Restoration and Retribution: How Including Retributive Components Affects the Acceptability of Restorative Justice Procedures

Two studies investigated people’s perceptions of the acceptability of restorative justice procedures for handling crimes that differ in severity. Results from Study 1 supported our hypothesis that as crimes increase in seriousness, people require a restorative justice procedure that also has a possible retributive component (i.e. a prison sentence). Study 1 also demonstrated that individuals assigned lower prison sentences for offenders who successfully completed a restorative procedure as compared to a traditional court procedure. The results from Study 2 replicated those from Study 1, as well as demonstrating that offenders who failed to successfully complete the restorative procedure received no reduction in prison sentence. These findings suggest that in order for citizens to view a restorative justice procedure as an acceptable alternative to the traditional court system for serious crimes, the procedure must allow for the option of some retributive measures.

INTRODUCTION

The criminal justice system in America faces many problems. The cost of maintaining the prison system is expensive for the government and its taxpayers to support. There is also the problem of “prisonization,” which refers to the (usually negative) changes that a person undergoes while in prison. People are released from prison with limited job opportunities (Pager, 2003) and with a strengthened criminal identity (Thomas et al., 1981). Further, many crime victims feel “left out” of their own proceedings, which causes them to feel frustrated and marginalized by the justice process (Barton, 1999; Hogan and Emler, 1981). These factors and others have inspired people to search for alternatives to the traditional court process. One alternative that has received considerable attention is “restorative justice.”

Restorative justice attempts to achieve restoration in three different, but interrelated, domains (Marshall, 2003). First, restorative justice aims to restore victims, materially and psychologically, to where they were before the crime occurred. The focus is on repairing the harms that have been caused by the offense. Second, restorative justice is concerned with restoring offenders to a law-abiding life style, by making them realize the harm that they caused the victim and by treating offenders with more respect than in the traditional court system. Finally, restorative justice attempts to repair the harms that have been caused to the community by the offense, as well as involve the community in the reintegration of the offender (Bazemore and Stinchcomb, 2004).

Restorative justice procedures usually involve a face-to-face meeting of the victim, the offender, and supporters for both sides. Community members are frequently present as representatives of the community interests. Before a restorative justice procedure can be initiated, offenders must admit their guilt. Once they do, a “conference” is scheduled, to be led by a trained facilitator. During the conference, offenders explain what they did and why they did it. Victims then have a chance to explain how the crime harmed them, and the offenders have an opportunity to apologize for the harm they caused. Together, the conference participants work out an agreement for what the offenders must do to repair the harm that followed from the offense. These agreements usually involve an apology, monetary compensation, some services that the offender does for the victim, community service, and the like.

Advocates of restorative justice claim that this form of justice is able to both decrease recidivism and increase victim satisfaction relative to the traditional court system. The benefits for both the victim and the offender are related to issues of procedural justice. People are more satisfied with their outcomes in legal proceedings when they feel that the procedures were fair and that they had a chance to be heard (Lind and Tyler, 1988). Both the victim and the offender are given a say in restorative justice processes, which should lead to more overall satisfaction with the procedures. Restorative justice advocates argue that victims feel more satisfied with restorative processes than traditional court procedures because they have an opportunity to confront the person who victimized them, receive an apology, and play a role in determining what the offenders must do to repair the harm. There is some empirical evidence to support the claim that restorative procedures are successful at increasing victim satisfaction. For example, Strang (2002) found that victims were emotionally better off and less afraid of future victimization after having participated in a conference rather than a traditional court process.

With regard to recidivism, advocates argue that the realizations and emotional experiences evoked by restorative justice conferences make offenders less likely to re-offend (e.g. Sherman, 2003). Offenders are treated with more respect than in the traditional court system, and they have an opportunity to see directly how their actions have harmed both the victim and community members. Offenders have a chance to both feel and express remorse for their actions. Although there is limited empirical support for these claims (Braithwaite, 2002), there is some evidence that for violent crime, offenders are less likely to commit another crime if they go through a conference rather than court (Sherman, 2003).

The Debate

There is an unresolved debate in restorative justice about what role punitive measures (i.e. retributive justice) should play. Braithwaite and his colleagues (Braithwaite, 1998, 1999, 2002; Braithwaite and Strang, 2001) argue that there is no place for punitiveness when the goal of the justice process is to repair the harm that the offense has caused. The focus is on the offense, not the offender. For example, Braithwaite and Strang (2001) write, “Restorative justice is about healing (restoration) rather than hurting. Responding to the hurt of the crime with the hurt of punishment is rejected, along with its corresponding value of proportionality—punishment that is proportionate to the wrong that has been done” (p. 1). This conceptualization depicts restorative and retributive justice as opposites that cannot be used together to achieve justice. Supporters of this view argue that only conferences and reparative sanctions, not courts and prison sentences, should be a part of restorative justice.

A number of scholars disagree with this conceptualization of restorative justice (Barton, 1999; Daly, 2002; Duff, 2003; Robinson, 2003), and they claim that not only can restorative and retributive justice be used together, but that in order to achieve justice, they must be. The proponents of this view argue that without retribution, restorative justice is incomplete. Duff (2003) contends that while a punitive-less restorative justice procedure does address the harm caused by a crime and achieve material reparation (e.g. paying for a stolen item), it neglects the wrong committed by the offender and the following moral reparation, which requires retribution. Robinson (2003) echoes this contention with his argument that atonement for a crime cannot be simply achieved by restoration alone, but must be accompanied by additional suffering on the part of the offender to demonstrate that the offender understands the wrongs he or she has committed. Achieving justice from the victim’s perspective is also important, and Barton (1999) points out that victims have a right to desire retribution, and that in order to achieve justice, this right cannot be taken away from them. These scholars and their supporters advocate a procedure that uses both restorative and retributive measures to achieve justice.

This debate also spills over into what is necessary for restoration to the community. Braithwaite (1999) contends that the punishment given to the offender only has to satisfy what the victim and the offender need to be “reconciled.” Advocates of the use of both restorative and retributive measures, however, contend that punitive measures are necessary to meet the needs of the community in which the crime took place. For example, Duff (2003) argues that a criminal code violation results in a public wrong, which requires not only that the victim be satisfied with the punishment given to the offender, but that the citizens of the community be satisfied with it as well. In addition, it has been argued that punishment of offenders provides an outlet for the community to demonstrate its disapproval of wrongdoing and reaffirm its values (Tyler et al., 1997). Therefore, even if the victim and the offender are satisfied with purely restorative sanctions, the community should still be able to assign punitive measures to the offender because the citizens have also been wronged and they will need a means of voicing their disapproval with the offender’s actions.

The debate over the compatibility of retributive and restorative justice has implications for the acceptability of restorative justice in American society. However, researchers have identified one important empirical question that has received insufficient attention (Roberts and Stalans, 2004): What do citizens believe will best achieve justice? The citizens of a community, state, or country surely should have some say in whether or not restorative policies will be implemented. Furthermore, if legislators believe that their constituents support restorative justice initiatives, then it becomes more likely that the legislators will vote to put restorative procedures into action (Roberts and Stalans, 2004). Therefore, the views of the ordinary citizen are of both theoretical and policy interest. Do the members of the public share the views of Braithwaite and others who contend that restorative justice should be without retribution, or do they agree with the opposing side, who argue that punitive measures should be included in restorative justice?

Beyond the implications this investigation has for the implementation of restorative justice procedures, there are also potential consequences for law and order concerns. People are more likely to obey the law if they view the law and the legal system as being legitimate (Tyler, 1990). If people do not view a procedure as being able to achieve justice, the implementation of restorative justice could cause a number of complications for the justice system, such as an increase in vigilante justice (Nadler, 2005; Robinson and Darley, 1995). One goal of this research is to gather some initial evidence regarding people’s perceptions about what role punitive measures should play in restorative justice.

