wright

This paper will consider what are the difficulties faced by the CJS, and why are the difficulties not being resolved?. It will then suggest what can be done about them. Firstly, there are conventional responses, and then a different response based on restorative principles will be proposed. Finally the possible effects of such an approach will be explored.

What are the difficulties faced by the CJS?

In many countries there are long delays in the courts, and prisons are overcrowded. Part of the reason is simply that the criminal justice agencies are doing their job. They act on the assumption that they should encourage people (usually victims) to report crimes to the police. The police detect as many offenders as possible, prosecutors charge as many of them as possible. If they are convicted, most of them then have to be supervised or imprisoned.

Another reason is the number and length of prison sentences; it is aggravated by the delays in the courts, which means that large numbers, who are ‘innocent until proved guilty’, are in prison awaiting trial. Some countries use few sanctions other than prison, and have high prison populations, especially if their prison sentences are relatively long. The greatest offenders in this regard are the United States, with nearly two million prisoners, 686 for every 100 000 of the population, and Russia with nearly one million, 638 per 100 000. A recent survey found that in Ukraine the rate is 406, in Poland 213, in the United Kingdom 139 and in the Faeroe Islands 21 (representing 9 prisoners in the population of 40 000 !) (Walmsley 2003). The introduction of community sanctions does not, however, necessarily lead to a reduction in the prison population. In Estonia, for example, with 4 723 prisoners (337 per 100 000) at the time of Walmsley’s survey, a probation service was created in 1993; but the prison population remains high, three new prisons are planned and in addition some 5 000 people were placed on probation (Hilborn, personal communication 2004).

It is generally accepted by criminologists that there is little correlation between the crime rate and the imprisonment rate; for example the crime rate increased during the 1990s both in Poland where the criminal policy was liberalized, in Romania and Bulgaria where the liberalizing tendency was much less evident, and in the former Soviet republics where imprisonment rates remained high (Krajewski 2004: 398). In England and Wales the prison population has been rising in recent years although crime rates have been falling. The average prison population was 45 817 in 1992 and 70 816 in 2002; in that year the cost was £4 821 million (Office for National Statistics 2004: tables 11.10,, 11.12). The British Crime Survey found approximately 12.3 million crimes in 2002/3, 36 per cent fewer than in 1995 (Simmons and Dodd 2003: 22; Independent 22.7.04). In the United States it was shown some years ago that that among the fifty states, those with high imprisonment rates were not necessarily those with high rates of recorded crime, but were more likely to be politically conservative and to have more black and poor citizens. A high crime rate was associated with factors such as unemployment and the level of urban population (William Nagel, cited in Wright 1982: 49). Similarly, the number of executions was not related to the homicide rate (Sellin 1980).

Why are the difficulties not being resolved?

The task that the criminal justice system has set itself is like that of the Danaïds, the young women in Greek legend who forever had to try to pour water into jars which had holes in the bottom. If the seriousness of crimes (or at least, of those crimes reported in the media because they are exceptional, and therefore ‘news’) appears to be increasing, there are demands for more and longer sentences, but these will never be enough – it will always seem that the pain inflicted on the offender does not match the harm which the offender caused to the victim or his or her relatives. If community sentences are used, such as probation or community service, there are demands for strict enforcement, which in turn leads to more people being sent to prison although their offences were not serious enough to justify prison in the first place.

There is a tendency for policy-makers to over-simplify. They see ‘criminals’ and ‘non-criminals’. ‘We’ are the non-criminals; we do not commit crimes, we are more likely to be victims of crime. ‘They’ are criminals, or at least anti-social. To control offenders, as the British prime minister Tony Blair said recently, ‘what has to happen is that the penalty they are paying for being a nuisance becomes more of a hassle for them than to stop being like that’ (Guardian, 1.9.2004). Psychological research, as well as everyday experience, has long shown that a crude behaviouristic method that emphasises punishments more than rewards is effective only for a short time and under specific conditions which the criminal justice system does not provide (Wright 1999: chapter 2).

