» Restorative Conferences with Serious Juvenile Offenders: An Experiment in Belgium (PDF)
From "Building a Global Alliance for Restorative Practices and
Family Empowerment," the Fourth International Conference on
Conferencing, Circles and other Restorative Practices,
28-30 August 2003, Veldhoven, Netherlands
Restorative Justice as a Goal-oriented Approach to Justice
While no generally accepted definition of restorative justice exists, I consider restorative justice as "an option on doing justice which is primarily oriented towards restoring the harm that has been caused by a crime" (Walgrave 2002: 192).
Restorative justice is thus characterized by its aim of restoration and not by the process which typically favours restorative outcomes. Such a goal-oriented definition is not commonly accepted. Many restorative justice proponents consider voluntary deliberative processes in an informal context as the key to restorative justice (Marshall, 1996; McCold, 2000; Boyes Watson, 2000). They all rightly promote informal voluntary settlements as being crucial for achieving restoration maximally. Nevertheless, restorative justice cannot be reduced to such process for two reasons. First, no process can be defined and valued without referring to the purpose it is undertaken for. Defining a process without referring to the goal it aims at is like pedalling in the air. A deliberative process is valued not because of the deliberation on its own, but because of the outcomes it helps to achieve. Why would a mediation or a conference be more "restorative"? Because the expressions of remorse, compassion, apology and forgiveness that it facilitates lead more easily to feelings of being respected, of peace and satisfaction. These feelings are outcomes aimed at, even if they are not explicitly written in the agreement.
Second, restricting restorative justice to voluntary deliberations would limit its scope and doom it to stay at the margins of the system. In the currently predominant model, the mainstream response to crime is coercive and punitive, but the criminal justice system can refer a selection of cases to deliberative restorative processes. As the system keeps the gate to possible restoration, it will probably refer less serious cases only, excluding the many victims who need restoration the most from its benefits. In many cases, agreement cannot be reached, or local deliberation may be insufficient, for a variety of possible reasons. Coercion may then be considered, which however must still primarily serve restoration as much as possible (Walgrave 2002). Such sanctions in view of restoration do not fulfill the potential of the restorative paradigm completely, because they are enforced and do not result from voluntary agreements. But restorative justice is not just a black-and-white option; it can be achieved in different degrees (McCold and Wachtel 2002, Van Ness 2002).
That is why, in my view, restorative justice must be defined through its restorative aims and not through its most important means, deliberative processes.
Deliberative Processes as Crucial Tools for Restorative Justice
The preference for deliberative processes is derived from the goal to restore. Voluntary, informal deliberation between the parties with a stake in the aftermath of an offence is by far the most appropriate to achieve reparation and genuine restoration. That is understandable.
The communicative potentials of mediation, circles or conferences indeed favour the authentic assessment of the harm suffered and may more easily lead to a true agreement on how it can be reasonably repaired or compensated. Ideally, the safe climate of respect and support offers all participants the opportunity to freely express in their own words what they have felt and still are feeling. The material harm, mental and relational suffering and social unrest become more genuinely perceptible. The atmosphere favours an open deliberation which leads to a better understanding of what should be done to repair the damage and to restore peace.
Such process is also crucial to enhance the offender's motivation. Probably most offenders originally seek simply to get away with the least possible unpleasant consequence. Many offenders agree to cooperate in a mediation or a conference because they think that this will be less severe than going to court (and under informal social pressure). But once they are involved in the process, a stream of moral emotions develops. Based on personal relations with the participants and on general empathy with other persons, including the victim, the offender is brought to feel intensely a mixture of all kinds of unpleasant emotions, like shame, guilt, remorse, embarassment and humiliation (Braithwaite and Mugford 1994, Maxwell and Morris 1999, Harris 1999, Morris 2000). It is this dynamic, developed in a positive atmosphere, that ideally favours his understanding and his personal commitment to repair (Harris, Walgrave and Braithwaite 2003). The offender's commitment is crucial to improve the "restorative calibre." It expresses his understanding of the wrongs committed and harms caused, as well as his willingness to make up for it, which communicates compliance with social norms. Through the recognition of the harm, the value and the rights of the victim are recognized as well. The restorative value of such gestures for the victim, the community and the offender is thus much greater than if the offender would comply only to avoid further trouble, or because he is forced to do so by a juridical sanction.
Altogether, such deliberative processes appear to be much more healing for all participants (Van Ness and Heetderks Strong 2001), as is also obvious through the empirical satisfaction studies with victims and offenders (surveys in Braithwaite 2002, Kurki 2003, McCold and Wachtel 2002).
