How restorative is restorative justice?

Before answering the question in the title, we need to have an idea what restorative justice is. Howard Zehr (2001a) has reduced the concept of restorative justice to two words: harm, rather than breaking the law, as the focus of attention, and engagement, involvement of the people directly affected by the harm in dealing with it. A widely quoted definition by Marshall (1999: 5) focuses primarily on the process of responding to an offence; the Restorative Justice Consortium has expanded it to mention the three key participants, victim, offender and community, and to include the recovery of the victim as an aim:

Restorative justice seeks to balance the concerns of the victim and the community with the need to reintegrate the offender into society. It seeks to assist the recovery of the victim and enable all parties with a stake in the justice process to participate fruitfully in it. (RJC 1999)

Zehr (2001a) has also pointed out what restorative justice is not. It is not a simple response to minor crimes, but can be helpful to victims of the most serious ones. It does not necessarily consist of mediation, and if it does, there may not be a face-to-face encounter, as we shall see. It does not necessarily lead to forgiveness or reconciliation, although the process may open doors to them. Nor is it a recent phenomenon: it is a ‘river fed by many streams’, from different ages and different continents.

What does this mean in practical terms? ‘A picture is worth a thousand words’, so I will begin with a word-picture of restorative justice in action.

A 13-year-old boy was arrested for a disturbing act of violence: he had delivered a karate kick to ‘Mr Robertson’, a disabled man who suffered from heart trouble. Mr Robertson knew ‘Dan’, the boy concerned, who had teased his daughter, also aged 13, and spread a rumour that she was having sex with her father. Mr Robertson spoke to him about this, telling him that it was dangerous to spread such rumours; a few days later, they met in the street, and Dan kicked him in the chest.

Mr Robertson, Dan and Dan’s mother were willing to take part in mediation. The mediator explained that Dan’s mother was only present because of his young age, and Dan himself should do the talking. Dan said that he was only repeating a rumour that he had heard from others, and that he and his friends had been sniffing glue; Mr Robertson explained what a trauma would have been caused to his family if Social Services had followed up such an accusation. Dan admitted that he had not thought about that; he also described how he had been worried about Mr Robertson, because he had a grandparent and an uncle who had both died of heart attacks.

Mr Robertson said that he thought Dan had been very brave to come to the meeting and talk as he had done; he seemed basically a ‘decent chap’ and should come and knock on his door next time he felt like sniffing glue. Dan said he had been frightened of the meeting, but was glad he had gone through with it; he was surprised that Mr Robertson had been so understanding, and not angry.

Dan received a formal caution, and reported that recently, when asked to join a glue-sniffing session, he told his friends that he had arranged to go elsewhere, and had visited Mr Robertson.

Marshall and Merry (1990: 55-7)

Before we go any further let us look at the word ‘restorative’. In English, it means restoring something which has been damaged to its former condition. My small dictionary does not give riparativo, but riparazione di torto o offesa can mean reparation for a tort or offence, and riparazione di danno is compensation for harm. Riparare includes the idea of rimediare offesa o gaffe (to make up for an offence or mistake), as well as aggiustare (to repair or mend) and proteggere (to protect). Aggiustarsi in turn can mean accordarsi (to come to an agreement). The English word ‘to restore’ can also mean restituire (to give back) or ripristinare (to revive an old custom or law). Already a single word, like a seed, has grown into a tree: all of these are useful ideas to bear in mind when thinking of the idea of la giustizia riparativa, restorative justice.

The word ‘justice’, giustizia, comes of course from the Latin ius, law, and should not be used as if it were synonymous with punishment (as is often implied in phrases like ‘he should be brought to justice’); in a restorative context the emphasis is less on the letter of the law, and more on the a peaceful, well-ordered society, encapsulated in the Hebrew word shalom (Zehr 1995).

We will consider eight branches of the tree, aspects of restorative justice, including the participants, the process, the safeguards, and finally, the use of experience to reduce the likelihood of similar harm in the future; but first, let us look at a description of ways in which restorative justice can be put into practice.

In the conventional system the victim and the offender are not able to tell the story of what happened in their own way, much less to communicate with each other. The damaging potential of the system is well known. It is aggravated by the fact that the system focuses on its objective – to determine guilt or innocence and decide on a sanction – and does not consider the effect of the process on the participants.

