The development of restorative justice

Paper for International Program on Victimization, mediation and restorative justice, Katholieke Universiteit Leuven, September 1997. Published in Juvenile offender-victim mediation, ed. by Beata Czarnecka-Dzialuk and Dobrochna Wójcik. Warsaw: Oficyna Naukowa, 1999.

A woman was threatened by a young robber with a knife when working in a convenience store. She was extremely traumatized, suffered nightmares, and her health deteriorated, but her family and friends were impatient with her, and colleagues even teased her because during the attack she had wet herself with fright. She was not invited to submit a victim impact statement because she was not identified as the victim – the store was. She wanted to be included in the process, and was terrified that the young man would come back to get her as he had threatened. She needed answers to the questions that had haunted her. The young man was sentenced to five years’ imprisonment, and when his parole hearing came up she travelled for four hours to attend it, but was not allowed to speak to him. Finally she was able to contact a victim/ offender mediation service, which arranged a meeting. The young man had no idea of the effect on his victim, and explained that all robbers say things like that, but don’t mean them. She told him that she forgave him and wished him well in the future. Since that meeting she has had no nightmares. Staff feel that it was a maturing experience for him, and there is a much better chance that he will respond to rehabilitative treatment. (much shortened, from Church Council on Justice and Corrections (1996: VII – IX)

This is an example of restorative justice in action, but not a `text-book example’: the victim was not involved in deciding how the offender would make reparation, and the offender was punished before he was given the oppor­tun­ity to make amends by meeting the victim. I have chosen it to suggest that restor­ative justice is not an all-or-nothing process: there can be degrees of `restorativeness’ even within the existing retributive system. Restoring does not have to be ordered through the criminal law. The ancient Israelites provided that if a man steal an ox he shall restore five oxen (Exodus 22:1); in some European jurisdictions a civil case is linked to the criminal one, and there have been suggestions (for example from Professor Louk Hulsman in the Netherlands and Sir Louis Blom-Cooper in England) that the criminal law itself should be `civilized’.

This paper will show some of the origins of restorative justice, and in particular the coming together of concern for the offender and for the victim. It will describe its early development in the form of victim/offender mediation, in Canada and the United States of America. It will take examples of the way the idea has taken root in Europe, from countries which have not changed their law (England and Wales), have modified it to facilitate mediation (Germany and Spain), or have incorporated it into (juvenile) law (Austria). It will then consider the re-birth of practices, restorative or communitarian or both, in countries which happen to be members of the British Old Commonwealth: New Zealand, Australia and Canada. (But perhaps this is no coincidence, because they are all countries where an indigenous culture, including ways of handling conflict, has been overlaid by the adversarial Western tradition.) Finally the paper will summarize features of restorative justice; some partially restorative measures; ways of promoting restorative justice; and conditions to be met if it is to fulfil its promise.


Mediaeval law provided for some forms of compensation by the offender to the wronged party, but it was, in England at least, superseded by punishment for many years (NB: punishment is here defined as the infliction of measures intended to be painful). There was a revival of interest in the congresses of the International Penal and Penitentiary Foundation; but it foundered on the problem of how to extract compensation from people who had no money (Tallack 1900). This was not resolved until the invention of community service as a sanction in the 1970s, and the introduction of victim/offender mediation which enabled victims to acknowledge the value of symbolic reparation. (For more detail see Wright 1996: ch. 1.)

The idea of restorative justice is to repair the harm caused by the crime; in particular to enable, or require, the offender to do so (when he or she is known), and to enable the victim (if he or she wishes) to take part in discussing what form this reparation should take. The community should play its part in making this possible; and relevant information gained about offences should be collected and passed to those responsible for crime prevention. The aim includes restoring what English law calls `the peace’, Americans `domestic tranquillity’, the Germans `Rechtsfrieden’ (peace under the law), and the Old Testament `shalom’ (a `profound sense of well-being’: Peake 1962: 474; a condition of `things being as they should be’: Zehr 1990: 130). Of course to put things back as they were is often not possible, nor can money compensate for the harm done, even if the offender could pay: we are talking about increased understanding, repaired relationships, making a new start. Anything the offender can do will be, for the victim, a symbol of that; it may consist in paying compensation, doing community service, or making an effort not to offend against anyone else, for example by taking part in a programme to tackle addiction, budgeting skills, improved literacy, or anger management.

Restorative measures for offenders have been around for a long time, especially for those who suffered deprivation in childhood, and further programmes to help overcome the after-effects of stigmatizing punishment – particularly of imprison­ment – such as unemployment, homelessness and the break-up of families. (It would be simpler not to send so many people to prison in the first place.)

The idea of helping victims was revived by Margery Fry, a leading penal reform­er; as a result of her efforts state compensation for victims of violence was introduced in New Zealand in 1963 and the United Kingdom in 1964. In some countries a civil action can be linked to a criminal case so that the victim can obtain compensation; in the UK the idea that the offender could be ordered by a criminal court to pay compensation to a victim was re-introduced in the Criminal Justice Act 1972. The same Act brought in community service orders, although it was not clear to what extent they were intended as a non-custodial penalty, an educative measure, or a way of making reparation (Wright 1984; Whitfield and Scott 1993).