Choosing an Adequate Model for Working with Serious Offenders in a Legalistic Context
Many versions of deliberative processes in view of restoration exist. Several versions of victim-offender mediation, different kinds of sentencing circles and various models of conferencing are currently being carried out all over the world. Literature and a personal visit to Australia and New Zealand led to the intention to begin a conferencing experiment in the Belgian juvenile justice context, more precisely in a few Flemish juridical districts.
For two reasons, the experiment addressed serious juvenile offenders. First, we expected that a succesful implementation for serious offenders would be an a fortiori argument for its possible extension to all offenders.
Second, Flanders has a well-developed mediation practice running for juveniles referred by the public prosecutor's office (Claes et al. 2003). These generally are less serious cases for which a succesful mediation may lead to non-prosecution. In the year 2002, for example, 1,587 juvenile cases have been mediated (OSBJ 2003)1. This programme is broadly available and applied with high success rates, so it would be unnecessary and unwise to compete with it. One may even wonder whether systematically implementing conferences for benign juvenile delinquency is not a kind of overreaction.
The option for working with serious offenders brought us to choose the original family group conference model as the basic model. There are several reasons for that.
The New Zealand family group conference is located more centrally in the juridical procedure than the other conferencing models. It may be used as a form of diversion, but it may also serve to prepare court decisions. Not limiting it to a form of diversion is crucial for working with serious offending, where public prosecutors will in any case want to prosecute. Moreover, submitting the outcomes for a court's approval may be necessary in a legalistic law regime like Belgium.
For serious offending, the role of the police as participants seems more appropriate than as organisers. The police's role is to inform about the facts of the case and to express and safeguard public interests in the settlement of the offence. This seems to be less the case when the police are involved as facilitators. When a police officer acts as the facilitator, he is in fact neutralised in his specific role as representative of public interests (that is why some practices do invite a "community representative," which sometimes appears rather artificial). Police-facilitated conferences, therefore, appear rather as cautioning sessions meant as diversion. One may even wonder whether the decisions are really left to the participants.
The model allows for the presence of advocates. This is important in discussing what will happen after serious offences, because it can include heavy obligations involving painful restraints for the offender. It is also better to include the advocates in the debates, because the outcomes will have to be submitted to the youth court, where the advocates will in any case have their say.
Setting Up Family Group Conferences in a Legalistic Context
An important difference in legal regime exists between the Anglo-Saxon countries, which are based on common law principles, and European continental countries, which are based on civil law principles. It has consequences for the ways in which restorative justice practices can develop. Common law regimes impose less strict procedural and sentencing rules. The police and the judiciary have considerable discretionary power, and large margins are provided for community input and debate. Victims and other community members play a role in the settlement, and there is room to experiment outside or at the margins of the judiciary. It is, therefore, no coincidence that most restorative justice experiments have emerged in common law countries.
The legality principle that dominates the civil law systems on the European continent prescribes strict legal rules and allows a limited scope only for exercise of discretion. In principle, the police only record offenses and must refer every case to the public prosecutor, a professional judge who oversees the criminal investigation and decides whether to prosecute. Possible restriction of freedom is very limited without a court's order. The input of nonprofessional citizens is limited. This results in less flexible systems that are less permeable to innovation but offer solid grounds for safeguarding legal rights. As a consequence, it is Europeans especially who raise procedural and jurisprudential questions about restorative practices, most of which were "invented" in the Anglo-Saxon countries.
In view of this fundamentally different legal context, it appears obvious that it is quite impossible to copy the New Zealand way of including family group conferencing (FGC) in the judicial procedure for implementation in a Belgian juridical context. It cannot be the police who will decide to refer to a conference. The role of the police in the conference is much less evident, because only the public prosecutor is competent to represent public interests. No decision in the conference can be carried out unless it is confirmed by a court's order.
We had to find a solution to that, which brought us to include FGC in the Belgian procedures as follows.
Cases are referred for possible conferencing by the youth court judge. The public prosecutor has thus already sent the juvenile to court in view of prosecution. This guarantees the exclusion of benign first offenders (who may be sent to mediation in an earlier phase). The conference is thus also not a diversionary option. The court's referral is done through a "provisional order," which in the Belgian juvenile justice system is provided for social investigation. The wording of the referral is more or less that "the potentials of an FGC must be investigated." In theory, no selection is made, except that the juvenile must not deny the facts. In practice, judges seem to make their own selections.
