Restorative legislation: not too little, not too much, but just right

final 27.12.2007

The development of mediation and restorative justice in Bulgaria and other countries in recent years is very welcome, and I believe that it will be used more and more widely as the benefits become better known. The new Bulgarian legislation, the Mediation Act 2004, mainly concerns civil mediation but also provides that mediation can be used in criminal cases (article 3(2)), and in order to consider how to put this into effect, this paper will look at selected developments in other countries. Some of these ideas may already have been implemented; some are worth considering; others fall short of restorative ideals. Similar considerations will apply wherever restorative justice is being introduced; its proponents in each country will have to decide which methods are suitable in their case. In some important ways they are very different from conventional notions of justice, but they should not be rejected merely for that reason.

 To begin, it may be worth briefly reviewing the aims and objectives of conventional crime policy. General crime prevention is principally a matter for social policy (social crime prevention) and physical measures (situational crime prevention). Increasing the severity of punishment has not been shown to have a significant effect, and has serious side-effects, especially when it involves the use of imprisonment. Insofar as deterrence is effective, it is achieved primarily by increasing the perceived probability of detection. Prevention also includes preventing those who have offended from doing so again; this is ideally done by persuading (and if necessary enabling) the offender. For the most serious crimes, with a serious likelihood of repetition, containment (detention) is necessary; but it is hoped that even that can be done in a more restorative way, as has been proposed by John Blad (2006).

 Restorative justice: a new dimension

Restorative justice adds a new dimension to these conventional aims: concern to restore the well-being of the person harmed by crime, as far as possible. This has three aspects: support for victims whose offenders are not convicted, including state compensation for victims of violence (this has developed separately in many countries, but its spirit is part of the same ideal: repairing the harm caused by crime); reparation by the offender; and psychological and emotional redress, which in many cases can be achieved by allowing the victim to take part in a mediated dialogue with the offender.

 Another feature of restorative practice in some countries is that members of the community are involved in the process, either as participants in a victim-offender meeting or ‘conference’; as ‘proxies’ representing the victims; as trained volunteer mediators; or as members of a non-governmental organization that provides the service.

 Lastly, the restorative process allows the background of the offence to be uncovered, and this understanding can help to formulate social crime prevention policies.

 The meaning of the word ‘restorative’ has become broader: it is now often used to mean processes in which decisions are made by a structured problem-solving discussion among the people involved, instead of being imposed by authority from above. Words like ‘restorative practices’ are being used, not only in the justice system but in schools, workplaces, communities and elsewhere.

 In England at present there is much concern about ‘anti-social behaviour’, usually by young people: it consists of actions which are not serious in themselves, but make everyday life unpleasant for a large number of people. Here is an example of how it was dealt with in Germany.

 A group of foreign young people repeatedly broke into an open-air swimming pool, threatened and insulted its users, and committed thefts. They were prosecuted several times, which had no effect on their behaviour. After a few weeks they were completely banned from entering the pool; they then threatened visitors outside, and demanded money with menaces, so that people were afraid to go in. They ignored a ban on going into the area, and broke in at night, causing considerable damage to the fence and buildings. Eventually they were brought to have individual meetings with some of their victims. What took place there is not described, but it must have had an effect on them, because they apologised, and then took part in a ‘round-table’ meeting, where it was agreed that they would make reparation by taking responsibility for various functions at the pool during opening hours. This showed local people that they now had a different attitude and wanted to be integrated. The manager took on the role of responsible adult, gave them tasks and supervised them and, when he was satisfied with their performance, took them behind the scenes to see the heating, filters and other technical equipment. This increased their interest and sense of responsibility. There have been no more incidents involving this groupi.

 Impact of legislation on restorative justice

Existing legislation

This paper will consider practices in selected jurisdictions, and the extent to which they fulfil the restorative ideal. There are four ways in which legislation can affect the introduction of restorative justiceii. You will be able to judge in which category the legislation in your country falls. Firstly, it may be possible to start restorative justice programmes within existing legislation. This was done with pioneering programmes in England, where the law allows 28 days after a person is found guilty, during which the probation service prepares a pre-sentence report to assist the court in deciding on the sentence. In a few places the probation service used this period to arrange a victim-offender mediation. The law also allows the court to defer (postpone) sentence for up to six months, to see whether the offender shows signs of changing his or her life; this too can be an opportunity for mediation. If the offender makes reparation, or at least begins to do so, the court will then pass a non-custodial sentence. Incidents arising out of a dispute, for example between neighbours or workmates, can be dealt with by community mediation services instead of being treated as criminal offences. The disadvantage of working within the existing system is that there may be difficulty in finding funds for the programme; and as the use of restorative justice is optional, there may be very few cases, so that the cost per case is high. Finland, which pioneered mediation with no new legislation from 1983 onwards, has now passed a law to provide a more systematic service; but the legislators have decided not to ‘over-regulate’ mediationiii. In South Africa, however, the Zwelethemba project (described below) even felt that specific legislation would kill its strong community base; it was enough that the project operates within the existing law (Wright 2004).

