Victim/offender conferencing: the need for safeguards

Paper presented under the title “Family group conferencing: is it restorative? Is it just?” at a workshop at the International Conference: Restorative justice for juveniles – potentialities, risks and problems for research. Leuven, May 1997.


Two relatively recent ideas are attracting interest among those who are seeking something better than criminal justice: the concept of restorative justice, and the practice of conferencing. Are they made for each other? Is conferencing the way to put restorative justice into practice? After outlining restorative justice and conferencing, this paper will examine how well they fit together, from the point of view of victims, offenders and the community; criteria for research assessment are proposed, and safeguards and points for concern are discussed. Methods of evaluation, follow-up, monitoring and feed-back are put forward. Potential problems and the future of restorative justice are reviewed.

 Restorative justice

At the heart of the idea of restorative justice is the recognition that crime is not merely law-breaking: it causes harm – to individual victims, the community, the state or the environment. The first response should, therefore, be to try to remedy that harm, as far as possible, or if necessary to make things better than before. Secondly, when offenders are known, they should be required to contribute towards making things right as much as they are able. This may be by paying compensation, meeting the victim, working for a charitable organization, or co-operating with a rehabilitative programme.

Thirdly, the community should play its part by offering support to the victim, for example through Victim Support, and by enabling offenders to make amends, for example by providing employment so that they can earn the money with which to pay compensation. What “the community” is will be discussed below.

The fourth leg of the restorative justice model is a specific strategy for crime prevention. This would replace the unsuccessful policy of reliance on the deterrent effect of punishing the small proportion of offenders who are caught and convicted. It would be implemented by a specific department of government, to which there would be feedback of the lessons to be learnt from crimes as they take place, so that the strategy can be improved. Thus it would be the crime prevention department that would be assessed on the basis of the crime rate; for the justice process the criterion would be its handling of individual cases. In the 1997 general election in the United Kingdom, two political parties, Labour and Liberal Democrats, have said that they will make local authorities responsible for “partnerships” to help prevent crime, and to set targets for crime prevention (Labour Party 1997: 23; Liberal Democrats 1997: 30).

Initially, restorative measures have been seen by many people as a new option available to courts, alongside traditional justice; but as they become more widely used they provide an opportunity to change the underlying philosophy in a restorative direction. The restorative philosophy has some different aims and hence different criteria by which to assess its success. Retribution for its own sake is not included, because a restorative response is held to be just as powerful and more constructive. Another traditional aim is general deterrence. The deterrent effect of the probability of being caught is at least as great when the system is restorative, and so is the opportunity to make the response proportionate to the offence (although this has complications which will be discussed below – see “Inequality”). Many of the inconsistencies of conventional sentencing arise precisely because it is trying to do several incompatible things at once: not only to deter, which means to try to frighten the general population into good behaviour, but also to resolve an individual case, and to symbolize, by the amount of punishment, the seriousness with which the court regards the offence (or with which the court believes that politicians and popular newspapers will regard it). Restorative justice uses a more constructive symbolism, recognizing that people are not motivated only by fear; their behaviour can be influenced by tangible incentives, and by intangible ones such as being well thought of by people who are significant in their lives, and hence having good self-esteem. It spreads understanding of the true reasons why people should not offend against one another: the reason why I should not commit an offence against you is not that I will be made to suffer if I am caught, but that you will suffer whether I am caught or not. But I probably shall not care much about you unless someone cares about me.

Restorative measures can be imposed by courts, but many advocates of restorative justice consider that it is most effective when the way to “make things right” can be decided by the people directly affected, the victim and the offender and, if conferencing is used, their families and supporters. The courts will then act as a safeguard by overseeing the process, as will be proposed below.

The foregoing is a description of the whole concept; parts of it can however be introduced alongside the existing system, e.g. Victim Support, or within it, e.g. compensation or community service orders. Advocates of restorative justice accept these as steps towards fuller adoption of the concept, but do not favour their use as token measures grafted on to a retributive system.

In this paper the words “restorative” and “reparative” will be used interchangeably. “Reparation” may mean an apology, money payment, work for a charitable organization, or co-operation with a rehabilitative programme. “Justice” will be used in a broad sense, to mean fairness in a legal context; “due process” will refer to the legal rules of procedure, generally intended to protect the defendant. “Punishment” means a sanction whose primary intention is to cause pain to the offender, unlike those sanctions which aim at rehabilitation, reparation or containment, and are not deliberately unpleasant.