The agency that will execute the FGC starts up the process and pays the preparatory home visits. If these lead to the facilitator's conclusion that an FGC is not appropriate—the case is not serious enough, the juvenile refuses to cooperate, an FGC risks being detrimental for the victim—he will send the case back to the judge, who will decide further as if this tentative planning did not occur. In other cases, the conference is organised as in New Zealand.
Contrary to the New Zealand situation, however, the outcome is not an "agreement," but a "common declaration of intention" signed by the participants. This declaration is sent to the judge, who will take it over in a court's judgement.
Whereas the judges can in principle, of course, decide not to confirm the declaration of intention, we have insisted to the judges that their referral to a conference should include their moral commitment to confirm the declaration, in respect for the emotional investment of the participants. So far, no accidents have occurred. In the same judgement, a date for a new court session in the future is decided, so that the period for carrying out the intentions (now "judgements") is fixed. This final session will then discharge the juvenile if he has succesfully complied.
When I proposed in the year 2000 to try to set up an FGC experiment, I got the enthusiastic support of two youth court judges, who had visited the New Zealand system in the frame of a programme of the King Baudouin Foundation 2. It was the vehicle to be taken seriously and to gain positive interest in the juridical world. Now FGC is possible in five juridical districts—Antwerp, Brussels (Dutch-speaking courts), Hasselt, Leuven and Tongeren—chosen for a mixture of pragmatic and principled reasons. From five NGO agencies that already offered mediation for juveniles, ten experienced mediators were selected to receive an intensive training in FGC, given by Allan Macrae, at that time an experienced youth justice coordinator in Wellington, New Zealand.
A organisational steering network was set up, composed of local advisory boards in every district and an overarching steering committee. In all these meetings, there are representatives from the judges, public prosecutors' office, police, advocates, victim support agencies, social services for juveniles, agencies that facilitate the conferences and the scientific team. All these groups meet, on average, every two months. They deal with the administrative follow-up of the experiment and discuss upcoming problems and themes, such as the role of advocates in the sessions, the participation rate of the victims, the relation between the conference and material compensation claims by the victims, the relation between the restorative approach and treatment needs in the offender, and so on.
A methodological workshop including all facilitators is organised every two months. Methodological questions are discussed, such as how to improve victim participation, what to do if parents refuse and the juvenile wants to participate, how to assure the restorative orientation of the session if the vicitm is not present nor represented, what is the relation between restoration and rehabilitation, how to cope with a victim advocate who apparently tries to play his very traditional role, what to do if the outcome is about to be disproportionately light for the offender (especially because the court will have to confirm the declaration of intention), and how to set up the follow-up of agreements. These sessions contribute decisively to the improvement of daily practice and development of a locally-based approach to monitoring FGC sessions.
The first FGC took place in January 2001, and at present, about 80 referrals have been made by nine different judges. About 50 sessions have been organised. The other cases are currently being prepared or have been considered not appropriate for several reasons.
Scientific support and evaluation is provided by the Research Group on Youth Criminology of the Katholieke Universiteit Leuven, funded by the Welfare Department of the Flemish Govenrment. The scientific methodology and provisional results will be presented in a session at this international conference given by Inge Vanfraechem.
Systematic scientific evaluation is being carried out, and its intermediate results will be presented elsewhere. The two years' experience, however, also allows for a few tentative comments.
Despite enthusiastic approval of the FGC principles by almost all actors in the field, the number of referrals is low. Judges tell us that they are so much under time pressure that they tend to rely on their daily routine practice and sometimes forget about the possibilities of FGC.
One other possible reason may be that judges and their social services have for many years been oriented to finding the "appropriate measure for the individual offender." But we do not advance any selection criteria for referral to FGC, based on the principle that all victims of all offences are in need of reparation and that a conference is the best possible process for this. The lack of selection criteria may cause some decision problems for the judges, who then try to develop their own intuitive selection criteria.
The option to position FGC in the centre of juridical processing and not as diversion has probably been a good one. It has allowed for broadening the scope to serious cases. Among the offences dealt with by our conferences are arson, carjackings, armed robberies, serious and repeated physical violence and aggravated thefts. Many juveniles were already known to the police and courts for previous offences. No doubt they would not have been sent to diversionary conferences, because the public prosecutor would in any case want to prosecute.