Prisons are inherently expensive, even when overcrowded and inhumane. To reduce their inhumanity costs even more: they require workshops, education and human relationships, and help in overcoming the effects of imprisonment itself. All of these needs are competing, within the law enforcement budget, with the needs of victims. They need to be kept informed, supported, and in some cases offered compensation or therapy. The cost is very small compared with the cost of prisons, but it is difficult to find even this amount when the prison system is making such great demands.

The existing system has other shortcomings which are also serious, although they do not actually increase its workload. For example, traditionally it has treated victims only as witnesses, and has not kept them informed about the progress (or lack of progress) in their case. Too often, especially in an adversarial system, the defending lawyer regards it as his or her task to undermine the evidence of the victim-witness, trying to make them contradict themselves in order to create ‘reasonable doubt’ in the minds of the jury and thus to secure an acquittal. The defendant has every incentive to deny, or minimize, the harm caused to the victim, in order to be acquitted or to receive a lighter sentence.

What can be done about them?

(a) Conventional responses

The conventional response, of course, is to demand ‘more of the same’: to try to find ways of making the court process faster and more efficient, with fewer cases delayed because papers have been lost, fewer witnesses called to prove points which are not disputed, and so on. If more courts and more prisons are needed, they should be provided. This approach is often accompanied by ‘tough’ political rhetoric, encouraging the courts to pass longer sentences, and sometimes changing the legislation to increase the permitted maximum, or to set a minimum. In an adversarial democracy, this can lead to an ‘auction’ in which politicians try to outbid each other in toughness; and they are encouraged by some sections of the media. For example a leading article in The Sun, the newspaper with the largest circulation in Britain, was headed ‘Prison works’ (a slogan used by the then Conservative home secretary Michael Howard) and applauded the high level of the prison population, saying ‘If the jails are full, … then build more’ (22 May 2003). Sentiments like these are based partly on retributive ideas, and partly on the behaviouristic psychology that has been mentioned above.

(b) A different response

The conventional view obviously has some foundation in everyday experience, but it is far from being the whole picture. Even people who have done bad things are capable of doing better. When someone harms another person, some response is required, but punishment is not the only possible response. A further unwanted side-effect is that punishment makes people think of themselves; they try to deny or minimize what they have done, and hence deny recognition to what the victim has suffered. Restorative justice, on the other hand, turns the offender’s attention to the other person and encourages him or her, as we say, to ‘come clean’, to admit everything, przyznać się, make a new start (under the English word ‘to clean’ the dictionary gives posprzątać or wyczyścić, and a related word is ‘to purify’, oczyścić się). Everyday experience also shows that fear is not the only way to induce people to behave well; another powerful motivation is to win the approval of those for whom they have respect, and in turn to earn the respect of the victim and other people by being willing to meet the victim and to make reparation.

Restorative justice approaches the problem from a different angle.

  • Focus on the harm (and sometimes dispute), not on the crime

  • Allow those affected by a crime to take part in the process, express their feelings and consider how to repair the harm

  • Hold the offender responsible, but also provide resources to enable him or her to make amends

  • Involve the community in the process.

Restorative justice does not encourage labelling. Firstly, it does not label people as ‘offenders’ but as individual human beings who have harmed someone else or the community. Many, though not all, are capable of experiencing remorse, especially after hearing the victim speak of his or her experience. Many are willing to make amends voluntarily, without being compelled to do so by the threat of punishment.

