Restorative justice: for whose benefit?

 Paper to European Forum for Restorative Justice Conference, Leuven, Belgium, October 1999

 In 1977 Professor Nils Christie, of Oslo University, published his seminal article ‘Conflicts as property’, in which he argued that the lawyers, probation officers and other professionals had ‘stolen’ conflicts, including crimes, which ‘belonged’ to the people directly involved. (Even mediators might do so if they become too professionalised.). Will they now restore the conflicts to their owners? In some countries restorative justice has been put into effect in a way which makes sure that the courts and their officials retain control; they make the decisions, even if the victims and offenders are then asked for their consent. Developing an idea suggested elsewhere (Wright, 1999: ch. 6), I will refer to this as ‘authoritarian’ restorative justice.


Many practitioners would, I believe, support a different approach, which might be called ‘democratic’ restorative justice. As a mediator, I do not want to see a confrontation between these two models. Courts will continue to be needed, to supervise the process and to deal with cases where the accused denies guilt or where the offender and victim do not agree about reparation. So at the end I will suggest how the ‘poachers’ could find a new role as ‘gamekeepers’. I should like to suggest that there is no hard-and-fast division between measures that are restorative and those that are not. Rather there is a spectrum, and to simplify I will take three points on the continuum: the above two, and ‘unilateral’ restorative justice.

1. ‘Unilateral’ restorative justice

1.1 Measures which are not based on punishment are intended to create benefits for either the offender or the victim and could be described as ‘unilaterally’ restorative. There have been measures to ‘restore’ or rehabilitate offenders throughout the twentieth century, as a reaction against the severity of the nineteenth with its solitary confinement, enforced silence and the deliberately useless labour of the crank and the treadwheel. Restorative measures for victims came later, with the introduction of state compensation for victims of violent crimes, in New Zealand in 1963 and the United Kingdom in the following year. Other countries such as France and Germany have allowed victims to join civil proceedings to the criminal process, to obtain compensation. In Britain civil proceedings are separate (and seldom used after a crime), but since 1972 the criminal courts have been able to order offenders to pay compensation. Compensation should take priority over fines and court costs, and can be used without any other sanction; since 1988 English courts must give reasons if they do not make a compensation order, and when the Crime and Disorder Act 1998 comes fully into force in 2000 they will also have to give reasons for not making a reparation order where they have power to do so. Compensation is usually paid in small, and often irregular, instalments, which are of limited use to victims, but the measure’s intentions are partly restorative (although it may also be used in a punitive spirit).

1.2 A more directly restorative development came from the community: Victim Support (first piloted in 1974) arranges for trained volunteers who are ordinary members of the community to offer support to people who have been victims

1 of a crime. In England the service has now been extended to give support to victims (and other witnesses) when they have to attend court.

1.3 Another partly restorative measure, introduced in England in 1972, is the community service order, in which the offender is required to do some form of work, usually for a non-governmental organisation (NGO, otherwise described as a non-profit, charitable or voluntary organisation); if possible the offenders are given the task of offering help to people who need it, for example because they are elderly, disabled or otherwise disadvantaged; sometimes they can work alongside members of the community who are working entirely voluntarily. Its aims have however been confused: it was also presented as being both rehabilitative, because it builds the self-confidence of offenders, and a punishment, because it deprived them of free time. The Conservative Home Secretary Michael Howard decreed that community service orders should include ‘hard manual work’ which is ‘physically, emotionally or intellectually demanding’ (Home Office et al., 1995). There is no objection to hard work in itself, but it is essential that punishment is not its primary aim, because that would give offenders entirely the wrong message. It would seem to say that working to help other people is unpleasant and should be avoided, whereas it should be experienced as rewarding and fulfilling. The more punitive it is, the less it is in anyone’s interest. In the early 1990s some young offenders helped to take aid to orphanages in Romania, which was helpful to the Romanian children (we hope), and probably increased the young people’s human understanding and self-esteem; but some short-sighted politicians objected, because the experience of a trip abroad might also have been enjoyable.

These are worthwhile initiatives, but they are not fully restorative because they aim to help either the victim or the offender, not both, nor do they promote communication between them.

2. ‘Authoritarian’ restorative justice

Authoritarian restorative justice is basically paternalistic; it assumes that people in authority know what is best for others. Three characteristics are: firstly, that decisions are made by the authorities (courts, probation officers, report writers), and these victim/offender mediation services are more likely to be organised by the authorities with little involvement of NGOs; secondly, that there is a narrow interpretation of reparation, with the emphasis on the outcome rather than on the process; and thirdly, that there is a tendency for it to focus on the offender more than on the victim, and to be applied in a punitive way.

2.1 In authoritarian restorative justice, decisions are made by courts or other people in authority; victims are merely asked for their consent, and are not involved in formulating solutions and carrying them out. Although the social worker who writes the report for the court may discuss the reparation informally with the victim, the latter is asked to decide before having had communication with the offender. The priorities are those of the government, which are not necessarily those of the participants. In England, for example, the government places great emphasis on speed. What happened was this. The conventional courts were very slow, and politicians decided that they must be speeded up. For persistent young offenders the average time from arrest to sentencing was 142 days; instead of trying to reduce the maximum, a target was set of reducing the average time to half that, or 71 days (Home Office 1997, paras. 7.2-7.3). Strong pressure has been placed on courts , and in some cases the time from the charge to the offender’s first court appearance can be as little as four working days. From the point of view of dealing with offenders this is commendable, but it gives victims very little time to consider whether they want to take part in mediation. According to workers in English restorative justice programmes, there are areas where courts adjourn (‘stand down’) the case for only two hours for reports for reparation and action plan orders to be written, so that victims are contacted very hurriedly by telephone or not at all. Where victims are involved, a different criterion should be used, such as the time from arrest to mediation.