Crime Severity and the Restorative Justice Models

One of the key determinants of whether people will accept non-punitive methods of restorative justice may be the seriousness of the offense. Past research has shown that people have a multifaceted conceptualization of “seriousness” that is closely tied to the moral wrongness of the offending act, which is in turn related to the deliberateness of the offender’s intent and the seriousness of the norm that was violated in the offense (Rosenmerkel, 2001; Warr, 1989). It is therefore no surprise that seriousness of the offense is a predictor of the severity of the prison sentence that a respondent would assign to the offense. Individuals want the punishment to be proportionate to the crime, and this entails escalating the punitiveness of the punishment as the crimes increase in severity (Carlsmith et al., 2002; Darley et al., 2000).

These findings demonstrate that people want to punish offenders for their transgressions, especially as crimes become more serious. This raises the issue of whether people will accept purely restorative procedures as an alternative to the traditional court system. If people are not willing to send more serious crimes to “pure” restorative justice procedures, then this severely limits the viability of restorative justice as an alternative to the traditional court system. Thus, a critical question with regard to the acceptance of restorative justice procedures is whether people believe that restorative sanctions are sufficient for serious offenses. Will people forgo punishment and feel that justice has been achieved with only restorative sanctions? Or, will they still want to punish offenders by adding more punitive measures to their sentencing decisions?

There is evidence that suggests that a restorative justice procedure that does not include any punitive measures may not be able to satisfy people’s desire for just deserts, as people are less likely to choose purely restorative procedures as the seriousness of the offense increases (Bilz, 2002; Doble and Greene, 2000). This effect may arise because individuals are not simply concerned with matters related to restoration for more serious offenses. Rather, people also require offenders to be punished for their actions, which may be due to the “symbolic moral significance” that is attributed to the violations caused by more serious offenses and the desire to express disapproval for wrongdoing (Miller and Vidmar, 1981; Tyler and Boeckmann, 1997; Tyler et al., 1997; Vidmar and Miller, 1980).

These findings raise the question of whether people will always find restoration inappropriate to handle serious offenses, or whether the paring of restorative motives with retributive ones can lead people to feel that justice has been achieved for crimes that are more severe in nature. Currently, there is a lack of research examining the acceptability of a procedure that uses both restorative and punitive measures (for exceptions, see Bowers, 1993; McGarrell and Sandys, 1996). Most studies investigating whether people find restorative measures appropriate have forced participants to choose between restorative and retributive measures, which may explain why people do not opt for restorative measures for more serious crimes (Roberts and Stalans, 2004).

Based on the evidence reviewed above, we expected that people would believe that restorative-only procedures were acceptable for crimes that were not serious offenses, but that for more serious crimes, people would want offenders to be assigned more punitive sanctions. We then expected that they would be willing to send these more serious offenders to restorative procedures only if those procedures included the possibility of punitive sanctions. Specifically, we expected that as people’s perceptions of the seriousness of the crime increased, the more they would move away from choosing a conference-only procedure and the more they would opt to use a procedure that includes both conferences and court, which would allow punitive measures to be taken in addition to restorative ones.

In the present research, we asked respondents to consider a number of brief descriptions of criminal incidents and to indicate for each case the justice procedure that they thought was appropriate for it. Respondents chose between two kinds of restorative procedures and a traditional court procedure. For the “pure” restorative justice procedure, we used a typical restorative justice procedure that does not include punitive measures (as outlined earlier). Offenders do not return to the courtroom, and they cannot be sentenced to prison following a successfully concluded conference. For the mixed (conference and court) procedure, we used a modified version of a procedure described by Barton (1999). This procedure required the same conference as the pure procedure, but after the conference had been completed and an agreement had been reached, offenders were returned to the courtroom for consideration of whether punitive sanctions would be imposed. The decision maker had the right to assign a prison sentence after the conclusion of the restorative justice conference. Therefore, in this mixed procedure, both restorative sanctions and punitive measures could have been inflicted on the offenders.

We also gave participants an option to send the cases to the traditional court process. We presented participants with this option because we did not want to force participants into choosing one of the alternative procedures if they felt that neither of the restorative options would achieve justice for a particular crime. In addition, people could prefer to use only restorative measures (the pure procedure) or only punitive measures (the traditional court system). The inclusion of the traditional court process as an option allowed us to assess whether people find any model of restorative justice acceptable by giving them an opportunity to select a punitive procedure that did not explicitly include restorative measures.

Reducing Prison Sentences

From a social policy perspective, another important issue arises. Citizens may well be enthusiastic about, and willing to send even serious criminal offenders to a mixed procedure that allows for both restorative and punitive measures. However, will this bring about a reduction in the prison sentence that an offender receives? That is, if the restorative aspects of a justice procedure are successfully concluded, will individuals be willing to grant a reduction in punitive sanctions (i.e. length of prison sentence assigned)? It is likely that the willingness of the court system to include the non-trivial burdens of adding a restorative component into the current justice system will be considerably increased if any corresponding reduction in the costly and socially undesirable aspects of the criminal detention system could be achieved. The overcrowding of prisons in the United States highlights this point, as there has been a 10-fold increase in the number of people in the prison system since 1971, with over 2,000,000 people currently incarcerated (Glaze and Palla, 2004). Therefore, one of the main attractions of a mixed procedure is that it would reduce the burden on the prison system by providing an acceptable way to lower sentences for offenders while still having the public feel that justice has been achieved.

The studies we report were designed to gather some preliminary evidence on these issues. First, if offenders complete a restorative conference, as prescribed by restorative justice procedures, do people think that the offender is entitled to a reduced prison sentence? Second, is this entitlement at all dependent on the success of the conference?

There is evidence that people will accept a shortened prison term for offenders who successfully complete a conference. If offenders offer an apology and appear remorseful, people will assign them less severe punishments than those offenders who do not (Felson and Ribner, 1981; Kleinke et al., 1992; Robinson et al., 1994; Rumsey, 1976). The offering of an apology and expressing of remorse may be taken as evidence that the offender understands that he has committed a wrong and is thus less likely to commit another offense (Gold and Weiner, 2000). Restorative justice conferences aim to provide a setting for these events to occur, as offenders are provided with an opportunity to realize the harm that they have caused and to apologize to the victim. It would follow, then, that people should reduce prison sentences for offenders participating in these conferences. We test this prediction in Study 1.

However, intuition suggests a limit on this prediction. It may be unlikely that citizens will reduce the prison sentences of offenders who do not successfully complete the restorative justice conference. We expect, then, that any sentence reductions granted to the offender after his participation in a conference would depend on the conference coming to a successful conclusion. We will test this prediction in Study 2.

STUDY 1

Our goal for the present research is to investigate people’s perceptions about restorative justice as a viable alternative to the traditional court system. In Study 1, we examine what role people believe punitive measures should play (if any) in restorative justice procedures to best achieve justice. Specifically, we are interested in whether the seriousness of the crime affects whether people require punitive procedures to be a part of the punishment for offenders. We are also investigating if, for the mixed model, the successful completion of a restorative justice conference leads to lower prison sentences for those offenders.

Method

Participants

Fifty-seven Princeton undergraduate psychology students (16 male, 39 female, 2 did not report their gender) participated in the study as part of a course requirement.¹ They ranged in age from 18 to 23 (M = 19.35, SD = 1.17). Among them 38 identified themselves as Caucasian, 1 as Hispanic, 5 as Asian-American, 7 as African-American, and 6 as “Other.”

Procedure

Participants were seated in front of a computer with Internet access. They completed the experiment online. At the beginning of the experiment, participants read detailed descriptions of what both the pure restorative and the mixed restorative and punitive procedures entailed (see Appendix A), and participants could return to these descriptions at any time during the experiment by clicking on an Instructions link. For experimental purposes, we called the pure restorative procedure “Victim Offender Mediation” (VOM) and the mixed restorative procedure “Restorative Justice Mediation” (RJM).