Secondly, restorative justice does not assume that everything that can be labelled ‘crime’ should be treated in the same way. The commission of a crime is a three-stage process: the legislature decides that a certain act is criminal, someone performs that act, and someone decides to report it to the police. (One could add further stages, for example the police must decide that it was indeed a crime, and a court must confirm this.) Thus one cause of the rising crime rate is parliament – by creating new offences. The number of harmful incidents has not necessarily changed. To-day thousands of people may be using inside information to make a profit on the stock exchange, or looking at pornography on their computers, or using dogs to chase foxes; the behaviour may be reprehensible but not criminal. To-morrow a new law may come into force, and all those actions become crimes. In England it is now an imprisonable offence to fail to send your children to school (Independent 15.4.2003, Daily Mirror 18.9.2004). Some Turkish politicians want to make adultery or unfaithfulness criminal (Independent, 15 and 17. 9.2004). Among less serious offences the problem is even more serious. Already in 1980 it was estimated by Justice, the British section of the International Commission of Jurists, that there were about 750 crimes, defined as having an intent of dishonesty, deliberate physical injury or sexual gratification, and some 6 500 breaches of administrative regulations, or actions of carelessness or omission not involving moral turpitude. It was suggested that the latter groups could be designated as ‘contraventions’ (where the offence is not disputed) (Wright 1982: 82-3). At a stroke the criminal statistics and the burden on the courts would be drastically reduced if these offences were dealt with administratively. Currently, however, they are being increased: according to a British newspaper, between 1997 and 2001 the British government presented 31 new law-and-order bills to parliament, and between 1997 and 2003 it created 661 new criminal offences (Cohen 2004).

If a crime is on the statute book, it is still necessary for an individual act or omission to be defined as criminal. In Britain, one agency that takes a pragmatic approach is Inland Revenue. It regards its primary task as collecting taxes, and not prosecution for its own sake; thus it has stated that during 2004/05 ‘We will be focusing on cases where prosecution will do the most to promote compliance and deter fraud’ (Inland Revenue 2004: 16). Similarly Customs and Excise, whose tasks include he reduction of money-laundering, is ‘extending existing powers to impose civil penalties as an alternative to prosecuting businesses that fail to operate satisfactory money-laundering controls’ (HM Customs and Excise 2004: 18).

We can explore some of these ideas by looking at three common types of case. In the first, a dispute between acquaintances leads to a fight; for example,. B makes an approach to A’s girl-friend, and A hits B. One possibility is to treat the incident as crime and prosecute A as an ‘offender’. He then gets a criminal record and punishment, court time is used, he may even be sent to prison. A and B will probably never speak to each other again. A second possibility is that B can sue A in a civil court, demanding damages for the injury suffered.

The alternative, being used more and more as community mediation services are created, is to treat the incident as part of a dispute. Instead of going to the police, B can go to the mediation centre, which invites A to meet B in the presence of mediators. They follow the normal mediation process, asking both to follow basic ground rules such as not interrupting and not using language that could cause offence; each of them describes what has happened in their own words, they express their feelings, and consider how they will behave to each other in future. It has been found in many different mediation services that if both parties are willing to meet, eight or nine cases out of ten will lead to an agreement about future conduct, a continuing relationship or at least a better understanding of the other person. Courts may like to consider how many cases fall into this category. I understand that the Polish Code of Criminal Procedure , as amended, allows police, prosecutors or judges to refer cases to mediation (Wójcik 2004).

A second common case is burglary1. Here too the conventional response is to prosecute. However, because of the pressure on the system, prosecutors (in common-law countries at least) may make a plea-bargain: they reduce the charge, for example from aggravated burglary (carrying a weapon) in favour of a lesser charge such as burglary: the offender pleads guilty to this, and receives a reduced sentence. This is disappointing to victims, because it fails to acknowledge the full seriousness of the crime that they have suffered.2

An alternative is to suspend the prosecution, so as to create an opportunity for mediation to take place. My understanding is that in Poland the Act for the Treatment of Juveniles 1982 (as amended on 15 September 2000) allows discretion to the family judge to apply the Code of Criminal Procedure or the Code of Civil Procedure, or to divert the case to a mediator, who can then interview offender and victim to assess suitability for mediation (Wójcik 2004)3. In some countries the case is referred to an accredited mediation service, which uses trained mediators. Some mediators are ordinary members of the community who have been trained; this fulfils another aim of restorative justice, namely to involve the community in the restorative process. Victim and offender are given an opportunity to meet. The victim can express feelings and ask questions, the offender can offer an apology and reparation, such as community service, a training course or anger management, and ‘the community’4 in turn provides opportunities for the community service and any training that the offender needs.