A further move towards restoring victims was the introduction of victim sup­port, in which volunteers, members of the public with basic training, contact victims to see if they need help. This was started as an experiment in 1974, in Bristol, England, and has spread to the rest of the United Kingdom and to seve­ral other countries. Other services offer specialized help to victims of rape or family violence, or to the relatives of victims of homicide and road deaths, as well as supporting victims and other witnesses when they have to attend court. In addition to the individual assistance offered, these services have drawn attention to the previous disregard of victims’ needs and wishes in the criminal justice process. Much more is now being done to keep them informed and to consider their convenience, and indeed their right to be involved in the process, although there is still a long way to go, and there are question marks over their involvement in punitive, as opposed to reparative or protective, aspects.

These measures did something to help overcome the effects of the crime or of criminal justice, but they did it for victims and offenders separately. Making the link was the fruit of the imagination of a young probation officer in Ontario, Canada, as we shall see in the next section.

Reparative ideas were beginning to be discussed. An American criminologist sug­gested `creative restitution’ through payments or service to the victim or the community (Eglash 1958) and a British Home Secretary reflected on the idea of reparation to the victim, though without doing anything about it (Home Office 1959: 7). The Home Office did, however, commission the Hungarian émigré Stephen Schafer (1960) to review the question of restitution to victims of crime, and he brought together information from a much wider historical and geogra­phi­cal perspective than was usual in the English-speaking world. Schafer dis­tinguished `restitution’, the reparation of the victim’s loss, from `compens­ation’, a civil method in which the victim applies to society for amends for the dam­age or injury suffered. He saw the need to do something for the victim, and to enable the offender to earn money from which to pay restitution; but from the offender’s point of view he saw restitution as punitive (Schafer 1968: 114-5), and hence some way from the restorative ideal as it was to develop later.

Other writers began to question long-held assumptions. Louk Hulsman in the Netherlands pointed out that many things that can happen to us are as traumatic as being the victim of a crime, or more so – loss of a job, being injured at work, and so on – so that there was no reason to treat certain arbitrarily defined acts (or people) in a different way (Hulsman 1981). Using similar logic, New Zealand created an Accident Compensation Corporation which would compen­sate people for all types of injury, of which violent crimes were only a small proportion. Gilbert Cantor (1976) in the United States pointed out that if the court system were `civilized’, the civil standard of proof could be adopted; pre­sumably then more people who harm others would be brought to justice. An­other American, Randy Barnett (1977), pointed to some of the moral prob­lems of punitive sanctions: to inflict harm on one person in the hope of in­flu­enc­ing others is not justifiable, and does not work; even if it did, the aim of the justice system should be not the suppression of crime but doing justice to victims.

Another sanction based on making things better, although not for the victim, is community service. The Canadian Ab Thorvaldson (1978) proposed requiring the offender to `give something back to society’ as an aim of sentencing; this makes reparation to society, vindicates the principles of justice which the crime violates, and stresses offenders’ accountability for their actions. Although its aim is not rehabilitation, that is a common side-effect: community service helps offenders to acquire skills and self-esteem and to show regret for their offences – in short, it threatens to `beat probation at its own game.’ (This account is based on Wright, 1996.)

So there were moves to restore the offender, and to restore the victim; to require the offender to pay compensation or carry out community service. What the theory did not encompass at this time was bringing both victims and offenders together as part of the process. That was introduced into the process by practitioners; as far as theory is concerned, the next leap forward was made by Nils Christie of the University of Oslo, speaking in 1976 at the opening of the Centre for Criminological and Legal Research at Sheffield (Christie, 1977). Looking at the Western system from the point of view of an uncomplicated society in rural Tanzania, Christie said that conflicts ought to belong to the disputants, but had been `stolen’ by lawyers and other professionals. In criminal cases, this meant that the injured party, the victim, had been deprived of the right to take part. Not only that, but society has been deprived of an opportunity for norm-clarification. That is what the legal process is supposed to do, but on the contrary the law is concerned to make each case fit the Procrustean bed of law and precedent, winner or loser, believed or disbelieved, and in criminal cases guilty or innocent. This can be seen in the way lawyers demand yes-or-no answers from victims, whose evidence they want to fit into the appropriate category of `facts’; and they try to prevent the victim from saying things which the victim feels to be important, but which from a legal point of view are irrelevant, or damaging to one side’s case. The victim is left feeling that he or she may have told nothing but the truth, but certainly not the whole truth. In the mediation process, in contrast, people can say not which law was broken, but how the crime hurt them and what it made them feel, and offenders have the opportunity to acknowledge that. In short, the victim and the offender can relate to each other as human beings – whether or not there was a relationship before, which of course is often the case. Compensation and victim assistance can be helpful, but it is at this personal level that restorative justice can realize its full potential.