Participants were told that they would be reading nine court cases and that for each case, they should think about how the offender should be treated. They were informed that all offenders had already admitted their guilt. They were instructed to act as impartial judges, and they were told that their task was to decide whether each case should go through the pure restorative procedure, the mixed, or the traditional court process. Participants were told that they should choose the process that they believed would best achieve justice for that crime. Participants were also informed that they would be answering general questions about each case.

After participants read through the instructions, they were asked to write down what the major difference was between the pure restorative and mixed procedures. This was to ensure that participants knew that the pure restorative procedure only used conferences (which did not include jail time as a punishment), and that the mixed procedure included both conference and court, allowing for jail time to be used as a punishment. All participants completed this successfully.

In the first part of the experiment, participants read through all nine cases in a randomly generated order. The crimes were as follows: vandalism, assault, property thefts (a bike theft and an oil drum theft), mugging, burglary, identity theft, attempted murder, and rape (see Appendix B). After reading each case, participants were asked to answer six questions about the crime and the offender. With regard to the crime, participants were asked how violent, how morally offensive, and how serious they thought the crime was. With regard to the offender, they were asked how likely they thought it was that the offender would commit another crime, how violence-prone they thought the offender was, and how well they thought that the offender could be rehabilitated. All six questions were answered on a scale that ranged from 1 (not at all) to 7 (very/completely). After answering these questions, participants were asked to indicate whether they would like to send the case to the pure restorative procedure, the mixed procedure, or the traditional court process.² Participants completed the questions for all nine cases before they moved on to the next section of the experiment.³

In the second part of the experiment, participants were told that the cases would come back to them for review. They were instructed that the cases that they sent through the traditional court process and the mixed procedure would be given to them again for sentencing, and that they would be informed of the outcome of the mixed procedure before they made their sentencing decision. They were instructed that they would also receive the cases they sent to the pure restorative procedure, but they would not be asked to provide sentences for them. Before answering any questions about a specific case, participants were given that case for review. For all cases, participants were asked why they thought the case was appropriate for the procedure they had chosen. For cases that went through the mixed procedure and the traditional court process, participants were told that a sentencing commission had researched what the standard sentence should be for the particular crime. The length of time of the standard sentence was not specified. They were then instructed that, as judge, they had the power to decrease, increase, or maintain the standard sentence as they thought was required by the overall circumstances of the case. Participants were asked what sentence they would like to assign. Sentences ranged from “extremely less than the standard sentence” to “extremely more than the standard sentence,” with an option to give no prison sentence at all. They were then asked to provide the reason(s) for their sentencing decision. Finally, participants indicated the exact amount of prison time that they thought the sentence should be for the particular offense.

For the mixed procedure, before participants were asked to provide a prison sentence, they were informed that the conference had taken place and that the offender was now awaiting sentencing. The participants were also given comments made by the facilitator and the victim. The facilitator stated, “The conference between the offender and the victim was successful. An agreement was reached, and, in my experience, this agreement was fair and what would be expected for this type of offense.”

The victim stated, “I am satisfied with the outcome of the conference, and I believe the agreement is fair.”

In order to investigate our hypothesis about whether people will assign lower prison sentences to offenders who completed a conference than to those offenders who went through the traditional court process, we needed to run an additional 18 participants who were only given the option to use the traditional court system.⁴ These participants, as before, were instructed to act as impartial judges and to give the prison sentence that they believed would best achieve justice for each case. They answered the same initial six questions about each offense. In the second part of the experiment, they followed the same procedure as the participants who opted for the traditional court process. At the conclusion of the experiment, all participants filled out a short set of demographic questions.

Results

Seriousness and the Send Decision⁵

In order to determine whether the seriousness of the crime predicted which procedure people would select for each case, we conducted both correlational analyses and a nominal logistic regression. We used the single-item measure of participants’ ratings of the seriousness of the crime for all of the analyses. The correlational analyses examined the relationship between the mean seriousness rating of each offense and which procedure the participants chose for each offense (Table I). We found that as offense seriousness increased, the tendency to send offenders to the pure restorative justice procedure dramatically decreased, r(7) = −0.97, p < 0.0001, whereas the tendency to send offenders to the mixed procedure (conference and court) dramatically increased, r(7) = 0.96, p < 0.0001. Both of these relationships approached unity (Fig. 1A and B).

Table I. Participants’ Mean Seriousness Ratings and Send Decision Proportions for Each Offense (Study 1)

Fig. 1.

The percentage of participants choosing the pure restorative procedure (A) and the mixed procedure (B) by the mean serious rating for each case.

For the nominal logistic regression, we regressed which procedure participants chose (i.e. restorative, mixed, or court) on participants’ ratings of the seriousness of each offense. There was a good model fit based on the seriousness of the offense, χ²(12, N = 351) = 160.95, p < 0.0001, Nagelkerke R ² = 0.42. Overall, the correct classification was 67.2%, with correct classification rates of 81.4%, 77.7%, and 0% for the cases sent to the pure restorative, mixed, and court procedures, respectively. The 0% classification rate for court is not surprising, as people were opting to use the pure restorative and mixed procedures far more than they were choosing the court process (Table I). Consistent with our expectations, the pure restorative procedure had crimes that were perceived as less serious (M = 3.38, SD = 1.17) than offenses that were sent to the mixed procedure (M = 5.43, SD = 1.32) or to the traditional court process (M = 5.02, SD = 1.68). Regarding our main interest in the difference between the pure restorative and mixed procedures, offenses that had a lower mean serious rating were more likely to be sent to the pure restorative procedure, while offenses that had a higher mean serious rating were more likely to be sent to the mixed procedure.

For presentational clarity, we also grouped the cases into three levels of seriousness (Table I) to further illustrate which procedures participants chose based on the seriousness of the crime. We divided the cases in the following way: The cases that participants rated below the midpoint of the seven-point scale were labeled as low-seriousness crimes (vandalism, assault, and bike theft). The cases that participants rated at or near the scale’s midpoint were labeled as being mid-seriousness crimes (mugging and the oil drum theft). Finally, the cases that participants rated well above the scale’s midpoint were labeled as high-seriousness crimes (burglary, identity theft, attempted murder, and rape).⁶

We then conducted a repeated measures ANOVA to determine if the procedure to which participants chose to send each case differed based on seriousness of the crimes (Table I). To perform this analysis, we used the mean proportion of cases from each seriousness group that were sent to the three different procedures. This analysis revealed that participants’ send decisions did differ based on the seriousness of the offense, F(8, 304) = 27.94, p < 0.0001, η² = 0.42. As shown in Fig. 2, the proportion of cases sent to the pure restorative procedure decreased as crimes increased in severity, while the proportion of cases sent to the mixed procedure (and the traditional court process) increased as the crimes became more serious. In addition, in each group, participants preferred at least one of the alternative procedures over the traditional court procedure. For low-seriousness offenses, most participants chose to send these cases to the pure restorative procedure. For mid-seriousness crimes, participants were predominantly split between the pure restorative and the mixed procedure. For the high-seriousness offenses, participants were more likely to send these cases to the mixed procedure.⁷

Fig. 2.

The proportion of cases (+SE) sent to the pure restorative procedure, the mixed procedure, or the traditional court process for low, mid, and high seriousness cases (N = 351).

Punitiveness⁸

We were interested in whether participants’ own judgments of the seriousness of each offense would predict the level of punitiveness that they would select for that offense. To this end, we coded participants’ decisions on where to send the cases as a continuous measure of punitiveness. The pure restorative procedure had the least possibility of punitive measures (all restorative, no punitive), and going to court had the greatest possibility of punitive measures (no restorative, all punitive), with the mixed procedure falling in the middle (part restorative, part punitive). Therefore, our coding scheme was: pure = 1, mixed = 2, and court = 3. A correlational analysis revealed that as participants’ ratings of the seriousness of the crime increased, the more likely they were to choose a procedure that had a greater potential to assign more punitive punishment, r(349) = 0.47, p < 0.0001.⁹

Perceptions of the Offense and the Offender

We were also interested in whether participants’ perceptions of both the offender and the offense correlated with the punitiveness of the punishments they assigned for each case. As shown in Table II, all of the offender-related and the offense-related features we measured were correlated with each other and with the punitiveness of the punishment. Regarding the offender-related questions, Table II shows that participants exhibited an increased tendency to opt for a higher punitive procedure the less they thought the offender could be rehabilitated, the more they thought the offender would commit another crime, and the more violence-prone they believed the offender to be. Regarding offense-related questions, the more serious, the more morally offensive, and the more violent the crime, the more likely participants were to choose a procedure with punitive punishments.