2.2. The second feature is a narrow interpretation of the idea of reparation. The British Home Secretary, Jack Straw, has spoken of reparation in rather limited terms: it might include victim/offender mediation, but this is not seen as a part of the process of agreeing what the reparation should be. The official guidance recommends social workers, writing a report before a reparation order is made, to suggest a type of reparative activity related to the offence, rather than to the offender; for example, washing a car which he had attempted to steal (Straw, 1998). These are all valid, as far as they go, but they do not include several aspects which are usually regarded as essential in restorative justice: They are determined by what someone in authority wants, not on what the victim wants or what the offender is best able to do.

In this way the offender is not given the chance to offer reparation, but is ordered to make it. Victim/offender mediation is a possibility, but only after the court has decided what the reparation should be. The victim is consulted about reparation, but is asked to decide before meeting the offender. So only a small part of the conflict has been given back to its ‘owners’; most of it is kept under the control of the criminal justice authorities.

2.3 When the focus is on the offender and the offence, not the victim, it is often a sign that an ostensibly restorative sanction is being applied punitively. Another sign is when its nature and length are related to the offence rather than to what the victim and the offender agree to be appropriate. For example: there have been cases where they had a good discussion and agreed on the reparation, but the court, thinking in terms of ‘proportionality’, increased this amount. This left both victim and offender frustrated because their agreement, and their feelings, were overruled. Where a victim does not wish to meet the offender, a court may decide that a reparation order should last for x hours, but the most relevant form of reparation could be taking part in a group of victims and offenders (Launay and Murray, 1989), even though it would not last so long, and it would also offer some support to victims whose ‘own’ offenders were not caught. The Home Office, as noted above, has stated that the reparative work should be related to the offence; but it is widely recognised that many offenders lack a sense of self-worth, and quite different work with people less fortunate than themselves (perhaps old, young, ill or suffering from disabilities) could give them a sense of achievement which would raise them in other people’s eyes and their own. To show respect to the offender, courts should allow flexibility about the nature of the work and the exact number of hours, work should if possible be of a kind which uses his abilities and strengths. It should ‘teach him a lesson’, but the lesson should be that he has a valuable contribution to make, and will be valued for it.

Restorative justice is authoritarian when it is described by politicians as a ‘tough’ process in which the offender is ‘confronted’ with the effects of his actions. It might involve shaming the offender (about which I shall say more later, in section 4). It tends to be reserved for relatively petty offences and for juvenile offenders; and often used in place of (or even in addition to) another non-custodial sanction, not in place of prison. It has a tendency to be unilateral, in that it focuses more attention on the offender than on the victim; there are often references to ‘the family’, meaning the offender’s family (as if victims’ families were less important), and programmes are often limited to young offenders or minor offences, or both, which means that no benefits are offered to other victims. The content of a reparative order is regarded as a matter for the courts; victim/offender mediation is seen primarily as a way of confronting the offender, and if conferencing

2 is proposed, its purpose is mainly to consider ways of supporting the offender while he makes the reparation which the court has ordered – both worthwhile objectives, but ones which should be balanced by benefits to victims, for example providing them with answers which only the offender can give, and letting them tell the offender directly how the crime has affected them. Authoritarian restorative justice does hope for such benefits for victims, but they appear to be seen rather as a by-product. English colleagues may recognise the above as a description of the reparation orders introduced in England and Wales under the Crime and Disorder Act 1998; several of the phrases I have used are taken from the Draft Guidance Documents on Reparation Orders and Action Plan Orders issued by the Home Office in 19983. Other problems, some of which could be classed as ‘authoritarian’, are discussed by Zehr and Mika (1998, 50). Colleagues in other countries may wish to make an ‘audit’ of their own system, to assess where on the spectrum it can be placed.

In whose interest is this authoritarian approach? Politicians presumably think that it will encourage people to vote for them. In Britain, the Labour Party made a number of promises before winning the general election of 1997, and wants to prove that it can keep them. Some of these promises have their origins in old ways of thinking, for example the idea of being ‘tough’ and speeding up the criminal process; this has its merits in the context of conventional justice, but it does not allow time for the victim to decide whether to take part in mediation, nor for the mediators to arrange a meeting. Criminal justice personnel are in the habit of being in control, to make sure that the system runs in the way they want it to. In a restorative system their role will be to provide safeguards, for example to ensure that the offender is not asked to make excessive or demeaning reparation. (I will say more about this in section 5.)

3. ‘Democratic’ restorative justice

What are the features of ‘democratic’ or community-based restorative justice? Firstly, it is operated as far as possible in, and by, the community; secondly, it includes a wider concept of reparation; and thirdly, it aims to benefit both the victim and the offender.


To explain the principle, I should perhaps begin by explaining how I believe social policy operates. (Once again, I am speaking from a British perspective; the picture, or at least the emphasis, will be different in different countries – see Faget, this volume.) There are three basic ways in which social policies can be put into effect. Individuals and businesses can pay national and local government to do what is necessary. Or they can give donations to non-governmental organisations (NGOs) to do it. Or they can do it themselves, as volunteers4. In fact, of course, there is a mixture of these methods. In some countries it may appear that there is not so much voluntary work, but many individuals are providing services, for example by looking after relatives who are young, elderly, or disabled, and it does not occur to them to describe themselves as ‘volunteers’. The local or national government may give funds to NGOs; state agencies and NGOs may use volunteers as well as paid staff to deliver their service. You could say that the ‘state’ is national and local government, and the ‘community’ is everyone else; but individuals pay the taxes so that the state can do what we want it to do (although what it does may not always exactly meet our wishes or expectations!).

We can look at it from two points of view: (a) money and (b) work.