When was the term `restorative justice’ first used? I cannot answer with certainty, but in an early Mennonite pamphlet Howard Zehr (1980) describes the Victim/Offender Reconciliation Program (VORP), sometimes using the word `restitution’, and saying that `the aim is to restore persons to community’ (p. 18). In 1985, Zehr wrote a pamphlet in the same series, with the title Retributive justice, restorative justice, containing the first version of his much-quoted table of the old and new paradigms; after that the term was used more often, especially after his book proposing that the whole system should be viewed through a `restorative lens’ (Zehr 1990).

A recent re-formulation of the idea is set forth in Balanced and restorative justice for juveniles (1995), which stresses the importance of balancing the needs and responsibilities of victims, offenders and community. A problem, perhaps reflecting the source of the study’s funding (the US Office of Juvenile Justice and Delinquency Prevention), is that measures limited to juvenile offenders exclude victims of adult offenders. But a start has to be made, and the project can be extended later.


Most of these theoretical articles date from the mid-1970s or later; but the practitioners were there before them. It was in 1974 that Mark Yantzi, in Kitchener, Ontario, suggested that two teenage offenders meet their victims: the court agreed, reparation was made, and with backing from the Mennonite community the Victim/Offender Reconciliation Programme (`VORP’) was born (the story has often been told, for example by Peachey, 1989; Wright, 1996: 100-102). This is apparently the first programme in which victims and offenders were able to meet. VORP began to spread, to the Mennonite centre in Elkhart, Indiana, and other places in Canada and the US, and further afield.

The idea of resolving conflicts through face-to-face negotiation guided by mediators was meanwhile spreading in other fields. The Society of Friends (Quakers) gave training in non-violent techniques to marshals of civil rights marches, and the Friends’ Suburban Project in Philadelphia started the Community Dispute Settlement Program in 1976; ten years later the experiment was described in a book (Beer, 1986).

Other initiatives operated on both sides of the line that divides civil from criminal law. James Hughes, a prosecutor in Columbus, Ohio, found that much court time was taken up by offences such as assaults, or threats as a result of disputes; he knew about dispute resolution in the Van Der Hoeven Kliniek in the Netherlands, and thought that if it could work for the disturbed patients in a maximum security clinic, it should also succeed in more ordinary cases. He had two other important insights: that court hours are inconvenient to many people with jobs, and that trained volunteers can be competent mediators (as well as making possible a high case-load without excessive staff costs). His pioneering project, started in 1973, was called the Night Prosecutor’s Program. A dispute centre was established in New York in 1975 by the Institute for Mediation and Conflict Resolution; it dealt with large numbers of cases, some of them criminal, such as felony charges where the victim and offender knew each other, and took some pressure off the overloaded criminal courts.

The reparative ideal has deep roots in Minnesota, where Burt Galaway helped to establish a pre-release centre for prisoners to go out to work and repay their victims. John Harding (1982) reported that by 1979 some 200 restitution projects were operating in the US; most of them focused only on monetary restitution, but some offered victim/offender mediation as well.

Despite what was said above about victim/offender mediation, there was some early academic interest in community mediation. A Harvard jurist, Lon L Fuller, pointed out how mediation helps to create norms to suit situations, unlike law which tries to fit situations into laws and precedents (Fuller 1971: 308); and Richard Danzig (1973) proposed community `moots’, a suggestion which led to the establishment of Neighborhood Justice Centers, initially in three American cities. About the same time in San Francisco the Community Boards were established; unlike the NJCs they distanced themselves from the formal justice system, with a philosophy of community empowerment. This was more in line with Danzig’s ideas; he also worked with an anthropologist, Michael Lowy, who had suggested that moots like those in villages in Ghana could be used even in complex Western communities. This re-learning from pre-Western societies has been a strong component in the development of mediation, most recently in family group conferences in New Zealand and circle sentencing in Canada, as will be seen below.

The next phase in the development of victim/offender mediation, which was to become restorative justice, is one of the exchange of ideas. A number of people from the United Kingdom and other European countries, and Australia, made the pilgrimage to Kitchener, Elkhart, New York, San Francisco, and on their return they spoke and wrote about the new experiment and in some cases established projects in their own countries. These contacts were mainly made by visits: fifteen years ago the Internet had not been invented and even fax machines were uncommon. But exchange of information was not complete even within the same country: when I visited the US and Canada in 1983 I found that I had been in contact, from England, with some people who did not know of each other’s work.

It may be interesting to look at differences in the way the idea developed in different countries.


Victim/offender mediation has been or is being introduced in at least twelve European countries, although in some cases still on a small scale. By way of example we will look briefly at one country where there has been no change in the law (England and Wales), two where there have been changes to assist mediation (Germany and Spain), and one where it has been fully incorporated into the juvenile law (Austria). Brief details of the others are given by Wright (1996: ch. 7).