 Enabling legislation

Secondly, a law may be passed which allows the use of mediation but does not require it. This allows mediation services more independence and flexibility, but (like existing legislation) it can mean that they have to work hard to persuade probation officers, prosecutors and judges to refer cases – and also to obtain funding. The Finnish Act of 2005 goes some way to resolving these problems by affirming the right of all citizens to have access to mediation and, crucially, by ensuring government funding (Kinnunen 2007) Another way of encouraging the use of mediation is to make sure that a mediated outcome is counted as an offence successfully brought to justice, just as much as a case that has been prosecuted.

 Less-than-restorative legislation

Thirdly, there may be legislation which limits the use of restorative justice or makes it more difficult. As one author has commented, ‘when restorative justice operates under the aegis of the criminal justice system, this may lead to restorative ideals being diluted and restorative justice forced to serve the objectives of the criminal justice system’ (Zernova 2006). The Crime and Disorder Act 1998 (sections 67-68) in England and Wales provided that the court could order a young person to make reparation to the victim; the court would decide the type of reparation, and the victims would be asked whether they consented, before any contact had taken place between the victim and the offenderiv. In practice, some youth offending services have ‘got round’ the Act by agreeing with their local court that the court would decide the length of the order, and the victim and offender could meet to agree on what form the reparation should take. But it should not be necessary to ‘get round’ the law.

 The ‘legality principle’ can create a difficulty when it prevents prosecutors from diverting cases out of the system(Miers and Willemsens 2004: 158). In Germany, for example, this was overcome by means of Article 153a in the Code of Criminal Procedure (StPO) which allows the prosecutor to suspend the prosecution (for offences carrying a minimum sentence of a fine or less than one year’s imprisonment, and subject to the agreement of the court and the accused) and to impose certain reparative measures or contributions to the community, if this would render prosecution no longer in the public interest. If the requirements are fulfilled, the prosecution cannot continue (StPO §153a, summarized from A Eser 1983: 25 n. 17).

 Normative legislation

Lastly, the law may provide for a restorative process to be the normal procedure, at least in certain categories of case. The first country to do this (but only for juvenile offenders) was New Zealand. The Children, Young Persons and Their Families Act 1989 introduced a system by which all offences by juveniles (except where a death occurs, or where the accused denies guilt) are dealt with in one of three ways: a warning by the police, possibly with an informal sanction (about three quarters of cases); referral by police to a youth justice co-ordinator, who arranges a family group conference (FGC) (about 12 per cent); or arrest and charge in a youth court, which must also normally refer the case to a FGC before deciding the sentence (about 12 per cent) (New Zealand 1989; Maxwell and Morris 1994; Morris and Maxwell 2001v).

 The framework is shaped by the purposes of sentencing, which may or may not be stated in the law. In England and Wales the Criminal Justice Act 2003 lists punishment as the first aim, followed by reduction of crime (including deterrence), rehabilitation of offenders, protection of the public, and reparation. In New Zealand, however, the Sentencing Act 2002, sec. 7(1), which likewise applies to adults, states that the purposes of sentencing (summarized) are:

(a) to hold the offender accountable for harm done;

(b) to promote in the offender a sense of responsibility for, and an acknowledgement of, that harm;

(c) to provide for the interests of the victim;

(d) to provide reparation for the harm;

The first three of these are new. Only then does it list the more conventional aims of denouncing the offender’s conduct, deterring the offender or others, protecting the community, assisting the offender’s rehabilitation and reintegration, or a combination of these (although it also says that the aims are not listed in order of importance).

The New Zealand courts must, among other things, ‘ take into account any outcomes of restorative justice processes, that have occurred or that the court is satisfied are likely to occur’ (sec. 8(j), emphasis added), including any offer of amends (financial or by performing a service), any agreement between the offender and the victim as to how the offender may remedy the wrong, loss or damage, or ensure that the offending will not continue or recur, and measures proposed by the offender’s family to apologise or make good the harm (sec. 10 (1)(a – d)). The court may adjourn the proceedings until compensation has been paid, work completed, the agreement fulfilled, and so on (sec. 10 (4)(a – e)). Where restorative principles are not applied, however, the Act reverts to retributive principles, with punishments that are intended to be proportionate and can be aggravated or mitigated (sec. 9).

 A pilot project was conducted in four courts, with built-in evaluation: 180 victims and 180 offenders were to be interviewed shortly after the conference and one year later, 90 conferences were to be observed and 90 more documented(Parker 2002). The research found, like almost every other research on restorative justice, that the majority of victims were still satisfied after a year (although almost a fifth said they would not attend a conference again). Offenders referred by a court to conferencing were compared with a comparison group. There were significantly lower reconviction rates after conferencing for violent offenders, traffic offenders (driving causing death or injury), and offenders who had committed theft or other offences, in other words all types except fraud and burglary offenders. Other conferencing groups who had a lower reconviction rate were: offenders with one or two previous proved cases; males; offenders aged 25 to 29 years or 30 to 39 years; and medium and high-risk offenders (i.e. offenders with a predicted reconviction rate of 25% or more) (Crime and Justice Research Centre 2005).

