Retributive and Restorative Justice PDF

Law Hum Behav (2008) 32:375–389 DOI 10.1007/s10979-007-9116-6 ORIGINAL ARTICLE Retributive and Restorative Justice Michael Wenzel Æ Tyler G. Okimoto Æ Norman T. Feather Æ Michael J. Platow Published online: 24 October 2007 Ó American Psychology-Law Society/Division 41 of the American Psychologic...

Law Hum Behav (2008) 32:375–389 DOI 10.1007/s10979-007-9116-6 ORIGINAL ARTICLE Retributive and Restorative Justice Michael Wenzel Æ Tyler G. Okimoto Æ Norman T. Feather Æ Michael J. Platow Published online: 24 October 2007 Ó American Psychology-Law Society/Division 41 of the American Psychological Association 2007 Abstract The emergence of restorative justice as an with due politeness. Whatever the exact nature of the alternative model to Western, court-based criminal justice transgression, the victims are not only deprived of some- may have important implications for the psychology of thing that is considered due to them (their life, property, justice. It is proposed that two different notions of justice respect, etc.), but also an agreed-on rule, law, or norm is affect responses to rule-breaking: restorative and retribu- violated. Therefore, even when it is possible to give back to tive justice. Retributive justice essentially refers to the victims what they have lost and undo the harm done to repair of justice through unilateral imposition of punish- them, there remains the fact that a rule-violation has ment, whereas restorative justice means the repair of occurred. The transgression itself, if intentional and justice through reaffirming a shared value-consensus in a blameworthy, is an injustice that victims and observers bilateral process. Among the symbolic implications of (perhaps also the offender) usually feel needs to be dealt transgressions, concerns about status and power are pri- with, beyond restitution of the victim’s losses. marily related to retributive justice and concerns about In the criminal justice system, in Western societies at shared values are primarily related to restorative justice. At least, the primary means of dealing with the injustice the core of these processes, however, lies the parties’ implied in a transgression is punishment. Courts impose construal of their identity relation, specifically whether or punishment on offenders; once a punishment is imposed, not respondents perceive to share an identity with the justice is often considered done. Of course, seeking justice offender. The specific case of intergroup transgressions is is not the only possible goal of punishment. Philosophers discussed, as are implications for future research on commonly distinguish between consequentialist and retri- restoring a sense of justice after rule-breaking. butivist justifications for punishment (see Duff 2001), which crudely correspond to the motives of behavior Keywords Retributive justice
Restorative justice
control (reducing the reoccurrence of the offense through Transgressions
Status/power
Values
Identity incapacitation, deterrence, or rehabilitation) and justice restoration, respectively (Vidmar and Miller 1980). These different punishment goals have also been confirmed Rule-violations and transgressions can take many forms, empirically among lay people (e.g., Carroll et al. 1987; De from the most heinous crimes to rather trivial breaches of Keijser et al. 2002; McFatter 1978, 1982). Yet, evidence norms, from mass killings to the refusal to greet somebody suggests that justice restoration, not behavior control, is the dominant motivation underlying people’s calls for pun- ishment (Carlsmith et al. 2002; Darley et al. 2000; M. Wenzel (&)
T. G. Okimoto
N. T. Feather School of Psychology, Flinders University, GPO Box 2100, McFatter 1982). The relevant research, however, considers Adelaide, SA 5001, Australia only one specific notion of justice restoration: retributive e-mail: [email protected] justice or just desert. According to this notion, an offender, having violated rules or laws, deserves to be punished and, M. J. Platow School of Psychology, Australian National University, Canberra, for justice to be reestablished, has to be punished in pro- ACT 0200, Australia portion to the severity of the wrongdoing. 123 376 Law Hum Behav (2008) 32:375–389 In recent decades, a movement emerged—first at grass As we will see, these are not only academic questions. root levels among practitioners disenchanted by the exist- People, under different circumstances, differ in their pre- ing criminal justice system, then taken up by academics— paredness to engage in restorative justice practices when which challenged the assumption underlying the existing these are offered as an alternative or complement to tra- criminal justice system that punishment of the offender is ditional court processes; and when people have engaged in sufficient, or even necessary, to restore justice after crim- restorative justice processes they often exhibit varying inal offenses. The alternatives that have been put forward levels of satisfaction (Strang 2002). However, in order to are now commonly referred to as restorative justice (Ba- open this question up to psychological inquiry, we want to zemore 1998; Braithwaite 1999). The various models, abstract from specific restorative versus retributive prac- heterogeneous as they are, typically regard transgressions tices, and even from the criminal domain altogether (to as conflicts that need to be given back to their rightful which restorative justice is by