Table II. Correlations of Offender and Offense Related Questions

We then conducted a linear regression to determine which of these factors would uniquely predict the punitiveness of the treatments that the participants assigned to each case. The only significant predictors retained in the prediction equation were the perceived seriousness of the crime, b = 0.45, t(346) = 9.35, p < 0.0001, and how likely it was that participants thought that the offender could be rehabilitated, b = −0.12, t(346) = −2.40, p < 0.02. These two predictors together explained 23% of the variance.

Effect on Sentencing

We predicted that when people sent cases to the mixed procedure, they would reduce the sentence given to the offender, as compared to cases that were sent through the traditional court system.¹⁰ Before we tested this prediction, we first checked whether the participants from the control group matched the other participants on the six offense and offender-related questions. Beyond the difference in the seriousness rating for the oil drum theft reported previously, there were no differences between the two groups on the remaining questions (all p’s > 0.05), except that participants in the control group (M = 1.39, SD = 0.50) found the bike theft to be less violent than did the other participants (M = 1.92, SD = 1.03), F(1, 55) = 4.29, p < 0.04.

In order to test whether participants reduced the sentences of those who successfully completed a restorative justice conference, we compared the sentences of the mid and high-seriousness crimes recommended by participants who sent the crimes to the mixed procedure to sentences for the same crimes as recommended by the control group participants who could only use the traditional court system. We did not include low-seriousness crimes in this analysis because of the low number of participants who opted to send these crimes to the mixed procedure. We conducted ANOVA’s on the mid and high-seriousness crimes, with type of procedure (the mixed procedure or the traditional court procedure from the control group) as the independent variable and sentence severity as the dependent variable. The ANOVAs revealed that for all mid and high-seriousness crimes, the severity of the sentences given to the offender were reduced when the cases were sent through the mixed procedure (Fig. 3). The participants who could only use the traditional court process generally assigned the standard sentence for these crimes. When these same cases were sent to the mixed procedure, participants always significantly reduced the jail time for offenders to below the standard sentence for the crime (all p’s < 0.02).

Fig. 3.

The severity of the sentence (+SE) given by participants who judged the cases going through the traditional court system and participants who sent these cases to the mixed procedure. *All pairwise p’s < 0.02.

Discussion

In Study 1, we expected that participants would send crimes that were low in seriousness to a purely restorative procedure, but they would require a procedure that also included the possibility of punitive measures for more serious crimes. The results confirmed our predictions. We found that participants did send a majority of the low-seriousness cases to the pure restorative procedure. As crimes increased in seriousness, however, participants no longer thought that this process was the appropriate justice procedure. Participants were split over whether to send mid-seriousness crimes to the pure restorative or mixed procedure, and they sent a majority of the cases high in seriousness to the mixed procedure, even though they had the option of the traditional court process. In addition, as the cases increased in seriousness, more participants opted to use the traditional court system. These results support our hypothesis that low-seriousness crimes can be handled justly (according to people’s perceptions) using only restorative measures, but that people desire the possibility of punitive punishments for more serious offenses.

In addition, we found that features of the offense and the offender affected the level of punitiveness that participants recommended in each case. There was a high degree of intercorrelation among these factors. The violence, moral offensiveness, and seriousness of the offense were correlated with each other and the possibility of punitiveness of the procedure that the participants chose. The same result was found for the offender-related questions (how violent-prone, likely to re-offend, and able to be rehabilitated the participants perceived the offender to be). These findings indicate that the respondents may be forming a global judgment about the seriousness and moral offensiveness of the crime, as well as a global judgment about the criminal character of the offender and then drawing on these judgments to determine their treatment assignments for him. The questions we asked about the crime and the criminal may have tapped closely related aspects of these two general constructs. In any event, we did find that only one of the measures from each of these two clusters (seriousness and rehabilitation) independently predicted the potential punitiveness of the procedure that participants chose. We will further investigate the importance of rehabilitation in Study 2.

The results from Study 1 also demonstrate that people are willing to reduce prison sentences for offenders who successfully complete a restorative justice conference. It appears that people thinks that what occurs during the conference and the resulting agreement should serve to lower the prison sentence for the offender.

STUDY 2

The findings from Study 1 demonstrated that individuals require punitive measures for serious crimes, and that they will lower prison sentences after an offender has successfully completed a conference. Like all justice procedures, however, restorative justice procedures are not always successful. Inevitably, there will be conferences in which agreements between the victims and the offenders will not be reached. When a conference is not successful, what happens to people’s desire to punish, and to their perceptions of the offender? One possibility is that an offender’s mere participation in a conference (regardless of the outcome) will serve as a mitigating factor on participants’ sentencing decisions. Another possibility is that individuals will act as if the conference had not taken place, or they may even choose to penalize the offender for the failure of the conference with a harsher sentence.

In this experiment, participants were presented with reports that some conferences from the mixed procedure had failed, either because the victim and the offender could not reach an agreement (no-fault conference outcome), or that failed because of a lack of effort on the part of the offender (offender-fault conference outcome), as well as other conferences that were successful (as in Study 1). In the no-fault conference outcome, there was no fault assigned to the offender, or anyone else, for the failure of the conference. In the offender-fault conference outcome, the offender was responsible for the failure of the conference. We included both of these types of conference outcomes to investigate whether offenders are sentenced and viewed differently depending on why the conference failed. We predicted that the reduction in sentencing that occurred for offenders who successfully completed a conference would not occur for those offenders whose conferences were not successful. It was possible that there would be an actual increase in severity of punishment with the cases in which the offender caused the conference to fail.

We also expected that people would view offenders who have unsuccessful conferences more negatively than those who have successful ones. Study 1 investigated individuals’ perceptions of the offender before they made their send decision. In the current study, we examined the effect that the success or failure of the mixed procedure had on participants’ views of the offender. Participants answered the offender-related questions (proneness to violence, recidivism, and rehabilitation) after they learned the outcome of the procedure they selected for each case (the pure restorative procedure, the mixed procedure, or the traditional court process). Having participants answer the questions after they knew the outcome allowed us to examine whether people’s views of offenders are affected when they fail at completing the restorative component.

We were especially interested in rehabilitation, as Study 1 demonstrated that how well people thought an offender could be rehabilitated partly predicted the potential punitiveness of the procedure that they chose. We predicted that people would view the offenders who were sent to the mixed procedure and then did not reach an agreement in the conference as being less likely to be rehabilitated than their counterparts who did reach an agreement. We will not present the findings for the other two offender-related questions, as neither was a significant predictor of punitiveness in Study 1.

Method

Participants

Forty-three Princeton undergraduate psychology students (24 male, 17 female, 2 did not report their gender) participated in the study for course credit. Twenty-three of the participants received the no-fault conference outcome first, while 20 received the offender-fault conference outcome first. The participants ranged in age from 18 to 22 (M = 19.51, SD = 1.03). Thirty-one identified themselves as Caucasian, 5 as Hispanic, 3 as Asian-American, 1 as African-American, and 3 as “Other.”

