Volume 1, Issue 1, June 1998

ISSN 1096-4886 http://www.westerncriminology.org/Western_Criminology_Review.htm
© 1998, The Western Criminology Review. All Rights Reserved.

Restorative Justice Through Victim-Offender Mediation:
A Multi-Site Assessment

Mark S. Umbreit

Citation: Umbreit, Mark S. 1998. "Restorative Justice Through Victim-Offender Mediation: A Multi-Site Assessment." Western Criminology Review 1(1). [Online]. Available: http://www.westerncriminology.org/documents/WCR/v01n1/Umbreit/umbreit.html


The relatively new and emerging practice theory of restorative justice emphasizes the need to provide opportunities for those most directly affected by crime (victims, communities, and offenders) to be directly involved in responding to the impact of crime and restoring the losses incurred by victims. Victim-offender mediation, a process which allows crime victims to meet fact-to-face with the offender to talk about the impact of the crime and to develop a restitution plan, is the oldest and most empirically grounded restorative justice intervention. This article reports on a study of victim-offender mediation in four sites with juvenile offenders and their victims, along with related studies. High levels of victim and offender satisfaction with the mediation process have been found, along with high successful restitution completion rates and reduced fear among crime victims.

Keywords: restoration, mediation, restitution, victims, accountability, reparation, restorative justice, balanced justice, balanced approach, victim services, community corrections, community justice

Restorative Justice Through Victim-Offender Mediation:
A Multi-Site Assessment

One of the most significant current developments in thinking about crime is the growing interest in restorative justice theory (Umbreit 1996, 1994a, 1989a; Umbreit and Coates 1992; Van Ness and Strong 1997; Wright 1991; Wright and Galaway 1989; Zehr 1990, 1985). At a time when the public debate about crime and punishment is largely driven by political leadership embracing the conservative or liberal solutions of the past, restorative justice offers a fundamentally different framework for understanding and responding to crime and victimization. Restorative justice emphasizes the importance of elevating the role of crime victims and community members, holding offenders directly accountable to the people they violate, restoring the emotional and material losses of victims, and providing a range of opportunities for dialogue, negotiation, and problem solving, whenever possible, which can lead to a greater sense of community safety, conflict resolution, and closure for all involved.

In contrast to the offender-driven nature of our current systems of justice, restorative justice focuses upon three client groups: crime victims, offenders, and community members. It represents a growing international movement with a relatively clear set of values, principles, and guidelines for practice. At this point restorative justice lacks a comprehensive plan for broad implementation as a new paradigm to fully replace current juvenile and criminal justice systems. As a relatively new practice theory (based upon old-fashioned principles) that is gaining support among a growing number of correctional policy makers and practitioners, victim advocates, court and law enforcement officials, it is important to examine the current development and impact of this movement. At its best, restorative justice truly represents a very different way of responding to crime through more active involvement of crime victims and the community. It goes far beyond the traditional liberal and conservative positions by identifying underlying truths and joint interests of all of those concerned about crime policy in a democratic society. At its worst, restorative justice could become another generation of correctional euphemisms to make current highly retributive practices look more effective or fair. Such a “window dressing” effect is clearly not the intent of restorative justice advocates, yet it could become the eventual reality of this movement if the underlying vision and values of the movement become lost.

The purpose of this paper is to point out a number of unresolved issues in the criminal justice system, present the underlying principles of restorative justice, and then to review the growing amount of empirical data on victim-offender mediation. There has been little evaluation of the full range of restorative justice polices and practices. However, the development of victim-offender mediation over the past two decades, which represents a very clear expression of restorative justice values, provides a rich source of empirical data. Finally, a number of implications for practice will be offered, along with identifying needs for further research into this important movement.


Many, if not most, criminal and juvenile justice systems in the U. S. are faced with numerous unresolved problems that have become pronounced in recent decades. With an ever-increasing emphasis upon retribution, there still exist contradictory impulses between punishment and rehabiliation among many correctional policy makers and practitioners. The basic purpose of sentencing is unclear. Is it meant to rehabilitate and change offender behavior? Are criminal sentences meant to deter others from committing crimes? Or, is the purpose of sentencing to incapacitate, or remove, the criminal from society? These and other goals contribute to ongoing confusion about what courts are trying to achieve.

Victims of crime feel increasingly frustrated and alienated by our current system of justice. Even though the justice system exists precisely because individual citizens have been violated by criminal behavior, crime victims have virtually no legal standing in the process of doing justice in American courts. The crime is against “the state” and state interests drive the process of doing justice. Individual crime victims are left on the sidelines of justice, with little or no input. Crime victims frequently feel twice victimized--first, by the offender and second, by the criminal justice system their tax dollars are paying for. For many crime victims their encounter with the justice system leads to increasing frustration and anger as they are largely ignored, and are not often even provided with information about the process, court date changes, and the final disposition of the case. Rarely do criminal justice professionals take the time to listen to the fears and concerns of crime victims and then seek their input and invite their participation in holding the offender accountable.

The failure of increasing punishments to reduce crime is another problem facing justice systems. If severe punishment and incarceration were effective, America would be one of the safest societies in the world. Despite the common citizen perception that the U.S. is too lenient on criminals, the fact is that more Americans are locked up in prisons per capita than any other developed nation in the world. Similarly sentences in the United States are more severe than those of other Western democracies. The United States is the only developed nation to routinely advocate and use capital punishment.

Finally, the skyrocketing cost of corrections, i.e., incarceration, is leading a growing number of legislatures and policy makers to reconsider the wisdom of the current retributive system of justice, which relies so heavily upon incarceration, while largely ignoring the needs of crime victims.