no means restricted; Brai- owners for them to resolve: offenders, victims, and their thwaite 2002; Roche 2006). The question for psychologists respective communities (Christie 1977). In practice this is whether different notions of justice guide responses to means the affected parties are directly involved in the transgressions and which psychological factors and con- justice process. In a deliberative interaction, they are given cerns determine people’s justice notion. We will take a voice to vent their feelings, present their side of the story, distinctly social-psychological perspective on this issue and ideally come to an agreement about the hurt the and assume that people’s notion of justice following a offense has caused, the offender’s responsibility, and what transgression depends on how they construe their rela- can be done to restore a sense of justice. Such measures can tionship to the offender and interpret the incident. include direct compensation to the victim, compensation or Ultimately, we will argue that people’s sense of identity in services to victims of similar offenses or to the wider the given context will affect their adoption of either a community, as well as meaningful punishments to the retributive or restorative notion of justice. offender that benefit the victim, the community and, The remainder of this article is organized as follows. We through facilitation of moral transformation, potentially the will first revisit the notion of restorative justice as it has offender (Bazemore 1998). While punishment can be, and been advanced in recent decades, specifically in criminol- often is, part of restorative justice practices, it is not cen- ogy and law, and review existing research. We will then tral. Crucial for proper restorative justice is a process of systematically develop our own framework: we will (a) deliberation that places emphasis on healing rather than distinguish between justice and behavior control motiva- punishing: healing the victim and undoing the hurt; healing tions following transgressions, (b) clarify what exactly the offender by rebuilding his or her moral and social justice has to accomplish after transgressions, (c) differ- selves; healing communities and mending social relation- entiate between status/power and value concerns ships (J. Braithwaite 1998, 2002). underlying justice motivation, (d) analyze how retributive If, as we assume is the case, restorative justice actually and restorative responses differentially address these two aims to rebuild a sense of justice (and not only modify concerns, and (e) identify how different notions of self and behavior), then it poses an interesting challenge to the identity could be implicated in the two justice processes. psychology of justice. Existing research on issues of justice We will then discuss implications of our conception for following rule-breaking has largely focused on the role of intergroup transgressions before concluding with implica- punishment, in an area that is called, quite tellingly, tions for further research. retributive justice research (Darley 2002; Feather 1999; Hogan and Emler 1981; Tyler et al. 1997; Vidmar 2000; Vidmar and Miller 1980). This domain of research com- Seeking Justice After Rule-breaking monly addresses issues of how and why people want to punish offenders, and it has been found that justice is a The Restorative Justice Movement prime motivation (e.g., Carlsmith et al. 2002). How does restorative justice fit in with these findings if it considers While our investigation will not be concerned with a par- punishment neither necessary nor sufficient to restore jus- ticular practice or tradition of restorative justice but rather tice? How does restorative justice serve a sense of justice seek to establish a psychological concept of restorative after transgressions if it does not operate solely through the justice, this concept will nonetheless have some grounding imposition of punishment? Can we distinguish different in the criminal justice practice and research. The modern psychological (or lay-philosophical) notions of justice, a Western criminal justice system strongly features princi- retributive and a restorative notion? What factors and ples of objectivity and consistency, in that an impartial conditions determine whether people subscribe more to a judge or jury deliberates a case on the basis of all available retributive or restorative notion of justice? evidence and applies the law based on its text, accepted 123 Law Hum Behav (2008) 32:375–389 377 interpretation, and common practice. However, because community, or take part in an educational program crime is considered the domain of the state, there is only a (Bazemore 1998). However, while these alternative pun- limited role for the stakeholders themselves; for victims ishments may be more conducive to its aims, the idea of and offenders to present their side of the story, express their restorative justice should not be reduced to their applica- feelings, ask and answer questions important to them, or tion alone. At the core of restorative justice is a dialogical offer an apology or forgiveness. Out of frustration with the process geared toward making offenders accept account- formal court-based justice system, practitioners began in ability for the harm they have caused (as well as its repair), the 1970s to experiment with alternative practices and show remorse, and offer an apology, while victims are, at ideas (e.g., Peachey 1989). In the 