(a) – Money

  • national and local governments provide funds to state services such as hospitals and social work departments, and to national and local NGOs; national government gives tax incentives to those who make donations to NGOs.
  • individuals and businesses pay taxes to national and local government (and individuals elect the representatives who will decide how the money is spent)
  • charitable trusts (often created by wealthy individuals or businesses) provide funding, usually for innovative projects; when these are well established, the government (ideally) funds most of the continuing running costs (but not the whole cost, so as to preserve the NGOs’ independence).
  • individuals and businesses make donations to NGOs to cover the rest of their costs; the state gives financial assistance to parents and other individual ‘carers’ (or reduces the amount of tax they have to pay). (b) – Work
  • national and local governments provide services
  • NGOs provide services; these are usually operated locally, supported and supervised by national ‘umbrella organisations’ such as Mediation UK or Victim Support
  • Some NGOs employ paid professional staff, and use volunteers only to help to raise funds.

individuals care for relatives or neighbours. If they want to volunteer in a more formal way, they join one of the NGOs in which members of the public, in their spare time, are trained and then work as volunteers to provide services. (sometimes they also work in government agencies, for example as classroom assistants in schools). In each of these two groups, the first item represents activity by the state, and the last is the most ‘democratic’ or ‘community-based’. There are many examples to show that NGOs with adequate funding, and lay people with adequate training and supervision, can deliver a professional service (although most NGOs also employ full-time professional staff). NGOs have an additional advantage over statutory agencies: because they are independent, they can criticise government policy. They have to do this with restraint, of course, because much of their work depends on funding and good relationships; but wise governments and civil servants will welcome informed comments and suggestions from NGOs and take notice of them.

This description goes some way towards defining that chameleon word ‘community’: it includes people as individuals (including victims, offenders, their families and friends), and those who support them, voters (who have some influence on government), community groups (youth clubs, NGOs), employers (as providers of jobs), and local authorities (as providers of services). The democratic state is an ox – if we feed it, it will pull our cart in the direction we want it to go; and it is a cow – if we feed it, it will provide milk for us and our needy brothers and sisters. (This is not an analogy that should be stretched too far!)

3.1 Democratic restorative justice benefits the community – but authoritarian restorative justice also claims to do that, so what is the difference? It is more democratic when as far as possible it is operated by the people directly involved. This is similar to the ‘New Careers’ concept of the Office of Economic Opportunity in the United States in the 1960s: that people involved in a social problem should be part of the solution to that problem. A word often used in the 1990s is ‘empowerment’; some people don’t like that word, but ‘enabling’ will do as well.

Systems which put democratic restorative justice into practice will be as community-based as possible. They will show respect for all human beings, including for example, powerless people such as children, members of ethnic and other minorities, illegal immigrants, mentally disturbed people and offenders (or accused persons). Obviously they will have to be prepared to use coercion in the last resort, but the principle will be to try to use voluntary methods first. The primary aim will be to persuade and enable, and to act as a model for the way they want their citizens to behave. They will operate as close as possible to what the British call the grass roots; they will be, as Germans say, bürgernah, close to the citizens. Perhaps there is a parallel with what the European Union calls ‘subsidiarity’, the principle of devolving power to the most local level that is appropriate. This includes the idea that as far as possible people should be encouraged to do things for themselves, rather than have them done to or for them.

Firstly, in contrast to the authoritarian tendency, democratic restorative justice allows victims and offenders and their supporters to make decisions about reparation as far as possible (but not about punishment

5), and mediation services are organised by NGOs rather than by state agencies. There will also be community involvement to enable offenders to make reparation: for example, businesses will be encouraged to provide jobs; NGOs will provide and supervise community service tasks; and the local authority will provide resources such as training in social skills, treatment for addictions, education for children with behavioural problems, and facilities for recreation, and will use the information gained to improve its. crime reduction strategy. Individuals in the local community can also contribute by being trained to become lay mediators, both for victims and offenders and to mediate in disputes between neighbours which sometimes lead to criminal acts.

Relatives of victims and offenders are also members of the community. They are assumed to be competent to make decisions, especially when the resources of the extended family are brought in; they are not, as sometimes happened in the past, presumed to be dysfunctional or deficient. (Weitekamp 1997). This is part of the advantage of conferencing (see footnote 2).

It is fair to point out that a community-based service may also be capable of acting in an authoritarian manner, and a system-based one may follow a ‘democratic’ ethos. The ultimate aim is to change attitudes in the community and the system. As Bazemore and Walgrave say (1999, 47, 52), ‘It is also possible to imagine a transformation of the formal system to make it operate in a more informal manner as well’; ‘an emphasis on both formal and informal justice processes seems necessary.’ The change can take place in either direction – ‘top-down’ or ‘bottom-up’ (Faget, this volume). I have assumed that it is more likely that pioneers in the community (for example, an independent mediation service) will gradually influence the system; but it is also possible for far-sighted individuals within the system to initiate change which can then spread to the community. It could even start in both places.

3.2 Secondly, democratic restorative justice takes a more holistic approach to reparation. It attaches importance not only to the outcome (the work, apology, compensation and so on) but to the process, the communication between the victim and the offender. Another way of making reparation is seen as co-operating with any rehabilitative programme that meets the offender’s needs and helps him to avoid re-offending (as I have suggested elsewhere: Wright 1998). This is obviously in the interests of the community and of the offender himself, but it is also what many victims want; it may not be possible to repair the harm they have suffered, but they want the offender to do something that will make him less likely to do it to anyone else. This is similar in some ways to rehabilitation, but it includes the additional idea that the offender is showing active regret (a similar idea in Austrian law is called tätige Reue). This overcomes the objection sometimes raised to rehabilitation – that it appears to reward offenders. Offenders are held accountable, and ‘their re-integration depends on their willingness to make efforts to restore victims and the community’, bringing together the ideals of restorative justice and rehabilitation (Levrant et al. p. 22-3). It could be said that offenders are entitled to a fair opportunity to make reparation.