Procedure

The procedure was identical to Study 1, except for the placement of the offender-related questions and the inclusion of the no-fault and the offender-fault conference outcomes for the mixed procedure. In Study 1, participants answered the offender-related questions (how likely they think it is that the offender will commit another crime, how violence-prone they think the offender is, and how well they think that the offender could be rehabilitated) in the first part of the experiment (before they made their decision about which procedure they wanted to use for the offense). In the current study, participants were asked these questions in the second part of the experiment (after they learned of the outcome of the conferences, if applicable, and had made their decision on what sentence they wanted to give for the particular case). This change from the previous experiment allowed participants to learn whether the conference had been completed successfully before answering these questions.

The other change from Study 1 was the inclusion of the mixed procedure conferences in which the victim and the offender could not reach an agreement. This conference either did not work because the victim and offender had tried but could not reach an agreement (no-fault) or because the offender had not cooperated (offender-fault). In both of the failed mixed procedure conferences, participants were told that an agreement had not been reached between the victim and the offender. In the no-fault conference outcome, the facilitator commented, “The conference between the offender and the victim was unsuccessful. Both the victim and the offender worked to come to an agreement, but, unfortunately, they were not able to come to agree about what the offender should do to repair the harm caused to the victim. The offender and the victim were both willing to work out an agreement, but they had different remedies in mind.”

The victim then stated, “I am disappointed with the outcome of the conference.”

In the offender-fault conference outcome, the facilitator commented, “The conference between the offender and the victim was unsuccessful. The offender was not putting much effort into the conference and seemed unwilling to work out a reasonable agreement. Thus, the offender and the victim were not able to come to an agreement about what the offender should do to repair the harm caused to the victim.” As in the no-fault conference outcome, the victim expressed his/her disappointment with the outcome of the conference with the same statement of disappointment.

In the “no-fault first” condition, the first case that participants sent to the mixed procedure was returned to them as a conference in which both the offender and victim tried to reach an agreement, but one was not reached. The next case that these participants sent to the mixed procedure was returned to them as a conference in which an agreement was not reached because of the offender (offender-fault). All of the rest of the cases that participants sent to the mixed procedure were successful (as in Study 1). In the “offender-fault first” condition, the order of the no-fault and offender-fault mixed procedure conferences was reversed.

Results

Crime Seriousness and the Send Decision

We first wanted to determine whether, as in Study 1, participants’ perceptions of the seriousness of the crime predicted which procedure they would choose for each case. Consistent with Study 1, as offense seriousness increased, the tendency to send offenders to the pure restorative procedure dramatically decreased, r(7) = −0.96, p < 0.0001, while the tendency to send offenders to the mixed procedure dramatically increased, r(7) = 0.94, p < 0.0001 (Table III).

Table III. Participants’ Mean Seriousness Ratings and Send Decision Proportions (Study 2)

A nominal logistic regression with seriousness as the predictor and the procedure chosen as the outcome confirmed this result. Again, there was a good model fit based on the seriousness of the offense, χ²(12, N = 387) = 113.70, p < 0.0001, Nagelkerke R ² = 0.29. Overall, the correct classification rate was 50.6%, with 74.6% correctly classified to the pure restorative procedure, 68.9% to the mixed procedure, and 0% to the court procedure. Again, participants chose court less frequently than they chose either of the restorative procedures. The pure restorative procedure had crimes that were perceived as less serious (M = 3.53, SD = 1.31) than offenses that were sent to the mixed procedure (M = 5.22, SD = 1.43) or the traditional court process (M = 5.19, SD = 1.59). Consistent with Study 1, this analysis demonstrates that for the pure and mixed procedures, offenses that had a lower mean serious rating were more likely to be sent to the pure restorative procedure, whereas offenses that had a higher mean serious rating were more likely to be sent to the mixed procedure.

To examine this finding more closely, we again divided the crimes into seriousness groups (see Table III). As in Study 1, vandalism, assault, and the bike theft were all rated below the mid-point on the seriousness scale, mugging and the oil drum theft were rated near the mid-point, and burglary, identity theft, attempted murder, and rape were all rated above the mid-point.¹¹ The no-fault first condition did not differ from the offender-fault first condition in its seriousness rankings (all p’s > 0.1).

We conducted a repeated measures ANOVA to determine whether the procedure to which participants chose to send each case differed based on the seriousness of the crimes (Table III). We used the mean proportions of cases from each seriousness group that were sent to the three different procedures as the dependent variable. As in Study 1, this analysis revealed that participants’ send decisions did differ based on how seriously they rated each offense, F(8, 336) = 17.18, p < 0.0001, η² = 0.29. As shown in Fig. 4, the proportion of cases sent to the pure restorative procedure decreased as the crimes became more serious. For the mixed procedure, the proportion of cases sent to this procedure decreased as the crimes became less serious. In addition, for the low and high seriousness crimes, one of the alternative procedures (the pure restorative or the mixed procedure, respectively) was the preferred option over the traditional court procedure. Most participants opted to send the cases that were low in seriousness to the pure restorative procedure, whereas most chose to send the high-seriousness cases to the mixed procedure. For the mid-seriousness crimes, participants were evenly divided among the three procedures.

Fig. 4.

The proportion of cases (+SE) sent to the pure restorative procedure, the mixed procedure, and the traditional court process for low, mid, and high seriousness cases (N = 387).

Punitiveness

We replicated the finding that participants’ own judgments of the seriousness of the offense would predict the level of punitiveness that they would select for each case. A correlational analysis demonstrated that as participants’ ratings of the seriousness of the crime increased, the more likely they were to choose a procedure that had more punitive punishments, r(385) = 0.41, p < 0.0001.

Effect on Sentencing¹²

Study 1 demonstrated that when offenders successfully complete a mixed procedure conference, their sentences were reduced, as compared to the sentences for those same cases when they went through the traditional court procedure. In the present experiment, we were interested in how the sentencing for a case is affected when conferences are not successful. Furthermore, we were interested in whether there was a difference between a conference that did not work simply because both parties could not come to an agreement (no-fault) and a conference that did not work because of the lack of effort by the offender (offender-fault). For the remainder of the analyses, we only used the four cases that were high in seriousness, as both the low and mid-seriousness cases were not sent frequently enough to the mixed procedure to permit these analyses. We collapsed across the high-seriousness cases for these analyses to increase our power and because participants in Study 1 had made similar reductions to the offenders’ prison sentences for all of the high-seriousness crimes. We also collapsed across both the no-fault first and the offender-fault first conditions, as the conditions did not differ from each other in their sentencing of offenses across the three different outcomes of the mixed procedure (all p’s > 0.05).

We found that there were differences in sentences among the no-fault conference outcome, the offender-fault conference outcome, the successful outcomes of the mixed procedure, and the traditional court system, F (3, 159) = 29.91, p < 0.0001, η² = 0.36 (Fig. 5). A post hoc Tukey HSD analysis revealed that offenders from both the no-fault (M = 4.48, SD = 1.06) and the offender-fault (M = 4.50, SD = 0.99) conditions received higher sentences than did offenders from the successful conferences (M = 2.67, SD = 1.41), both p’s < 0.0001. The sentence severity did not differ between the two unsuccessful conferences, p > 0.1.

Fig. 5.

The severity of the sentence given to offenders in the no-fault unsuccessful outcome of the mixed procedure (n = 29), the offender-fault unsuccessful outcome of the mixed procedure (n = 26), the successful outcome of the mixed procedure (n = 42), and the traditional court process (n = 66).

We also examined how the sentences given in the successful and unsuccessful mixed procedure conferences compared to sentences that came from the traditional court system. Tukey’s HSD revealed that, as expected, only the offenders who successfully completed conferences received lower sentences than did those offenders who went through the traditional court system (M = 4.68, SD = 1.04), both p’s < 0.0001. Offenders who went through the no-fault and the offender-fault mixed procedures were assigned very similar sentences to those who went through the traditional court process, both p’s > 0.1. For the successful mixed procedure cases, offenders received a reduction from the standard sentence, whereas offenders who were in unsuccessful mixed procedures were treated similarly to the offenders who went through the traditional court system, as they received (approximately) the standard sentence for that crime.