1990s, these were con- least implicitly, encouraged to overcome their resentment ceptualized more programmatically as restorative justice. and offer forgiveness (Retzinger and Scheff 1996; Roche Due to the post-hoc nature of the concept, restorative jus- 2003; Strang 2002; Zehr 1985). tice subsumes a number of different notions and there is not But does it work? As far as its crime-control potential is necessarily consensus about its definition. concerned, the evidence suggests that, relative to court One early intellectual basis of restorative justice is processes, restorative justice programs tend to reduce re- Christie’s (1977) view that offenses are to be considered as offending, but with considerable variation between offence conflicts that rightfully belong to victims and offenders, types and contexts (Braithwaite 2002; Latimer et al. 2005; and that these parties ought to participate in its resolution. Strang et al. 2005). Clear conclusions are often hampered Criminal justice institutions and law professionals steal by the problem of self-selection, when offenders and vic- those conflicts from the affected parties and rob them of tims are given the choice to participate in restorative justice their opportunity, their right and duty, to learn and grow programs in addition, or as an alternative, to a court process through their conflicts. Notably, Christie (1977, p. 8) (Latimer et al. 2005). Methodological rigor would demand contends that the ‘‘big loser is us – to the extent that society a random allocation to experimental conditions (e.g., is us. This loss is first and foremost a loss in opportunities restorative justice versus court process) and to analyze the for norm-clarification’’. This understanding of offenses as participants’ data accordingly, irrespective of whether they conflicts is a prevailing motif in restorative justice. It is actually completed the allocated treatment (Sherman et al. reflected in what has become a frequently cited definition 2005; Strang et al. 2005). For our discussion, however, of restorative justice as ‘‘a process whereby all the parties self-selection is not so much a problem but rather a ques- with a stake in a particular offence come together to tion of interest. Why is it that some victims and offenders resolve collectively how to deal with the aftermath of the are willing to participate in restorative justice programs offence and its implications for the future’’ (Marshall 1999, while others are not? And once they have experienced the p. 5). In line with this definition, prototypical restorative restorative justice alternative, why do some fail to com- justice practices involve assembling the parties affected by plete the program? Why are some victims and offenders the offense, face to face, to discuss and resolve the injus- more satisfied than others? tice. For example, this can take the form of victim–offender Generally, the evidence suggests that victims feel more mediation, a mediated discussion between victim and satisfied with restorative justice programs compared to offender in a safe and structured setting; family confer- court processes (Latimer et al. 2005; Sherman et al. 2005; ences that also involve family, peers, or supporters of Strang 2002; Strang et al. 2005). The high level of satis- victim and offender; or circle sentencing, where other faction seems to be a result of victims’ greater level of members of the community furthermore represent wider participation in the process rather than their being satisfied interests (Bazemore and Umbreit 2001). with the reparation (Beven et al. 2005). Victims seem to Beyond this procedural element ‘‘of bringing together place less importance on material restoration than ‘‘emo- all stakeholders in an undominated dialogue about the tional restoration’’ (Strang 2002). However, there is also consequences of an injustice’’ (Braithwaite 2002, p. 12), evidence showing a consistent minority of victims who feel restorative justice has also a distinct value frame that, dissatisfied with the restorative justice process (Strang foremost, places emphasis on healing rather than punishing 2002). Similarly, victims’ willingness to take part in (Braithwaite 2002; Braithwaite and Strang 2001). While in restorative justice varied between 36 and 92% in the dif- practice it can also involve punishment of the offender, ferent conference programs reported by Strang et al. restorative justice promotes more constructive punishments (2006). So far, there is little known about the factors that as alternatives to the mere infliction of suffering on the predispose victims to restorative justice processes. When offender such as incarceration. Restorative punishments are are victims favorably oriented toward dealing with an more constructive and meaningful in that they oblige the injustice by restorative means? Similar questions could be offender to do something for the victim (or, alternatively, asked for the perspectives of offenders and community to victims of similar offenses), provide some service to the stakeholders (see Braithwaite 2002; Roberts and Stalans 123 378 Law Hum Behav (2008) 32:375–389 2004); however, our focus will largely be on the victim’s individual harms society by violating its rules in some perspective. normatively unallowable way, the scales of justice are out Although we have thus far traced the notion of restorative of balance, and sanction against the individual restores this justice only in the criminal justice arena, it has in fact found balance....