 In Austria, the law gives the prosecutor discretion to waive prosecution. If certain conditions are met, the case must be diverted if this promotes the injured party’s interests to the greatest extent possible. If the prosecutor does not divert the case, the court may propose an out-of-court resolution (aussergerichtlicher Tatausgleich). Prosecutors also have discretion in Norway, under the Criminal Procedure Act 1998.

 Victim-offender mediation in Northern Ireland was started by two groups based in the Nationalist and Loyalist communities respectively. Mediation is often seen as an alternative to the official justice system; in this case it was established as an alternative to punishment beatings by paramilitary gangs. The idea was officially adopted, and mediation, or ‘conferencing’, is now part of the juvenile justice system, since the Justice (NI) Act 2002 established the referral of cases by youth courts and the Public Prosecution Service to youth conference co-ordinators. Extension of the programme to adults is under consideration. The community-based programmes, however, are now finding great difficulty, because the authorities are imposing conditions which they are unable or unwilling to meet (O’Mahoney and Doak 2006).

 In England and Wales, youth courts must make a ‘referral order’ referring certain young people who admit their guilt to a youth offending panel, which is restorative insofar as it focuses on making an action plan rather than imposing punishment, it may include reparation, the victim may attend, and the panel includes trained volunteers. This does not apply to offences which are too trivial or too serious, and it is limited to young people who are making their first appearance in court. The proportion of victims who attend is low, although efforts are being made to improve this. The Youth Justice Board is currently carrying out a consultation about referral orders, which will give an opportunity for their ethos, training and practice to become more restorative; any such moves will however be limited by the fact that the panels are embedded in a fundamentally retributive system.

What should and should not be legislated?

In some countries, such as Finland, it was possible to introduce restorative justice under existing legislation; this of course depends on the legislation, and it does not provide a firm base: the programme is dependent on persuading the relevant agencies to refer cases to it, and to provide funding. Let us look at some of the points which have to be considered when introducing victim-offender mediation. In each case, there is an additional question: should the legislation order this, or merely permit it? Does any existing legislation need to be amended, in order to make it allowable, or to clarify the relationship between restorative justice and the criminal justice system? How much should be specified in detail in the law, and how much can be delegated, for example to an NGO? Some of the key questions are:

Relationship with criminal justice system

At what stage of the criminal justice process should it be used?

For which types of case would it be suitable?

Do the parties have the final word on the agreement?

What should happen if an agreement is not reached, or not kept?

What safeguards are needed for participants?

Voluntary participation

Must the accused admit involvement or guilt?


Time limits

Using the process to improve crime prevention

Arrangements for funding

Legal framework for the mediation service

For which types of victim or offender would it be suitable?

Which model(s) of mediation should be used?

By whom should the mediation service be delivered? Who trains the mediators, and how are they supervised and supported?

Let us consider these in more detail.

 Relationship with criminal justice system

At what stage of the criminal justice process should it be used? As the Council of Europe (1999) recommends, mediation should be available at all stages of the criminal justice process (Principle no. 4). If the incident involved a dispute between people who know each other, for example neighbours or work colleagues, they could choose to go straight to a mediation centre, rather than to the police. If it is reported to the police, ‘in places like Austria, New Zealand and Norway, restorative justice is not an alternative to the system, it is the system. It is the norm and the court system is only there as a fall-back’ (Liebmann 2007: 30). There are however limitations on the types of case treated in this way; for example in Austria a case may be diverted to mediation if the maximum punishment is not more than five years’ imprisonment (e.g. serious bodily injury, burglary), and in New Zealand the system was originally for juveniles only, although it is now being extended. In Finland, on the other hand, there are almost no limits: any type of offence, regardless of the age of the offender, may be referred, at the discretion of the police and the prosecutor, with the exclusion only of vulnerable victims, and sexual offences against children. In practice domestic violence is also excluded (Kinnunen 2007: 6)

 The Justice (Northern Ireland) Act 2002 gives the prosecutor has discretion to ‘divert’ a case involving a juvenile to the youth conference co-ordinator, and is expected to do so in most cases (O’Mahoney and Doak 2006). If the case is prosecuted in court, and the accused is convicted, the court must refer the case to a conference before deciding the sentence, unless the offence carries a maximum sentence of life imprisonment or can only be tried on indictment (in the higher court with a jury), and it may refer these cases. Only terrorist offences cannot be referred. In cases where the offender disputes the facts but is found guilty, the case may proceed to a conference if the young person consents.

 Post-sentence. In some countries, such as Belgium, restorative methods are used in prisons. It can be helpful to both victims and offenders to reach an understanding before the offender is released. In addition it can be used to handle disputes between prisoners, and as the effects and benefits become known, between prisoners and staff (Willemsens 2004; Liebmann 2007: ch 9-10).