Restorative justice provides an entirely different way of thinking about crime and victimization. Rather than the state being viewed as the primary victim in criminal acts and placing victims and offenders in passive roles, as is the case in the prevailing retributive justice paradigm, restorative justice recognizes crime as first and foremost being directed against individual people. It assumes that those most affected by crime should have the opportunity to become actively involved in resolving the conflict. Restoration of losses, allowing offenders to take direct responsibility for their actions, and assisting victims in their journey of moving beyond their frequent sense of vulnerability by means of achieving some closure, stand in sharp contrast to focusing on past criminal behavior through ever-increasing levels of punishment (Umbreit 1996, 1995b, 1994a, 1991a; Wright 1991; Zehr 1990). Restorative justice attempts to draw upon the strengths of both offenders and victims, rather than focusing upon their deficits. While denouncing criminal behavior, restorative justice emphasizes the need to treat offenders with respect and to reintegrate them into the larger community in ways that can lead to lawful behavior. It represents a truly different paradigm. Restorative justice

  1. is far more concerned about restoration of the victim and victimized community than costly punishment of the offender.
  2. elevates the importance of the victim in the criminal justice process, through increased involvement, input, and services.
  3. requires that offenders be held directly accountable to the person and/or community that they victimized.
  4. encourages the entire community to be involved in holding the offender accountable and promoting a healing response to the needs of victims and offenders.
  5. places greater emphasis on the offender accepting responsibility for their behavior and making amends, whenever possible, rather than on the severity of punishment.
  6. recognizes a community responsibility for social conditions which contribute to offender behavior.

In a very real sense, the theory of restorative justice provides a blueprint for moving into the next century by drawing upon much of the wisdom of the past. Dating all the way back to 12th century England, following the Norman invasion of Britain, a major paradigm shift occurred that turned away from the well established understanding of crime as a victim-offender conflict within the context of community. William the Conqueror’s son, Henry I, issued a decree securing royal jurisdiction over certain offenses (robbery, arson, murder, theft, and other violent crimes) against the King’s peace. Prior to this decree crime had been viewed as conflict between individuals. The traditional emphasis was upon repairing the damage by making amends to the victim.

Restorative justice also draws upon the rich heritage of many recent justice reform movements, including community corrections, victim advocacy, and community policing. The principles of restorative justice are consistent with those of many indigenous traditions, including Native American, Hawaiian, Canadian First Nation people, Aborigines in Australia, and the Maori in New Zealand. These principles are also consistent with values emphasized by nearly all of the world religions.

Many of these principles can also be seen in the pioneering work of an Australian scholar who addresses the issues of crime, shame and reintegration. Braithwaite (1989) argues for "reintegrative shaming," a type of social control based upon informal community condemnation of wrongdoing, but with opportunities for the reintegration of the wrongdoer into the community. He states that the most effective crime control requires active community participation "in shaming offenders, and, having shamed them, through concerted participation in...integrating the offender back into the community." Braithwaite notes that societies with low crime rates consist of people who do not mind their own business, where clear limits exist to tolerance of deviance; and where communities have a preference for handling their own problems.

While Braithwaite (1989) does not specifically address restorative justice or victim-offender mediation, he argues for principles of justice that emphasize personal accountability of offenders, active community involvement, and a process of reconciliation and reaffirmation of the offender. These relate directly to the restorative justice paradigm, with its emphasis upon mediation and dialogue whenever possible.

Restorative justice is expressed through a wide range of policies and practices directed toward offenders and crime victims, including victim support and advocacy, restitution, community service, victim impact panels, victim-offender mediation, circle sentencing, family group conferencing, community boards that meet with offenders to determine appropriate sanctions, victim empathy classes for offenders, and community policing. Little empirical data is available on most restorative justice policies and practices, although a growing number of studies are being initiated. As the oldest, best documented, and most broadly used expression of restorative justice, victim-offender mediation has been the subject of numerous studies in North America and Europe over the past two decades.

The distinction between the old paradigm of retributive justice and the new paradigm of restorative justice has been developed by Zehr (1990). Whereas retributive justice focuses on punishment, the restorative paradigm emphasizes accountability, healing, and closure.

Table 1
Paradigms of Justice




Crime defined as violation of the state

Crime defined as violation of one person by another

Focus on establishing blame, on guilt, on past (did he/she do it?)

Focus on problem solving, on liabilities, and obligations, on future (what should be done?)

Adversarial relationship and process normative

Dialogue and negotiation normative

Imposition of pain to punish and deter/prevent

Restitution as a means of restoring both parties; goal of reconciliation/restoration

Justice defined by intent and process: right rules

Justice defined as right relationship; judged by outcome

Interpersonal, conflictual nature of crime obscured, repressed; conflict seen as individual versus the state

Crime recognized as interpersonal conflict; value of conflict is recognized

One social injury replaced by another

Focus on repair of social injury

Community on sideline, represented abstractly by state

Community as facilitator in restorative process

Encouragement of competitive, individualistic values

Encouragement of mutuality

Action directed from state to offender
-victim ignored
-offender passive

Victim and offenders roles recognized in problem/solution
-victim rights/needs recognized
-offender encouraged to take responsibility

Offender accountability defined as taking punishment

Offender accountability defined as understanding impact of action and helping decide how to make things right

Offense defined in purely legal terms, devoid of moral, social, social, economic, or political dimensions

Offense understood in whole context - moral, economic and political

"Debt" owed to state and society in the abstract

Debt/liability to victim recognized

Response focused on offender's past behavior

Response focused on harmful consequences of offender's behavior

Stigma of crime unremoveable

Stigma of crime removable through restorative action

No encouragement for repentance and forgiveness

Possibilities for repentance and forgiveness

Dependence upon proxy professionals

Direct involvement by participants

Source: Zehr (1985).



As communities move more toward a fully developed restorative justice system of responding to crime and victimization, juvenile and criminal justice practice would include the following characteristics, some of which are already in place:


The initial conceptualization of restorative justice began in the late 1970's and was first clearly articulated by Zehr (1985). At that time the discussion of this new paradigm was based largely in North America and with a small network of academicians and practitioners in Europe. Restorative justice was not then being considered seriously by mainstream criminal and juvenile justice policy makers and practitioners.

By 1990, an international conference supported by NATO funds was convened in Italy to examine the growing interest in restorative justice throughout the world. Academicians and practitioners from numerous countries (Austria, Belgium, Canada, England, Finland, France, Germany, Greece, Italy, Netherlands, Norway, Scotland, and Turkey) presented papers related to the development and impact of restorative justice policies and practices. International interest in restorative justice has continued to grow. In 1995, the New Zealand Ministry of Justice issued a working paper on restorative justice for serious consideration as a federal policy. In May of 1997, the first International Conference on Juvenile Restorative Justice was convened in Leuven, Belgium, bringing together a large group of practitioners and scholars from North America and Europe. Finally, a working subcommittee of the United Nations is currently examining the concept of restorative justice and will be preparing a draft resolution for presentation at a conference of the United Nations in the year 2000.