[T]he perpetrator deserves to be punished in applicability and appeal in a variety of transgression con- proportion to the past harm he or she committed’’. This is texts, from bullying at school to corporate regulation, post- the notion that mainly underlies the conventional Western conflict intra-national reconciliation, and international criminal justice systems and has been the focus of most relations (Braithwaite 2002; Roche 2006). We will therefore psychological research on justice after transgressions address our research question at a more general level, where (Darley 2002; Feather 1999; Tyler et al. 1997; Vidmar all forms of transgressions—whether criminal or interper- 2000). It is important to understand that, for retributive sonal—can be responded to in restorative or punitive ways. justice, the punishment per se, or the suffering and Our theoretical discussion is not limited to current restor- humiliation it implies for the offender, restores the justice. ative practices but abstracts from them. Nor are we only The offender, having violated accepted rules and disturbed concerned with those criminal transgressions that have the moral balance, deserves to be punished. The punish- typically been considered for restorative programs, but we ment is considered necessary and sufficient for the accept that the type and meaning of a transgression could be restitution of justice. The punishment can be imposed a relevant factor. We will conceptualize restorative justice as unilaterally. The offender does not have to agree to it nor a psychological or lay-philosophical understanding of jus- show contrition or remorse about her actions. Indeed, in his tice and investigate when it, rather than retributive justice, philosophical-normative account of just desert, Von Hirsch motivates responses to rule-breaking. (1993) argues that punishment must not seek to elicit such internal states in the actor. While empirically there is some evidence that an offender’s remorse can lead to more Two Psychological Conceptions of Justice lenient punishment (see Vidmar 2000), conceptually remorse is not essential for retributive justice. People have a variety of goals that motivate responses to In contrast, restorative justice represents a more con- rule-breaking (e.g., De Keijser et al. 2002). Generally, it structive form of justice restoration (Walgrave 1995). As can be argued that any response to rule-breaking can have discussed, restorative justice understands an offense as a two objectives (Vidmar and Miller, 1980): (1) behavior conflict between victim, offender, and community that control, in the sense of securing future compliance with the needs to be resolved in interaction between those parties. In rules; and (2) justice restoration—the term we prefer over practice, both victim and offender are therefore given a Vidmar and Miller’s (1980) ‘‘retribution’’ that is rather voice to express their views and emotions. Based on a one-sided in light of the present paper. proper understanding of what the extent of the harm is, Both objectives can be pursued by different means, restorative justice is geared toward making the offender referred to here as punitive and constructive means. For take responsibility and accept accountability for his behavior control, deterrence (specific deterrence of the actions, and express a sincere apology to the victim. The offender, or general deterrence of the wider public) and victim is (at least implicitly) encouraged to express will- incapacitation (confining the offender) are clearly punitive ingness to forgive the offender and show respect to the means; offenders are unilaterally penalized or deprived of offender as a human being generally capable of redemption liberties. In contrast, rehabilitation (training and re-educa- and moral transformation (Govier 2002). Further, both tion of the offender) means that, while offenders are parties then decide together, bilaterally, on what is a suit- expected to take steps for their better adjustment to society, able punishment for the offender and/or a suitable they are provided with assistance and support to learn new compensation for the victim. Consistent with this, we argue skills, expand their behavioral repertoire, and change atti- the main issue in restorative justice, as a constructive form tudes. Rehabilitation can thus be characterized as a more of justice restoration after rule-breaking, is gaining a shared constructive means toward behavior control, where both understanding of the harm the offense has done and the parties (offenders and reactors) work together. values it violated. Justice is restored when the relevant Aside from behavior control, justice restoration aims at principles and values that have been violated by the offense re-establishing some moral order and sense of justice. are re-established and re-validated through social consen- Again, however, we contend this can take a punitive or a sus (see Sherman et al. 2005). constructive form. First, justice could be re-established by It should be emphasized again that we have now redefined the mere unilateral meting out of punishment. This is retributive and restorative justice as two psychological, or retributive justice or the notion of just desert, succinctly lay-philosophical, notions of justice. We contend that, captured by Carlsmith et al. (2002, p. 284): ‘‘[W]hen an depending on the circumstances, people may have different 123 Law Hum Behav (2008) 32:375–389 379 notions of justice—different