For the third example, let us take an offender which is a firm, a ‘legal person’. As a result of its inadequate safety procedures, the local water supply has been poisoned, and many people have become ill. Criminal prosecution often leads such a company to spend large amounts of money on lawyers who attempt to prove that the company was not legally liable, or if it was to reduce the amount of compensation that it is ordered to pay. There have also been cases where the victims have brought a civil action for damages, which has the advantage that there is a lower standard of proof, but is also likely to be resisted in lengthy and costly legal proceedings.

There have however been cases which were dealt with in a restorative way. One Japanese company discharged waste containing mercury into the sea; people who ate the contaminated fish suffered paralysis, nervous disorders and even death. The company agreed to pay each victim substantial compensation, and to meet the cost of dredging the bay; but instead of being made bankrupt the company was enabled to continue trading so that it could meet its liabilities (Wright 1982: 264). Examples of nursing homes, mining companies and nuclear power installations that have improved their performance as a result of restorative approaches are given by Braithwaite (2003: 17-18, 63-66). Here too, the restorative process does not necessarily work by itself; it can be reinforced by public pressure from employees, customers, and actual or potential victims.

If however it is felt that, in the current penal climate, a prison sentence is unavoidable, a restorative approach can at least reduce the length of the sentence. In a recent judgement by the Lord Chief Justice of England, Lord Woolf, an armed robber had been sentenced to seven years’ imprisonment, but this was reduced to five years, because he had taken part in a ‘conference’ at which the offender, the victim and members of both families were present. He wrote a letter of apology to the victim, promised to tackle his .drug addiction, and applied to be transferred to a prison where treatment was available. Advocates of restorative justice would have preferred this to be a substitute for a prison sentence, but at least a restorative element for both the victim and the offender was introduced. (R. v. Collins (2003), Times 14 April)

In England and Wales, a new measure (introduced in the Youth Justice and Criminal Evidence Act 1999) which has some restorative features is the referral order. Magistrates’ courts are now required to refer all young offenders (with few exceptions) aged 10 to 17, who plead guilty and are appearing in court for the first time, to a youth offending panel. This consists of two trained volunteers and one member of staff. Other people such as victims can be present, but this has not happened as often as hoped, largely because workers are not accustomed to working with victims. These panels have the potential to conduct a restorative process and reach a restorative outcome; they also involve members of the community.


Effects on the system

What does restorative justice achieve? Diversion of cases from the system before the start of prosecution or preliminary proceedings obviously relieves the pressure on courts; but in addition, when courts refer cases to mediation before sentence, courts are commonly presented with a ‘package’ of measures which make custody unnecessary. In New Zealand, mediation (family group conferences) was introduced in 1989 for juveniles for all types of offence except homicide. The conferences were used either as diversion from court or, for more serious cases, before sentence. The number of cases in the youth court fell from 8 193 in 1989 to 2 352 in 1990 (from 63 per 1 000 of the population in 1987-89, the three calendar years preceding the new Act, to 16 per 1 000 in 1990), a reduction of 71 per cent. The figure rose in later years, but remained at a much lower level. Court time was also saved because the new procedure encouraged young people to accept responsibility, rather than denying it and challenging the prosecution to prove it. Custody was not abolished; there is still ‘supervision with residence’ and ‘corrective training’, but the number of young people sent to custody fell from 173 in 1989 to 64 in 1990, a fall of 63 per cent (McElrea 1998).

The role of judges, prosecutors and defence lawyers is as important as before, but it is changed. Prosecutors act as ‘gatekeepers’ to decide which cases to refer to mediation5, and judges oversee the process to ensure that it is carried out in accordance with restorative principles; both judges and prosecutors are then left with more time to devote to the most serious cases and those where the accused denies guilt. Similarly there is a different role for defence lawyers: they no longer represent the accused but advise them; their task is not to deny or minimize the offence but to safeguard the rights of the accused. Ideally, similar protection should be available to the victim.