Among the English pioneers was John Harding. He had been in charge of one of the first community service order projects in the 1970s; in 1983, as Deputy Chief Probation Officer of the West Midlands, he commissioned a feasibility study (Wright 1983) which led to one of the first victim/offender mediation ser­vices, in Coventry. At first it was independent, but later it was taken over by the probation service, which reduced its community links, although it does have an advisory committee. Its running costs were somewhat high because it did not use volunteer mediators. Another project managed by the probation service was in Leeds, later extended to the whole of West Yorkshire. There the mediation is done by members of the community, who are trained but are not required to have previous qualifications; they are paid a sessional fee. Only about one in five of victim/offender contacts are face-to-face; it is commoner for the medi­ator to act as a go-between (indirect mediation). Mediators in that county have also taken on the work of contacting victims before the release of a prisoner, work which was entrusted to the probation servie by the Victim’s Charter of 1990 and 1996. It is felt that mediators, as neutral people, are well placed to under­take this work, and to enable victims to express concerns about the offen­der’s release; often parole conditions are added as a result, though the date of parole is not affected. When doing this work mediators do not expressly en­cour­age mediation, but facilitate it when victims or offenders spontaneously request it (West Yorkshire 1997). Currently some 25 services in England and Wales are operating or being established; most have about 100-400 cases referred per year.

The only area with substantial numbers of cases (about 1500 a year, most of them referred by the police) is the county of Northamptonshire, although it works with pre-court juvenile offenders only. This is probably because the decision to offer mediation is integrated into the criminal justice process, and it is significant that the service is managed by a committee representing all the main criminal justice agencies, as well as health, education and Victim Support. Lay mediators are not used. The name, “Diversion Unit”, suggests a focus on keeping young offenders out of the criminal justice system, but it takes account of the needs of victims; a survey found that 76% of victims were satisfied or very satisfied, and 92% of offenders (Northamptonshire 1995).

In England and Wales there have been changes in the law to accommodate compensation by offenders, and by the state, as we have seen, but none specifically to encourage victim/offender mediation. The same is true in Scotland, which has a different legal system; but there, as in other European countries, prosecutors (procurators fiscal) have deferred the decision to prosecute to allow mediation to take place (Warner 1992).

In an integrated system one more aspect needs to be available, but seldom is: arrangements for the many victims whose offenders are never found, and also for those offenders whose victims do not wish to take part. This has the further advantage of making participation in mediation genuinely voluntary for victims, because they know that if they do not do so, the offender can still show remorse in another way. Victim/offender groups of this kind have run in one or two places, for example in Rochester, Kent (Launay 1985, Launay and Murray 1989); a pilot programme has been run for armed robbers and victims in Long Lartin maximum security prison, which it is hoped to extend.

The German term is Täter/Opfer Ausgleich: the word Ausgleich (evening out – in the sense of making things level, not a social occasion!) implies successful mediation, and can perhaps be translated `reconciliation’. The word `Wiedergutmachung‘ (making good again, or redress) is also used; it conveys not only material but personal and interactive reparation by the offender (Netzig and Trenczek 1996: 247). German experts also studied (and visited) the American pioneering programmes in the 1980s. It was possible to introduce victim/offender mediation in Germany under existing law; a few youth workers had used it by 1984, but the first two projects, in Braunschweig and Reutlingen, started in 1985, followed in 1986 by the `Scales’ (die Waage) in Cologne (Schreckling 1990). One commentator saw V/OR as a move away from the `war against crime’ towards the `disarmament’ of the criminal justice system; but it was therefore regarded by some hardliners as a `trojan horse’, bringing in a change of paradigm, a criminal justice process aiming primarily at restoration and preservation of the peace (Pfeiffer 1992). The victims’ organization Weisser Ring also viewed it with some reservations, and warned that V/OR services should clarify their aims, and involve victims and victim organizations. They should not use mediation to improve the offender’s position in the criminal justice process, because of the danger of pressuring the victim to make an agreement (a point recognized in Austria – see below); and referrals should be accepted from victim organizations as well as criminal justice agencies, or at least the victim organizations should be brought in at an early stage (Eppenstein 1991). Given the tendency there has been in some quarters to operate V/OR services from the offenders’ point of view (Davis et al. 1988), these guidelines do not seem unreasonable.

A survey in 1996 found that V/OR is now available in 368 places in Germany, with about 9 000 offenders and 8 000 victims referred in 1995. Distribution, however, is uneven, with about 150 services receiving less than 50 cases a year. A quarter of the services are run by private non-profit agencies, but these account for half the cases. More detailed research is in progress on services and individual cases (Hartmann 1997).