Restorative justice, instead of taking the whole matter out of their hands, offers both victim and offender the chance to discuss it and decide how to resolve it, through victim/offender mediation, or an extension of this known as ‘conferencing’, in which victims and offenders are invited to include their extended families. (In this paper ‘victim/offender mediation’ will be used as a general tem to include conferencing, which is similar but with members of the extended family present (as described below). The term ‘penal mediation’ (mediazione penale) will not be used, because mediation is seen as a restorative, not a punitive, process.)

There are at least five ways of doing this, and it is too early to say that any of them are ‘better’ or ‘more restorative’ than others; besides, there should be at least a degree of choice for the participants.

  • The first level, which should be used in all cases, is the initial contact with the victim and offender, to listen to them, explain the process and offer them the opportunity to take part. In some cases there will be no need to continue beyond this point; the offender may send a message to the victim, who may accept it, and the victim may not want to commit time to a meeting.

  • This may require the mediators to make more than one visit back and forth; this is described as indirect mediation, or ‘shuttle diplomacy’. Research suggests that it does not produce such high levels of satisfaction in the participants, but they are entitled to choose it.

  • Thirdly, there may be direct victim/offender mediation, with a face-to-face meeting. This has produced many successful outcomes; it does however depend on the skill of the mediators in conducting the meeting, and especially in using appropriate techniques to compensate for any imbalance of power between the parties. Sometimes either party may be accompanied by family members or supporters, so that there is an unclear dividing line between victim/offender mediation and:

  • conferences. These were first used in New Zealand under the name family group conferences. Based on traditional Maori practice, they bring together the extended family of the offender, and some other significant people in his life, to work out how he should make reparation and keep out of trouble. The victim is also invited to attend, and bring a supporter. Concerns have been raised about this process: Could the offender feel intimidated by a roomful of adults, or the victim by all the offender’s family? At one stage in the procedure the offender and his family are allowed ‘private time’ with no officials or victim present; does this make the victim feel excluded?

Another version is known as a community conference; which can bring in more supporters of the victim, and other members of the community for whom the offence had some relevance. This raises a different concern: if the offender’s family are having problems only indirectly related to the offence, should outsiders be present while they are resolving it?

  • Finally, there are sentencing circles, as developed on the basis of a different indigenous tradition, in Northern Canada. Here relevant members of the local community take part – for example an alcohol counsellor, a representative of the chamber of commerce, neighbours and friends – but also the judge, prosecutor and defending advocate. This means that the circle has the authority of a court of law, and the judge retains his or her powers to impose sentences, including punitive ones, in addition to any reparative or rehabilitative measures that the circle may recommend. These too have possible disadvantages; for example care has to be taken to ensure that one local family or group does not dominate the circle. To assemble a larger number of people takes time, and sometimes travelling expenses, so this method tends to be reserved for more serious offences. However, some observers (e.g. Braithwaite and Strang 2000; Morris and Young 2000) are tending to favour conferences or circles over one-to-one mediation because it is more likely that if anyone tries to dominate, someone will stand up to them, and that the larger number of people is more likely to come up with constructive ideas, and if necessary make them happen – for example, the offender could go and live with them.

One underlying idea is that the crime has taken away some of the victim’s self-determination. The criminal justice process does not help, because it either ignores victims or uses them as a tool to secure a conviction. Restorative justice should help them to regain their autonomy. This depends, however, on the way in which it is applied. In England, under the Crime and Disorder Act 1998 the court can make a reparation order, which may include reparation, and also victim/offender mediation; but the mediation is not seen as apart of the process of agreeing on the reparation. It is true that the victim has to be consulted as to whether he or she wants reparation, but this takes place at an early stage before the victim and offender have had an opportunity to meet, and therefore does not give them back the power to decide what they want. Any country considering legislation would do well to consult closely with those who have studied or practised restorative justice, to avoid this kind of mistake.