 For which types of case would it be suitable? We need to distinguish suitability according to the type of offence, and according to the individuals involved (see below). Research shows that restorative justice is most effective in the more serious cases, although it will then generally be combined with another sentence (Sherman and Strang 2007). It is usual to introduce it at first for juvenile offenders and relatively minor offences, but as it becomes better known and accepted, it can be extended. In defining seriousness, it is common to use the maximum punishment as a yardstick This has the advantage of simplicity. It would however be a welcome sign of recognition of the victim, if harm suffered were the criterion of seriousness, for example theft to the value of less than x months’ income, injury resulting in less than y months’ absence from work, not entering a house without consent and with intent to steal, no sexual contact without consent, or other offences with comparable effects on the victim.

 Similarly, it is often introduced at first for juvenile offenders, which is more acceptable to public and professional opinion; but that excludes the victims of adult offenders, so these should be included as soon as possible. In Norway, mediation can be used for adults, but mostly for juveniles, and for civil disputes as well as crimes in about equal proportions.

 It is important to remember that restorative justice is not limited to ‘street crimes’ like theft and robbery, but can be applied to ‘white-collar’ offences. The New Zealand judge Stan Thorburn (2005) reports the following casevi:

 A large intensive pig-farming company polluted a river when a leaking pipe allowed a huge quantity of pig effluent to flow into drains. When it was prosecuted by Waikato District Council it took full responsibility, and initiated a restorative justice process. It apologised, contributed to a local tree planting project, and paid the legal costs of the council and of the restorative justice facilitator. The judge considered that it had also repaired the damage to its relationships with local peoplevii.

Regulations are often derided as the invention of power-hungry bureaucrats; but failure to follow regulations on such matters as fire escapes, unsafe machinery, and nutritional standards in institutions can lead to illness, injury and death. In England, the Better Regulation Council, which is responsible for enforcing legislation on health and safety and similar matters, is very interested in approaching enforcement using a restorative approach rather than a punitive one (Macrory 2006) Professor Braithwaite has shown how a restorative approach can be effective in regulating nursing homes. In Australia, inspectors threw away the old rule books, with hundreds of regulations which were not kept; after discussing the essentials with management, staff and residents they replaced them with just 31 standards. They spent time talking to residents and staff about how the quality of care could be improved, even in nursing homes with the sickest residents. Performance against each of the 31 standards was discussed at a conference of the inspectors, managers, residents and relatives, and sometimes someone from an outside advocacy group similar to a Helsinki human rights committee). Thus instead of prosecution the regulation moved away from the criminal model, and there was an improvement both in the quality of life for the residents and in compliance with the law (Braithwaite 2002: 17). This restorative approach might also be helpful in children’s homes where there have been reports of extremely low standards of care.

 Do the parties have the final word on the agreement? In New Zealand, when a case is diverted instead of going to trial, it appears that the parties generally reach an agreement that is acceptable to both. When it goes to court, about 80 per cent of agreements are endorsed by the judge. In the most serious cases, however, as mentioned above, it appears that judges think first in terms of a punishment, but if a restorative process has taken place they can reduce the punishment in accordance with section 8(j) of the Sentencing Act 2002, sometimes even replacing it with a non-custodial one such as community service. Judge Thorborn maintains that ‘tension between retributive and restorative principles can be blended.’ In Norway, it appears that the sentence comes before the mediation: mediation can be added to a community sentence or a suspended sentence, if both parties agree.

 What should happen if an agreement is not reached, or not kept? It appears that agreements are more likely to be kept if they are agreed by the victim and offender rather than imposed by a court. But if they are not, this should also be handled in a restorative way. The first step would be to discuss with the offender the reasons why he or she had not kept the agreement: had there been a change in his or her circumstances, or had he or she not received the support that was needed? . In a restorative system the law should allow some discretion; it is possible, for example, that so many conditions had been imposed on the offender that he could hardly remember them all, let alone keep to them. Otherwise the law may allow a new conference to be held, or may send the case back to court; in Northern Ireland the court may then refer the case back to the Youth Conference Co-ordinator. In Finland, victims can enforce the agreement; but victims’ advocates might think that this is too much of a burden to place on them.

 What safeguards are needed for participants? the first safeguard is the training, support and supervision of the mediators, who should know and understand the codes of principles and ethics of mediation such as the Council of Europe’s .recommendation, referred to above, and the Draft guidelines for a better implementation of the existing recommendation concerning mediation in penal matters (CEPEJ(2007)13prov2) issued by the European Commission for the Efficiency of Justice (CEPEJ) in November 2007. The law may require the mediation service to draw up the training programmes and codes, but if they are not part of the law, it is easier to revise them in the light of experience. The Zwelethemba programme in South Africa, for example, has created its own Code of Good Practice (Cartwright and Jenneker 2005).