Family group conferencing

If restorative justice were limited to reparation, it could simply be imposed by a court. But the concept includes not merely the outcome but the process: offering those involved the opportunity to communicate, and to agree about how to make things better. This process is commonly called victim/offender mediation, although some prefer the word dialogue; in Germany it is offender/victim settlement (Täter/Opfer Ausgleich), in Austria out-of-court offence resolution (aussergerichtlicher Tatausgleich). It generally involves a one-to-one meeting between the victim and offender in the presence of the mediators, or indirect communication with the mediators as go-betweens, but either party may bring family or supporters. In family group conferences (FGCs) the families or other supporters of both parties are generally present, although in New Zealand for minor offences the young person is often accompanied only by his or her parents (Stewart 1996: 67). FGCs have been presented as a radically new method; but it is hard to see a clear dividing line between them and victim/offender mediation, and much of what will be said about either of them in this paper can be applied to the other. The word “conferencing” may be used as a more general term, and “facilitators” for those who conduct the sessions.

FGCs were designed to incorporate features of Maori traditional methods, but New Zealanders were also aware of the development in North America of Victim/Offender Reconciliation Programs (VORPs), not least because of visits by Burt Galaway from the mid-80s onwards. Galaway found that public opinion in New Zealand was receptive to the idea: most people would be likely to accept a reduction in the use of imprisonment for property crimes if there were an increase in the use of reparation (Galaway 1984:11, quoted in New Zealand 1995). A variant of the process, sometimes called “community accountability conferencing”, was developed in New South Wales, Australia, and has also been adapted for use in schools (Maroochydore State High School (n.d.)). Conferencing could become the starting point for a process of change within the system.


A distinguishing feature of restorative justice is that the victim should benefit from it at least as much as the offender, although early practice in victim/ offender mediation and FGCs did not always achieve this. It is therefore necessary to ask what victims want. This has been summarized by Teresa Reynolds, policy and information manager of Victim Support, as follows. I have italicized the needs which tend to be met better by a restorative process.

Help with the practical/emotional effects of crime

To be treated sensitively and with respect

– To know what’s happening in the case

To understand what’s happening in the case

To be heard and taken seriously

– Public acknowledgement that wrong has been done

– To feel it is worth pursuing the case

Quick resolution of the case

To know “Why me?”

– To know that the offence will not be repeated

– Some want an apology from the offender

– Some want compensation

– Some want retribution (most do not)

– Most want to be free of responsibility for decisions about the offender

(Reynolds, 1997)

The first of these is also provided by Victim Support in the United Kingdom and by similar organizations elsewhere, and efforts are being made within the British system to achieve some of the others, through the Victim’s charter and related initiatives (Home Office 1996).

The traditional system does not serve victims well. A survey in England in 1991 found substantial proportions, sometimes the majority, who would have liked more information about court procedures, how to get to court, and what to do on arrival. They felt that there was insufficient consultation over setting the date of the hearing, and insufficient information about time of case; sometimes cases did not even take place on the scheduled day. No one told them what to do in the courtroom, nor explained the roles of different officials, and they felt nervous, intimidated, worried or frightened in the witness box. A common complaint was that they had to wait in the same area as the defendant (J Raine and R Smith, The Victim/Witness in Court Project: report of the research programme (1991), cited in Victim Support 1995). Many of these are now being addressed in England and Wales, through the Crown Court Witness Service and initiatives introduced by the Victim’s Charter, but some, such as a chance to be heard, asking “Why me?” and apologies, are difficult within traditional structures; whereas in victim/offender mediation and conferences, most of them are routinely taken care of.

Restorative justice for victims

Traditional court procedure is regarded as merely a means to an end, whereas mediation and conferencing take account of the effects of the process. But if they are to help the victim’s recovery, they need to make sure that restorative practices are adhered to. For the victim this means that the criteria for selection of cases should relate to victims as well as to offenders. The location of the meeting should be neutral, and the time and place convenient. Victims should have the opportunity to describe the crime in their own words, and to tell the offender the effects it had. They should be able to be involved in working out how to “make things right”, but not in deciding on punishment. They should be able to bring their family and/or supporters, and their participation should be voluntary.

The particular needs of victims, and the families and supporters of both victims and offenders, must be considered. They may need to be reimbursed for the expenses of travel, accommodation, child care, and loss of earnings; this should be as much a normal part of the process as it is for witnesses in a traditional trial. Qualified interpreters should be available when necessary.

Safeguards for voluntary participation of victims

Although mediation and conferencing are officially voluntary, a criticism is that victims and offenders might feel under pressure to take part, knowing that if they refuse, or do not accept the agreement, they would be responsible for the young person’s being taken to court and perhaps sentenced more severely. This has been raised as a concern in England by Victim Support, although in Scotland and other European countries, where prosecutors discontinue cases if mediation is successful, it does not appear to be seen as a problem. There are several ways of minimizing this effect.

1. The possibility of mediation can be raised, perhaps by a Victim Support worker, before the offender is known (provided of course that he was not known to the victim already), to allow the victim time to reflect.

2. Mediation may be offered after the decision to caution or sentence; thus the victim’s choice about whether to take part cannot influence those decisions.