Reduction of crime

A frequently asked question is whether restorative justice reduces re-offending. In addition to being desirable in itself this reduces the burden on the system still more. The results must be monitored, although advocates of restorative justice maintain that the increased satisfaction of victims would be sufficient justification in itself, even if the re-offending rate was unchanged.

If there were less crime, this would ease the pressure on the criminal justice system (unless it compensated by occupying itself with more petty offences). There are three different ways of reducing the number of crimes. One is to reduce the number of actions defined as criminal in the Statute Book (Penal Code). In fact, the number is more likely to increase, as we saw earlier, but at least legislators should think carefully before adding to the list of actions that are criminalized.

The second is to reduce the amount of re-offending. The number of studies of this is growing rapidly. For example, Maxwell and Morris (1999) found that reconviction rates were ‘certainly no worse and may be better in some respects’ than the closest available comparisons, and that cautious verdict could apply to the research findings generally. They found that other factors in the young people’s lives were more significant predictors of re-offending, such as feeling that they were unwanted by their parents, having no adult at home when they returned from school, being more often hit by parents, and so on: in other words, factors which are outside the power of the courts. Well conducted family group conferences, which induce remorse without making offenders feel bad about themselves, were associated with a lower level of re-offending.

An extensive review by Kurki (2003) also found encouraging but not overwhelming results. For example, one report combining four victim/offender mediation studies found a recidivism rate of 19 per cent for those who participated in mediation and 28 per cent for those who went through the traditional criminal justice process, but the analysis did not control for the individual characteristics of offenders. The ‘RISE’ study in Australia tried to overcome this lack by allocating cases randomly to conferencing or courts. For juveniles who had committed violent crimes, the re-offending rate after conferencing fell by 49 per cent, and for those who went to court by 11 per cent; but for drunk driving the rate was slightly higher after conferencing. On other measures, such as satisfaction, however, conferencing scored significantly higher .

The third way is to reduce offending in general. Advocates of restorative justice believe that the task of responding to individual crimes should be kept separate from the reduction of crime in general. The latter belongs to those responsible for crime reduction policy, and ultimately social policy in general. The conventional system operates on deterrence, but it is generally accepted that the risk of being caught is a more powerful deterrent than the severity of punishment. Despite the intervention of the courts, in England and Wales the reconviction rate for adult males sent to prison, who were discharged in 1999, was 55 per cent. For young makes aged 14 to 20, it was 74 per cent, and for those aged 14 to 17 it was 80 per cent (Home Office 2003: tables 9.3, 9.4). Crime reduction is a combination of situational measures, which make crime harder to commit, and social measures, which aim to reduce the pressures towards crime and increase social integration. Restorative justice has a potentially useful role, which has not been developed as much as it could be. Unlike the conventional system it encourages discussion of the offence and its background in a non-adversarial atmosphere, and consequently it can point to types of situation where offending behaviour is more likely, so that action can be taken if there is the political will. One example comes from Zwelethemba and other townships in South Africa, where local people have formed ‘peacemaking committees’ which meet to ‘heal, not hurt’: to deal with both civil conflicts and criminal cases by reaching a consensus about repairing the harm that has been caused. They are linked to peace-building committees, which use information from the mediation meetings to identify local needs, for example by building a children’s playground and providing loans to enable people to start small businesses (Shearing 2001, Roche 2003: 264-6).

A final example comes from Germany. The pioneering victim/offender mediation project ‘Die Brücke’ (The Bridge) in Munich, noticed that an increasing number of criminal offences, including bodily harm and sexual harassment, were committed in or near schools. Some victims said that their complaints were not taken seriously by the teachers; in other cases, however, the offenders were excluded and felt stigmatized. Die Brücke therefore created a plan to teach children to be peer mediators, to explain the idea to teachers and parents, and other activities (Nothhafft 2003). This can not only enable pupils themselves to deal with conflict among their peers, but encourages them to respect each other through active listening and prejudice reduction, which form an essential part of mediation training.