Restorative justice can be introduced at various stages in the criminal justice process, and in many ways, the earlier it is offered, the better for both victims and offenders (not to mention the saving of court time); early diversion, however, tends to be limited to the less serious cases. In England the police have discretion to impose a reprimand or a final warning on a young offender in certain cases. Prosecutors have power to discontinue a case when they consider that it is ‘not in the public interest’ to proceed, but this is little used at present; in continental Europe, on the other hand, prosecutors are the main source of referrals to victim/offender mediation. In Italy, the trial of a juvenile can be suspended, with probation, to allow the young person to do voluntary work or other activities of social utility, or victim/offender mediation (art. 28, Decree of the President of the Republic 448/89, cited by Baldry et al. 1998: 377-8). In Italy, the legality principle (mandatory prosecution) may make it difficult for the case to be diverted out of the system in this way before it has reached the court. Prosecutors, however, can make enquiries about a young offender during investigation, and in Turin there is a mediation service to which they can send the case for evaluation. If the outcome is positive, the judge dismisses the case, with the advantage that the offender has had to face the consequences of his act, and the victim has had the opportunity to express their feelings and obtain answers to questions (Baldry et al. 1998: 379-80). Some European countries have modified their legislation to facilitate restorative justice, especially by modifying the legality principle (Trujillo 2000); this is an encouraging development, but care needs to be taken to avoid creating difficulties. The English Crime and Disorder Act has already been mentioned; another example is Poland, where mediators are answerable not to a mediation service but to a judge, and the Committee for the Introduction of Mediation has had to campaign to persuade judges to appoint only trained mediators.

Much of the foregoing implies that the primary purpose of restorative justice is to achieve reparation; but it has also been suggested that the value lies in the transformative power of the process itself, as we shall see (Section 4).

The eight branches of restorative justice are:


1. Restorative justice begins by helping the victim, the person harmed by a crime, to recover.

2. It holds the offender accountable for the harm s/he has caused, and enables him or her to do something towards putting it right, and to earn re-acceptance into society..

3. R J also involves the community in the process and its aftermath


4. Those directly involved in the process are empowered because the outcome is for them to determine, provided that it is restorative; often the offender as well as the victim will need assistance.

5. When the outcome includes reparation, this may take various forms, such as compensation to the victim, work for the community, or co-operation with a rehabilitative programme


6. Courts continue to handle cases which are not suitable for mediation or conferencing. In the ideal restorative justice system they will also have the role of overseeing the process.

7. A restorative process, like any other, requires standards for practice and administration, and hence for training.


8 Finally, crime reduction: information about factors that precipitate crime will emerge from mediation and conferences; this should be passed to those agencies whose work can contribute to crime reduction, for use in formulating policy.

Now let us look at these in more detail. In each case a note about problems with the existing criminal justice process will be followed by some features of restorative justice, and comments on the ‘restorativeness’ of particular practices. Most of the examples are drawn from English experience; listeners may wish to make comparisons and contrasts with current practice and future proposals in Italy.


1. The victim

In the English criminal justice process victims have no special status except as witnesses, although steps have been taken recently to keep them better informed, and treat them better when they come to court. In Italy too they were often ‘treated as a “piece of evidence”’, although, as in some other European jurisdictions, they may have the right to sit in court as a civil party and ask for restitution (Baldry et al. 1998: 377).

People whose lives have been disrupted by a crime have various needs, which the conventional criminal justice system is only now beginning to address: information about the process, acknowledgement of the harm they have suffered, action to put things right as far as possible, action to make it less likely that other people will suffer the same experience, and a process for responding to the crime doing so which does not cause them more pain.

Restorative justice begins by trying to meet these needs of the person who has been harmed by a crime. Sometimes the word ‘healing’ is used, or ‘putting things right’. In some cases, of course, complete restoration is not possible: when something with great sentimental value has been stolen, or a permanent physical injury has been caused. This paper will be primarily about individual victims and offenders; but there can also be corporate victims, ‘legal persons’, for example shops and banks. If a bank is robbed, the bank is sometimes listed as the ‘victim’, but the real victim is the member of staff who faced the robber. There can also be corporate offenders, for example large companies which cause pollution or neglect the health and safety of their workers. Whether their conduct is criminal depends on the law of the country concerned; even if it is, the criminal law may not be the best way to secure redress for the injured and bereaved, and improvement in the company’s future behaviour (Braithwaite 1989).

An essential part of the restorative concept is that it is to help the victim; it is not merely a new way of dealing with offenders. (This means that restorative justice should not be limited to young offenders, because if there is an benefit for the victim, this should not be dependent on the offender’s birthday, although there may be practical and political reasons for introducing restorative justice for juveniles as a first step). As regards relationships between victims and offenders, there are three possibilities: the offender is not detected, the victim and offender know each other, or the crime is committed by a stranger.