views about the essence of victim for the loss of not only the material but also perhaps justice after rule-breaking and therefore how it ought to be the emotional value of those goods. The stolen jewelry may restored (Wenzel and Thielmann 2006). Retributive justice have had particular emotional value to the victim, and if the is restoration of a sense of justice through the imposition of thief has already sold the stolen item, additional compen- punishment, in form of adjudication or revenge; restorative sation may be required to replace that emotional loss. In justice is restoration of a sense of justice through renewed these cases, the victim experienced some deprivation, loss, value consensus. or harm that cannot be undone but that may be balanced out by some equivalent gain. That is, the harm is somehow valued (usually in monetary terms) and distributive justice Restitution, Compensation, and Censure or equity is reinstated in kind, so to speak. However, while transgressions can be violations of In order to further clarify our conceptualization of retrib- distributive and procedural norms, people may find the utive and restorative justice, it is worthwhile investigating mere re-establishment of distributive and procedural justice what exactly it means when people seek justice after rule- insufficient to restore justice. This is particularly so when breaking. This is because, occasionally, we find the term intention, or gross negligence, and blame are attributed to restorative justice used in a sense different from ours, the transgressor (Darley and Pittman 2003). When the namely as the restoration of an earlier state of affairs: offender is considered to have knowingly and intentionally restitution, the undoing of a harm, or compensation (van violated a rule, norm, or principle, without any other jus- Prooijen 2004; Schroeder et al. 2003). Indeed, these are tification, blame is attributed to the offender (Shaver 1985), also elements of restorative justice practices that are often further contributing to the experience of injustice beyond cited as distinguishing it from the punitive, traditional the violation of distributive or procedural entitlements criminal justice model. However, as discussed, it is not that (Mikula 2003). Mere restitution may not be adequate even punishment cannot be part of restorative justice practices as a response to negligent acts that result in a distributive or (unless perhaps in some abolitionist understanding of it; procedural injustice (Okimoto and Tyler 2007). Hence, the see Duff 2001). We want to emphasize that restorative restitution of a pre-offense state of affairs may not be justice is not to be confused with compensatory justice regarded as sufficient. Rather, the rule-violation is itself an (Darley and Pittman 2003). injustice that needs to be addressed, beyond any distribu- Transgressions against rules, norms, or principles usu- tive and procedural injustices implied (Tyler and Smith ally (except perhaps in victimless offenses, such as self- 1998). People usually feel justice demands that the offen- mutilation) involve a violation of entitlements that are der be punished, above and beyond a possible restitution or implied in the rule, norm, or principle. These can include compensation to the victim (Darley and Pittman 2003). As entitlements to certain outcomes, or distributive entitle- Heider (1958, p. 267) put it: ments in the widest sense (see Lerner 1991; Wenzel 2004); What is necessary is that the deeper sources of [the or they can be procedural entitlements to fair decision- offender’s] actions, the sources that impart the full making procedures and fair treatment (see Cropanzano and meaning to the harm and that most typically have Ambrose 2001; Lind and Tyler 1988; Sunshine and Heuer reference to the way [the offender] looks upon [the 2002). For example, a thief may have attempted to steal victim], should be changed. jewelry that legitimately belongs to the victim (according to accepted property rights), thus disregarding and violat- Generally, we can say what is deemed necessary is an ing the victim’s entitlement to own and dispose of property undoing of the moral-symbolic meanings of the offense as he wishes. Or, an employer may have used discrimina- (we will discuss two possible meanings in the next section). tory hiring practices that violated an applicant’s entitlement In the widest sense, these are undone through censure. In to equal opportunity or bias-free consideration of her job declaring the rule violation as wrongful, as something that application. One possible response to these transgressions should not have happened, censure in effect annuls the is to demand that the harm to the victim be undone and the moral-symbolic meaning of the offense. victim’s entitlements be reinstated. In other words, resti- Both retributive and restorative justice entail censure, tution is required that reestablishes distributive or but with quite different understandings of its operation. In procedural justice. The jewelry may be returned to the retributive justice, censure takes the form of punishment victim of theft; the candidate’s job application may be that is unilaterally imposed on the offender. In contrast, reconsidered and the employer may be asked to revise her restorative justice involves self-censure of the offender, recruitment procedures. who accepts the harm done, takes responsibility, and In some cases, an appropriate response to the trans- expresses remorse. More