Interest in the United States has grown extensively during the past five years. Representing one of the oldest and most visible expressions of restorative justice, the practice of victim-offender mediation, which began in the late 1970's, is now occurring in more than 290 communities throughout the United States and a considerably larger number in Europe, as noted in the following table:

Table 2

International Development of
Victim-Offender Mediation Programs


Number of Programs




Available in all jurisdictions













New Zealand

Available in all jurisdictions



South Africa




United States


Source: Umbreit (1994), Wright (1991), and Umbreit and Greenwood (1996).

The American Bar Association (ABA) has played a major leadership role in the area of civil court mediation for over two decades. After many years of little interest in criminal mediation, if not skepticism, in 1994 the ABA fully endorsed the practice of victim-offender mediation and recommended its development in courts throughout the country.

The Center for Restorative Justice and Mediation at the University of Minnesota has sent out hundreds of restorative justice information packets to correctional officials, policy makers, and practitioners throughout the world. In 1996 the U.S. Department of Justice convened their first national conference on restorative justice, bringing together policy makers and practitioners from throughout the country. Perhaps one of the clearest expressions of the growing support for restorative justice is seen in the National Organization for Victim Assistance’s monograph endorsing “restorative community justice.” During the early years of this movement, most victim advocacy groups were quite skeptical; many still are. However, there is a growing number of victim support organizations actively participating in the restorative justice movement.


In contrast to many previous reform movements, the restorative justice movement has major implications for system-wide change in how justice is done. In fact, restorative justice places a heavy emphasis upon systemic change. As a result of the Balanced and Restorative Justice (BARJ) project supported by the Office of Juvenile Justice and Delinquency Prevention of the U.S. Department of Justice, numerous county and state jurisdictions throughout the country are examining the merits of restorative justice. There are already fifteen states that have drafted and/or introduced legislation promoting a more balanced and restorative juvenile justice system. The BARJ Project has been working extensively with six juvenile justice county systems (Deschutes and Lane in Oregon; Travis in Texas; Dakota in Minnesota; Allegheny in Pennsylvania; and Palm Beach in Florida). These are actively involved in implementing restorative justice polices and practices. Mission statements are being reexamined and rewritten, job descriptions have changed, policies are being revised to include more victim and community involvement, resources are being redeployed, new restorative interventions are being initiated, and a far greater awareness of victim needs for involvement and services is being developed.

In 1994, the Vermont Department of Corrections embarked on one of the most ambitious system-wide restorative justice initiatives. Following a public opinion poll, which indicated broad dissatisfaction with the criminal justice system and openness to more restorative and community based responses to nonviolent crime, the Department “took a wrecking ball” and demolished a one hundred year old correctional system built upon the options of either prison or probation. They identified up to fifty percent of the probation caseload that they believed could be held accountable by Reparative Probation Community Boards made up of citizen volunteers. Instead of traditional probation supervision, a wide range of property offenders were directly referred to appear before a Reparative Community Board. In dialogue with the offender, the Board determines a community-based restorative sanction, often times including victim-offender mediation, community service, or a meeting with a victim panel. The Department is now encouraging crime victims to be represented on each Reparative Probation Community Board. No other known restorative justice initiative represents such major structural change--it clearly elevates the role of community volunteers and crime victims in the process of holding offenders accountable to the community they violated.

Individual restorative program initiatives are much more widely dispersed throughout the country than system-wide ones. In addition to the more than 150 victim-offender mediation programs throughout the United States there are numerous other programs (such as creative community service, neighborhood dispute resolution, financial restitution with victim input, victim/offender dialogue groups or panels) that incorporate many or all of the principles of restorative justice. While hard figures are difficult to obtain, a conservative estimate is that at least 200 to 300 of these programs are developing in urban and rural communities throughout the country.


In light of the growing interest in and support of restorative justice theory and practice, the question still remains, "Is the larger public really interested?" The data that have emerged from examination of a number of individual programs, as noted below, are rather persuasive. Yet is there evidence of public support for the principles of restorative justice? The strong "law-and- order" and "get-tough" rhetoric that dominates most political campaigns would suggest not. After all, how often have we heard ambitious politicians or criminal justice officials state that "the public demands that we get tougher with criminals"? This perception--or some would argue, misperception--fuels the engine that drives our nation toward ever-increasing and costly criminal punishments.

There is, however, a growing body of evidence to suggest that the general public is far less vindictive than portrayed and far more supportive of the basic principles of restorative justice than many think, particularly when applied to property offenders. Studies (Clark 1985;Gottfredson and Taylor 1983; Public Agenda Foundation 1987; Public Opinion Research 1986; Thomson and Ragona 1987) in Alabama, Delaware, Maryland, Michigan, Minnesota, North Carolina, Oregon, and Vermont have consistently found that the public is deeply concerned with holding offenders accountable while being quite supportive of community based sanctions which allow for more restorative outcomes.

A study in Minnesota is illustrative. A statewide public opinion survey, conducted by the University of Minnesota (Pranis and Umbreit 1992), challenges conventional wisdom about public feelings related to crime and punishment. A probability sample of 825 Minnesota adults, demographically and geographically balanced to reflect the state's total population, were asked three questions with implications for restorative justice. 1 The first question was: "Suppose that while you are away, your home is burglarized and $1,200 worth of property is stolen. The burglar has one previous conviction for a similar offense. In addition to 4 years on probation, would you prefer the sentence include repayment of $1,200 to you or 4 months in jail?" Nearly three out of four Minnesotans indicated that having the offender pay restitution was more important than a jail sentence for a burglary of their home.

To examine public support for policies that address the underlying causes of crime, a concern that is closely related to restorative justice, the following question was asked: "For the greatest impact on reducing crime, should additional money be spent on more prisons, or spent on education, job training and community programs?" Spending on education, job training, and community programs to reduce crime was favored by four of five Minnesotans rather than spending on prisons.

The third and final question related to restorative justice addressed the issue of interest in victim-offender mediation. The question was as follows: "Minnesota has several programs which allow crime victims to meet with the person who committed the crime, in the presence of a trained mediator, to let this person know how the crime affected them, and to work out a plan for repayment of losses. Suppose you were the victim of a nonviolent property crime committed by a juvenile or young adult. How likely would you be to participate in a program like this?"