Firstly, a central aspect of the restorative idea is to enable the victim and the offender to communicate, as we shall see, but in the majority of cases this is impossible because the offender is not detected; in that case there should be support from members of the community, and professional help if needed. Another possibility is to arrange for groups in which victims can meet other offenders who have committed similar offences against other victims, so that they can get answers to at least some of their questions.

Secondly, however, in many of the more serious crimes, especially violence, the clear-up rate is fortunately higher, not least because many victims and offenders knew each other previously. Mediation then offers the possibility of resolving the dispute and maintaining the relationship, if that is what both parties want. Indeed, if there is a community mediation service in the locality, the case could be taken there instead of to the police and the courts. There is a theory that everything which could be classified as a criminal offence, should be; but everyone knows that if it were, the whole system would collapse.

Thirdly, in the case of a detected crime by a stranger, restorative justice offers the chance for the victim and the offender to meet. But it will not fulfil its restorative potential if it is not done well. To take three examples: the general focus of the system, the person who contacts the victim, and the way in which they are contacted.

A problem in England, under the Crime and Disorder Act 1998, is that the stated primary aim is not the recovery of the victim but the prevention of repeated offences; in other words, it is focused on the offender. In addition, the government has given a pledge that it will halve the time taken to bring persistent young offenders to court; the result has been that many victims have been contacted hurriedly or not at all, since the timing is dictated by the court’s timetable and not by whether the victim has recovered sufficiently to consider mediation.

Procedural details can be important: for example, Who makes the contact? The first official agency to contact the victim is usually the police; but in England there is a Data Protection Act, intended to protect privacy, and this has been interpreted as meaning that the police must not pass the victim’s name and address to any other agency, without their express permission. It is therefore a police officer who must ask the victim whether he or she would like to know more about mediation. In some places police officers have been trained either to explain, or preferably to say that a mediator would be the best person to do so, and victim take-up rates of up to 87 per cent have been reported; but elsewhere the take-up has been very low, which suggests that a better way of making contact is needed.

How well is the process explained? The explanation should be based on exploring what the victim wants, not what the mediator thinks they want; and it should avoid raising expectations. Two common reasons for wanting to take part in mediation are: for the victim’s own benefit, to allow them to ask questions, tell the offender how they feel, and so on; and for the offender’s benefit, when the victim as a responsible citizen is willing to give some time in the hope of helping to change his attitudes and find better ways of fulfilling his potential in life (which will benefit the community too). In either case the victim should not be encouraged to hope for too much, and should understand for example that the offender may not even make a satisfying apology: disappointed hopes are worse than none, and sometimes re-victimize the victim. The visitors should also explain to victims that, if they do not wish to meet the offender, indirect mediation is available (see below).

2. The offender

As for the offender, in the conventional trial, in England at least, the accused may have to say nothing but to confirm his name and address and to say whether he admits guilt; anything else is said on his behalf by his lawyer. Defence lawyers commonly see it as their task to minimize the punishment of their client. This may be done by securing an acquittal, if necessary by exploiting procedural safeguards against wrongful punishment; or if a conviction cannot be avoided, to belittle the crime and the offender’s culpability. This is unhelpful to the victim, whose need is to hear the seriousness of the offence acknowledged; and it focuses the offender’s thoughts on the harm the punishment will do to him, not on the harm he did to the victim. Restorative justice challenges the notion that punishment is an effective response to wrongdoing; at the very least, it should be recognized that the more severe the punishment, the less the chance of arriving at the truth, and vice versa.

If there is an expression of apology, this too is made by the lawyer, and it is in the context of the plea for reduced punishment, which detracts from its credibility. In England, the Crime and Disorder Act 1998 falls short, because the offender is ordered to make reparation, with no opportunity to do so voluntarily first, still less to discuss with the victim what form it should take. The sequence of events does not encourage restorative processes: an official asks the victim about reparation before there has been an opportunity for victim/offender mediation to take place.