precisely, whereas in retributive gression may demand that the offender compensate the justice censure is a one-sided affair, in restorative justice 123 380 Law Hum Behav (2008) 32:375–389 censure is a collective effort shared between victim, Namely, transgressions can be regarded as threats to status/ offender, and community. In retributive justice, moral power on the one hand, and threats to shared values on the meaning is restored through assertion against the offender; other hand (see Tyler et al. 1997; Vidmar 2000). in restorative justice, it is restored through consensus with First, it may be argued that offenders, through violating the offender. The latter reflects what Hudson (1998) called, a rule or law, take advantage of their victims and the wider with reference to Habermas’ discourse ethics, a ‘‘dialogic’’ community; they put themselves above others and assume morality in restorative justice. It reflects a communitarian a position of superiority and power. Offenders disrespect view of restorative justice that morality is essentially a victims and their rights, and they disrespect the community social product of a shared community and interdependence and its laws. They express low regard for victims and (Braithwaite 1989). It may also reflect postmodern posi- communities who therefore feel humiliated; they exploit tions of restorative justice that reject the idea of universal victims and communities who therefore feel disempow- and objective truths, instead assuming that morality and ered. Hence, one major symbolic meaning of transgressions community need to be open to different views and be is the implied status/power relation. As Murphy and continuously negotiated anew (Pavlich 2001). Hampton put it (1988, p. 25): Our discussion focuses on transgressions where the One reason we so deeply resent moral injuries done intent and responsibility of the offender is clear. In this to us is not simply that they hurt us in some tangible case, restitution or compensation are generally not suffi- or sensible way; it is because such injuries are also cient to restore justice after transgressions; rather, a messages – symbolic communications. [...] Inten- response is required that undoes the moral-symbolic tional wrongdoing insults us and attempts (sometimes meaning of the rule violation itself. However, this is not to successfully) to degrade us – and thus it involves a mean that restitution and compensation are dispensable kind of injury that is not merely tangible and sensible. here. Failure to repair distributive or procedural injustices (if they are repairable) could be considered to perpetuate In psychology, Heider (1958) similarly emphasized the the transgression, and this might annul or discredit any symbolic implications of a transgression in terms of power form of censure. Moreover, restitution and compensation and status. As a response to harm or injury, in particular might, under certain conditions, have additional symbolic when they regard the implied power/status relation as meaning beyond the material restoration of distributive and illegitimate and undeserved, victims seek revenge to procedural justice. Indeed, Okimoto (2007) demonstrated restore their honor and their self-image (see also Bies and that compensation efforts by representatives of the group to Tripp 1996; Vidmar 2000). More recently, Miller (2001) which the offender belonged reaffirmed the victim’s argued that feelings of disrespect were often at the heart of membership value and identification with the group. Where experiences of injustice, and retaliation was the victims’ compensation is expressly imposed against the offender’s attempt to reassert themselves in order to restore their self- will it might have punitive meaning, and where it is agreed image and identity. Victims’ anger as a response to their on bilaterally it might have restorative meaning, in each demeaning treatment also ‘‘serves a self-presentational case carrying communications of censure. function... [by establishing an] identity as a strong and determined person who demands respect and does not tolerate unjust treatment by others’’ (Miller 2001, p. 541). Symbolic Implications of Transgressions: Status/Power In short, transgressions symbolically (if not actually) imply and Values an offender’s usurpation of power and status, and the dis- empowerment and degradation of victim and community. As we have seen, the quality of the censure response differs Responses to the transgression may, in turn, be attempts to between retributive and restorative justice, and with it, the restore the power and status relation, through taking power/ implied constitution of morality as quasi-objective and status away from the offender and reasserting the power/ neutral versus subjective and social. We argue further that status of victim and community. censure in retributive versus restorative justice primarily Second, it can be argued transgressions are a violation of addresses different moral-symbolic implications of trans- values that underlie the rules or laws that have been bro- gressions. What exactly are the moral-symbolic implications ken. Specifically, transgressions are a violation of values of transgressions that require a response beyond the resto- that are expected to be shared between offender and victim ration of distributive and procedural fairness? In the due to their shared membership in a relevant community. philosophical, sociological, and psychological literature, we The deliberate