Rehabilitation

We predicted that participants’ perceptions of how well an offender could be rehabilitated would be affected by whether or not a conference was completed successfully. We also expected that offenders whose cases were sent to the pure restorative procedure would be viewed as the most favorable candidates for rehabilitation, whereas offenders sent to the traditional court system would be viewed as the least favorable. For this analysis, we again used only the high-seriousness cases and collapsed across the two different unsuccessful outcomes of the mixed procedure, as their ratings did not differ from each other, p > 0.1. We found that participants’ ratings for the rehabilitation potential of offenders did differ based on which procedure was selected for the case and the outcome of the conference, F(3, 168) = 12.55, p < 0.0001, η² = 0.18 (Fig. 6). As expected, offenders who did not successfully complete the mixed procedure (M = 4.30, SD = 1.19) were viewed as less likely to be rehabilitated than were those who had a successful mixed procedure conference (M = 4.98, SD = 1.19), p < 0.04. Offenders whose cases were dealt with through the traditional court process (M = 3.71, SD = 1.26) received lower rehabilitation scores than did those whose cases were handled via one of the alternative procedures (regardless of the success of that procedure), all p’s < 0.05. Participants rated the offenders whom they sent through the pure restorative (M = 5.56, SD = 1.13) and mixed processes (who completed them successfully) as most likely to be rehabilitated (all p’s < 0.05), and these judgments did not differ from each other, p > 0.1.

Fig. 6.

Participants’ ratings of how well they thought the offender could be rehabilitated for the pure restorative procedure (n = 9), the successful outcome of the mixed procedure (n = 41), the unsuccessful outcome of the mixed procedure (n = 56) and the traditional court process (n = 66).

Discussion

Study 2 replicated the Study 1 result that the seriousness of the crime predicts which procedure participants will assign for each crime. As before, people primarily opted for the pure restorative procedure for low-seriousness crimes, but increasingly used the mixed procedure as offenses became more serious. Although participants in this study chose to use the traditional court system slightly more than the participants in Study 1 did, the overall pattern of results was the same. In addition, we also replicated the result that as people’s perceptions of the seriousness of the crime increased, they tended to choose a more punitive procedure.

More importantly, in the present experiment, we found that when offenders did not successfully complete a conference in a mixed procedure, they received more severe sentences post-conference than those offenders who did come to an agreement. It did not make a difference whether the conference failed because of lack of agreement between the parties or lack of effort on the part of the offender. The offenders whose conferences failed were assigned slightly higher than the standard court sentence for their crimes, which was similar to the sentences received by those who went through the traditional court system.

Participants also assigned similar sentences for offenders who had failed conferences as for offenders who went through the traditional court process. Participants may have blamed the offender for the failure of the conference regardless of whether he was directly implicated or not. Indeed, one participant commented about the offender in a no-fault case, “Alex probably does not feel the remorse that he should.” In addition, instead of seeing the failed conference as a reason to increase the severity of the sentence, participants may have simply judged the case as if it were sent through the traditional court system. One participant stated, “The [mixed procedure] conference was unsuccessful, so I think the case should be regarded as a standard instance of this crime.” These findings suggest that regardless of the reason for the lack of agreement between the offender and the victim, people do not want to assign much more than the standard sentence for offenders who fail to successfully complete their conferences.

We also found support for the hypothesis that people’s views of offenders’ abilities to be rehabilitated were influenced by the outcome of the conference for offenses that were high in seriousness. The offenders who were sent to the mixed procedure and successfully completed a conference were viewed as having the same potential for rehabilitation as those offenders whom participants sent to the pure restorative procedure. The offenders who did not successfully complete a conference, either because of lack of agreement between parties or because of their lack of effort, were seen as being less likely to be rehabilitated than the offenders who did complete the conference. However, the unsuccessful offenders were still considered more likely to be rehabilitated than the offenders whom participants chose to send through the traditional court process. Overall, these results demonstrate that when a conference is not completed successfully, for whatever reason, there are negative consequences for participants’ judgments of the severity of the sentence and their perceptions of the offenders.

GENERAL DISCUSSION

In two studies, we investigated people’s perceptions of the acceptability of restorative justice procedures, as well as the effects of both successful and unsuccessful conferences on people’s sentencing decisions and views of the offender. The results from Study 1 revealed that people are willing to resolve crimes that are low in seriousness through purely restorative procedures, but they require the possibility of assigning punitive measures to send more serious offenses to a procedure with restorative components. This latter result suggests that people are willing to allow restorative procedures when the offense is of a serious nature, but they also want the ability to inflict a retributive option. There were respondents who chose to send some of the offenders directly to the regular court system, perhaps because they saw little use in allowing restorative procedures to be engaged in at all. In addition, we found that perceptions of the seriousness of the offense and the ability of the offender to be rehabilitated predicted the punitive versus restorative character of the procedure that they chose. The less serious the offense, and the more potential they saw for the offender to be rehabilitated, the more likely they were to send the offender to a strictly restorative process.

The cases that participants sent to the restorative justice process that retained the possibility of also assigning punitive treatment (in the form of prison sentences) were of special interest to us. When these cases were returned to the participants for the punitive judgment, their task was to recommend the duration of the prison sentence that should be imposed. They were given a report that the restorative conference had been successfully concluded and that the conference facilitator and the victim were both satisfied with the outcome. Given this information, the participants made considerable reductions in the sentences to be imposed.

The second study added a component designed to determine the reactions of participants when they learned that in some cases the restorative conference did not come to a successful resolution. The results showed that people did not lower the prison sentences for offenders whose conferences were unsuccessful; rather, they gave these offenders the standard sentence for the crime, as they did with those who were sent through the traditional court system. Nor did people punish offenders for failed conferences by assigning more severe prison sentences. What is especially interesting is that respondents who were told that the conference failure was due to the behavior of the offender did not add to the standard punishment for that crime. Whether participants would have assigned more severe prison sentences to offenders who were actively disrespectful to the victim during the conference remains to be studied.

Our findings indicate that people do not always feel the need to assign prison punishments when a law is broken and the crime is rather minor. It may be that they do not think that a retributive punishment is necessary for low seriousness crimes, or they may think that some of the activities imposed by the restorative conference, such as hours of community service, are sufficiently punitive to accomplish that goal. However, as the crimes become more serious, people seek to impose the sort of retribution that prison inflicts on the offender, in addition to restoration.¹³

Certainly, these results would require replication beyond the limits of the small samples of college students that we tested. But assuming for the moment that our findings are generalizeable, what would be their bearing on the debate concerning restorative justice? They would suggest that most people do not share the view of Braithwaite and his colleagues (Braithwaite, 1998, 1999, 2002; Braithwaite and Strang, 2001), who argue that restorative procedures should be administered without possibilities of punitive measures. We are not arguing that our results, especially given the limitations of our studies, should cause people to abandon Braithwaite’s hope that punitive proportionality can be abandoned in restorative justice procedures. However, our data support the view of those who argue that to achieve perceptions of justice, at least for serious offenses, both restorative and retributive measures should be available within the system (Barton, 1999; Daly, 2002; Duff, 2003; Robinson, 2003). This is consistent with research that has found that people will opt for both restorative and retributive measures for death penalty cases when given the option to do so (Bowers, 1993; McGarrell and Sandys, 1996). As others have suggested, restoration alone may satisfy only part of what people seek in dealing with an offense once a certain level of severity has been reached (Roberts and Stalans, 2004).

One reason that people require the possibility of assigning retributive punishments to serious crimes may be that they seek to give a just deserts punishment to those offenses. Prior research identifies just deserts as a consideration that guides what information people want to obtain about a crime in order to make a sentencing decision (Carlsmith, 2006; Carlsmith et al., 2002; Darley et al., 2000). Furthermore, these studies have shown that a concern with just deserts heavily determines the sentences that respondents assign to serious crimes. In addition, several nations such as New Zealand, Canada, England, and Wales have codified the objectives of sentencing and explicitly included just deserts considerations (Roberts, 2005), which provides further evidence of the strong societal support for just deserts in sentencing.