3.3 Thirdly, the importance of aiming to benefit both victim and offender will be recognised. Democratic restorative justice treats everyone equally, but victims ‘more equally than others’; in other words, the first question it will try to answer, when deciding on new measures in the field of criminal justice, will be: What will be the effect on victims? In Britain the Crime and Disorder Act 1998 puts the emphasis on general crime prevention strategies carried out by local authorities and the police, rather than the courts, and if successful this will mean that there are fewer victims. But when a crime is committed, it considers first the action to be taken in regard to the offender, such as the requirements to be included in a reparation order, and only then does it say that the attitude of the victim(s) should be ascertained. Victim/offender mediation is seen as one of these requirements; but in democratic restorative justice it will be part of the process of deciding what the requirements should be (provided that the victim and the offender are willing to take part). To show respect to the victims, the potential benefits to them should be at least as important as the effects on the offender. Preferably the initial interviews with both should be carried out by the same person, who should be a trained mediator.

Conventional justice aims to be fair to everyone, but in practice there is a tendency for people to be treated less favourably if they are poor, inarticulate, do not speak the language of the country, or come from ethnic minorities. Restorative justice could fall into the same error, and should be monitored to make sure that it does not do so.

4. Comparisons and contrasts

Some criteria are beginning to emerge for assessing where on the spectrum a particular version of restorative justice lies. As we have already seen, decisions may be taken mainly by the authorities or by the participants. Reparative measures may be proportionate to the seriousness of the offence, or to the wishes of the victim and the abilities of the offender. Some versions are limited to young offenders or certain types of (usually petty) offence, at an early stage in the criminal justice process, which means that other victims are excluded; others are open to all (subject to individual assessment), before, during or after the official process. In some, there is pressure on the offender to take part, and the victim may not be given full information nor the time to consider it before deciding; others recognise that the process should be voluntary for all concerned. In some it is seen as just another programme which may be used in the context of a conventional, retributive justice system; others, although they recognise that at present restorative measures have to be used within the existing framework, are working towards a system which is increasingly restorative in its whole philosophy. Bazemore and Walgrave (1999, 64) make a similar point. Other differences, already mentioned, are that reparation to the community (community service) may be intended to be primarily punitive, rehabilitative, or reparative; reparation to the victim may emphasize work or monetary compensation, or may broaden the concept of reparation to include the process and any action by the offender to improve his chances of avoiding re-offending.


 Comparisons and contrasts

 Authoritarian <————————————-> Democratic

 authorities, courts decide              participants decide; community involved                         
proportionate to offence                related to victim’s needs/wishes, offender’s abilities
young offenders                                 all ages
petty crimes                                        all crimes
offender compelled                         offender can offer reparation first
 pressure on victim, speed important
                                                                informed consent, time for reflection
add-on to retributive system     working towards restorative system
punitive community service      reparative community service
narrow reparation                        reparation includes process and rehabilitation
                                                                    (courts supervise)

 One concept that may be found in either type of restorative justice, depending on how it is interpreted, is ‘reintegrative shaming’. John Braithwaite, who coined the term, saw the process as a basically constructive one; the offender’s act is vigorously condemned, for example in a family group conference while recognising his (and everyone’s) human worth and potential to reform (Masters, 1998: 125-6). The process encourages the offender to think of the victim, which in turn encourages the victim to think of the offender as a person, distinct from his behaviour. The victim describes the effects of the crime; this may make the offender feel ashamed, but it does not cause a lasting stigma.. It is intended only for the purpose of helping the offender to understand, and hence to feel regret, apologise, and often to offer reparation. This can cut the fuel supply from the victim’s anger, helping him or her to overcome the effects. In some cases it may lead to forgiveness, which is seen as a benefit for the victim as well as the offender.

It is the word ‘shaming’ that makes many people uneasy. The principle is clear: that the offender should be encouraged to understand how his act has harmed another person. He should feel sorry for that reason, not because he is going to be punished, and when he has shown that he is sorry, he should be reintegrated. But in our culture which is still predominantly retributive, it is too easy to concentrate on the shame and to forget the reintegration. I have been told by a campaigner for restorative justice in Australia, Braithwaite’s home country, that a television programme there showed a confrontation between the parents of a young murder victim (now separated and each present with their new partners) and the two young men convicted of his murder, who were supported only by the very distressed mother of one of them. The police officer acting as ‘mediator’ did not use the important material that was exchanged between the participants; there was little apparent structure and no outcome, but at the end the deceased boy’s mother placed his ashes at the feet of the convicted men; this did not seem to help any of those involved to put the event behind them. It was described as ‘restorative justice’, but it shamed the offenders without producing any agreements about how they could express their regret or rebuild their lives. Much of it seemed to at least one viewer to be a demonstration of the destructive aspects of a punitive system (Benjamin, 1999). A Swedish colleague, who worked for many years reforming the system from within, says: ‘I think it would be better not to use the words shame and shaming at all: … a sort of vocabulary not known to the lips of Socrates or Christ’ (Marnell, 1999).

Of course this may have been an isolated example, but it points to he risk that the term ‘reintegrative shaming’ can be misunderstood. This intuitive idea is supported by empirical research: offenders are more likely to re-offend if they (or their parents) felt shamed at a family group conference, but re-offending is less likely if the conference is conducted in a way that avoids stigmatic shaming but leads them (inter alia) to show remorse and feel that they have repaired the damage, they are less likely to do so (Maxwell and Morris, 1999). It might help if a new term could be found, such as ‘restoring respect’; this implies a temporary loss of respect, but it emphasises the positive end-result of the process; including growth in both the offender’s self-respect and the respect of others for him – not least the victim’s recognition of the offender’s courage in facing him and apologising. Allison Morris has suggested the term ‘reintegrative remorse’.