The German Code of Criminal Procedure (sec. 153a) had already been amended in 1975 to enable the prosecutor, with the agreement of the court, to require the offender to pay a sum to the victim or a voluntary organization, or the state, or to perform community service, `if thereby the public interest in criminal pro­secu­tion would be removed’. The German Penal Code made some references to making good the damage and reaching agreement with the injured party (Dünkel and Rössner 1989), but a new sec. 46a was introduced in 1994: repar­ation or mediation is not mandatory, but courts must take account of it or give reasons why not (Hartmann 1997: 2). As in other countries practitioners have found that it is not always clear what is a criminal case or a civil one, or even which person is the offender and which is the victim (Netzig and Trenczek 1996: 246).

In Spain, it is traditional to put the interests, and education, of juvenile offenders first; from their point of view taken in isolation this could be regarded as moderately restorative. More recently, legislation such as the Act 35/1995 gives power to central government, in association with the 17 Autonomous Communities, to promote assistance to victims and state compensation for victims of violent and sexual crimes. Judges, prosecutors and others are given a duty to inform victims about state compensation and how to obtain reparation and the assistance of a lawyer at the time of trial and sentence; in `public crimes’ the prosecutor also conducts civil action for compensation in the name of the victim. For adults, the Penal Code of 1995 introduces community service (with the person’s consent and with respect for his or her dignity), but there are no provisions relating to V/OM.

Victim/offender mediation for juvenile offenders was introduced in Catalonia in 1990 with no new legislation, and Professor Antonio Beristain has proposed the use of yet another adjective: `recreative’ justice. Other criminologists have said that penal law could be `mercantilized’, and if (as in civil law) the reparation were focused primarily on deciding amounts of monetary compensation that would be a risk. Justice based on fully restorative principles would not, however, fall into that trap. In 1992 the Tutelary Courts for Juveniles Act provides three possibilities of reparation by juveniles to victims in all Autonomous communities of Spain: the prosecutor can propose that the judge discontinue the case if the juvenile has repaired the harm to the victim or agreed to do so; the judge can suspend the sentence for up to two years if the juvenile (with his parents or lawyers) and the victim agree on reparation; or the court can order community service. The first two of these provide scope for mediation but do not necessarily involve victims, who are assumed to agree if, when summoned, they do not object. Catalonia remains the centre of most V/OM activity; since 1996, all cases go to the mediation service for a decision whether mediation is appropriate. If the victim does not want to participate, reparation can involve community service (the foregoing is based on Varona, 1996 and personal communication).

Austria, like Germany, traditionally follows the `legality principle’ of mandatory prosecution; and like Germany, has felt the need to modify this. In Austria it is, as Christa Pelikan explains, the juvenile law that has been changed, in 1988, to allow an indictment to be dropped after conflict resolution has succeeded, or at least an effort by the young person has been recorded by the court. This followed a pilot project from 1985 to 1987, in Linz, Salzburg and Vienna, inspired by the idea of `reappropriation of the conflict’. Conflict resolution is seen as encouraging and strengthening the capacity of people to deal with the material and emotional damages and grievances created by people against the integrity and interests of their fellow human beings – not a panacea but a step towards more personal autonomy. This could be a `mission state­ment’ for restorative justice, although that term is not used. The Austrian ex­press­ion is aussergerichtlicher Tatausgleich (out-of-court offence resolution); it is controlled by prosecutors and is seen as strengthening their role. It was felt that criminal law was unable to restore the equilibrium disturbed by a criminal act; and in any case, research showed that only 15% of conflicts lead to a sanct­ion, including only 5% involving social control agencies, and 1% criminal jus­tice. So it was felt that inviting the victim to express what kind of reparation he or she wants serves the victim’s interest better than a criminal trial, as well as sparing the young offender the stigma of a criminal record (Pelikan 1991).

It was felt that the `need to punish’ is much overestimated: victims mainly want reparation, concern and information, according to Dr Udo Jesionek, vice-president of the Weisser Ring in Austria, and president of the Vienna Juvenile Court. One important feature of the Austrian process is that the consent of the victim is not required; this is at first sight paradoxical, in a process that is intended to empower both victims and offenders, but the offender is allowed, to make reparation in another way if the victim does not take part, and this happens in practice. Jesionek stresses the great importance of this provision, because it removes from the victim a possible, unacceptable pressure to accept an agreement (contrast the New Zealand family conferencing procedure, below). Advice to the participants is available from social workers/probation officers, and from Weisser Ring, respectively. Conversely, participation is voluntary for the juvenile; if he does not take part, there will be a trial, unless he receives a warning or the case is discontinued, and if he does take part there may still be a trial if the prosecutor or judge does not consider the reparation adequate. The juvenile is required to acknowledge the act, but not to make a formal confession of guilt. There are leaflets for juveniles in all police stations stressing the importance of making good damage or loss; this may be done without state intervention, and may lead to discontinuance (Jesionek 1992). The Penal Code also allows discontinuance when adults have made reparation, and this is done especially when a juvenile and a young adult have offended jointly.