A question is if this position influences the deliberation. Knowing that the outcome will be submitted to court, do the participants really speak freely about their feelings and about what they propose as possible reparative actions? We would think so, but some indications point in the other direction. Facilitators have mentioned that they felt somewhat uneasy in one or two sessions because the victims were muchless demanding for serious crimes (burglary and serious vandalism in a restaurant, for example) and because they feared that the judge would not approve the declaration of intention. It may indicate some external juridical pressure on the conference, but it may also show that the participants take their own options without concern for what the judge will decide. So far, all declarations of intention have been confirmed by the court's decision. Twice, however, a judge observed that she confirmed the declaration because she "wanted to play the game" of the experiment, while she found that the intentions were focussed too exclusively on the victims' harms and the offenders' reintegration and insufficiently addressed the public aspects of the crime.
Placing the police in the role as we have done has been an important choice. The police's opening of the conference with reading the facts at stake makes clear immediately to all participants that this is not going to be a "soft" talking hour, but that there is public concern about a serious event and a public pressure to find a solution. While police cannot legally represent public interests (which in the Belgian legal context is the monopoly of the public prosecutor), they can provide a crucial symbolic expression of the authorities' commitment in the meeting. Moreover, police are available to inform participants if some discussion emerges about what really happened, and they invoke a sense of security, which may help to convince some victims especially to participate.
In practice, not all police officers do completely understand their role. They must be there to make clear the facts at stake and to express public concern, but at the same time, leave the process and the decisions to the victims, offenders and their communities of care, as much as possible. It is not easy to find adequate balance in a role that is unusual to them. Most officers do, in fact, "underact" due to their uncertainty. General briefings are organised, and a debriefing with the facilitators takes place after each conference, but the responses of individual police officers differ. As experience builds, one may wonder whether some specialisation would not be preferable.
Participation of lawyers in the conference had been seriously discussed before the experiment. The fear was that they would take over the debate. It was nevertheless decided to take them on board, partly because of practical reasons: if the declaration of intention is to be submitted to a court session, it is better to engage the lawyers in the process that leads to the declaration. This is especially true because the conferences are dealing with serious crimes, including possibly intensive pressure on the offender and leading to time-consuming commitments. If the advocates were not involved right from the beginning, there is a considerable risk that they would intervene between the FGC meeting and the court session to dissuade their clients from what has been accepted.
This is now not the case. On the contrary, in fact. Advocates of the juveniles3 appear especially to be among the best partners. Many of them fully understand the opportunities of an FGC, and some ask the judge to refer their client to an FGC. So far, the great majority play their role of advisor very well during the sessions, while staying in the background.
There have been, however, some problems with the advocates of the victims. They are present only in a minority of conferences, but then tend to approach the meeting in a traditional way. Instead of trying to make peace, which is the spirit of a conference, some of the victims' advocates try to win the war. It is difficult for the facilitator to cope with that. Often it is the juveniles' advocate who informs his colleague on the spot about the specific FGC context. Except in one case, it has more or less worked, but a better solution is to be found. Plans are made to write a more general information brochure for all advocates.
As said hereabove, we have opted for the New Zealand FGC model as the basis for our own conferencing method. That means the use of a less strict script than in the Real Justice approach, for example, and the inclusion of a private time for the juvenile and his supporters to prepare the proposal of a reparation plan.
To script or not to script? That is a question. We are using the more scripted model in another experiment with conferencing for severe disciplinary problems in schools, based on a training given by the Netherlands Echt Recht organisation4. Comparison between the less-scripted and more-scripted approaches remains difficult because we apply them to different problems in different contexts. The differences may be exaggerated. The FGC model may be less scripted, but it provides a sequence of phases and themes, so that the session is not completely without a pre-planned concept. The Real Justice model is scripted, but it is hard to imagine that an experienced facilitator will literally repeat what is in the handbook and never deviate from it. Scripts seem to be very helpful for beginners, but experienced facilitators should transcend it and be more flexible about what goes on in the conference. The more the facilitator sticks to the script, the less margin there is for input by the nonprofessional participants. Provisionally, I would suggest that if one would opt for expanding the conferencing approach as much as possible, it may be useful to provide good scripts to avoid "accidents." But the more the facilitators are operating in a very specialised context, and are themselves experienced and professionalised, the less need there would be for a strict script. Experienced FGC facilitators in fact develop their own script, which is often only partly written down.