Restorative justice does not attempt to reduce crime by general deterrence; instead, there should be a crime reduction strategy. In England and Wales the Crime and Disorder Act 1998 (section 6) requires local authorities and chief police officers to formulate and implement such a strategy. This could be taken one step further by encouraging mediation services (or a probation officer or community member present at the conference) to pass on, without breaching individual confidentiality, information about possible criminogenic influences and remedial action, as indicated above (section 3.1).

5. For whose benefit?

5.1 The participants

The benefits claimed for participants in victim/offender mediation have often been repeated; here it may be worth remembering that they are conditional upon good practice.

5.1.1 Victims can benefit, provided that the system takes as much account of them as of offenders; that their safety (and if necessary anonymity) is assured; and that they are not encouraged to have too high expectations. Some may be helped to come to terms with deep feelings; others may simply feel that they have done what they could for the community and the offender. The more serious the crime, the greater the potential benefit; limited restorative resources should be concentrated on offences which victims regard as having had a great impact on their lives.

Advocates of restorative justice should however be cautious when explaining the idea to victims: not all of its advantages will be experienced by all of them. If they have unrealistic expectations, they are likely to be disappointed. There is evidence that many victims find it important to receive answers from the offender, to tell him or her the impact of the offence, and to receive an apology; a smaller proportion of victims wanted compensation from the offender (in a study of English victim/offender mediation services: Umbreit and Roberts, 1996). Other surveys have found that it is common for victims to want to encourage offenders, especially young ones, to make better use of their lives – either from a sense of public duty, or from empathy with the offender (Marshall and Merry, 1990); many victims, or members of their families, have themselves been in trouble with the law. All of these aspects of victim/offender mediation and conferencing offer victims advantages (provided that they take part willingly, and do not find it burdensome) which the conventional procedure cannot provide, even if it has been improved to make it more ‘victim-friendly’.

5.1.2 Offenders benefit by being treated as responsible and required to accept responsibility, provided that they are not stigmatised, and that they also accept the opportunity to make reparation (in the wide sense referred to above, section 3.2).

5.1.3 The community benefits because offenders are more likely to admit their acts, to understand their effects on others, and to speak openly about the pressures that led them towards crime, so that lessons can be learnt. Offenders are also more likely to make reparation if they have agreed to do so; but the community must, in fairness, make the reparation possible. This means that employers have to be willing to employ them, social and educational services have to enable them to acquire the life skills and vocational qualifications they need, and taxpayers have to make this possible. Individuals should be involved as much as possible, for example as mediators and mentors. The community should also take note of information relevant to crime reduction strategy, which is passed on by the mediation service

6; in this context, as I suggested earlier, the ‘community’ means the local authority and in some cases national government, which will need to address the structural inequalities in society that can increase pressures towards crime (Levrant et al., 1999, p. 14-15).

5.2 Justice officers

The development of restorative justice will depend on securing its acceptance by criminal justice officers – judges, prosecutors and police. This will not be obtained if they believe that their work will be more time-consuming or less interesting, or if they consider that it endangers the rights of victims or offenders. They are accustomed to making decisions that affect people’s lives; even probation officers, in England at least, are required to exercise an increasingly controlling role. What will be the effect on them of a new way of working which enables victims and offenders to make their own decisions? Will it diminish their authority? On the contrary, in each case the officers can be reassured that they would keep at least one of their principal existing functions; in other ways, their role will be enhanced, and their workload may be reduced as cases are diverted out of the system. They will no longer simply make day-to-day decisions, but will oversee the system to ensure that it operates in a genuinely restorative way. They can be shown that restorative justice achieves many of the goals that they would agree to be desirable, such as looking after victims and holding offenders accountable. The structure of the criminal justice system puts the state and its officials in a parental role, and offenders in the position of disobedient children, and this attitude extends into authoritarian restorative justice. To work in a democratic system should be more rewarding, because it allows and encourages offenders to behave in a more responsible and adult way, and to think of others as well as themselves; it also involves members of the community.

5.2.1 In New Zealand courts, Judge McElrea has found that since the introduction of restorative justice, offenders are less likely to deny what they have done and challenge the prosecution to prove it; ‘the family group conference process encourages an admission of responsibility’ (McElrea, 1998). Where the accused denies involvement in the crime, judges retain the essential function of deciding guilt or innocence, but where the act is not denied, the courts’ workload has been reduced, while they take on the new role of acting as a safeguard for victims and offenders. They would also ensure legal protection against unwarranted intrusion by the state (Bazemore and Walgrave, 1999, 53). They satisfy themselves that conferencing was carried out in accordance with accepted standards; for example, if a court considers that a conference did not involve all the relevant members of the offender’s or the victim’s family, it can require the service to hold another one. In the few cases which are so serious that the offender must be restricted or detained for the protection of the public, this decision also continues to be made by the courts.

5.2.2 How does restorative justice affect the police? In New Zealand, where the police have discretion to refer cases direct to a conference rather than to a court, over 80 per cent of cases are handled in this way (McElrea, 1994, p.95). In England and Wales police discretion has been restricted by the Crime and Disorder Act 1998, but, as indicated below, prosecutors could refer cases if mediation services were more widely available and prosecutors were more aware of them. An important principle in a democratic system is to use the minimum necessary intervention. The processes of justice, including restorative ones, require time and money – another reason for concentrating on the most serious cases (see also 5.1.1 above). In some places, such as New South Wales, Australia, and Thames Valley, England, the police have been the ones who introduced restorative justice; this broadens the horizons of police officers, but that should not be the primary reason for introducing a new way of dealing with offenders, and it has been criticised for concentrating too much power in the hands of an already powerful agency (Sandor, 1994).