We also found that people consistently lowered prison sentences for offenders who successfully completed restorative conferences. When people are able to satisfy their goal of restoration for the harm of the crime, they may not need to be as punitive as they are for offenders who either have failed conferences or have only been dealt with through the traditional court system. This result replicates previous findings that people assign less severe punishments when offenders demonstrate remorse (Felson and Ribner, 1981; Kleinke et al., 1992; Robinson et al., 1994; Rumsey, 1976 ) or provide some form of restitution to the victim (Shultz et al., 1986).

Our findings raise the question of why people opted to assign offenders to serve prison sentences when they had successful restorative conferences, as the participants were informed that the victim was satisfied with the outcome of the conference. Participants’ comments suggest that this result may be due to their concern with more than one justice goal (Daly, 2002). A number of participants remarked that they wanted to see both restoration of the victim and punishment of the offender. For example, one stated, “The purpose of the meeting was to help the victims come to terms with the incident. Although they reached a satisfactory agreement, the offender still needs to be punished for her crime.” It appears that people may distinguish between two goals when attempting to achieve justice: to repair the harm caused to the victim, and to punish the offenders for the wrongs they have committed. Other respondents suggested a third goal when they expressed concern with what the offender would gain from the conference. One stated, for the case of attempted murder, “Jake needs to serve time, but maybe less than the standard sentence because his real rehabilitation will come from the agreement he made with Sam.” Future research should more systematically investigate individuals’ lay theories about what happens in restorative justice conferences for both victims and offenders, and whether what they think transpires can affect their decision to send cases to a restorative justice procedure.

The concern that people have with the effect of the conference on the offender highlights the importance of the concept of rehabilitation for the acceptability of restorative justice. In Study 1, we demonstrated that how well people thought an offender could be rehabilitated was associated with the potential for punitive sanctions in the procedure they selected. And people’s perceptions of how well an offender could be rehabilitated were affected by whether or not a conference was successful in Study 2. Both of these findings provide evidence that the concept of restorative justice and rehabilitation are closely associated with each other. This is consistent with Bilz’s (2002) finding that people who are most supportive of sending a case to restorative justice procedures are the ones who believe in the possibility of rehabilitating offenders. Future investigations may want to further explore the link between restorative justice and rehabilitation. There may be apparently “un-rehabilitatible” offenders, who regardless of the seriousness of their crime, will not be considered for restorative conferences. On the other hand, influencing people’s thoughts about the likelihood of offender rehabilitation in general may affect their willingness to accept restorative justice procedures for specific offenders.

POLICY IMPLICATIONS

Our results are to some extent about procedures that create citizens’ acceptance of the use of restorative justice procedures. They suggest that in order for citizens to view restorative justice as an acceptable alternative to the traditional court system for serious crimes, the procedure must allow for the option of retributive measures. Although participants assigning cases to justice procedures opted for purely restorative procedures when the crimes were low in seriousness, they no longer found this pure restorative procedure acceptable for more serious offenses. Thus, unlike purely restorative procedures, the mixed restorative and retributive procedure (the favored selection for more severe crimes), can accommodate crimes that differ in severity. It allows the option to assign punitive punishments, but does not require it. If justice has been achieved through a successful conference, then a judge or jury can decide not to assign the offender a jail term or any other punitive sanction. Essentially, the mixed procedure would become identical to a purely restorative procedure because the punitive option would not be used. As crimes increase in their severity, the system can exercise the option to send offenders to prison in addition to their fulfilling the conditions of the conference agreement. This finding is especially important because restorative justice procedures have been shown to be most effective for crimes that are high, rather than low, in seriousness (Sherman, 2003), and most current restorative procedures have been limited to handling crimes that are low in severity (Barton, 1999). These results suggest that a mixed procedure holds the promise of meeting people’s expectations for justice, which should lead to more compliance with the law in general (see Tyler, 1990).

At the same time, considerable thinking and research needs to be done about how the justice system might implement a mixed restorative and retributive system. If the punishment was administered before the restorative elements, then the conference might be less reintegrative for the offender and less meaningful for the victim. However, particularly for the more serious crimes that receive longer prison sentences, there may be practical issues if the restorative elements are completed first. The community could become concerned that the offender is now a flight risk, or that a recent serious offender is temporarily at large. These are genuine difficulties with the mixed system, and they would have to be weighed against the gains achieved by the increase in the number of cases that would be sent to a mixed system as opposed to one that only allowed for restorative procedures. But one such gain is the incorporation of victims and their concerns in the justice process, which has been absent from the traditional court procedure (Barton, 1999; Hogan and Emler, 1981).

Another encouraging policy-relevant finding is that people chose to lower prison sentences for offenders when they successfully completed a restorative justice conference. These findings demonstrate that a mixed restorative and retributional procedure could result in a decreased reliance on the prison system for handling offenders. In addition, these results suggest that people are open to other avenues for achieving justice besides sentencing offenders to prison terms. This is promising not only for the implementation of restorative justice, but for other approaches that can aid the criminal justice system in reducing the current reliance and overuse of prisons in the United States.

The findings from these two studies indicate that the marriage of restorative and retributive measures appears to satisfy people’s expectations for justice, as well as reducing the strain on the prison system. These results are especially promising with regard to forming a blueprint of how restorative justice can be implemented in the American criminal justice system. The current studies suggest that restorative justice can present a viable alternative to the current criminal justice system for a range of serious offenses when there is still an option to assign retributive punishments. The combination of restorative and retributive justice may provide an improved justice process that provides victims, offenders, and members of the community a way to address both the harms and the wrongs that result from crime.

Appendices

Appendix A: Instructions

This study is about possible alternatives to the currently standard criminal court processes. People who think about our criminal justice process are constantly thinking of ways to improve the current system, so that it will function better and make all those involved in the process feel that justice has been achieved. A major problem with the current system is its reliance on prisons, which are overcrowded and expensive to run. One set of innovations, known broadly as restorative justice, seeks to bring the interests of the victim into the forefront, while still imposing treatments on the offender. Crime victims often feel much better after going through such a process, and this process also provides an opportunity for the offender to understand vividly the consequences that his or her crime had on the victim and the community at large.

Today you will be considering two alternative processes. One of these processes is known as Victim Offender Mediation (VOM) and the other is known as Restorative Justice Mediation (RJM). The following is a description of these two processes. We will begin with the VOM procedure.

After an offender has been arrested for a crime and admitted his guilt, a judge will decide whether or not the case is appropriate for VOM. If the victim agrees to it, the offender will be given the option of attending a VOM instead of a traditional sentencing before a judge. The VOM is a face-to-face meeting between an offender and his victim, with a facilitator (a neutral third party who has prepared both sides for the meeting beforehand) present. Sometimes, there may be multiple victims, or even members of the community or friends/relatives of the parties present.

During this meeting, the victim is allowed to ask the offender any question he or she wishes. During the meeting, the victim and offender may work out an agreement outlining what the offender must do to atone for his wrongs and make the victim whole again. The terms of the agreement may include an apology, monetary compensation, some services that the offender does for the victim, community service, and the like.

During the conference, the offender may come to understand just what anguish the offenses caused for the victim, may feel remorse, and may express that remorse to the victim. Victims have rated these experiences very positively and have reported that these exchanges go a long way toward helping them recover. Victims have reported positive experiences with the VOM procedure for all types of crimes, even ones that may initially seem surprising.

Although there are varieties of possible outcomes of a VOM, imprisonment is not one of the possibilities. As an expert in restorative justice states, “Restorative justice is about healing (restoration) rather than hurting. Responding to the hurt of the crime with the hurt of punishment is rejected, along with its corresponding value of proportionality-punishment that is proportionate to the wrong that has been done.”