If mediation or a conference is offered before the final decision in the criminal justice process:

3. The offender can do community service instead of direct reparation to the victim; this enables him to show in a tangible way that he regrets the action, even when the victim wants to have no contact with him.

4. The offender can take part in a victim/offender group (Launay 1985; Launay 1987; Launay and Murray 1989) if the victim does not want a personal dialogue. This provides an opportunity to discuss the offence, for victims whose “own” offenders are not caught, and for offenders whose victims do not want to meet. It has not, however, been much used so far.

5. Reparation and mediation can be considered separately. Mediation can be seen as a personal communication between the victim and offender as individuals, which does not influence the criminal justice process; either victim or offender can therefore choose not to take part, without affecting the sentence. If however the offender makes some reparation, it is fair that the court should know about this, so that it can be taken into account in sentencing.

6. Victim/offender mediation can be indirect: the mediator can act as a go-between. At a conference, the victim need not be present at all: he or she can be represented by a supporter. This may not give such good results (Marshall and Merry 1990: 243-244; Umbreit and Roberts 1996: 27), nor offer victims so much opportunity for empowerment, but may place less pressure on them.

In all cases, preparation of the victim and the offender before the mediation or conference is essential, so that they have realistic expectations and can make an informed choice (Carroll 1994: 177). Also critical is the facilitators’ assessment of who should be invited to take part. This has implications for the training of mediators and facilitators, and, because it takes time, for the adequate funding of the service.


What should be the criteria for assessing restorative justice from the victim’s point of view? Those that are central to the concept of restorative justice are not the easiest to measure. Many victims would like to know that the process makes offenders less likely to re-offend; but as we have seen, this is not a primary aim, and it is a lot to expect of one meeting lasting perhaps an hour or two. The agreed reparation will take longer, and perhaps have additional influence on the offender’s attitude and behaviour.

A basic measure of whether victims like mediation is the rate of attendance at sessions; to discover how they feel after taking part, some form of follow-up should (with their permission) be built in to the process, to monitor their satisfaction; this should also be allowed for in the project’s budget. In-depth research can be carried out from time to time to assess other factors such as the reduction of fear as a result of meeting the offender. Other yardsticks are suggested by the Balanced and Restorative Justice Project (1995: 17, 39-41), such as the number and proportion of victims satisfied, whether victims received sufficient information, and whether the outcome adequately reflected the severity of the offence (but see “Inequality” below).

Potential problems for victims with conferences

The original New Zealand model of FGCs was evolved from a welfare perspective, in addition to the Maori influences: when a young person needed support, the extended family would be invited to attend a conference, including “private time” with no social workers present, to arrive at a plan that would be in the young person’s best interests. This model has been adopted in some English social service departments (Marsh and Crow 1996). The New Zealand Children, Young Persons and their Families Act 1989 also applied the idea to cases where the young person had committed an offence and there was a victim; but this led to three problems. The first was that too little account was taken of the convenience of victims in arranging the time and place of the conference. Secondly, victims could feel unsupported in the presence of the young offender and several members of his or her family.

The third problem concerned the procedure at the conferences. After a general discussion of the offence, the offender and his or her family have “private time” during which they formulate a plan for the young person to make amends and keep out of trouble. They then return and explain it to the victim. If the victim is happy with it, there is no difficulty; but if the victim feels that it is inadequate, and amendments to the plan cannot be agreed, he or she is put in the position of having to object, and to be responsible for the young offender’s return to court. This combination of factors is the probable explanation of the results of the first research into the New Zealand experiment, which is the only programme involving victims and offenders where only half the victims were satisfied with the outcomes and a quarter said they felt worse (Maxwell and Morris 1996: 100). The 1989 Act was amended in 1994 and some of these concerns have been addressed; victims have the right to bring their children, all the residents of the house in the case of burglary, and supporters, and to be consulted about the time, date and venue. They can apply for travel costs and reimbursement of lost wages (Stewart 1996: 68).

The picture is complicated because there are different conferencing models in use. Similar ones are used in New Zealand and parts of Australia; but the one adopted in Wagga Wagga, New South Wales, has some differences, in addition to its somewhat controversial location in the police service. Three of these are worth considering. Firstly, the conference may begin with a social worker reading out what the young person is alleged to have done (NZ) or the offender may be asked to describe what took place (NSW). A further possibility would be to offer the victim the option of speaking first. Secondly, the offender’s family and the victim and supporters may withdraw to separate rooms to discuss the action plan (NZ), or the whole discussion may take place in a single group (NSW), which goes some way to meeting the concern about pressure on the victim, referred to in the previous paragraph. Thirdly, the conference may be regarded as finished when agreement has been reached (NZ) or importance may be attached to informal contacts afterwards, before everyone goes home (NSW). The pros and cons of these options need to be considered.