5.2.3 In countries which follow the ‘legality principle’ (mandatory prosecution), it may not be appropriate to increase the discretionary powers of the police; but perhaps those of the prosecutor could be extended, as in Scotland. There is a potential, especially in countries like Germany which have modified the principle of mandatory prosecution, to allow prosecutorial discretion to divert cases to mediation (§153a, German Code of Criminal Procedure; 216 ter in the Belgian Code). Scottish procurators fiscal already have this power. This reduces the workload of courts; the offender does not have the stigma of a criminal conviction, but is required to make reparation, which often includes referral to counselling for problems involving drugs, alcohol or violence. This would also be possible for English prosecutors under the Code for Crown Prosecutors introduced in 1994, but they have not yet done so (Wright 1994).

5.2.4 Social workers, including probation officers and youth justice workers (social workers who work with young offenders) have traditionally had a dual function (in addition to writing reports for courts)

7. Firstly, they worked directly with offenders as counsellors; probation officers would ‘advise, assist and befriend’ their clients. But, secondly, they also had to try to control the probationers, and bring them back to court if they did not do what they were ordered to do; in the 1990s this aspect has been reinforced, and in authoritarian restorative justice they are still the people who decide, or at least make recommendations to the court. In democratic restorative justice the victim and the offender agree on reparation (with their families and supporters in the case of conferencing); where does this leave the social workers? They would no longer exercise a ‘diagnostic’ function, which in practice can be a controlling or authoritarian one, with which many social workers are not comfortable; the object is to empower participants by allowing them to work out solutions. (Some social workers and other justice officers, of course, may choose to be trained as mediators.)

Instead, they would have three functions. Firstly, they would help to establish the victim/offender mediation service and be represented on its management committee

8. They would help to arrange its liaison with the justice system. They would monitor the performance of the victim/offender mediation or conferencing service, and would, for example, ensure that a follow-up system was in place, so that the service could contact victims and offenders after a few weeks (having asked, at the time of the mediation, for their permission to do so) to ask about their level of satisfaction with the process. Secondly, in the case of conferencing, they would have a direct role, giving the participants information about available services which could help the offender to make reparation and, where appropriate, providing them. Thirdly, they act as a link between the victim and offender on one hand, and the community as represented by the local authority on the other. If mediations and conferences show a need for, say, accommodation for homeless offenders, adult literacy programmes, mentors for young people, or detoxification units, the social workers should try to ensure that these are provided. Similarly, if they find a recurring pattern of circumstances leading to increased criminality, such as schools which exclude troublesome pupils, employers unwilling to employ school leavers, or lack of recreational facilities, they can draw this to the attention of the staff of the chief police officer and the chief executive of the local authority, who are now (in England) responsible for crime reduction strategy (as mentioned above, in section 4).

5.3 Other contributors to the process

Others who have a stake in the process include defence lawyers, victim assistance workers and mediators, who are employed by criminal justice or independent agencies, or work as volunteers.

5.3.1 Defence lawyers will still be available to advise their clients, and to represent those who deny the act with which they are charged. In New Zealand, the normal procedure when a young person has been arrested and appears in court is to appoint a youth advocate (paid by the court without means testing), to ensure that the legal rights of the alleged offender are protected (Stewart 1996, 71), and this seems a good model to follow. Where the offender accepts responsibility for the act, however, they will not be put in the position of trying to avoid a conviction by looking for legal technicalities, or to obtain a reduced sentence by making excuses for his behaviour, sometimes by trying to place blame on the victim; these practices are not uncommon under the Anglo-American system. The defending lawyer will be able to make sure, for example, that everyone understands that the accused has not lost any legal rights if he or she does not deny the act; if the mediation or conference fails to reach agreement, the case will go to court and legal defences (provocation, honest mistake, etc.) can be used.

5.3.2 For the victim assistance worker there are two possibilities. In countries like Britain where victim assistance is already well developed, they may support the victim through the mediation process; otherwise the mediators and the victim assistance workers may be the same people, and mediation may be seen as a form of support.

5.3.3 In the most democratic model of restorative justice the mediators themselves (like the victim assistance workers) are drawn from a wide cross-section of the public, reflecting different social and ethnic backgrounds. The function of the professional managers of mediation services (in England often called co-ordinators), is to recruit, train, supervise and support the lay mediators

9. In some places, however, the work is entirely done by paid mediators, who are required to take a longer and more detailed training, in other words to join a new profession. There are several reasons for questioning this trend. Firstly, it misses an opportunity to give power back to ordinary people, who have shown that they can handle the work competently. As Bazemore (1997, 11, italics in original) has suggested, ‘a victim/offender mediation program which uses only paid staff and does not consult neighborhood residents on their priorities [may be] simply in rather than of the community’. These concerns might be addressed by using a professional with a lay co-mediator who would be seen to be neutral, to work in pairs. Secondly, if mediation is undertaken by people who are also social workers or police officers, they are not seen to have the neutrality which is essential to mediators; this certainly applies when they are involved in the same case in a different capacity10. Thirdly, members of the public can spread their insights into the origins and effects of crime among their friends and colleagues (subject to the requirements of confidentiality). Fourthly, there is the question of cost. This should not be the primary factor; nevertheless, it is unlikely that the state will employ enough full-time professionals to handle more than a small fraction of the total number of cases, especially as it is widely regarded as good practice for mediators to work in pairs when visiting and mediating. It was one reason why Victim Support in Britain decided to rely on volunteers11.