Restorative justice, in contrast, holds the offender answerable for the harm he has caused. To be ‘answerable’, in English as in Italian, means both essere responsabile (to be responsible) and dover rispondere o render conte (to have to respond or give an account). What effect does R J have on the offender? There is a paradox: in some ways it could be seen as ‘softer’ than punishment, but facing a victim can be much harder. ‘Softness’ and ‘hardness’ are not the point: what restorative justice offers is more understanding. The law states the fact that an offence is prohibited; restorative justice shows how and why harming another person is wrong. The act is condemned, but the actor is not humiliated, so that his self-esteem is not still further damaged and his reintegration into society is less difficult. In this atmosphere it is more likely that he will accept responsibility for his act, whereas the threat of punishment encourages him to try to evade it by exploiting procedural safeguards which have been introduced to protect the accused against wrongful conviction. In some places the methods followed under the name of restorative justice contain strong elements of punishment, which have been described as ‘authoritarian’ rather than ‘democratic’ restorative justice (Wright 2000), and they are likely to lead to the same adverse side-effects as punishment does.

A restorative process should safeguard the offender’s rights, however, as much as a conventional one. Mediation should not take place unless he admits the alleged facts; if he denies them, they should be tested in court. It is not however necessary for him to admit legal guilt, so that if the mediation process cannot continue or is unsuccessful for any reason, he retains the right to opt for a court trial at any time and to plead Not Guilty. A further safeguard is required because of the victim’s involvement, or rather their right to choose whether to be involved or not. Suppose an offender admits the act and is willing to make amends but the victim does not want to take part; the offender should still be able to make reparation through some form of service to the community. This also places an obligation on the community to provide suitable work, as we shall see later.

All offenders have rights, and restorative justice aims to make them face their responsibilities; in addition many also have needs, which need to be addressed if the offender is to be reintegrated into society. Restorative justice does not aim directly to prevent the individual from committing further offences; it seeks to achieve this by persuading and enabling rather than by threats of punishment. Under restorative justice the community offers to meet offenders’ needs, and asks the offender to co-operate as part of his reparation, because his needs do not excuse him for harming another person. This will be referred to below (section 5).

3. Involving the community

Since restorative justice aims to heal the community as well as the individuals directly implicated in a crime, it is sometimes described as communitarian, and can involve members of the community at all levels: individuals, such as employers and landladies, including ‘legal persons’; the victims’ and offenders’ ‘communities of care’, their families, friends, and others whose opinion they value; people who live locally; and NGOs. All of these can offer help and support in different ways, to both victims and offenders. In a different sense the ‘community’ means local and national government, to whom people pay taxes to pay (among other things) for the provision of services.

In the context of restorative justice, community involvement may be arranged at the organizational level: rather than keep the whole process in the hands of statutory agencies, the organization of mediation and conferencing can be entrusted to voluntary organizations (NGOs). Secondly, individuals may be trained to act as mediators, either as volunteers, or as the same sort of ‘ordinary’ people who become volunteers, but paid a fee for each case they undertake. Another possibility is for a professional and a volunteer to work in pairs. In some countries, including the United Kingdom, the work of contacting victims to offer support is largely carried out by volunteers.

Finally, the community al all levels has an important role through making it possible for the offender to make reparation, whether through the efforts of individuals (employers, landladies) in providing jobs and accommodation; clubs and faith groups in offering support to a person in spite of his past; NGOs and statutory agencies for such initiatives as providing rehabilitative services such as Alternatives to Violence Projects, addiction treatment programmes, and opportunities for reparative work with people, such as taking people with disabilities to the shops in their wheelchairs, teaching disabled children to ride horses).

The criminal justice system does not rely primarily on volunteers. In England the largest category of volunteers consists of the magistrates who preside over the lower courts. In some countries, such as Poland, there are volunteer probation officers; and throughout the world much humanitarian voluntary effort, which could be better used, is engaged in mitigating the harmful effects of penal measures imposed by the state itself, through organizations for the assistance of prisoners, ex-prisoners and their families.

However attractive community working is, there clearly has to be some involvement of statutory agencies, and it is worth looking at the experiment in the English Crime and Disorder Act 1998, which created ‘Youth Offending Teams’, bringing together local authority agencies; not only those such as Police and Probation which have some experience of working together, but also Education and Health, which are relevant to ‘bullying’ (which sometimes amounts to assault or robbery) and to drug addiction and mental health. It has not been entirely easy to build these new teams (Holdaway et al. 2001), but it is generally thought to have been a sound innovation.

Those who believe in the maximum involvement of the community would recommend that as much as possible should be done by NGOs, and the NGOs themselves should use trained volunteers., as part of the process of healing and empowering the community.