violation of these values can imply that the often find two themes. These are not always clearly sepa- offender questions them, challenges their validity, or rated and, indeed, may be inextricably intertwined; undermines the consensus that lends validity to the values. however, at least analytically, they can be distinguished. In his classic text, Durkheim (1964) argued similarly that 123 Law Hum Behav (2008) 32:375–389 381 crime is contrary to what he called common conscience: upon themselves, the punishment might effectively be presumably shared sentiments that constitute and derive considered self-imposed and its status/power reduction from the social cohesion in a collective. Or, as Vidmar effect might be neutralized. Hence, if a transgression is (2000, p. 42) puts it: ‘‘An offense is a threat to community primarily viewed as the usurpation of power by the consensus about the correctness – that is, the moral nature offender and as a humiliation of victim and community, – of the rule and hence the values that bind social groups then the imposition of a punishment on the offender may be together’’. As a transgression can be regarded as a chal- a means of demonstrating and restoring the power/status of lenge to essential identity-defining values of a relevant victim and community. This argument is consistent with social community, responses to the transgression may aim empirical evidence suggesting that the humiliating nature to restore and reaffirm the validity of those values. of a provocation leads to greater retaliatory aggression While symbolic implications of transgressions in terms (Murugesan and Sears 2006). Likewise, norms of honor of status/power and value can be conceptually distin- have been found to explain cultural (and subcultural) dif- guished, they are not necessarily independent from each ferences in aggressive responses to insults (Nisbett and other. Specifically, when a transgression expresses disre- Cohen 1996), while personality differences in one’s pre- gard or contempt for the values held by the victim and occupation with one’s self-image (narcissism) help explain supposedly shared within the community, the transgression individual differences in aggression-proneness (Baumeister may also be interpreted as communicating low regard or et al. 2000). disrespect for the victim and community, inasmuch as these In contrast, restorative justice is the reestablishment of values define the identity of victim and community. Con- justice through a renewed value consensus, through the versely, when a transgression is perceived as lowering the reaffirmation of values violated by the transgression. victim’s status and/or power, this may also be seen as Obviously, this justice notion speaks directly to the value violating shared values of equality or respect; that is, val- implications of the offense. If a transgression is primarily ues that prescribe how much status and power people are seen as the violation, questioning, or undermining of values due. In fact, it can be noted that Tyler et al.’s (1997) that are presumed to be shared in the relevant community, relational model, which similarly emphasizes the impor- then the most effective undoing of the incident should be tance of the symbolic meaning of transgressions for the offenders’ re-endorsement of those values. The offen- people’s motivation to respond to an injustice, subsumes der should be brought back into a consensus about under the notion of ‘‘group value’’ both the values held by a community values. The renewed consensus provides social relevant group and the value (i.e., status, respect, reputa- validation of those values (Turner 1991). Indeed, the tion) of a group and/or the members within the group offender’s re-endorsement in the form of remorse and (Tyler et al. 1996; Tyler and Smith 1999). However, we apology could contribute more to the validation of those argue that status/power and value implications of a trans- values than the preaching of the righteous who ‘‘knew gression can be theoretically differentiated and that this is better all along’’ (Alexander and Staub 1956; Vidmar useful for the explanation and prediction of different 2000). Of course, this should be the case only, or in par- notions of justice invoked in responses to transgressions. ticular, when the apology is sincere and not coerced. Restorative justice, which engages the affected parties in a dialogue, a process of mutual respect and equal voice, Retributive and Restorative Justice Addressing encourages offenders to take responsibility for their actions Symbolic Implications and offer an apology, toward the reestablishment of a consensus between the parties. It should therefore be an Retributive justice is the reestablishment of justice through effective response to a transgression that is primarily seen unilateral imposition of punishment on the offender con- in terms of its value implications (see Sherman et al. 2005). sistent with what is believed the offender deserves. Such a Hence, in general, the interpretation of a transgression as response could be sufficient for the restoration of a status/ a threat to status and power should lead to a preference for power relation that a transgression has disturbed. A uni- the retributive justice notion, whereas its interpretation as a lateral punishment means that offenders have to endure threat to shared values should lead to a preference for suffering against their will. The imposition of punishment restorative justice. However, we grant