This does not mean that there is no place for professional mediators; on the contrary, it enhances their role. They are responsible for recruiting, training, supporting and supervising the lay mediators. They administer the service and its relationship to the justice process, and have a vital role in regard to standards, for example by ensuring that mediation is fully explained to both victims and offenders, to enable them to make an informed decision whether to take part. They will be responsible for maintaining adequate records, including a regular follow-up of participants (with their consent) to assess their satisfaction with the process. All this makes the role of co-ordinators a highly responsible and professional one; and they will also probably want to undertake some mediations themselves, to maintain their own skills and to observe the performance of their co-mediators. It is likely that some people will begin as lay mediators, and later will become professional co-ordinators; this is another kind of ‘New Career’ (see above, section 3.1).

6. Conclusion

At the national level it is in the interests of everyone that

  • the mediation or conferencing process has clearly formulated aims and objectives
  • standards are drawn up, in consultation with practitioners. This is already being done in various countries (see for example Mediation UK 1995, Restorative Justice Consortium 1999, Servicebüro 1998) .
  • the standards are incorporated into training, and followed in practice; in addition, local and national mediation services will need to maintain working groups to consider new situations as they arise, and decide what should be the best practice.
  • systems for accreditation of services, mediators and trainers are in operation.
  • information about the pressures towards crime are fed back to those responsible for crime prevention.

Would restorative justice be in the interests of politicians? Would they feel able to obtain support for it from a supposedly hard-line public? The public is not as hard-line as is often boelieved (Wright, 1989; Weitekamp, this volume), there are punitive attitudes, but these are partly due to the encouragement of populist politicians and media. I do not believe that politicians would lose votes by setting new criteria by which the success of their policies could be judged, such as the percentages of victims satisfied with the restorative justice process, and agreements fulfilled by offenders.

Perhaps I may finish by suggesting a speech which a Home Secretary or Minister of Justice from any part of the political spectrum could confidently make when seeking re-election?

 “Our government has transformed the system of justice in this country by introducing victim/offender mediation and conferencing. Over 80 per cent of victims who took part have said they were satisfied or very satisfied with the process, not least because of the high proportion of agreements which the offenders completed. For those whose offenders were not caught or did not take part, victim/offender groups are now organised in most parts of the country, in which victims can meet offenders who committed similar crimes.

As regards offenders, the move towards replacing punishment by constructive measures has brought a shift from the culture of excuses, denials, and attempts to secure acquittal on legal technicalities. More offenders now accept responsibility for their actions, recognising that they have not merely broken the law, but have harmed another individual; but if they take the opportunity to make reparation they will be given support as they make a fresh start. This has reduced the work-load of courts, and hence also the time taken to deal with cases. 

We are steadily reducing the proportion of cases that take over 28 days to reach mediation or sentence (as the case may be), partly by targeting resources on the most serious cases. The cost of the programme is more than covered by savings in expenditure on courts and prisons. These measures do not put the public at risk: offenders who present a serious threat to the safety of the public are still deprived of their liberty while they make their reparation.

It is too early to speak of a reduction in recidivism, but unlike many politicians we are taking a long-term view and we have every reason to believe that this will be seen within a few years, for three reasons. Firstly, the reduction in the working week, in addition to reducing unemployment, has made it easier to recruit large numbers of lay mediators, and this has increased public understanding that action to reduce crime has to be taken by local communities themselves, and should not be left to ‘the state’. Secondly, the mediation process has produced a flow of information to the local government and police officers responsible for crime prevention strategy. Thirdly, the reduced numbers in custody make it possible for prison staff to organise more constructive regimes.

All this represents, I believe, the growth of a more mature society. It marks a move from retributive justice (crime followed by punishment of the offender) towards restorative justice (crime followed by reparation by the offender). This is a challenge to all who think about the law: do we agree with the radical Dutch Professor Louk Hulsman, who said12 that a move from criminal law towards civil law makes both our justice and our country more ‘civilised’?

I am grateful to Ann Warner Roberts, Ursula Smartt, Gunnar Marnell and Ivo Aertsen for commenting on drafts of this paper.





Bazemore, Gordon (1997) ‘What’s “new” about the balanced approach?’ Juvenile and Family Court Judges, 48(1), 1-22

Bazemore, Gordon, and Lode Walgrave (1999) ‘Restorative juvenile justice: in search of fundamentals and an outline for systemic reform.’ In: Gordon Bazemore and Lode Walgrave, eds. Restorative juvenile justice: repairing the harm of youth crime. Monsey, NY: Criminal Justice Press.

Benjamin, Carmel (1999) Personal communication.

Christie, Nils (1977) ‘Conflicts as property.’ British Journal of Criminology, 17(1), 1-17.

Home Office (1997) No more excuses: a new approach to tackling youth crime in England and Wales. (White Paper) Cm 3809. London: Stationery Office.

Home Office, Department of Health and Welsh Office (1995) National standards for the supervision of offenders in the community. London: Home Office Probation Division.

Launay, Gilles, and Peter Murray (1989) ‘’Victim/offender groups.’ In: Martin Wright and Burt Galaway, eds. Mediation and criminal justice: victims, offenders and community. London: Sage.

Levrant, Sharon, Francis T Cullen, Besty Fulton and John F Wozniak (1999) `Reconsidering restorative justice: the corruption of benevolence revisited?’ Crime & Delinquency, 45(1), 3-27.

McElrea, F W M (1994) ‘Justice in the community: the New Zealand experience.’ In: Jonathan Burnside and Nicola Baker, eds. Relational justice: repairing the breach. Winchester: Waterside Press.

McElrea, F W M (1998) ‘The New Zealand model of family group conferences.’ European Journal on Criminal Policy and Research, 6, 527-543.

Marnell, Gunnar (1999) Personal communication.

Marshall, Tony, and Susan Merry (1990) Crime and accountability: victim/offender mediation in practice. London: HMSO.

Masters, Guy (1998) ‘The importance of shame to restorative justice’. In: Lode Walgrave, ed. Restorative justice for juveniles: potentialities, risks and problems for research. Leuven: Leuven University Press.