New ideas in the Old Commonwealth

The Maori people of New Zealand were asking, in the 1980s, for greater recognition of their traditions, not least in criminal justice: not only were they over-represented in courts and prisons, like many ethnic minorities, but the adversarial procedure was foreign to their culture. When a young person was in trouble they would assemble the family, clan and tribe to discuss what to do for the best; when the young person had harmed someone, the injured party also took part. There was already a feeling that social workers should involve families more in decisions that affected them; and there was an interplay of ideas with other parts of the world – Burt Galaway, for example, visited more than once and carried out a public opinion survey confirming, as elsewhere, that many people, when asked, support reparative sanctions (Galaway 1984; see also Lee 1996). The Children, Young Persons and Their Families Act 1989 introduced the new procedure using family group conferences both for welfare cases and for those involving a crime (and hence usually a victim). There was also official interest in the concept of restorative justice, and a discussion paper was issued (NZ Ministry of Justice 1995); but with the change of coalition government after the 1996 election this has not yet been taken further.

One significant innovation was the involvement of the offender’s family and the victim; another was the way the procedure was combined with the criminal justice process. Most 14- to 16-year-old suspects, provided they do not deny the offence, are first referred to an FGC; this normally deals with the matter, but if the case is too serious, or a satisfactory agreement cannot be reached, it can recommend that the young person be summoned to court. The other route to court is by arrest, which is only allowed if the young person is likely to abscond, re-offend or interfere with evidence or witnesses; but if the young person does not deny the matter, the court must refer the case to an FGC, rather as in England it asks for a pre-sentence report. In the great majority of cases the FGC’s agreement is endorsed by the court; but because the court oversees the process, crimes of any gravity (except homicide) can be dealt with in this way. The number of court hearings for this age-group went down from 63 to 16 per 1 000. Before the Act there were 200 residential places for this age group; after it, only 76 were needed.

Because the idea evolved primarily from a welfare perspective, however, there was at first a tendency for the process to be offender-centred; FGCs were arranged without consulting victims about convenient times, and in some cases without even informing them. After this was brought to light by research, practice improved; also, victims were allowed to bring relatives or other supporters when it was pointed out that they were outnumbered by the offender’s extended family. One point in the procedure continues to cause unease, however: because it is welfare-based, it includes the `private time’ of the offender’s family, during which they produce a plan for reparation and other action. The victim can veto this (and so can the police); but if the victim does so, he or she is in the invidious position of holding up the proceedings and possibly making it necessary for the offender to go to court. Victims can avoid this pressure by acquiescing, or by not taking part (but they can ask someone else to represent them). Despite these problems, 60% of victims who attended FGCs found them `helpful, positive and rewarding’, but about a quarter felt worse. The researchers found the service suffering from `funding starvation’ which limited the facilities available for offenders who had agreed to attend them, such as anger management courses and job training programmes (Maxwell and Morris 1996). It has been suggested that the distinction between FGCs concerned with the young person’s welfare, and those where victims are involved, should be marked by giving another name to the latter, such as `victim/offender conferences’.

It was a coincidence that the Australian criminologist John Braithwaite was working on his theory of `reintegrative shaming’ at about the same time (Braithwaite 1989). Its basis is not so much restorative as communitarian, stressing individual autonomy and limitation of state power, but the theory fitted the FGC practice well. An overhaul in the New South Wales police gave an opportunity to adapt the idea, and it was pioneered in a town called Wagga Wagga. The process was modified, especially by including the victim’s family, and by having everyone present in the room throughout, so that all can contribute to finding solutions agreed by everyone. It has however been criticized for being based in the police force; and although Braithwaite distinguishes `reintegrative’ shaming from traditional stigmatization, some observers fear that this distinction will not be understood by the public and politicians, and perhaps even some practitioners. One of the originators of this method, Senior Sergeant Terry O’Connell, has visited Britain and the US to promote it. Another distinguishing feature of FGCs is that, by involving the families, they constitute a communitarian approach to justice; this could be enhanced by the involvement of lay mediators, subject to adequate training and supervision.

Even more communitarian is the idea, pioneered in Canada, of sentencing circles. These, too, have roots in the traditions of First Nations; they were first implemented by a judge, Barry Stuart, in 1992. Members of the community, including the victim, can join the circle and are consulted, but it is the judge who makes the sentencing decision (LaPrairie 1995). The discussion in the circle provides an opportunity for those involved to tell their stories, and for the community to find constructive and healing ways forward; Judge Stuart has said that the test of success is the effect on the community. But the agenda is not entirely restorative, because sentences can still include punitive sanctions, albeit often less harsh ones than the judge might otherwise have imposed. To this extent, although circle sentencing gets high marks for community involvement, it falls short of the fully restorative ideal, because for those who take part it legitimates the acceptability of punitive sanctions.


At the Conference on Restorative Justice for Juveniles held in Leuven in May 1997, several speakers complained that there was no agreement about what restorative justice is. At such an early stage in the evolution of the concept, some diversity is hardly surprising. Work is still in progress, and may continue to be for some time to come. A description, if not a definition, of the idea of restorative justice is attempted at the beginning of this paper (see section on Origins). To conclude, it may be worth considering four aspects of the concept.