It is not claimed that restorative justice is a panacea, but it could help to reduce the burden on the criminal justice system, by developing an approach based on repairing harm, diverting some cases from the system, increasing participation by victims, offenders and members of the community, and treating them with respect. To encourage people to respect each other, and to handle conflicts restoratively when they occur, are two of the most important ways of developing a just and humane society.


Cohen, Nick (2004) ‘Turning right to wrong’. Observer, 1 August, p. 27.

Her Majesty’s Customs and Excise (2004) Spring departmental report 2004. Cm 6224. London: The Stationery Office.

Home Office (2003) Prison statistics England and Wales 2002. Cm 5996. London: The Stationery Office.

Inland Revenue (2004) Spring departmental report: the government’s expenditure plan 2004-6. Cm 6225. London: The Stationery Office.

Krajewski, Krzysztof (2004) ‘Crime and criminal justice in Poland.’ European Journal of Criminology, 1 (3), 377-407.

Kurki, Leena (2003) ‘Evaluating restorative justice practices.’ In: Andrew von Hirsch et al., eds. Restorative justice and criminal justice: competing or reconcilable paradigms?. Oxford: Hart Publishing.

McElrea, F W M (1998) ‘The New Zealand model of family group conferences’. Paper to international symposium ‘Beyond prisons: best practices along the criminal justice process’, Kingston, Ontario, Canada.

Maxwell, Gabrielle, and Allison Morris (1999) Understanding re-offending. Institute of Criminology, Victoria University of Wellington, New Zealand.

Nothhafft, Susanne (2003) ‘Conflict resolution and peer mediation: a pilot programme in Munich secondary schools.’ In: Elmar G M Weitekamp and Hans-Jürgen Kerner, eds. Restorative justice in context: international practice and directions.. Cullompton, Devon: Willan Publishing.

Office for National Statistics (2004) Annual abstract of statistics. London: The Stationery Office.

Roche, Declan (2003) Accountability in restorative justice. Oxford: Oxford University Press.

Sellin, Thorsten (1980) The penalty of death. Beverly Hills, CA: Sage.

Shearing, Clifford (2001) ‘Transforming security: a South African experiment.’ In: Heather Strang and John Braithwaite, eds. Restorative justice and civil society. Cambridge: Cambridge University Press.

Simmons, Jon, and Tricia Dodd, eds. (2003) Crime in England and Wales 2002/3. Home Office Statistical Bulletin 07/03.

Walmsley, Roy (2003) World prison population list (fourth edition).. Findings 188. London: Home Office Research, Development and Statistics Directorate.

Wojcik, Dobrochna (2004) ‘A letter from Poland: mediation in Polish penal and juvenile law.’ Crime prevention and community safety: an international journal, 6 (2), 61.66.

Wright, Martin (1982) Making good: prisons, punishment and beyond. London: Burnett Books and

Wright, Martin (1999a) ‘Key questions on victim/offender mediation.’ In: Juvenile offender-victim mediation, ed. by Beata Czarnecka-Dzialuk and Dobrochna Wójcik. Warsaw: Oficyna Naukowa.

Wright, Martin (1999b) Restoring respect for crime. Winchester: Waterside books.

Warsaw CJS burden 14pt spkg 049.doc

1 Many burglaries could be included in the previous example, because the Home Office’s British Crime survey has found that about half of burglary victims know the offender (Independent, 4.5.2004)

2 The problem could be avoided by reducing the sentence without reducing the charge, as has been proposed by the Sentencing Guidelines Council in England and Wales (Daily Telegraph 20.9.2004)

3 There is a possibility that victims may feel that they are placed under pressure by having to decide whether to accept mediation, because if they refuse, the offender may receive a punitive sentence instead of a restorative one. This can be avoided by informing both the victim and the offender that if the offender is willing to make reparation, but the victim does not want to take part, the offender will still receive a restorative sentence such as community service (Wright, 1999a).

4 ‘The community’ is placed in quotes because there is some discussion about what the term means. Here, it may refer to individuals who offer help or employment, NGOs which offer opportunities for community service or for treatment or training, or local authorities which provide services as necessary.

5 In England and Wales, police can also reprimand or give final warning, which may be combined with mediation.