Maxwell, Gabrielle, and Allison Morris (1999) ‘Restorative conferencing for young people and re-offending.’ Paper at III International Conference on Restorative Justice for Juveniles, Leuven, Belgium.

Mediation UK (1995) Practice standards for mediation practice and the management of mediation services. Mediation UK, Alexander House, Telephone Avenue, Bristol BS1 4BS. (Revised edition published in 1998.)

Restorative Justice Consortium (1999). Standards for restorative justice. RJC, National Council for Social Concern, Montague Chambers, Montague Close, London SE1 9DA.

Sandor, Danny (1994) ‘The thickening blue wedge in juvenile justice.’ In: Christine Alder and Joy Wundersitz, eds. Family conferencing and juvenile justice: the way forward or misplaced optimism? Canberra, Australia: Australian Institute of Criminology.

Servicebüro für Täter-Opfer-Ausgleich und Konfliktschlichtung (1998) TOA Standards: ein Handbuch für die Praxis des Täter-Opfer-Ausgleichs. Das Servicebüro, Aachener Strasse 1064, D-50858, Köln, Germany.

Stewart, Trish (1996) ‘Family group conferences with young offenders in New Zealand.’ In: Joe Hudson, Allison Morris, Gabrielle Maxwell and Burt Galaway, eds. Family group conferences: perspectives on policy and practice. Annandale, NSW, Australia: Federation Press; Monsey, NY: Willow Tree Press.

Straw, Jack (1998) ‘Calling young offenders to account: UK applications of restorative justice.’ Speech at conference of Office of Public Management and Thames Valley Police, 29 October.

Umbreit, Mark, and Ann Warner Roberts (1996) Mediation of criminal conflict in England: an assessment of services in Coventry and Leeds. Center for Restorative Justice and Mediation, University of Minnesota, 386 McNeal Hall, 1985 Buford Avenue, St Paul MN 55108, USA.

Walgrave, Lode, ed. (1998) Restorative justice for juveniles: potentialities, risks and problems. Leuven: Leuven University Press.

Weitekamp, Elmar (1997) ‘The paradigm of restorative justice: potentials, possibilities, and pitfalls.’ Paper at 9th International Symposium on Victimology, Amsterdam, The Netherlands.

Wright, Martin (1989) ‘What the public wants.’ In: Martin Wright and Burt Galaway, eds. Mediation and criminal justice: victims, offenders and community. London: Sage.

Wright, Martin (1994) ‘New prosecution code may open way for mediation referrals.’ Mediation, 10(4), 8-9; see also ‘After mediation and reparation, prosecution may no longer be in the public interest.’ Mediation, 1999, 15(4), 13.

Wright, Martin (1998) ‘Restorative justice: the basic idea, and practice in the United Kingdom’. Paper to Seminar ‘Victim/offender mediation in adult cases’, Popowo near Warsaw, 2-4 September.

Wright, Martin (1999) Restoring respect for justice: a symposium. Winchester: Waterside Press.

Zehr, Howard, and Harry Mika (1998) ‘Fundamental concepts of restorative justice.’ Contemporary justice review, 1, 47-55.

9700 conf/leuven EFRJ 99A.doc


1 For the purposes of this paper, individual victims of crimes will be considered, which could include, for example, an employee of a bank that has been robbed; different considerations may apply where the victim is an organisation or there is no individual victim, or where the offender is an organisation, for example a company causing pollution.

2 ‘Conferencing’ is similar to victim/offender mediation, but the extended family of the offender, and often of the victim also, are invited to be present. The term ‘restorative conferencing’ covers both. Victim/offender mediation may be direct, where they meet in person, or indirect, where the mediators act as go-betweens; for simplicity, this paper will refer mainly to direct mediation.

3 The British government has set up pilot schemes to try out the new legislation, with evaluation by the Universities of Sheffield and Hull, and hopefully the system will be modified in the light of their findings.

4 For simplicity I will omit one other possibility: the individual can pay a professional person or profit-making organisation to provide the service. This may also be done through an insurance scheme, in which the individual pays premiums to the insurance company, which then pays for medical or other services in certain closely defined circumstances.

5 We have much to learn about reparation from traditional communities, but would not want to reintroduce the punitive practices of some of them, for example of some Aboriginal peoples where, after the community conference, the offender may be pinned to a tree with a spear through his thigh.

6 This information would be passed on in general terms, without reference to specific cases, because the mediation process must maintain confidentiality.

7 In some countries probation officers are not strictly speaking justice officers, although they work closely with the system.

8 In saying this, I am assuming that the mediation service will have a management committee, with representatives of different agencies (probation, education, victim assistance and to on) and also of mediators and of the local community.

9 In some places, the mediators are recruited in the same way, but are paid a fee for each case. It is argued that this helps in maintaining standards, for example by requiring the lay mediators to take part in regular in-service training; and it may make it easier for people with low incomes to volunteer. On the other hand, many volunteers feel that their relationship with victims and offenders is enhanced when it is known that they are working for the common good, and not for money.

10 ‘Mediators shold maintain clear boundaries between mediation and other forms of intervention’ (Mediation UK , 1995, 5); ‘Agents of criminal justice involved in formal action in a particular case should not normally act as impartial mediators or conference facilitators in the same case’ (Restorative Justice Consortium, 1999, I (b) 3); ‘Keine Überschneidung von Vermittlungsarbeit und Betreuungs- und Kontrollfunktionen für eine Partei im selben Fall’ [No overlap of mediation work and control or care functions for a party in the same case] (Servicebüro, 1998, 4.3).

11 Victim Support now has some 15 000 volunteers and nearly 1 000 staff in England, Wales and Northern Ireland (VS Annual Report 1999).

12 In a speech to the Howard League for Penal Reform, October 1976