Distinguishing features of measures having a place in a restorative system of justice A restorative measure is based on healing, not on punishment (as defined at the beginning of this paper), and hence any restriction of liberty is not for punishment but for public protection – and possibly for the enforcement of agreed reparation, although there is a limit to what people can be compelled to do if they are determined in their refusal. It shows the offender the human consequences of his act, but also enables him to make amends for it. It provides opportunities for both victim and offender to communicate, and to agree a reparative action plan. The community is involved in the process.

Partially restorative measures include compensation and community service orders, and state compensation to victims of violence, as well as community assistance such as Victim Support. Measures such as keeping victims informed of the progress of their case, and supporting witnesses in court, though welcome in the existing system, are not truly restorative because they palliate the effects of the criminal justice system, not of the crime itself. Measures usually described as rehabilitative have some restorative aspects as regards the offender, and to some extent the community, but little from the victim’s point of view.

Ways of promoting the introduction of restorative justice include raising aware­ness of the benefits to victims, and of the fact that many of them want some good to come out of the harm they have suffered. There should be recog­ni­­tion that many members of the public support the idea of reparation by offenders, but that the process also holds the offender accountable. For practical reasons, however, there should also be awareness of the benefits to the agents of criminal just­ice, such as police, prosecutors, courts, and probation and youth justice workers, including reduc­ed cost and workload, and increased job satisfaction.

Conditions to be met if it is to fulfil its promise Those who are used to the conventional system are, not surprisingly, sceptical of proposals for change, especially a change of most of the current value system. But they should not demand higher standards for the new than they tolerate in the old. Restorative services should not, for example, be dismissed if they fail to lead to lower reconviction rates than other measures; if they lead to similar reconviction rates, but have advantages that conventional measures lack, then they deserve to be used. It is unreasonable to expect that a one- or two-hour mediation session will change a person’s whole life, unless it is backed up by adequately resourced supportive measures in the community. Services must concentrate on the benefits to victims, and not be oriented more to offenders. There must be high standards; good training; safeguards for both victims and offenders; and clarity about aims, which are the criteria for evaluation.

It has been said that criminal justice cannot be fair in an unjust society. Adversarial procedures and punitive sanctions tend to add to social frag­mentation and injustice. The case for restorative measures is that they have the potential to help to re-create communities and make society a little more fair.

c:rjleuvlect (rev)


Balanced and Restorative Justice (1995) Balanced and restorative justice for juveniles: a national strategy for juvenile justice in the 21st century. School of Public Administration, Florida Atlantic University, Fort Lauderdale FL 33301-1905.

Barnett, R (1977) `Restitution: a new paradigm of criminal justice.’ Ethics, 87(4), 279-301.

Beer, Jennifer E (1986) Peacemaking in your neighborhood: reflections on an experiment in community mediation. Philadelphia, PA: New Society publishers.

Cantor, Gilbert (1976) `An end to crime and punishment.’ The Shingle (Philadelphia Bar Association), 39(4), 99-114.

Christie, Nils (1977) `Conflicts as property.’ British Journal of Criminology, 17(1), 1-15.

Church Council on Justice and Corrections (1996) Satisfying justice: a compendium of initiatives, programs and legislative measures. Co-published by the Correctional Service of Canada. Church Council on Justice and Corrections, 507 Bank Street, Ottawa, Ontario K2P 1Z5, Canada.

Davis, Gwynn, Jacky Boucherat and David Watson (1988) `Reparation in the service of diversion.’ Howard Journal, 27(2), 127-134.

Dünkel, Frieder, and Dieter Rössner (1989) `Law and practice of victim/offender agreements.’ In: Martin Wright and Burt Galaway, eds. Mediation and criminal justice: victims, offenders and community. London: Sage.

Eglash, A (1958) `Creative restitution.’ Journal of Criminal Law, Criminology and Police Science, 48(6), 619-22.

Eppenstein, Dieter (1992) `Täter/Opfer Ausgleich: Zwischenbilanz und Perspektiven.’ (V/OR: interim balance and outloook). In: Bundesministerium der Justiz. Täter/Opfer Ausgleich: Zwischenbilanz und Perspektiven. Symposium 19-21 Juni 1989. Bonn: Forum-Verlag Godesberg.

Fuller, Lon L (1971) `Mediation: its forms and functions.’ South California Law Review, 44, 305-339.

Galaway, Burt (1984) Public acceptance of restitution as an alternative to imprisonment for property offenders: a survey. Wellington, NZ: Ministry of Justice.

Harding, John (1982) Victims and offenders: needs and responsibilities. London : Bedford Square Press.

Hartmann, Arthur (1997) `The development of victim/offender reconciliation in Germany.’ Unpublished paper.

Home Office (1959) Penal practice in a changing society. Cmnd 645. London: HMSO.