Restorative justice for offenders

The informality of restorative justice may allow offenders to speak more freely than in court, but it must be balanced by safeguards (see below). It can increase their understanding, by showing that what they have done is wrong not merely because it is against the law, but because it has harmed another person. It enables them to make things right (at least partly) in an active way, rather than merely submit passively to punishment (or rebel against it). A conference may also bring together their family in a setting which potentially enables them all to work together in the interests of the young person.

Safeguards for offenders

Restorative justice may give the appearance of being “soft”, although it can be seen as more demanding than punishment because it makes offenders face up to the real effects of their actions; as a result, concern is often raised that young offenders could feel under pressure to admit guilt when they may have a defence in law (see for example Warner 1994; Sandor 1994: 159). A partial answer is that the same objection can be raised in relation to any form of pre-trial diversion from the criminal process, although Warner (1994) suggests that the risk may be greater with conferences. In fact, young people are not required to admit guilt before an FGC in New Zealand; it is enough that they do not deny the act. In Austria, the basis is acceptance of civil liability, Einstehen für die Tat (Braithwaite 1994: 205). A distinction should also be made between the person who denies committing the act at all, and the one who admits the act but claims that it was not a crime, for example because he was entitled to take the goods, he was acting in self-defence, or the victim consented. In such cases, the legal dichotomy “Guilty/Not Guilty” leaves an unfinished conflict between the victim and the offender, which can be better resolved in a mediation or conference session than in court. If the accused has wrongly admitted guilt, mediation or conferencing is a good way of bringing this into the open. Finally, if the worst happens and an innocent person does plead guilty, it is perhaps less of an injustice to make them do something constructive than to inflict punishment on them.

A related concern is that offenders might be pushed, by the victim or their own punitive family, into agreeing to make reparation out of proportion to the seriousness of the offence or to their ability to deliver.

Both of these dangers require that offenders should have access to lawyers; these, however, would have to be in sympathy with restorative aims, and not bury them in rigid legal precedents. Police should be required to ensure that a parent or guardian is present during questioning (Warner 1994). New Zealand provides that a youth advocate may attend if requested by the young person, paid by the court without means testing (Stewart 1996: 71). A further safeguard against disproportionately large (or small) action plans is that for the more serious offences, the courts have to approve plans that have been agreed at a FGC, and may alter them if they appear too burdensome (or insufficient). This is especially necessary because in New Zealand the action plan can include punishment; in true restorative justice punishment (in the sense defined at the beginning of this paper) would not be an option, but there would still be a need for court supervision of preventive measures such as curfews and especially residential programmes, to ensure that these were not disproportionate to the offence and that particular groups (girls, ethnic minorities) are not subject to excessive intervention (Warner 1994: 149). Compulsory detention would be reserved for cases where the protection of the public required it; institutional regimes would be reparative, not punitive. Some of the money saved by sending fewer people to prison should be used to provide reparative programmes in the community.

Courts would also be necessary to decide the small proportion of cases where the offender denies the alleged act altogether; but under a fully restorative system the court’s sanctions would be reparative, not punitive.


Question of proportionality and consistency are consistently raised in discussions of conferences (especially by lawyers). What happens if the offender cannot conceivably compensate the victim fully? Inevitably, different victims and offenders will arrive at different action plans for outwardly similar offences, and the variation will probably be even greater than in the courts. Does this matter? If the each victim and offender (and their families and supporters) have agreed on what appears right for them, does it matter if others have resolved things differently? Warner (1994: 148) thinks it does: offenders could become aggrieved when they find out. But perhaps the converse is the problem: courts try (not very successfully) to impose consistent sentences when the impact of the crimes on the victims, and the impact of the punishments on the offenders, are disparate. And the fundamental difference remains: courts are aiming at consistency of punishment, where inequalities may indeed cause resentment; but conferences aim at reparation, tangible or symbolic, which is rightly determined by the needs and wishes of the victim and the capabilities of the individual offender.


Since under restorative justice the primary responsibility for crime control would be with a crime prevention agency, the reconviction rate would not be the main measure of “success”, although it will need to be monitored to make sure that re-offending does not significantly increase (but it could also ascertain whether the frequency and severity of offending decline in the longer term – on which little longitudinal research with control groups has been done so far). The criterion used in assessing victim/offender mediation has been whether the offender felt fairly dealt with; perhaps an unexpected choice at first sight, but if the system aims to lead the offender to respect the law, rather than merely fear it, this is appropriate. One stated aim of conferences is to strengthen the young person’s family in supporting and controlling him or her. Other factors which are relatively easy to measure could be used, such as the amount of compensation or community service, but these depend on individual agreements between victims and offenders, so that more does not necessarily mean better. Another measurement is whether the agreed compensation or community service is actually completed; this may be regarded as a measure of the reparation made by the offender, or of the victim’s satisfaction.