More than four of five Minnesotans expressed an interest in participating in a face-to-face mediation session with the offender. This finding is particularly significant because criminal justice officials and program staff who are unfamiliar with mediation often make such comments as, "there is no way in the world that victims in my community would ever want to confront the offender" or "only a small portion of victims would ever be interested." The finding is especially important since the vast majority of crime is committed by either juveniles or young adults. Some would suggest that the victim-offender mediation process is likely to be supported only for crimes involving juvenile offenders. This is certainly not the case in Minnesota. Eighty-two percent of respondents indicated they would be likely to participate in a program that would allow them to meet the juvenile or young adult who victimized them.

In sum, the survey results paint a picture of a far less vindictive public than is portrayed in the media. Respondents indicate greater concern for restitution and prevention strategies that address underlying issues of social injustice than they do more costly retribution. Holding an offender personally accountable to their victim is more important than incarceration in a jail. Public safety is understood to be more directly related to investing in job training, education, and other community programs than incarceration.

While it might be tempting to suggest that this public opinion survey simply reflects the rather unique liberal social policy tradition of Minnesota, its findings are consistent with a growing body of public opinion research across North America (Bae 1991; Galaway 1994; Gottfredson and Taylor 1983; Clark 1985; Public Agenda Foundation 1987; Public Opinion Research 1986; Thomson and Ragona 1987). These studies have found broad public support for payment of restitution by the offender to their victim instead of incarceration for property crimes, and support for crime prevention strategies instead of prison strategies to control crime. The studies did not explicitly ask respondents if they supported "restorative justice." The questions asked, however, addressed important underlying principles that are fundamental to the theory of restorative justice.


Victim-offender mediation is a process which provides interested victims of primarily property crimes the opportunity to meet the offender, in a safe and structured setting, with the goal of holding the offender directly accountable for their behavior while providing important assistance and compensation to the victim (Umbreit 1995c). With the assistance of a trained mediator, the victim is able to let the offender know how the crime affected them, to receive answers to questions they may have, and to be directly involved in developing a restitution plan for the offender to be accountable for the losses. The offender is able to take direct responsibility for their behavior, to learn of the full impact of what they did, and to develop a plan for making amends to the person(s) they violated. Some victim-offender mediation programs are called "victim-offender meetings" or "victim-offender conferences."

While many other types of mediation are largely "settlement driven," victim-offender mediation is primarily "dialogue driven," with the emphasis upon victim healing, offender accountability, and restoration of losses. Contrary to other applications of mediation in which the mediator would first meet the parties during the joint mediation session, in victim-offender mediation a very different process is used based upon a humanistic model of mediation (Umbreit 1995c). Such a model involves reframing the role of the mediator from being settlement driven to facilitating dialogue and mutual aid; scheduling separate premediation sessions with each party; connecting with the parties through building rapport and trust, while not taking sides; identifying the strengths of each party; using a non-directive style of mediation that creates a safe space for dialogue and accessing the strengths of participants; and recognizing and using the power of silence.

Most victim-offender mediation sessions do result in a signed restitution agreement. However, this agreement is secondary in importance to the initial dialogue between the parties. Dialogue addresses emotional and informational needs of victims that are central to their healing and to development of victim empathy in the offender, which can lead to less criminal behavior in the future. Several studies (Coates and Gehm 1989; Umbreit and Coates 1993; Umbreit 1995a) have consistently found that the restitution agreement is less important to crime victims than the opportunity to talk directly with the offender about their feelings regarding the crime.

From its inception in Kitchener, Ontario when the first victim-offender mediation program was established in 1974, many criminal justice officials have been quite skeptical about victim interest in meeting the offender. Victim-offender mediation is clearly not appropriate for all crime victims. In all cases, practitioners are trained to present it as a voluntary choice to the victim. With more than twenty years of mediating many thousands of cases throughout North America and Europe, experience has shown that the majority of victims presented with the option of mediation choose to enter the process. The statewide public opinion poll in Minnesota (Pranis and Umbreit 1992) found that 82 percent of a random sample of citizens from throughout the state would consider participating in a victim offender program if they were the victim of a property crime. A multi-state study (Umbreit 1994a) found that, of 280 victims who participated in victim-offender mediation programs in four states, 91 percent felt their participation was totally voluntary.


As the oldest and most well developed restorative justice intervention, the practice of victim-offender mediation with juvenile and adult offenders has been the subject of fifteen studies in the United States (Coates and Gehm 1989; Gehm 1990; Nugent and Paddock 1995; Umbreit and Coates 1992; Umbreit 1994a, 1991a, 1989a), Canada (Collins 1984; Fischer and Jeune 1987; Perry, Lajeunesse, and Woods 1987, Roberts 1995; Umbreit 1995a), and England (Dignan 1990; Marshal and Merry, 1990; Umbreit and Roberts 1996).

A small but growing amount of research in the field of victim-offender mediation provides further insight into how the process works and the impact it is having on participants and the justice system. All of these studies have found that the vast majority of victims and offenders benefit from the process of meeting each other, talking about the offense and its impact upon all involved, and developing a plan for restoring losses. While not the cure-all for crime and delinquency, victim-offender mediation programs also provide a number of benefits for the justice system. The essential findings that have emerged follow.

Program Characteristics

The majority of more than 280 victim-offender mediation programs in the United States continue to be administered by private community based nonprofit agencies (Umbreit and Greenwood 1997). A growing number of probation departments and other public agencies, however, are beginning to develop these programs, frequently with the involvement of community volunteers to serve as mediators. Most programs employ a four phase process consisting of: (1) case referral and intake; (2) preparation for mediation, at which time the mediator meets with the parties separately prior to the mediation session in order to listen to their stories, explain the program, invite their participation, and prepare them for the face-to-face meeting; (3) mediation, at which a trained third party mediator (most often a community volunteer) facilitates a dialogue that allows the victim and offender to talk about the impact of the crime upon their lives, provide information about the event to each other, and work out a mutually agreeable written restitution agreement; and (4) follow-up, which monitors restitution agreements; follow-up mediation sessions are scheduled if problems arise.

Table 3 describes the characteristics of four programs for juvenile offenders in different parts of the United States. All four programs were part of the first large multi-site (four states) evaluation of victim- offender mediation, employing a quasi-experimental design with pre- and post-intervention measurements and two different comparison groups. The complete methodology and findings are available in Victim Meets Offender: The Impact of Restorative Justice & Mediation (Umbreit 1994a). Three of the programs were managed by community based agencies and one was administered by a probation department. While these four programs had between 30 and 40 percent of their case referrals result in a face-to-face mediation session, many other programs report rates of 50 to 60 percent, or even higher (Coates and Gehm 1989; Galaway 1988, 1989; Gehm 1990; Marshal and Merry 1990; Umbreit 1988, 1989a 1991a; Wright and Galaway 1989).