  • We help to create a safe and secure environment in our community

  • We respect the South African Constitution

  • We work within the law

  • We do not use force or violence

  • We do not take sides in disputes

  • We work in the community as a co-operative team, not as individuals

  • We follow procedures which are open for the community to see

  • We do not gossip about our work or about other people

  • We are committed in what we do

  • Our aim is to heal, not to hurt

Other countries have drawn up their own codes of standards, principles and ethics. They show that restorative justice is indeed bringing about a revolution in thinking: it is replacing the ancient lex talionis. The old axiom said that when a person has broken the law, he or she should be punished; the new one says that when a person has caused harm, he or she should as far as possible put it right (Wright 2007).

 The mediation service should also be required by law to provide a system for support and supervision of mediators, although the details can be left to the mediation service. If a mediator performs badly, or is the subject of a complaint, the supervisor should discuss this with them, and if necessary require them to take additional training before continuing to mediate. One safeguard is a follow-up process, by which participants are invited, a few weeks later, to say how the mediated agreement is working, and also to comment on the mediation process itself. This should be a normal part of the management system.

 Mediation services commonly exclude lawyers from taking part in mediation, because their training tends to promote adversarial attitudes, although the accused has the right to consult one before and after. They should understand the principles of restorative justice, and if they are allowed into the mediation session, they are only there to advise, not to represent the offender.

 The mediation service should be required by law to have a procedure by which participants are informed of the standards and can complain if the mediators or the mediation service do not appear to have followed them, and this process should itself include an opportunity for independent mediation at an early stage.

 Voluntary participation. There is universal agreement that victims should not be pressured into taking part in mediation against their wishes. On the other hand it has been argued that they have a right to be informed of this service so that they can decide, and affirmation of this point in legislation would be welcome. .

 As regards offenders, it is commonly stated (for example in the Principles of Restorative Processes of the Restorative Justice Consortium (2004)), that Participation [is] to be voluntary for all participants and based on informed choice’. The Bulgarian Mediation Act 2004 (Art. 5) makes the same point. The United Nations says that ‘Restorative processes should be used only … with the free and voluntary consent of the victim and the offender’, although later it qualifies this: ‘Neither the victim nor the offender should be coerced, or induced by unfair means, to participate …’ (emphasis added). Where a restorative process is part of the system, as in New Zealand, the offender’s choice may be very limited: take part willingly or go to court. However, this is a choice; it can be argued that since they did not give their victim a choice, it is fair that they should have to face the consequences; and they should be given credit for doing so. If the offender shows reluctance, the victim should be informed, and of course there should be no meeting if there is any danger of physical or psychological harm to either party.

 Must the accused admit involvement or guilt? Some countries, such as England and Wales, require a full admission of guilt, in the legal sense, before a restorative process goes ahead. This is presumably intended as a safeguard against self-incrimination, but it means that if the accused think they can escape conviction, perhaps through lack of sufficient evidence or because of a procedural mistake by the prosecution, they may well refuse. Others, such as New Zealand, only require that the accused ‘does not deny’ being involved. This means that there is something to discuss with the victim; but if the mediation is not successful or the reparation not completed, he or she can still have a legal defence if the case is returned to the court.

 Confidentiality The United Nations Basic Principles (2002) state (No. 14) that ‘Discussions in restorative processes that are not conducted in public should be confidential, and should not be disclosed subsequently, except with the agreement of the parties or as required by national law.’ Thus where the case is diverted from the criminal process, discussions in the victim-offender meeting would not be revealed (unless there was information about a serious crime), on the grounds that without this confidentiality, victims and offenders night not feel able to speak freely. If the case goes to court after mediation, the same principle applies, but a statement agreed by both parties could be given to the judge.

 Time limits Governments regard speedy justice as the ideal, and from some points of view they are right; but those which have a concern for victims recognise that victims may require a little time to decide that they are ready to meet their offender, and after that it may take some more time to find a convenient date and time, especially if conferencing is used, because it involves a larger number of people.

Using the process to improve crime prevention

In England and Wales the involvement of members of the community as volunteers makes them aware of some of the background, and they may use this understanding to try to bring about change, although there is no in-built encouragement for them to do so. For example, one member of a youth offender panel was so concerned about the number of offenders coming from a particular school that he went to speak to the school governors to persuade them to take action (Crawford and Newburn 2003: 152). This may involve situational crime prevention: for example, one large retail store changed its security system after young offenders explained how easy it had been to shoplift there. A restorative approach encourages offenders to co-operate in this way in order to make things right.

 The greatest potential, however, is for social crime prevention. This has been built in to the Community Peace Programme (Cartwright and Jenneker 2005) in about twenty communities modelled on one in Worcester, a town of about 75 000 inhabitants in Western Cape, South Africa, known in the Xhosa language as Zwelethemba, meaning ‘a place of hope’. It was established in 1997, with the strong encouragement of the Minister of Justice. It was designed as a system of local governance; dealing with conflicts and crime is incidental. Each community forms a Peace Committee of five to twenty persons. When a complaint is brought to a Peace Committee, it organizes a PeaceMaking Gathering of the disputants and other people who may be able to help, facilitated by three or more people. They read the Code of Good Practice (see above) , and help the parties to resolve the dispute. They agree a plan of action to reduce the likelihood that the same thing will happen again. At the end they normally do something to symbolize their commitment to what has been decided. By 2005 about 11 000 Peace Gatherings had been held.