4. Empowering the parties

Criminal justice, as we have seen, takes control out of the hands of the parties who are directly affected. It dictates that there will be certain outcomes (conviction or acquittal, sentence); the process of reaching them is very formal, and takes account of its correctness, not its effect on the parties. Restorative justice invites the parties to deal with the matter themselves, except that if they do not wish to, or cannot agree, the court, on behalf of the state, remains available, as we shall see.

But if the power has genuinely been handed over to them, can ‘we’ (the state) say what the outcome should be? If we do that, we immediately take back some of the power we have given them. Reconciliation, or an agreement on reparation, may seem desirable to us, but is it what the parties themselves want? One belief of restorative justice is that, unlike the criminal justice system, it recognizes the value of the process, as well as the result. Thus it has been argued that a discussion which fails to reach agreement, but leads the parties to understand each other better, is more satisfactory than one in which the mediators push the parties to an agreement which they do not fully accept. There have been cases where a victim has accepted an agreement in which the offender would not pay compensation for the value of a stolen article, but would for example give them a present as a sign that he is sorry, or would do some work for a charitable organization such as the Red Cross, or co-operate with a rehabilitative programme which will help him to stay out of trouble..

This raises questions, which cannot be dealt with in detail here, but some answers will be briefly indicated as a basis for further discussion::

  • Should victims be able to require punishment as part of any agreement? – No, because they have neither authority nor training to do so, and punishment is damaging, not restorative.

  • Should they be able to require deprivation of liberty, not as punishment but for the protection of the public against a substantial risk that the offender will commit additional offences? – No, for the same reason. Since it is not punishment (defined as the infliction of pain for its own sake), it is not ‘unrestorative’, but any such decision should be made by a court.

  • If the offender appears to need treatment or training, and this is not included in the agreement, should the court impose it? – This has been part of the rehabilitation debate for a long time. There is a case for saying that it is justified (a) provided that it is not disproportionate to the seriousness of the offence (which is of course an arbitrary assessment), and (b) in cases such as addiction, where the individual is not fully in control of his actions. In England the new Drug Testing and Treatment Orders work on this principle. Restorative justice does not only restore the victim: often the offender will also need support.

It is easy to criticize the existing criminal justice process; but restorative justice, like anything else, can be done well or badly. The first requirement is to clarify and prioritize the aims and objectives of the service. One controversial way of making sure that they are followed is to provide mediators with a script, giving the exact wording of every question they ask, but many practitioners are resistant to this method. If the aim is to enable the participants to regain control, this does not appear to be the way to do it.

There is no doubt that poor practice can be found: inadequate screening of cases to ensure the psychological or physical safety of participants, revealing information about the victim to the offender, failure to employ power balancing techniques, or to offer shuttle mediation where a victim could not face direct mediation, or to make sure that the parties were aware of alternatives to mediation (Landau and Landau 2000, writing in the context of domestic violence, but applicable to other types of offence).

The American author and practitioner Mark Umbreit (1998) has described some of the ways in which practice can be more or less restorative. It is more restorative if both victim and offender have at least one preliminary interview to explain the process and allow them to decide whether to take part, and if the victim is offered some choice as to where and when the mediation session will take place. The way in which the mediators conduct the meeting should be non-directive, and they should tolerate both the expression of emotions and periods of silence. A session which gives due space for a healing process will probably last for at least an hour; where many people are present, for example in a sentencing circle, the typical duration will be from four to seven hours, with a break for lunch. Where the converse of these conditions is found (short meetings, pushing the participants towards a rapid agreement, and so on), it will have less of a restorative effect on them.

5. Reparation

In the English legislation, compensation orders (money payments) and reparation orders (work for the victim or the community) come close to being seen as punishment, because they are ordered by courts, and indeed the former community service order has, since 1 April 2001, been re-named the ‘community punishment order’ – a clear retreat from restorative ideals. It could be argued that the practice has become punitive, and the re-naming has merely reduced the level of hypocrisy; but conversely, to abandon the more idealistic name may make it harder, in the future, to return to the idealism of the original concept. The length of a reparation order must be commensurate with the seriousness of the offence, rather than with any agreement between the victim and the offender. But if it is accepted that the participants should be given carte blanche, within limits, to use their opportunity for communication in the way that seems best to them, we can no longer assume that there will be a tangible outcome labelled ‘reparation’. The offender’s participation in the process may itself be the reparation. But if they do agree on reparation, to complete the process, it can take several forms. It may be made directly to the victim: an apology, monetary compensation, a present, or some hours of work (the latter is more likely in the case of a corporate victim (‘legal person’). Some victims want nothing for themselves but prefer the work to be done for the community; in Germany a payment of a sum of money to a charitable cause is also an option. Finally, many victims want to feel that the offender is sincere when he says he intends to stay out of trouble, and will be reassured if he takes part in a suitable rehabilitative programme.