Restorative measures require active participation by the offender, and this should be assessed; but often offenders cannot complete what they have agreed to do unless facilities are available: community service opportunities, occupational training, social skills programmes, or simply employment. The extent to which these have been made available should be part of the monitoring (BARJ 1995: 21).

Potential problems for offenders

The offender may be alienated from his or her family, or not wish a certain parent to be present. Efforts are then made to find a “community of care”, or a peer, but this is not always possible (Wundersitz and Hetzel 1996: 123-4); in that case a volunteer mentor or befriender could be. The young person could be overwhelmed by the number of people taking part in the conference. If the conference shows that the offender has specific needs, he or she can be referred to a separate care-and-protection FGC for the family only, but in practice this seldom happens (Maxwell and Morris 1996: 98).

Participation in mediation or conferences should be voluntary for offenders, as for victims; the safeguards for victims, listed above, will also protect offenders. They should have a right to make reparation to the community, even if the victim does not want to accept anything from them. But although they can choose not to take part in mediation or a conference, they cannot insist on it if the well-being or safety of the victim is considered to be at risk.

When supervision or detention is necessary for the protection of the public, there will be a problem in determining when the offender may safely be released; but this is the case with any indeterminate sentence, and it will be simpler in a restorative system because public safety will be the only criterion.


There is some discussion about what the community actually is; in large cities people often belong to several communities, of people who do the same work, or belong to the same clubs or voluntary organizations, or come from the same ethnic or religious background, and so on. People who live in the same street or block of flats do not necessarily form a community. So how does restorative justice affect “the community”? Firstly, the families and supporters of victims and offenders, who attend a conference, are the members of the community most affected by the particular crime. Secondly, if the mediation service uses lay facilitators, they will form another mini-community. In both these ways, mediation and conferencing can make a small contribution towards re-creating a sense of community in urban areas by being what Germans call bürgernah (close to the citizens). Thirdly, the local government of a town or part of a city acts on behalf of the whole community of local citizens; they benefit when an offender is reintegrated into society, but the local authority has a responsibility to provide the resources to make this possible, such as educative programmes, detoxification, and above all work and accommodation.

The question has been raised, whether victim/offender mediation ought to take place in public. In fact court proceedings are often attended by very few people, and (in England at least) the public is not admitted to youth courts; so a conference, attended by several people from the families of the victim and the offender, may be considered to provide an acceptable degree of openness to the public.

The involvement of the community should help to educate people in the need for preventive social policies, to which they themselves can contribute. This would help to build what Walgrave (1995: 236) has called an “emancipatory society”. It could help people to recognize each other’s human needs, and perhaps, at a more general level, point the way to substantial social reforms which could reduce pressures towards crime. Many victims show sympathy and understanding to offenders (Marshall and Merry 1990: 157). Replies to a small survey of people working in mediation or in criminal justice suggested that meeting an offender might “lead the victim to be a little more charitable”, but probably not to significantly change their attitudes to social policy, or their life-style (Wright 1996: 130). Kay Pranis, restorative justice planner in the Minnesota Department of Corrections, has suggested as a more modest goal that “The ultimate measurement for the system should be: Is the community stronger after the criminal justice intervention than it was before the crime was committed?” (Pranis 1997: 40).

Evaluation, follow-up, monitoring and feed-back

This would not, however, be easy to measure. Nor would the broad measure of success suggested by Maxwell and Morris (1996: 105): the operation of a fair, humane and acceptable methods of dealing with young offenders (and, the authors might have added, supporting victims), or promoting independence and self-sufficiency of individuals and families (Carroll 1994: 169). Approximate criteria can be devised, however. Observing a process and modifying it in the light of experience gained is far from being unique to restorative justice, but it should form part of good practice. In traditional justice much relevant information is not collected at all, for example the cost of the sentences imposed by individual judges; in the United Kingdom performance indicators are being introduced, but in a way which could encourage agencies to modify their recording practices in order to appear more efficient. An inquiry has, for example, been announced into alleged “massaging” of crime figures in Nottinghamshire (Independent, 30 April 1997). Good practice could be divided into four parts.


After each process, such as a conference, those concerned should allow time to review it and consider whether they handled it as well as possible. Their training should include the ability to give and receive constructive criticism. The use of two co-facilitators assists this process; in addition an outside independent evaluator could observe actual meetings (with the permission of all participants) to assist in evaluating how the facilitators handled the meeting and how well it fulfilled restorative aims.


A normal part of the process should be to ask the participants’ permission to contact them after a time to ask how the situation has developed, and how the process worked from their point of view.


Routine record-keeping should include basic information relevant to providing good service, insofar as this can be collected without being intrusive: for example, ethnic backgrounds, languages spoken, time intervals between stages in the process. This information should be published in the annual report, with suggested explanations of any discrepancies, for example between the ethnic background of participants and facilitators and of the local population, and proposed action to make improvements.