 Conflicts mainly concern money, insults, threats or gossip (similar to complainant offences) or property offences. More serious offences are referred to the police; those which cannot be resolved by consent are also referred, because the Committees do not use force to solve problems. As in Northern Ireland, part of the intention is to avoid vigilante action. The police, in turn, may offer complainants the option of taking suitable cases to the Peace Committee.

 For every gathering, a payment is made to the Peace Committee, from which two thirds goes to the facilitators and – this is the most distinctive feature – one third goes into a PeaceBuilding Fund (about 2 500 000 rands, or approximately 250 000 euros, had been paid into the fund by 2005). This is used to fund projects in the community to help to deal with problems that have come to light in the gatherings; as far as possible it is spent within the community. Examples are loans to small businesses (to increase employment), building a children’s playground, and repairs to an old people’s home (Roche 2003: 264-6).

 Arrangements for funding

When cases are diverted out of the criminal justice system, costs are saved: Prison sentences may be shorter, or replaced by community sentences. If cases are diverted before the court hearing, court time will be saved, and the accused will not spend so much time in prison waiting for trial. The costs of administering restorative justice, however, usually come from a different part of the budget. A way needs to be found, therefore, to transfer the savings of prison costs to fund the restorative justice services. It has been suggested (Wright 1999: 18-19) that a system should be introduced by law by which, for every thousand by which the prison population is reduced, a sum equivalent to, say, three quarters of the average annual cost would be transferred every year to the restorative justice agencies, and the remaining quarter to improving education and training in prisons.

 Legal framework for the mediation service

For which types of victim or offender would it be suitable? The suitability of the process for victims and offenders should be assessed when they are visited by mediators; if they are willing to proceed, and it does not appear that a meeting would make matters worse, or risk physical or psychological danger to anyone, the meeting should go ahead. This is a matter for professional judgement rather than legislation. It is therefore suggested that the decision as to the seriousness of the offence should be made by a prosecutor or judge, while the assessment of the individuals could be delegated to mediation services and trained mediators.

 Which model(s) of mediation should be used? Methods are developing all the time; if legislation is too detailed, it may be an obstacle to progress. It is probably best for the law to provide the framework by laying down the general principle that meetings can take place between victims and offenders (or disputants, in cases where it is not clear who is the offender and who is the victim). Within those limits, the details of the model are probably best left to the mediation service. It could be said that requirements for safeguards should be guaranteed by law; differences in procedure, where there may be differing views about which is preferable, can be left to guidelines which mediation organizations produce in the light of experience and research.

 To some extent the models are a continuum: one-to-one mediation, the same with a relative or other supporter, or ‘conferencing’ meetings with several members of the extended family. In the ‘New Zealand model’ of ‘family group conferences’, after the preliminaries, the offender and his or her family are left alone to consider their proposals for an action plan. This usually includes reparative and rehabilitative measures, but may also include a formal caution or appropriate penalties. In cases that are returned to court, the court may add other conditions such as a community work order, a supervision-with-residence order, or for offenders aged 15 or over, may refer the case to the District Court for sentence (New Zealand 1989: sec. 260, 263, 306.

 ‘Community conferences’ may also include representatives from community organizations, drug and alcohol treatment programmes, community-based work programmes, and so on. In Northern Ireland, 83 per cent of conference plans include activities, usually provided by the voluntary (NGO) sector (O’Mahoney and Doak 2006). One-to-one mediation may be indirect, that is, the parties do not meet but the mediators ‘shuttle’ backwards and forwards between them; but these visits can be regarded as preliminaries to explain the process and prepare the parties for the meeting. There should be no pressure to take part, although face-to-face meetings tend to be more successful, and some services have found that if they do not offer the alternative of indirect mediation, the take-up of the offer of face-to-face meetings is higher. Others, however, have experimented with contacts by telephone, written statements, letters, and video or audio tape recordings, as well as using proxy victims.

 By whom should the mediation service be delivered? Who trains the mediators, and how are they supervised and supported? The mediation or ‘conferencing’ may be fully integrated into the state system, as in New Zealand and Northern Ireland. In New Zealand, the youth justice co-ordinator (YJC) has responsibility for ensuring that the principles of the Act are being met (New Zealand 1989; Maxwell and Morris 1994). YJCs are officers of the Children and Young Persons Service.