Case Referral

Referrals of both juvenile and adult cases to victim-offender mediation programs have been made by judges, probation officers, prosecutors, police, and, at times, defense attorneys or victim advocates. Referral can occur at both a pre- and post-adjudication level. The majority of programs in the United States serve as a diversion from the juvenile justice system primarily for first or second time offenders and most receive referrals from the local probation department. The number of referrals can range from as few as a hundred a year or less to as many as 1,000 or more. Table 4 describes the referral characteristics of four programs for juveniles.

Table 3
Program Characteristics (1991)






Start date





Primary referral source










Total 1991 budget





Number of staff

1.5 FTE

3.5 FTE

3.6 FTE


Use of co-mediators





Number of volunteer mediators





Length of mediation training

40 hours

40 hours

25 hours

30 hours

Total 1991 case referrals





Total mediations in 1991





Proportion of mediations to case referrals in 1991





Source: Umbreit (1994a).


Table 4
Referral Characteristics
(Two Year Period, 1990-91)







Cases referred


















Individual victims






Individual offenders






a. against property

b. against people


























7. Most frequent property offense






8. Most frequent violent offense






Source: Umbreit (1994a).

Immediate Outcomes of Mediation

There are a number of immediate outcomes of the victim-offender mediation process. In the four programs identified below, a total of 1,131 face to face mediation sessions occurred and in 95% of these cases a mutually agreeable restitution plan was negotiated. These restitution plans primarily included financial restitution, although community service or personal service for the victim were also included in many agreements.

Table 5
Immediate Outcomes
(Two Year Period, 1990-1991)







Number of Mediations






Successfully negotiated restitution agreements






Agreements with:
a. Financial restitution
b. Personal service
c. Community service





















Total financial restitution






Average financial restitution






Total personal service

1,028 hrs.

439 hrs.

508 hrs.

585 hrs.

2,560 hrs.

Average personal service

18 hrs.

21 hrs.

16 hrs.

16 hrs.

18 hrs.

Total community service

1,073 hrs.

4,064 hrs.

1,937 hrs.

588 hrs.

7,662 hrs.

Average community service

37 hrs.

31 hrs.

18 hrs.

15 hrs.

25 hrs.

Source: Umbreit (1994a).

Victim Participation

The majority of property crime victims are quite willing to participate in a mediation session with their offender when given the opportunity (Coates and Gehm 1989; Gehm 1990; Galaway 1988; Marshall and Merry 1990; Umbreit 1985, 1989a, 1991a, 1993a). A study of the VORP program in Orange County, California, which represents the largest victim-offender mediation program in North America with over 1,000 referrals a year, found that 75 percent of victims of minor property and personal offenses were interested in participating in the mediation process (Niemeyer and Shichor 1995). In a large multi-site study (Umbreit 1994a), 70 percent of victims who were never even referred to mediation indicated their interest in meeting with the juvenile offender if the opportunity were presented to them. In the Minnesota statewide public opinion poll, cited earlier, 82 percent of citizens (many of whom were crime victims) indicated that they would be likely to consider participating in a mediation session with a juvenile or young adult offender if they were the victim of a nonviolent property crime (Pranis and Umbreit 1992).

While the possibility of receiving restitution appears to motivate victims to enter the mediation process, after mediation victims report that meeting the offender and being able to talk about what happened was more satisfying than receiving restitution (Coates and Gehm 1989; Umbreit 1988, 1991a, 1994a, 1995c, 1996).

Offender Participation

Offenders involved in mediation programs, while anxious about a confrontation with their victim, report that meeting the victim and being able to talk about what happened was the most satisfying aspect of the program (Coates and Gehm 1989; Umbreit 1991a, 1994a).

Juvenile offenders do not seem to perceive victim-offender mediation as a significantly less demanding response to their criminal behavior than other options available to the court. The use of mediation is consistent with the concern to hold young offenders accountable for their criminal behavior (Umbreit 1994a).

Voluntary Participation in Mediation

Mediation is perceived to be voluntary by the vast majority of victims and juvenile offenders who have participated in it. In the largest multi-site study (Umbreit 1994a) of victim offender-mediation in the United States, 91 percent of victims indicate they participate voluntarily and 81 percent of juvenile offenders indicate voluntary participation. A particularly interesting finding that emerged from this large multi-state study concerned the comparison group of victims and offenders who did not actually participate in mediation. Had they been given the opportunity, 72 percent of juvenile offenders (a sample matched with offenders in mediation) indicate that they would have chosen to participate in mediation and 70 percent of victims would have chosen mediation.

Client Satisfaction

Victim-offender mediation results in high levels of victim and offender satisfaction with the mediation process and outcome (Coates and Gehm 1989; Dignan 1990; Marshall and Merry 1990; Umbreit 1988, 1991b, 1993b, 1994a, 1995a, 1995b, 1996; Umbreit and Coates 1992, 1993). The victim-offender mediation process has a strong humanizing effect on the justice system response to crime, victims, and juvenile offenders.

The following comments by victims of juvenile crime are illustrative of how they experienced mediation. “I was allowed to participate and I felt I was able to make decisions rather than the system making them for me.” “The mediation made me feel like I had something to do with what went on...that justice had been served.” “I liked the personal quality of mediation...it made me feel less like a victim, but still a victim.”

Comments by juvenile offenders about their experience in mediation are represented by the following statements. “I liked the fairness of it.” “To understand how the victim feels makes me different...I was able to understand a lot about what I did.” “I realized that the victim really got hurt and that made me feel really bad.” “I had a chance of doing something to correct what I did without having to pay bad consequences.”

After participating in a mediation session, victims and juvenile offenders in four states were far more likely to indicate that they were satisfied with the manner in which the justice system disposed of their case through mediation than a matched sample of similar victims and offenders who were not able to participate in mediation (see Table 6).

Table 6

Most Important Mediator Tasks

In a large multi-site study of victim-offender mediation (Umbreit 1994), the most important tasks of the mediator were identified by the juvenile crime victims as:

  1.          Mediator provided leadership
  2.          Mediator made us feel comfortable
  3.          Mediator helped us develop a restitution plan
  4.          Mediator allowed us (victim and offender) to talk.