In addition there should from time to time be independent in-depth research assessment, as has been suggested above.


The three previous points relate to the operation of the service; there should also be feed-back from the justice process to the wider society. If it is found from conferences that many young people who commit offences share certain characteristics (truancy from school, lack of skills, unemployment, family discord, and so on), this information should be passed to the crime prevention agency, so that social policy can be adapted to reduce this pressure towards crime. (Information would not of course be revealed about individual cases without the person’s consent, because proceedings in mediation and conferences are confidential.) For example, a survey of 14 boys aged 15 to 16 in Belgium who had stolen about 50 handbags from women aged 60 to 80 found that most of them had no elderly people in their family, and little or no empathy for elderly people, whom they saw as always grumbling – “ça râle tout le temps” (Grandjean 1993); this suggests that there could be a preventive function in projects where teenagers would meet elderly people. More general problems, such as the association between crime and drug-taking, may be approached similarly.

Points of concern

Restorative justice will have to consider a number of problems, which cannot be examined in detail here. It should be remembered that the traditional system gives rise to many legal conundrums as well. Some of them are legal in nature, for example how to deal with attempts: if no harm was caused, is there anything to make reparation for? Is it acceptable to base the system on harm done rather than, as hitherto, on criminal intent (mens rea)? Is there a risk that when mediation was unsuccessful, the offender would face double jeopardy, being punished for the failure of the conference as well as for the offence (Warner 1994: 150-151)?

Thought will also have to be given to the ultimate sanction for non-compliance, and whether even this can be restorative rather than punitive; to dealing with the offender who shows no remorse; and to the perennial problem of mediation – imbalance between the parties, especially in cases of domestic violence and racial harassment. Ways have to be found for the maintenance of high standards of training and support for facilitators, including complaints procedures for victims or offenders who felt that the procedure was not fairly or competently carried out. Another problem will be to find appropriate and acceptable restorative responses to the most serious crimes; Mark Umbreit in Minnesota, David Doerffler in Texas, and others, have shown that restorative processes between perpetrators and the relatives of homicide victims can take place alongside the traditional system, but could the response to the most serious crimes ever be wholly restorative?

Other questions centre on the differences between the restorative and the retributive philosophies. Howard Zehr, a pioneer of both the theory and the practice of victim/offender mediation, has expressed concern that the criminal justice “industry” make take over the name and use it for un-restorative, punitive practices, with the result that the idea would be discredited (Zehr 1997). The same action, for example paying compensation or doing community service, may be experienced as reparative or punitive, depending on how it is presented; advocates of restorative justice will have to work hard to spread understanding of its benefits.

It would be possible to misapply the philosophy by using it to restore the status quo in an unjust society, and to shift the whole responsibility on to the offender without sufficient regard for the social pressures which at least partly shaped his or her actions (Sandor 1994: 159). What is the balance to be restored: public order and security, or solidarity and social justice (Walgrave 1995: 229)? Should the justice system uphold freedom for the majority with exclusion of the undeserving and feckless, or an “inclusive” society which emphasizes respect for human dignity and a sense of social responsibility, recognizing “the capacity and will of individuals to change – to improve if they are given guidance, help and encouragement; to be damaged if they are abused or humiliated” (Faulkner 1996: 5, 6). A balanced approach will recognize that there has to be a two-way process: offenders should make amends to those they have harmed, but also society should assist their reintegration, and should act on crime-reduction implications which become evident from cases brought to mediation or court. As Braithwaite (1994: 200-1) points out, social unfairness certainly exists, but the criminal justice system is itself a major part of the injustices suffered, especially by ethnic minorities, and needs reform as much as other aspects of society.

The question is not only social, however, but personal; mediation, by focusing on the process rather than the outcome, can be “transformative”: that is, it can help people to recognize each other’s humanity, and this can be more important than merely achieving a settlement (Bush and Folger 1994).

Since Braithwaite (1989) proposed the concept of “reintegrative shaming” it has attracted much controversy. According to Walgrave and Aertsen (in press) Braithwaite has said that he will use the term “restorative shaming” in future; but he has changed the uncontroversial word. There may well be different kinds of shame – Leibrich (1996), for example, distinguishes public humiliation, personal disgrace and private remorse – and good conference practice may promote the latter, which Leibrich identifies as the most influential. But even conference co-ordinators may not all keep on this side of the dividing line, and some sections of the general public are all too ready to aim for public humiliation. In England, the Labour Party (1996), then in opposition, proposed to allow courts to lift the ban on publicly naming some 16- and 17-year-olds; this was headlined in a tabloid newspaper as “Spotlight of shame for the tearaways” (Daily Mail, 21 May 1996). It is reported from Texas, USA, for example, that ex-offenders are being required to place signs on their doors with messages such as “A person released on parole after sexually molesting children lives here” (Le Soir (Brussels), 3-4 May 1997, p. 26). Ex-offenders have to live somewhere, and in a restorative society they could meet their new neighbours, who might say “We know about your past offences, and we will help you to avoid repeating them”. But the judge in Texas made it clear that he believes in the preventive effect of humiliation.