Here again, the way in which it is done can make a difference as to how restorative it is. It is less restorative if the emphasis is on how hard it is for the offender to face his victim, on the amount of money paid or the number of hours of work done; more so, if there is dialogue and negotiation. If the work is unskilled, involves no contact with its beneficiaries, and especially if it is done in a public place, it is not merely less restorative but comes close to being punishment disguised as reparation.

The guidance accompanying the English legislation recommends that the work should if possible be related to the offence; there is a case for this, but it should be borne in mind that unpleasant work for its own sake has no place in restorative justice: the more important consideration is that many offenders lack self-esteem, and the overriding objective should be to arrange for them to complete a task satisfactorily and to know that the beneficiaries have valued their contribution.


6.. Courts

Courts have often been criticized for their rules, their formality and other characteristics; but they do have a necessary role. This paper has emphasized the role of voluntary individuals and organizations in delivering restorative justice; but courts would have two main functions. Firstly, they would continue to handle cases which are not suitable for mediation or conferencing. These may include those where the accused denies any involvement, or where one party exercised their right to leave the proceedings; and those where an additional sentence of restriction of liberty has to be imposed for the protection of the public. As far as possible such sentences would be restorative; that is, persons undergoing restriction or deprivation of liberty would be able to make reparation as described above (section 5). Secondly, they would have a new function: to oversee the process to ensure that it puts restorative principles into practice, as they do in New Zealand.

Lawyers too would have a different role. There has been some concern that their adversarial methods could undermine the mediation process, and it has therefore been proposed that they should be kept out of it. But in New Zealand the post of youth advocate has been made attractive to lawyers, and applicants are closely questioned to make sure they understand that their job is to ensure that the restorative process is carried out well, in addition to safeguarding defendants’ rights (Zehr 2001b).

There is hope that a restorative process can be potent enough to influence the courts in a restorative direction (Wright 2000: 26), and to create pressure to extend it from juveniles to adults. The New Zealand and Austrian experience has shown that this is possible, but the government of that country has been active in promoting the idea, for example by producing videos, and according to David Carruthers, Principal Youth Court Judge on New Zealand, such continuing promotion is still necessary even after ten years (Carruthers 2001).

7. Quality assurance

Courts are not noted for encouraging monitoring and research, except on such superficial factors as the throughput of cases; and indeed their multiple aims make it all but impossible to evaluate their work, or to say whether a particular judge is a good sentencer (Wright 1999). As regards mediation, as we have seen, it depends on the process, and therefore the process should be recognized as the primary aim. A combination of routine monitoring and evaluation, and periodic in-depth research, should help to establish what are good practices; these should then be incorporated into standards and training covering such practice questions as confidentiality and the conduct of the mediation session, and administrative ones such as a complaints procedure for participants and an equal opportunities policy. There should also be a procedure for revising them in the light of experience and of new situations that arise. In the United Kingdom standards have been drawn up by Mediation UK (1998) and the Restorative Justice Consortium (1998), and other countries have done or will do likewise; more general principles have been compiled by the Council of Europe (1999) and the United Nations (2000). These will form a useful basis, although care has to be taken, especially in international documents where it can be difficult to reach an agreed text, to ensure that less-than-perfect formulations do not become ‘set in stone’; either they should be general enough to permit flexibility, or there should be a procedure for revising them. .