 There may be overall state regulation, with some autonomy at local level (Fig. 1). But it is more usual to insert a specialist mediation body, which insulates the mediation service from the direct control of the government (Fig. 2). In Norway, mediation is a separate governmental service, under a Conflict Resolution Council; it is integrated into the justice system, with 22 local mediation services corresponding to police districts. There are about 700 mediators; they are volunteers, aged 18 or over, requiring no specific professional background; They must be enfranchised within the municipality, and must not have received a custodial sentence. They undergo four days’ training,, and are appointed for four years. They receive a small hourly fee and expenses (Bolstad 2004).

 Another example which leaves overall regulation in the hands of the state, but delegates the practical operation of the service to an intermediate body, is in Finland, where there is a strong tradition of local provision of mediation services; overall responsibility lies with the Ministry of Social Affairs and Health, but national guidance and monitoring is in the hands of a new Advisory Board on Mediation in Criminal Matters. Finland also uses volunteers (about 900), receiving only nominal reimbursement for their expenses. They are co-ordinated by about 90 to 100 staff in 25 regional offices, and are expected to handle about 10 000 cases annually (Kinnunen 2007).

Fig. 1

Fig. 2

Mediation for both juveniles and adults is administered in Austria by NEUSTART (New Start), formerly called the Association for Probation and Social Work. It is nominally an NGO, and has its own management, which gives it some independence from government, but its independence is limited by the fact that it is subsidised by the Ministry of Justice. Mediators must have a professional qualification in social work, or a degree in law or psychology plus practical experience of social work; and mediation training lasts three to four years. They normally work only as mediators, and do not work also as probation officers.

 In Poland mediation began as an initiative of an NGO, the Polish Centre for Mediation, which conducted pilots and evaluated them, and lobbied for changes in the law to enable victim-offender mediation to take place. Here the detailed control by the state appears to be less; the price paid is lack of adequate funding.The law lays down that individuals can be mediators if they are 26 years old or more, have Polish citizenship, have no criminal record, and have experience of social work or similar work. They must not however be currently employed in the criminal justice system.  The Bulgarian Mediation Act (Art. 4) also makes this requirement; mediators must be included in the Ministry of Justice register (Art. 8), but otherwise there are no restrictions, This apparently leaves the way open for the involvement of members of the community as mediators.

 In addition to the organization that administers the mediation service, there may be an association of mediators. The volunteers who work in the youth offender panels in England and Wales have formed an association, and I understand that an umbrella organization of mediators is being formed in Bulgaria (Chankova 2005).


 Life is simpler for conservatives than for progressives, because there is only one way of standing still but many different ways of making progress. This is certainly the case with restorative justice. At its most basic it is no more than another way of dealing with offenders; but the more it is practised, the more clearly we can see that it is much more than that: it is a different philosophy of how people interact, and how to persuade people to act in an acceptable way towards each other. This means holding discussions among those who are affected, and the restorative movement is developing better ways of structuring and facilitating these ‘conferences’ so that everyone is listened to with respect. It can begin in schools, by showing children how to resolve conflicts by seeking common ground. It has been found that the introduction of restorative practices can transform the ethos of a whole school, and there is reason to believe that making mediation generally available could have a similar effect on communities and society as a whole.

 Some incidents arising from inter-personal conflicts do not need to be treated as criminal at all, but can be taken directly to a mediation centre. Others may be diverted out of the criminal justice system; or the criminal justice process can include a restorative meeting. This interaction between criminal justice and restorative processes needs to be clarified in the law; but this paper has suggested that some aspects of mediation do not need to be ordered by legislation, they may either be contained in subordinate legislation, or, as in some countries, left to an approved non-governmental body to decide – and to change as necessary in the light of experience.

 Restorative justice will need further development, but it offers the promise of helping victims to overcome the effects of the harm they have suffered, and offenders to make amends for the harm they have caused. It offers a kind of alchemy, by which the shame which the offender feels at what he or she has done is not weighed down by additional stigma, but is converted into something like pride at the way he or she has responded, by making reparation for the harm caused to the victim. Punishment adds to the stigma of the wrongdoer, and does nothing for the victim; restorative justice gives the offender the opportunity to overcome it, and the victim, as well as the offender, the opportunity to be involved in the process. Greater understanding of pressures towards crime leads to better crime reduction policies. Rejection is transformed into re-acceptance in the community. This is shown in the two case histories: the young hooligans who showed that they could take responsibility for the swimming bath, and the company which cleared up the pollution which it had caused. In both cases, and many others like them, people who damaged the community became people who benefited the community.



Aertsen, I, et al., (2004) Rebuilding community connections: mediation and restorative justice in Europe. Strasbourg: Council of Europe.

Blad, J R (2006) ‘The seductiveness of punishment and the case for restorative justice: the Netherlands.’ In: D J Cornwell, Criminal punishment and restorative justice: past, present and future perspectives. Winchester: Waterside Press.