The future of restorative justice

The worst future for restorative justice would be that its vocabulary would be taken over by the retributive system, using the wrong sort of shaming, with degrading and unpleasant community service tasks, and other punitive sanctions. Or it could be inadequately funded, adopted as an appendage to the traditional system, used only for minor cases, perhaps being added to punitive sanctions rather than replacing them.

Can mediation and conferences be used routinely and yet retain their ideals? Much depends on the quality of the implementation of restorative justice methods (Walgrave 1995: 237; Wright 1995), and this in turn depends to a large extent on the avoidance of “funding starvation”, as leading researchers have pointed out (Maxwell and Morris 1996: 102, 105, 108; Morris et al. 1996: 231). Good practice depends on adequate resources to meet the participants’ expenses in attending conferences; for the services needed to enable offenders to make reparation; for the training and support of staff and lay workers; and for independent assessment. These are particularly important in the present experimental phase. It will be better to process a smaller number of cases to a higher standard than vice versa.

The more hopeful future is that restorative measures will be introduced alongside traditional ones, adequately resourced and assessed; and that, provided that the research assessment was favourable, it would be used for an increasing proportion of cases until it became the accepted response to criminal acts. Attitudes could be transformed among participants in the process and, by degrees, in society as a whole.

I am grateful to Ann Warner Roberts for comments and suggestions on the draft of this paper.


Balanced and Restorative Justice Project (1995) Balanced and restorative justice for juveniles: a national strategy for juvenile justice in the 21st century. By Gordon Bazemore and Mark Umbreit. Fort Lauderdale: School of Public Administration, Florida Atlantic University; St Paul, MN: Center for Restorative Justice and Mediation, University of Minnesota.

Braithwaite, John (1994) “Thinking harder about democratising social control.” In: Christine Alder and Joy Wundersitz, eds. Family conferencing and juvenile justice: the way forward or misplaced optimism? Canberra: Australian Institute of Criminology, 1994.

Bush, R A B, and J P Folger (1994) The promise of mediation: responding to conflict through empowerment and recognition. San Francisco: Jossey-Bass.

Carroll, Milt (1994) “Implementational issues: considering the options for Victoria.” In: Christine Alder and Joy Wundersitz, eds. Family conferencing and juvenile justice: the way forward or misplaced optimism? Canberra: Australian Institute of Criminology, 1994.

Faulkner, David (1996) Darkness and light: justice, crime and management for today. London: Howard League for Penal Reform.

Grandjean, Isabelle (1993) Le vol à l’arraché: “Derrière un sac, retrouver une personne” – expériences de médiation victime/jeune délinquant menées dans le cadre des prestations éducatives et philanthropiques. Paper presented to Criminology Congress, Budapest, 1993. Charleroi, Belgium: Centre GACEP.

Home Office (1996) The victim’s charter: a statement of service standards for victims of crime. London: Home Office.

Labour Party (1996) Tackling youth crime: reforming youth justice. A consultation paper … By Jack Straw and Alun Michael. London: Labour Party.

Labour Party (1997) New Labour: because Britain deserves better. (Election manifesto.) London: Labour Party.

Launay, Gilles (1985) “”Bringing victims and offenders together: a comparison of two models.” Howard Journal, 24, 200-212.

Launay, Gilles (1987) “Victim offender conciliation.” In: B McGurk, D Thornton and M Williams, eds. Applying psychology to imprisonment: theory and practice. London: HMSO.

Launay, Gilles (1989) “Victim/Offender groups.” In: Martin Wright and Burt Galaway, eds. Mediation and criminal justice: victims, offenders and community. London: Sage.

Liberal Democrats (1997) Make the difference. (Election manifesto.) London: Liberal Democrats.

Maroochydore State High School (n.d., c. 1995) Community accountability conference. Maroochydore, NSW: the School.

Marsh, Peter, and Gillian Crow (1996) “Family group conferences in child welfare services in England and Wales.” In: Joe Hudson, Allison Morris, Gabrielle Morris and Burt Galaway. eds. Family group conferences: perspectives on policy and practice. Leichhardt, NSW, Australia: Federation Press; Monsey, NY, USA: Willow Tree Press.

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

Maxwell, Gabrielle, and Allison Morris (1996) “Research on family group conferences with young offenders in New Zealand.” In: Joe Hudson, Allison Morris, Gabrielle Morris and Burt Galaway. eds. Family group conferences: perspectives on policy and practice. Leichhardt, NSW, Australia: Federation Press; Monsey, NY, USA: Willow Tree Press.