For juvenile offenders the mediator’s most important tasks were:

  1.          Mediator made us feel comfortable
  2.          Mediator allowed us (victim and offender) to talk
  3.          Mediator helped us develop a restitution plan
  4.          Mediator was a good listener.

Top of page

Restitution Agreements--Successful Completion

Restitution agreements that are perceived as fair to both parties are negotiated in 9 out of 10 cases that enter mediation. A number of programs report successful completion of restitution agreements, ranging from 79 to 98 percent (Coates and Gehm 1989; Galaway 1988, 1989; Gehm 1990; Umbreit 1986a, 1988, 1991a, 1994a).

A study (Umbreit 1994a) of victim-offender mediation programs in Albuquerque and Minneapolis found that offenders were significantly more likely to complete their restitution obligation to victims compared to similar offenders in a court administered restitution program without mediation. As shown in Table 7, 81 percent of offenders in mediation successfully completed their restitution obligation compared to 58 percent who were referred to a court administered restitution program without mediation.

Table 7

Reduction of Victim Fear and Anxiety

Victim-offender mediation programs in Albuquerque, Minneapolis, and Oakland also significantly reduced fear and anxiety among juvenile crime victims (Umbreit 1991a, 1994a, 1994b). Prior to meeting the offender, 23 percent of victims were afraid of being revictimized by the same offender. After actually meeting the offender and talking about the offense and its impact on all involved, only 10 percent of victims were still fearful of being revictimized. Similarly, prior to mediation, 67 percent of victims were upset about the crime, while only 49 percent were afterwards. These findings are consistent with similar studies in England (Umbreit and Roberts 1996) and Canada (Umbreit 1995a).

Table 8


        A multi-site study by Umbreit (1994) found that somewhat fewer additional crimes were committed by juvenile offenders in several victim-offender mediation programs in the United States (18 percent), as compared to similar offenders who did not participate (27 percent). This finding is consistent with two English studies (Marshal and Merry 1990; Dignan 1990) that examined programs working with adult offenders. The results of these studies were in the expected direction but did not achieve a difference that was statistically significant. However, a more recent single-site study by Nugent and Paddock (1995) does find a significant reduction in recidivism following mediation.

Cross-National Comparisons on Key Outcomes

Consistent and positive outcome data have emerged from the first cross-national assessment (Umbreit 1996) of victim-offender mediation programs in four states of the United States, four Canadian Provinces, and two cities in England. Specific outcome measures examined include victim and offender satisfaction with the referral of their case to mediation; victim and offender satisfaction with the outcome of mediation; victim and offender perception of fairness in the criminal justice system response to their case through mediation; and victim fear of revictimization by the same offender following mediation.

Table 9
Comparison of English, Canadian, and U.S. Studies
of Victims and Offenders Participating in Mediation (in Percent)

Combined English Sites (2)

Combined Canadian Sites (4)

Combined U.S.
Sites (4)

Victim satisfaction with referral of their case to mediation




Offender satisfaction with referral of their case to mediation




Victim satisfaction with mediation outcome




Offender satisfaction with mediation outcome




Victim fear of revictimization by same offender, following mediation

16% (50% less than victims who were not in mediation)

11% (64% less than victims who were not in mediation)

10% (56% less than prior to mediation for same victims)

Victim perceptions of fairness in the referral of their case to mediation




Offender perceptions of fairness in the referral of their case to mediation




Source: Umbreit and Roberts (1996).


What Have We Learned About Victim-Offender Mediation for Crimes of Severe Violence?

During the early years of development in the field of victim-offender mediation, the mid to late 1970s, most practitioners believed that mediation was primarily, if not exclusively, for property crimes and minor assaults. In recent years, however, a growing number of victims of severe violence, and at times offenders, are requesting a mediated dialogue to talk about the impact of the crime and seek a greater sense of healing. These cases involve offenses such as sexual assault, attempted homicide and murder (involving surviving family or friends), which require a far more intense and lengthy mediation process. It is common in these cases for the mediator to meet with each party 3-5 times over a 10-12 month period, in addition to coordinating this intervention with other caregivers, such as therapists or support people. A highly skilled mediator trained in advanced victim-sensitive offender dialogue is required. The only such advanced training currently is offered at the Center for Restorative Justice and Mediation at the University of Minnesota.

While the number of severely violent cases in mediation/dialogue is still relatively small, the demand for them is clearly going to increase in the future. The Victim Services Unit of the Texas Department of Criminal Justice is the only state agency that currently offers this service exclusively to any victim of severe violence. This program is currently being evaluated by the author. Several other state Departments of Corrections, through their Victim Services Units, are considering similar mediation/dialogue services to victims and offenders in severe violent crime cases. These sessions are nearly always held in a maximum security prison.

Victims and offenders often speak of their participation in a mediated dialogue as a powerful and transformative experience which helped them in their healing process. Parents of murdered children have expressed their sense of relief after meeting the offender/inmate and sharing their pain. They have also been able to reconstruct what actually happened and why. One mother whose son was murdered stated, “I just needed to let him see the pain he has caused in my life and to find out why he pulled the trigger.” A school teacher who was assaulted and nearly killed commented, after meeting the young man in prison, “It helped me end this ordeal...for me, it has made a difference in my life, though this type of meeting is not for everyone.” An offender/inmate who met with the mother of the man he killed stated, “It felt good to be able to bring her some relief and to express my remorse to her.” A doctor in California whose sister was killed by a drunk driver was initially very skeptical about meeting the offender. Following the mediation session, the victim stated, “I couldn’t begin to heal until I let go of my hatred...after the mediation I felt a great sense of relief...I was now ready to find enjoyment in life again.”

Only two small studies involving four cases each have been conducted in the United States. The first (Umbreit 1989a) found that a mediated dialogue session in several very violent cases, including a sniper shooting, was very beneficial to the victims, offenders, and community members or family members. Three of these four cases (all adult offenders) were handled by a police department in upstate New York (Genesee County) that operates a comprehensive restorative justice program. The other study (Flaten 1996), which involved four cases of severe violent crime committed by juvenile offenders, found very high levels of satisfaction with the process and outcomes, for both victims and offenders. The offenders were inmates in a juvenile correctional facility in Alaska.