8. Crime reduction

As we have seen, in conventional criminal justice, the focus is on establishing the facts: discussion of the context is not encouraged, either because it might incriminate the accused, or because it is not regarded as relevant. In a mediation or conference, in contrast, background information about factors that precipitate crime will emerge; this should be passed to the strategic crime reduction for use in formulating policy. It may appear for example that a particular area needs more recreational facilities for young people, or more information about the dangers of drugs, or simply more employment and better housing. The involvement of volunteers will help to spread understanding of these issues throughout the community. The volunteers, however, should represent a cross section of the community, or rather of the communities which exists in most countries; if the NGOs and their volunteers are made up largely of, for example, well educated and middle-to-high income strata of society, they may not heed the signs that society needs to change. Restorative justice does not have to be confined to the criminal justice process. As always, there are advantages in starting with young people. Mediation by children, for children (known as ‘peer mediation’, mediazione fra pari), has been shown to work in two ways. Firstly it can be effective in stopping bullying (a word which can be used to describe acts such as robbery or assault); secondly it can teach children how to discipline themselves, so that teachers have less need to impose disciopline upon them. A pilot study in Rome found that 44 per cent of children who had been bullied said they had not spoken to their teachers, but 56 per cent had spoken to their own friends, which shows the confidence they have in their peers (Baldry 1997: 163). Crime prevention is closely linked to social policy; and by drawing attention to deficiencies in the social fabric, restorative justice can help to plan and to create a more just and pleasant society for everyone.

I am grateful to Margarita Zernova for commenting on an earlier draft; but I am responsible for the views and any errors.



Baldry, Anna Costanza (1997) ‘Bullismo a scuola e mediazione fra pari.’ In: Gianvittorio Pisapia and Daniela Antonucci, eds. La Sfida della mediazione. Padua: CEDAM.

Baldry, Anna Costanza, Gaetano De Leo and Gilda Scardaccione (1998) ‘Victim/offender mediation in the Italian juvenile justice system: a first attempt at definition.’ In: János Boros, Iván Münnich and Márton Szegedi, eds., Psychology and criminal justice: international review of theory and practice. Berlin: Walter de Gruyter.

Braithwaite, John 1989) Crime, shame and reintegration. Cambridge: Cambridge University Press.

Carruthers, David (2001) ‘A mainstream restorative justice system: New Zealand youth justice at work – the theory and the reality.’ Address to conference, ‘Restorative and community justice: inspiring the future’, Winchester, 28-31 March 2001.

Holdaway, Simon, Norman Davidson, Jim Dignan et al. (2001) New strategies to address youth offending: the national evaluation of the pilot youth offending teams. London: Research, Development and Statistics Directorate, Room 201 Home Office, 50 Queen Anne’s Gate, London SW1 H 9AT.

Landau, Barbara, and Niki Landau (2000) ‘Domestic violence policy: lessons to be learned.’ Interaction (The Network: interaction for conflict resolution, Waterloo, Ont.), 12 (2), 4-5 (see also pp. 1, 3)

Marshall, Tony F (1999) Restorative justice: an overview. Home Office, Research, Development and Statistics Directorate, Rm 201, 50 Queen Anne’s Gate, London SW1H 9AT.

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

Mediation UK (1998) Practice standards.: Mediation UK, Alexander House, Telephone Avenue, Bristol BS 1 4 BS.

Restorative Justice Consortium (1998) Standards for restorative justice. RJC, c/o Society of Black Lawyers, Room 9, Winchester House, 11 Cranmer Road, London SW9 6EJ.

Restorative Justice Consortium (1999) Manifesto. RJC, as above..

Trujillo Matias, Francisco Jesus (2000) Mediation: would it work in Spain too? Unpublished dissertation, Master of European Criminology, Catholic University of Leuven, Belgium.

Umbreit, Mark (1998) Victim/offender mediation comtimuum: from least to most restorative impact and Quality restorative justice practice: grounding interventions in ket restorative justice values. Handouts from Center for Restorative Justice and Peacemaking, University of Minnesota.

United Nations. Economic and Social Council. Commission on Crime Prevention and Criminal Justice (2000) Basic principles on the use of restorative justice programmes in criminal matters. Draft resolution. JUSTICE_UN_BPbody.htm

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

Wright, Martin (2000) ‘Restorative justice: for whose benefit?’ In: European Forum for Victim/Offender Mediation and Restorative Justice, ed.: Victim/offender mediation in Europe: making restorative justice work. Leuven, Belgium: Leuven University Press.

Zehr, Howard (1995) Changing lenses: a new focus for crime and justice. Scottdale, PA: Herald Press.

Zehr, Howard (2001a) ‘Offenders, victims and community: justice of the future.’ Paper presented to conference, ‘Restorative and community justice: inspiring the future’, Winchester, 28-31 March 2001.

Zehr, Howard (2001b) Talk at London Mennonite Centre, 2 April 2001.

Conf papers/Rome how restorative 014xx