If an agreement is reached, it is put into a written contract, which both parties sign. Upon successful completion of the agreement, the process is concluded. The offender does not return to the courts. However, if the parties cannot come to an agreement, or if the offender fails to live up to the terms of the contract, then the offender is routed back into the regular criminal justice system for sentencing before a judge, who will know that the VOM was unsuccessful.

The RJM procedure is similar to that of the VOM, but there are a number of important distinctions that you must keep in mind. The following is a description of the RJM procedure.

As in the VOM procedure, after an offender has been arrested for a crime and admitted his guilt, a judge will decide whether or not the case is appropriate for RJM. If the victim agrees to it, the offender will be given the option of attending RJM. The RJM conference is very similar to the VOM conference. What differs is that in RJM (unlike VOM in which the case does not return to the courts) the judge retains the right to add any of the standard sentencing options after the RJM process, if he or she feels that they are warranted. In RJM, imprisonment remains as one of the options for the punishment of the offender.

Like VOM, victims have rated these experiences very positively and have reported that these exchanges go a long way toward helping them recover. Victims have reported positive experiences with the RJM procedure for all types of crimes, even ones that may initially seem surprising.

The agreement is put into a written contract, which both parties sign. Once the agreement is reached, the offender is then returned to the courts, where he or she will stand before a judge for possible prison sentencing. The judge considers this issue based on the knowledge that the RJM was successful. A successful completion of the conference does not mean that the offender avoids a court sentence; if the judge feels that justice has not been done unless such a sanction is added, he or she will add the sentence. A successful RJM process does not even necessitate a lowering of the sentence that the judge would have given with no RJM process. If the judge believes that the offender should receive the standard prison sentence for the crime (or even a longer sentence), then he or she is still free to assign such a sentence. In sum, the judge may choose to decrease (including not giving any jail time), increase or maintain the standard sentence as he or she believes is called for by the circumstances of the case.

If the parties cannot come to an agreement, or if the offender fails to live up to the terms of the contract, then the offender is routed back into the regular criminal justice system for sentencing before a judge, who will know that the RJM was unsuccessful.

Your task is to read nine scenarios; each of which describes a crime that occurred. While you are reading, we ask that you think about how the offender in each case should be treated. We are interested in your view; we are not looking for one answer in particular. You will be asked to act as the impartial judge for each crime, in which each offender has already admitted their guilt and each victim is willing to participate in the alternative procedures. It will be your first task to decide whether the crime goes through the traditional court process, VOM or RJM. In the traditional court process, the case will remain in the court and the judge alone will make the decision on the punishment for the offender. If you choose either VOM or RJM, then the case will first be routed through a conference. For cases that go through RJM, they will then return to courts for sentencing. When you are considering each case, you should choose the process that you feel will best achieve justice for that particular crime.

If you need to reread the descriptions of either VOM or RJM, you can return to them at any time throughout the study by clicking on the Instructions link.

You will also be asked some general questions about each case. Once you have worked through all of the cases, there will be additional questions for you to answer regarding the handling of the offenders. Again, there is not one correct answer; we want to know your view!

Appendix B: Cases

Vandalism: It was Halloween. Timothy and Rob, both eighteen years old and seniors in high school, dressed up in “ghost” costumes (sheets with holes cut in them for eyes) and headed out at 11:00 p.m. They strung toilet paper on the trees and bushes of several houses on Timothy’s girlfriend’s street, and shoe-polished several windows with the names of their school’s sports team. They went too far in their antics, however, when they scraped their school’s initials in the doors of Jerry’s car, causing Jerry several hundred dollars worth of paint damage. The next day, they were arrested for vandalism and destruction of private property, for the car damage.

Bike Theft: For several weeks, Tina had admired the very expensive bicycle of Julia, a person who lived several blocks away. Julia, who was an avid bike rider, had purchased the bike with money that she had been saving for over a year. Julia left the bike well-locked in the sidewalk bike stand outside her building. One morning, Tina walked over to the bike with a bolt cutter, to steal it. She found that it was pretty easy to ride off on this bike without being detected. Two days later, a police officer appeared at her door and arrested her, as Julia had spied the bike locked up outside Tina’s apartment.

Assault: Dwayne and his college buddies were at a local sports bar, watching the close, final game of the NBA playoffs. Another group of men was watching the same game, but rooting for the opposite team. Dwayne’s team missed a crucial three-pointer, and the other group cheered and began taunting Dwayne’s group. Dwayne asked them to “shut up,” at which point Matt, one of the men from the other group, told Dwayne to “come over here and make me.” They exchanged words and, at this point, Dwayne punched Matt in the face, breaking his nose. Dwayne was pulled away from the man before he could do any more damage. The police were called and Dwayne was arrested for assault.

Mugging: Andrew was walking down a busy street. He passed Jim, who was wearing a business suit and talking on his cell phone. Jim was in the middle of a heated exchange with another member of his firm over the case of one of their most important clients. Figuring that Jim was adequately distracted, Andrew slipped his hand into Jim’s back pocket, removed his wallet, and quickly disappeared into the crowd. Jim did not notice and Andrew escaped undetected. The police were able to apprehend Andrew later that day when he attempted to use the stolen credit cards.

Oil Drum Theft: Rich knew that construction companies will pay significant amounts of money for empty oil drums. He planned to steal several drums and selected one of the few independently owned gas stations in the area, where he knew no one would come around during his theft. Charles, the owner of the gas station, had recently suffered a huge cut in his profits when a competing franchised gas station has opened down the street. Money was tight, and he was concerned about the livelihood of his business. After the gas station was closed and everyone had gone home, Rich stole four oil drums. These oil drums were required to be returned to the manufacturer, and their theft resulted in Charles paying a significant amount of money in fines. Rich was later arrested for the theft while trying to sell the oil drums to a construction company.

Burglary: Lisa had been scouting out an upscale neighborhood and noticed that one family, the Beckers, appeared to leave for a vacation. The Beckers had just moved to the neighborhood over the summer, and they recently had finished redecorating their home. She broke into the house in the middle of the night, armed with a gun. Lisa managed to remove over $100,000 worth of property from the house before a neighbor noticed the burglary and called the police. Lisa was apprehended while driving away from the house in her van, where all of the stolen property was discovered.

Identity Theft: On a trip to the supermarket, Eliza’s wallet fell out of her bag while she was loading groceries into her car. Allison found the lost wallet after Eliza had driven away. The wallet contained many pieces of identification, including Eliza’s social security card. Allison used this information to obtain a driver’s license that had her photograph but Eliza’s information. Allison then used the driver’s license to take out mortgages on Eliza’s property, as well as opening a number of credit cards. The police finally caught Allison when she attempted to obtain a passport with Eliza’s information.

Attempted Murder: Jake had been sitting in traffic for two hours when the road started to clear up. He went to change lanes when he was cut off by another car. Sam, the driver that cut him off, then proceeded to drive well below the posted speed limit. This enraged Jake, who began flashing his high beams and honking his horn at Sam. Sam pulled over and Jake followed him to the side of the road. Both men got out of their cars and started arguing. The argument became more and more heated, and eventually Jake pulled out a knife and stabbed Sam in the chest. A passing motorist saw the stabbing and called the police. Jake was charged with attempted murder as Sam was seriously wounded but survived the attack.

Rape: Pam and Alex worked together at a local restaurant. He is a 28-year-old recently divorced cook; she is a waitress. Pam and Alex had dated for a few months two years before, but Pam had broken up with him because, as she told her friends, he had started to act “creepy.” Pam and Alex had slept together only for the last month they were dating. One day, Alex asked Pam if they could meet for drinks. She agreed. The next evening, after their date, Pam invited Alex to her apartment for coffee. They sat on the sofa and talked about old times. Suddenly, Alex reached out and began kissing her. She protested, but he pressed on. While she cried, he had sexual intercourse with her. After Alex left, Pam called the police, and Alex was arrested for rape.