The only study (Roberts 1995) that has examined a large number of cases studied the Victim Offender Mediation Project in Langley, British Columbia. This community based Canadian program pioneered the early development of victim-offender mediation and reconciliation with property offenses and minor assaults. A new project, initiated in 1991, applied the mediation process to crimes of severe violence with incarcerated inmates. Prior to starting this project, a small study (Gustafson and Smidstra 1989) assessed whether victims and offenders involved in severe violent crime would be interested in meeting with each other in a safe and structured manner, after intensive preparation, if such a service were available. A very high level of interest was found in such meetings.

In the study conducted by Roberts (1995), virtually all of the 22 offenders and 24 victims who participated indicated support for the program. This support included their belief that they found considerable specific and overall value in the program, felt it was ethically and professionally run, and that they would not hesitate to recommend it to others. The overall effects of the mediation session expressed by victims included:

        - they had finally been heard
        - the offender now no longer exercised control over them
        - they could see the offender as a person rather than a monster
        - they felt more trusting in their relationships with others
        - they felt less fear
        - they weren’t preoccupied with the offender any more
        - they felt peace
        - they would not feel suicidal again
        - they had no more anger.

For offenders, the overall effects of a mediated dialogue with the victim included:

        - discovering emotions, feelings of empathy
        - increasing awareness of impacts of their acts
        - increasing self-awareness
        - opening eyes to the outside world, rather than closed institutional thinking
        - feeling good about having tried the process
        - achieving peace of mind in knowing one has helped a former victim.


The restorative justice movement is having an increasing impact upon policy makers and practitioners in the criminal justice system. While it remains a movement and is not a fully developed new system, encouraging data have emerged from studies of the uniquely restorative intervention of victim-offender mediation. There are more than 600 victim-offender mediation programs in North America and Europe. These offer many opportunities for victims, offenders, families, and other community members to be actively involved in a restorative process of justice. Offenders learn of the real human consequences of their behavior and can be held directly accountable, through making amends, to the person(s) they violated. Victims are invited to play a more active role in holding the offender accountable, letting offenders know how the crime affected them, and working toward some form of resolution. Family members and community volunteers can provide support and assistance. The twenty year experience of victim-offender mediation breathes life into the emerging practice theory of restorative justice. Mediation is clearly leading the way toward greater implementation of restorative justice principles. However, it is not the exclusive expression of restorative justice. Nor does this intervention fully address the larger systemic change implications of restorative justice as a fundamentally different paradigm.

Far more experimentation is needed as the restorative justice movement matures. Several studies are being designed to assess the broader system-wide impact of this new paradigm. Additional interventions that maximize the unique qualities of restorative justice need to be developed and assessed. Family group conferencing and circle sentencing are currently being studied. Both represent very promising restorative interventions developing in North America; they are based upon indigenous traditions in New Zealand and North America, respectively. In addition to examining the immediate outcomes for the individuals involved and the justice system, longitudinal studies are needed to assess the long term effects of the positive outcomes we are beginning to see.

The restorative justice movement, and victim-offender mediation specifically, also face a number of important risks. Perhaps the greatest risk is that of “window dressing,” in which criminal and juvenile justice systems redefine what they have always done with more professionally acceptable and humane language while not really changing their policies and procedures. A few pilot projects may be set up on the margins of the system, while the mainstream of business is entirely offender-driven and highly retributive with little victim involvement and services, and even less community involvement.

The other major risk facing the movement is that of being so focused on restorative justice interventions that we ignore the larger issue of the tremendous overuse of costly incarceration. Unless the issue of overuse of incarceration is ultimately dealt with, there will simply not be the financial resources available to move toward a truly restorative justice model. Similarly, concern for the overrepresentation of people of color in our juvenile and criminal justice systems could easily be lost with a hasty and exclusive focus on restorative interventions.

As a relatively young reform effort, the restorative justice movement, and the practice of victim-offender mediation as its oldest empirically grounded intervention, holds a great deal of promise as we move toward the next century. By drawing upon many traditional values of the past, and from many different cultures, we have the opportunity to build a far more accountable, understandable, and healing system of justice that can lead to a greater sense of community through active victim and citizen involvement in restorative initiatives.


1. The sampling error is plus or minus 3.5 percentage points.


Bae, I. 1991. “A Survey on Public Acceptance of Restitution as an Alternative to Incarceration for Property Offenders in Hennepin County, Minnesota.” In Restorative Justice on Trial, edited by H. Messmer and H. Otto. Dordrecht, NETH: Kluwer Academic Publishers.

Braithwaite, John. 1989. Crime, Shame and Reintegration. New York, NY: Cambridge University Press.

Clark, P. 1985. Perceptions of Criminal Justice Surveys: Executive Summary. Lansing, MI: The Michigan Prison and Jail Overcrowding Project.

Coates, R.B. and John Gehm. 1989. "An Empirical Assessment." Pp. 251-263 in Mediation and Criminal Justice. M. Wright and B. Galaway, London: Sage Publications.

Collins, J.P. 1984. Evaluation Report: Grande Prairie Reconciliation Project for Young Offenders. Ottawa,CAN: Ministry of the Solicitor General of Canada, Consultation Centre (Prairies).

Dignan, J. 1990. Repairing the Damage. Sheffield, UK: Centre for Criminological and Legal Research, University of Sheffield.

Fischer, D.G. and R. Jeune. 1987. "Juvenile Diversion: A Process Analysis." Canadian Psychology 28:60-70.

Flaten, C. 1996. "Victim Offender Mediation: Application With Serious Offenses Committed by Juveniles." Pp. 387-402 in Restorative Justice: International Perspectives, edited by B. Galaway and J. Hudson.

Galaway, B. 1994. “Manitoba Public Views about Restorative Justice.” Winnipeg, CAN: Faculty of Social Work, University of Manitoba.

Galaway, B. 1989. "Informal Justice: Mediation Between Offenders and Victims." Pp. 103-116 in Crime Prevention and Intervention: Legal and Ethical Problems, edited by P.A. Albrecht and O. Backes: Berlin: Wlater de Gruyter.

Galaway, B. 1988. "Crime Victim and Offender Mediation As A Social Work Strategy." Social Service Review 62:668-683.

Gehm, J. 1990. "Mediated Victim-Offender Restitution Agreements: An Exploratory Analysis of Factors Related to Victim Participation." Pp. 177-182 in Criminal Justice, Restitution and Reconciliation, edited by B. Galaway and J. Hudson. Monsey, NY: Criminal Justice Press.