Bolstad, T (2004)‘Norway’, in: D Miers and J Willemsens, eds. (2004) Mapping restorative justice: developments in 25 European countries. Leuven: European Forum for Victim-Offender Mediation and Restorative Justice

Braithwaite, J (2002), Restorative justice and responsive regulation. New York: Oxford University Press, p. 17

Cartwright, J and M Jenneker (2005) ‘Governing Security: a working model in South Africa – the Peace Committee’. Paper to conference Consolidating transformation: a new decade, Western Cape, South Africa, 7-8 Feb 2005

Chankova, D (2005) ‘Positioning mediation in Bulgarian justice system: the new mediation act significance.’ Paper to Fifth International Conference of the World Mediation Forum, Crans-Montana, Switzerland, September 2005

Council of Europe. Committee of Ministers. (1999) Recommendation No. R (99) 19 … to member states concerning mediation in penal matters. Reprinted in I Aertsen et al., 2004..Rebuilding community connections: mediation and restorative justice in Europe. Strasbourg: Council of Europe.

Crawford, A and T Newburn (2003), Youth offending and restorative justice: implementing reform in youth justice. Cullompton, Devon: Willan Publishing.

Crime and Justice Research Centre, Victoria University of Wellington (2005)New Zealand Court-Referred Restorative Justice Pilot: Evaluation. Prepared by the … Centre … with Sue Triggs for Ministry of Justice. 2005.

Eser, A (1983), Einführung in das Strafprozessrecht. München: C H Beck.

Kinnunen, A (2007) , ‘Restorative justice and mediation in criminal cases in Finland.’ Newsletter of European Forum for Restorative Justice, 8(3), 4-6.

Liebmann, M (2007) Restorative justice: how it works. London: Jessica Kingsley Publishing.

Macrory, R B (2006) Regulatory justice: making sanctions effective. Final report. London: Better Regulation Executive.

Miers, D and J Willemsens, eds. (2004) Mapping restorative justice: developments in 25 European countries. Leuven: European Forum for Victim-Offender Mediation and Restorative Justice.

New Zealand (1989) Children, Young Persons and Their Families Act 1989 base=pal_statutes.nfo&jump=a1989-024%2fs.272&softpage=DOC#JUMPDEST_a1989-024/s.272 ;

Maxwell, G M and A Morris (1994). ‘The New Zealand model of family group conferences’, in: C Alder and J Wundersitz, eds, Family conferencing and juvenile justice: the way forward or misplaced optimism?. Canberra: Australian Institute of Criminology.

Morris, A and G M Maxwell (2001) ‘Implementing restorative justice: what works?’. In: A Morris and G M Maxwell, eds, Restorative justice for juveniles: conferencing, mediation and circled. Oxford: Hart Publishing.

O’Mahoney, D and J Doak (2006), ‘The enigma of “community” and the exigency of engagement: restorative youth conferencing in Northern Ireland.’ British Journal of Community Justice, 4(3), 9-24.

Parker, L (2002)‘New Zealand expands official recognition of restorative justice.’ (accessed 29.12.2007)

Restorative Justice Consortium (2004) Principles of restorative processes. London: the Consortium.

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

Sherman, Ł W, and H Strang (2007) Restorative justice: the evidence. London: Smith Institute. Accessible on

Thorburn, S A (2005) ‘Observing the application of restorative justice in courts of New Zealand (a brief survey of cases over 10 years).’ Paper for International Symposium, Shenzhen City, China.

United Nations Office on Drugs and Crime (2002) Basic principles on the use of restorative justice programmes in criminal matters. Reprinted in UN ODC (2006) Handbook on restorative justice programmes. New York: United Nations.

Willemsens, J (2004) ‘Belgium’, in: D Miers and J Willemsens, eds. (2004) Mapping restorative justice: developments in 25 European countries. Leuven: European Forum for Victim-Offender Mediation and Restorative Justice.

Wright, M (1999) Restoring respect for justice. Winchester: Waterside Press,

Wright, M (2004) ‘Peace committees in troubled communities’. An account of a seminar by John Cartwright. , 2004, accessed 16.11.2007

Wright, M (2007)‘‘Punishment and restorative justice: an ethical comparison.’ In: R Mackay et al., eds. Images of restorative justice theory. Frankfurt am Main: Verlag für Polizeiwissenschaft, 2007

Zernova, M (2006) ‘Working under the aegis of the criminal justice system: implications for restorative justice practice.’ British Journal of Community Justice, 4(3), 67-79.


i Thanks to Professor Arthur Hartmann and Frank Winter, TOA-Bremen e.V. for this example.

ii Miers and Willemsens (2004) distinguish three types, permissive, coercive and mandatory. Most of the information in this paper comes from this source unless otherwise stated.

iii Act on mediation in criminal and certain civil cases (1015/2005). See A Kinnunen (2007).

iv A similar problem seems to be developing in the application of the conditional caution, under the Criminal Justice Act 2003.

v Police referrals have fallen to 8 per cent, with 16 per cent going to youth court, according to M Liebmann, 2007: 266.

vi . Waikato Regional Council v. PIC New Zealand Ltd., 2004, cited in S A Thorburn, (2005)

vii This leaves out of account the morality of intensive farming of animals, which is a separate issue.