Morris, Allison, Gabrielle Maxwell, Joe Hudson and Burt Galaway (1996) “Concluding thoughts.” In: Joe Hudson, Allison Morris, Gabrielle Morris and Burt Galaway. eds. Family group conferences: perspectives on policy and practice. Leichhardt, NSW, Australia: Federation Press; Monsey, NY, USA: Willow Tree Press.

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

Pranis, Kay (1997) “From vision to action: some principles of restorative justice.” Church & Society, March/April, 32-42.

Reynolds, Teresa (1977) “The victim’s perspective.” Paper to ISTD/Mediation UK conference “Repairing the damage: restorative justice in action,” Bristol, 20 March 1997.

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: Australian Institute of Criminology, 1994.

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. Leichhardt, NSW, Australia: Federation Press; Monsey, NY, USA: Willow Tree Press.

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, School of Social Work, University of Minnesota, 386 McNeal Hall, 1985 Buford Avenue, St Paul MN 55108, USA.

Victim Support (1995) The rights of victims of crime: a policy paper. London: VS.

Walgrave, Lode (1995) “Restorative justice for juveniles: just a technique or a fully fledged alternative?” Howard Journal, 34 (3), 228-249.

Walgrave, Lode, and Ivo Aertsen (in press) “On `reintegrative shaming’ and `restorative justice’: interchangeable, complementary or different?”

Warner, Kate (1994) “Family group conferences and the rights of the offender.” In: Christine Alder and Joy Wundersitz, eds. Family conferencing and juvenile justice: the way forward or misplaced optimism? Canberra: Australian Institute of Criminology, 1994.

Wright, Martin (1995) “Victims, mediation and criminal justice.” Criminal Law Review, March, 187-199.

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

Wundersitz, Joy, and Sue Hetzel (1996) “Family conferencing for young offenders: the South Australian experience.” In: Joe Hudson, Allison Morris, Gabrielle Morris and Burt Galaway. eds. Family group conferences: perspectives on policy and practice. Leichhardt, NSW, Australia: Federation Press; Monsey, NY, USA: Willow Tree Press.

Zehr, Howard (1997) “Restorative justice hits the big time: but will it remain true to its vision, values?” Accord (Mennonite Central Committee Canada), 15(3), 11-12.

What victims want

Help with practical/emotional effects of


To be treated sensitively and with respect

To know what’s happening in the case

To understand what’s happening in the case

To be heard and taken seriously

Public acknowledgement that wrong has

been done

To feel it is worth pursuing the case

Quick resolution of the case

To know “Why me?”

To know that the offence will not be


Some want an apology from the offender

Some want compensation

Some want retribution (most do not)

Most want to be free of responsibility for

decisions about the offender

(Reynolds, 1997)

(Items provided by V/O M and conferencing are underlined)

Restorative justice for victims

Convenient time and place of hearing;

neutral location

Criteria for selection of cases relates to

victims as well as to offenders

Opportunity to tell story in own words

Opportunity to tell offender effects of crime

Involvement in “making it right”

No involvement in punishment

Involvement of family and/or supporters if


Involvement to be voluntary

Safeguards for voluntary participation of victims

1. Possibility of mediation raised before offender is known

(to allow time to reflect)

2. Mediation offered after decision to caution or sentence

If mediation/conference offered before final decision in criminal justice process:

3. Offender can do community service instead of direct reparation

4. Offender can take part in victim/offender group if victim does not want dialogue

5. Sentence can be independent of result of mediation

(Mediation seen as personal, sentence as public; but unfair to offender who makes reparation? Separate reparation from mediation?)

6. At conference, victim need not be present

7. Complaints procedure

Court problems for victims and witnesses

Would have liked more information about:

court procedures

how to get to court

what to do on arrival

Insufficient consultation on date of hearing

Cases did not take place on scheduled day

Insufficient information about:

time of case

what to do in courtroom

roles of different officials in courtroom

Waiting in same area as defendant

Felt nervous, intimidated, worried or

frightened in witness box

J Raine and R Smith, The Victim/ Witness in Court Project: report of the research programme (1991)

Restorative justice for offenders

Opportunity to speak

Increased understanding

Opportunity to make things right

(Conference:) Brings family together

Assistance in changing behaviour

Safeguards for offenders

Not admit guilt, but not deny act

Access to lawyers

Supervision by court

Complaints procedure

Criteria for assessment



Attendance rate

Amount of reparation agreed/completed


Fairly dealt with

Family strengthened

Reparation agreed/completed

(including rehabilitative)

Availability of community support


The above

Reconviction rate

Points of concern


Harm not mens rea

Double jeopardy

Ultimate sanction

Very serious crimes

Reinforcing unjust status quo