Gottfredson, S. and R. Taylor. 1983. The Correctional Crisis: Prison Overcrowding and Public Policy. Washington, D.C.: U.S. Department of Justice, National Institute of Justice.

Gustafson, D.L. and H. Smidstra. 1989. Victim Offender Reconciliation in Serious Crime: A Report on the Feasibility Study Undertaken for The Ministry of the Solicitor General (Canada). Langley, B.C.: Fraser Region Community Justice Initiatives Association.

Marshall, T.F. and S. Merry. 1990. Crime and Accountability. London: Home Office.

Niemeyer, M. and D. Shichor. 1995. A Preliminary Study of a Large Victim Offender Reconciliation Program. Orange, CA: St. Vincent de Paul Center for Community Reconciliation.

Nugent, W.R. and J.B. Paddock. 1995. “The Effect of Victim-Offender Mediation on Severity of Reoffense.” Mediation Quarterly 12 (4):353-367.

Perry, L., T. Lajeunesse, and A. Woods. 1987. Mediation Services: An Evaluation. Manitoba, CAN: Research, Planning and Evaluation Office of the Attorney General.

Pranis, K. and M. Umbreit. 1992. Public Opinion Research Challenges Perception of Widespread Public Demand for Harsher Punishment. Minneapolis, MN: Citizens Council.

Public Agenda Foundation. 1987. Crime and Punishment: The Public’s View. New York, NY: The Edna McConnell Clark Foundation.

Public Opinion Research. 1986. Report Prepared for the North Carolina Center on Crime and Punishment. Washington, D.C. Public Opinion Research.

Roberts, T. 1995. Evaluation of the Victim Offender Mediation Program in Langley, B.C. Victoria, BC: Focus Consultants.

Thomson, D.R. and A.J. Ragona. 1987. “Popular Moderation Versus Governmental Authoriarianism: An Interactionist View of Public Sentiments Toward Criminal Sanctions.” Crime and Delinquency 33(2):337-357.

Umbreit, Mark S. 1985. Crime and Reconciliation: Creative Options for Victims and Offenders. Nashville, TN: Abingdon Press.

________ 1986a. Victim Offender Mediation and Judicial Leadership. Judicature, December.

________ 1988. “Mediation of Victim Offender Conflict.” Journal of Dispute Resolution. University of Missouri School of Law, Columbia.

________ 1989a. Victims Seeking Fairness, Not Revenge: Toward Restorative Justice. Federal Probation, September.

________1989b. "Violent Offenders and Their Victims." In Mediation and Criminal Justice, edited by Martin Wright and Burt Galaway.

________ 1991a. "Minnesota Mediation Center Gets Positive Results." Corrections Today Journal (August):194-197.

________ 1991b. “Mediation of Youth Conflict: A Multi-System Perspective.” Child and Adolescent Social Work 8 (2):141-153.

________1993a. "Crime Victims and Offenders in Mediation: An Emerging Area of Social Work Practice." Social Work.

________ 1993b. "Juvenile Offenders Meet Their Victims: The Impact of Mediation in Albuquerque, New Mexico." Family and Conciliation Courts Review 31(1):90-100.

________ 1994a. Victim Meets Offender: The Impact of Restorative Justice & Mediation. Monsey, NY: Criminal Justice Press.

________ 1994b. “Crime Victims Confront Their Offenders: The Impact of a Minneapolis Mediation Program.” Journal of Research on Social Work Practice.

________ 1995a. Mediation of Criminal Conflict: An Assessment of Programs in Four Canadian Provinces. St.Paul, MN: Center for Restorative Justice & Mediation, University of Minnesota.

________ 1995b. “The Development and Impact of Victim-Offender Mediation in the United States.” Mediation Quarterly 12 (3):263-276.

________ 1995c. Mediating Interpersonal Conflicts: A Pathway to Peace. West Concord, MN: CPI Publishing.

________ 1996. “Restorative Justice Through Mediation: The Impact of Offenders Facing Their Victims in Oakland.” Law and Social Work.

________ and R.B.Coates. 1992. The Impact of Mediating Victim Offender Conflict: An Analysis of Programs in Three States. Juvenile and Family Court Journal. Reno, Nevada: National Council of Juvenile and Family Court Judges.

________ and R.B. Coates 1993. "Cross-Site Analysis of Victim Offender Mediation in Four States." Crime and Delinquency 39(4):565-585.

________ and A.W. Roberts. 1996. Mediation of Criminal Conflict in England: An Assessment of Services in Coventry and Leeds. St. Paul, MN: Center for Restorative Justice & Mediation, University of Minnesota.

Van Ness, D. and K. Strong. 1997. Restoring Justice. Cincinnati, OH: Anderson Publishing Company.

Wright, M. 1991. Justice for Victims and Offenders. Philadelphia, PA: Open University Press.

Wright, M. and B. Galaway. 1989. Mediation and Criminal Justice. London: Sage.

Zehr, H. 1990. Changing Lenses: A New Focus for Crime and Justice. Scottsdale, PA: Herald Press.

Zehr, H. 1985. Retributive Justice, Restorative Justice. Akron, PA: Mennonite Central Committee.

Mark S. Umbreit

Mark Umbreit is Director of the Center for Restorative Justice & Mediation and the National Restorative Justice Training Institute and is an associate professor at the University of Minnesota School of Social Work, St. Paul Campus.

Dr. Umbreit is an internationally recognized trainer, mediator, author, and researcher in the field of restorative justice and mediation, with over twenty-five years of experience as a practitioner. He has been a featured speaker on restorative justice and mediation throughout the United States and in Canada, England, Belgium, Italy, Denmark, and China.

He has completed the first cross-national studies of victim-offender mediation in four United States cities, four provinces in Canada, and two cities in England.

Dr. Umbreit is the author of two leading books in the field, Victim Meets Offender: The Impact of Restorative Justice (1994) and Mediating Interpersonal Conflicts: A Pathway to Peace (1995). He has also authored numerous articles and monographs on restorative justice and mediation practice. He is a practicing mediator with expertise in facilitating victim sensitive mediation/dialogues with offenders in crimes of severe violence.

Contact information: Center for Restorative Justice and Mediation, School of Social Work, University of Minnesota.

Top of Page

Return to Western Criminology Review Archive

Go to Western Society of Criminology Home Page