Hulsman, Louk (1981) `Penal reform in the Netherlands: Part I – Bringing the criminal justice system under control.’ Howard Journal, 20(3), 150-159.

Jesionek, Udo (1992) `Der aussergerichtliche Tatausgliech im neuen österreichischen Jugendrecht.’ (Out-of-court conflict resolution in the new Austrian juvenile law.) In: Kriminalitätsopfer im Spannungsfeld der Interessen (Tensions around the interests of crime victims). Mainz: Weisser Ring.

Launay, Gilles (1985) `Bringing victims and offenders together: a comparison of two models.’ Howard Journal, 23 (3), 200-212.

Launay, Gilles and R Murray (1989) `Victim/offender groups.’In: M Wright and B Galaway, eds. Mediation and criminal justice: victims, offenders and community. London: Sage.

Lee, Angela (1996) `Public attitudes towards restorative justice.’ In: Burt Galaway and Joe Hudson, eds. Restorative justice: international perspectives. Monsey NY: Criminal Justice Press, and Amsterdam: Kugler Publications.

LaPrairie, Carol (1995) `Altering course: new directions in criminal justice. Sentencing circles and family group conferences.’ Australian and NZ Journal of Criminology, 28, 78-99.

Maxwell, Gabrielle, and Allison Morris (1996) “Research on 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. Leichhardt, NSW: Federation Press, and Monsey, NY: Willow Tree Press, 1996.

Netzig, Lutz, and Thomas Trenczek (1996) `Restorative justice as participation: theory, law, experience and research.’ In: Burt Galaway and Joe Hudson, eds. Restorative justice: international perspectives. Monsey NY: Criminal Justice Press, and Amsterdam: Kugler Publications.

New Zealand Ministry of Justice (1995) Restorative justice: a discussion paper. Ministry of Justice, POBox 180, Wellington NZ.

Northamptonshire Diversion Unit (1995) Annual Report 1994-1995. The Diversion Unit, 75a London Road, Kettering, NN15 7PQ.

Peachey, Dean (1989) `The Kitchener experiment.’ In: M Wright and B Galaway, ed. Mediation and criminal justice: victims, offenders and community. London: Sage.

Peake’s Commentary on the Bible (1962). Wokingham: Van Nostrand Reinhold (UK).

Pelikan, Christa (1991) `Conflict resolution between victims and offenders in Austria and the Federal Republic of Germany.’ In: Frances Heidensohn and Martin Farrell, eds. Crime in Europe. London: Routledge.

Pfeiffer, Christian (1992) `Täter/Opfer Ausgleich: das Trojanische Pferd im Strafrecht?’ (V/OR: the Trojan horse in criminal justice?’ In: Kriminalitätsopfer im Spannungsfeld der Interessen (Tensions around the interests of crime victims). Mainz: Weisser Ring.

Schafer, Stephen (1960) Restitution to victims of crime. London: Stevens.

Schafer, Stephen (1968) The victim and his criminal: a study in functional responsibility. New York: Random House.

Schreckling, Jürgen (1990) Täter/Opfer Ausgleich nach Jugendstraftaten in Köln. (Victim/offender reconciliation after juvenile offences in Cologne.) Bonn: Bundesministerium der Justiz.

Tallack, W (1900) Reparation to the injured. LondonL Wertheimer, Lea.

Thorvaldson, S A (1978) `The effects of community service on the attitudes of the offenders.’ Unpublished doctoral thesis, University of Cambridge.

Varona, Gema (1996) Restorative justice: new social rites within the penal system? Master’s Tesina, Oñati International Institute for the Sociology of Law.

Warner, Sue (1992) Making amends: justice for victims and offenders. An evaluation of the SACRO reparation and mediation project. Aldershot: Avebury.

West Yorkshire Probation Service (1997) Victim/Offender Unit annual report 1996/7. Oxford Place Centre, Leeds LS1 3AX.

Whitfield, Dick, and David Scott (1993) Paying back: twenty years of community service. Winchester: Waterside Press.

Wright, Martin (1983) Victim/offender reparation agreements: a feasibility study in Coventry. Birmingham: West Midlands Probation Service.

Wright, Martin (1984) In the interests of the community: a review of the literature on community service orders. University of Birmingham, Department of Social Administration.

Wright, Martin (1996) Justice for victims and offenders: a restorative response to crime. Winchester, England: Waterside Press. 2nd ed.

Zehr, Howard (1980) Mediating the victim/offender conflict. Akron, PA: MCC Office of Criminal Justice.

Zehr, Howard (1985) Retributive justice, restorative justice. Elkhart, IN: MCC Office of Criminal Justice.

Zehr, Howard (1990) Changing lenses: a new focus for crime and justice. Scottdale, PA: Herald Press.

Author’s address: 19 Hillside Road, London SW2 3HL, England.

Tel/fax (+44)/(0) 181 671 8037

14 October 1997 c:rjleuvlect (rev)