So far, we have found the inclusion of a private time very useful and sometimes even crucial. Mostly, crime problems of a juvenile have to do with relational or personal problems in families. Many are brought up in the conference, but some personal problems, tensions or events in the past may be too intimate to be dealt with in the presence of the other participants (i.e., victim, police, social services), who have no affairs with it. People have their pride and right of privacy. These problems may be crucial for the family to come to grips with in the aftermath of the crime and to think of a possible proposal. Therefore, the opportunity should be offered to the family to speak about them among each other. It also appears that some members of the family or other supporters may have some ideas for possible reparation but would like to test them in the inner circle. It is good that they may speak out and collaborate in a private circle, before proposing a plan to the others. It is a guarantee that the proposal will be more genuinely supported by all members of the community of care, and that it will be a better base for further deciding on what will be done.
Altogether, the FGC programme is working well, leading to high levels of agreement and satisfaction among the various nonprofessional and professional participants. This has been demonstrated by systematic empirical follow-up and evaluation, undertaken in parallel with the action programme. The description of this research and its intermediate results are presented in a separate paper by Inge Vanfraechem.
1 The importance of this is obvious if one knows that, in the Flemish districts, less than 3,000 juveniles are yearly judged for offences.
2 The King Baudouin Foundation is a private organisation, founded at the initiative of the former Belgian King Baudouin, which supports all kinds of renovating initiatives in a broad scope of social life in Belgium.
3 Though this is not legally provided, we have in Belgium a tradition of advocates who specialise in working in juvenile justice cases, which has been most helpful in this experiment.
4 Based on the Handleiding 2002.
Boyes-Watson, C. (2000), Reflections on the Purist and the Maximalist Models of Restorative Justice. Contemporary Justice Review 3:441–50.
Braithwaite, J. (2002), Restorative Justice and Responsive Regulation, Oxford, UK, Oxford University Press.
Braithwaite, J. and S. Mugford (1994), Conditions for a succesful reintegration ceremony. British Journal of Criminology, 34, 2: 139-171.
Claes, M., F. Spiesschaert, C. Van Dijck, S. Van Grunderbeek and I. Vanfraechem (2003), Implementation and Development of Restorative Justice for Juveniles in Flanders. In L. Walgrave (ed.), Repositioning Restorative Justice, Cullompton, UK, Willan Pulishing, forthcoming.
Handleiding voor het leiden van Echt Recht conferenties (2002), Voorhout, Netherlands, OKS/WESP.
Harris, N. (2001), Shaming and Shame: Regulating Drink-Driving. In E. Ahmed, N. Harris, J. Braithwaite and V. Braithwaite (eds.), Shame Management through Reintegration, Cambridge, UK, Cambridge University Press.
Harris, N., L. Walgrave and J. Braithwaite (2003), Emotional Dynamics in Restorative Conferences. Paper presented for publication in Theoretical Criminology.
Kurki, L. (2003), Evaluating Restorative Practices. In A. von Hirsch, J. Roberts, A. Bottoms, K. Roach and M. Schiff (eds.), Restorative Justice and Criminal Justice: Competing or Reconcilable Paradigms? Oxford, Hart: 293-314.
Marshall, T. (1996), The Evolution of Restorative Justice in Britain. European Journal of Criminal Policy and Research 4:21–43.
Maxwell, G. and A. Morris (1999), Understanding Reoffending: final Report, Wellington, New Zealand, Victoria University, Institute of Criminology.
McCold, P. (2000), Towards a Holistic Vision of Restorative Juvenile Justice: A Reply to the Maximalist Model. Contemporary Justice Review 3:357–414.
McCold, P. and T. Wachtel (2002), Restorative Justice Theory Validation. In E. Weitekamp and H.J. Kerner (eds.), Restorative Justice. Theoretical Foundations, Cullompton, UK, Willan Publishing: 110-142.
Morris, A. (2002), Shame, Guilt and Remorse. Experiences from Family Group Conferences in New Zealand. In I. Weijers and A. Duff (eds.), Punishing Juveniles. Principle and Critique, Oxford, Hart: 157-178.
OSBJ (2003), Jaarverslag (Annual Report), Leuven, Belgium, Ondersteuningsstructuur Bijzondere Jeugdzorg.
Van Ness, D. and K. Heetderks Strong, Restoring Justice, Cincinnati, OH, USA, Anderson (2nd).
Walgrave, L. (2002), Restorative Justice and the Law: Socio-Ethical and Juridical Foundations for a Systemic Approach. In L. Walgrave (ed.), Restorative Justice and the Law, Cullompton, UK, Willan Publishing: 191-218.
IIRP Graduate School
531 Main St.
PO Box 229
(610) 807-9221 | Email