the issue could be in the name of justice means that offenders are powerless more complicated. First, it is possible that punishment, as and the suffering means that their status is lowered. The part of retributive justice, could also serve to restore values. punishment can be unilateral; the offender need not be For instance, from a punishment-as-communication per- persuaded by the punishment and need not show any spective (Duff 2001), it has been argued that the imposition remorse or contrition. In fact, if offenders accepted their of punishment communicates to offenders (and the wider punishment as deserved, as something they have brought community) that the offense was wrong and violated 123 382 Law Hum Behav (2008) 32:375–389 important principles or values. As Kahan (1998, p. 615) are less likely to show forgiveness (Exline et al. 2004). puts it, ‘‘What a community chooses to punish and how And, whereas an apology and a request for forgiveness can severely tell us what (or whom) it values and how much.’’ be ambiguous in their meaning (are offenders sincere or are This corresponds to the Durkheimian social-functional they only mocking victims even further?), the meaning of view that punishment serves to re-assert the collective punishment and imposed suffering is unambiguous. Hence, conscience and cohesion in a group (Vidmar 2000). Yet, punishment might be considered a necessary response to a we contend that punishment per se is unlikely to be an transgression that is primarily viewed in terms of its status/ effective means for re-establishing value consensus. Except power implications. in the improbable event that the offender accepts the Further, it has been argued that because of the voice ‘‘lesson’’ of the punishment and becomes a ‘‘better’’ person restorative justice practices give to victims and the wider because of it, the social consensus the punishment may community (to express their hurt and to suggest ways of express is only suboptimal. Punishment amounts to an dealing with the incident), restorative justice entails an attempt to enforce community values which may only element of empowerment (Braithwaite 2003; Zehr 1985). increase the offender’s resolve to resist (Brehm and Brehm So, again it would appear that aspects of a restorative 1981). Considering the power and status implications of justice process are also capable of addressing status and punishment, offenders are unlikely to agree with those who power concerns. However, this empowerment through a humiliate or disempower them (Braithwaite 1989). Hence, restorative process really applies all around to victims, there remains at least one party that does not share in the communities, and offenders; it sets restorative justice (but consensus supposedly expressed by the punishment: the also other forms of mediation and dispute resolution out- offender. The uncertainty persists as to why people like the side the courts) apart from more traditional criminal justice offender do not appear to share in the community’s values, systems where the power lies mainly in the hand of a third potentially continuing the threat to the values and identity party or legal authority. This has less to do with an of the group. However, this uncertainty can be reduced if empowerment of victims and communities relative to the punishment is understood as a psychological exclusion offenders in order to reverse a power differential caused by of the offender from the community (e.g., as apostates; see the transgression. Vidmar 2002; see also the black sheep effect; Marques and Consistent with this discussion, we predict that the Paez 1994). If offenders are no longer regarded as members restorative justice notion is primarily salient when the of the community (symbolically, by withholding from them transgression is regarded as a violation of community rights members typically have, or physically, by locking values, whereas the retributive justice notion is primarily them away), their dissent no longer causes uncertainty or salient when the transgression is regarded as disempow- threat to the value consensus (see Turner 1991), although erment or humiliation of victim and community. Although the consensus has then a reduced range. Hence, in terms of some elements of restorative justice might also address the a reaffirmation of values toward the rest of the group, there status/power imbalance brought upon by the transgression, is some scope for retributive justice to address value and the mere punishment of retributive justice might also implications of a transgression. go some way to address an implied value threat, retributive Conversely, there could be other options than punish- justice more typically (and more effectively) restores a ment for restoring a status/power relation. For example, in status/power balance while restorative justice more typi- a restorative process, offenders could offer an apology and cally (and more effectively) restores a values consensus. ask for forgiveness. The offenders’ admission of wrong- doing signals that their appropriation of power was illegitimate and annuls it. Their concession that they owe the victims/community an apology amounts to an Retributive and Restorative Justice as a Function acknowledgment of rights and an expression of respect for of Social Identity them. As forgiveness can only be granted by victims (perhaps by the wider community on their behalf; see Concerns about status/power versus values may be a Govier 2002), the offenders’ request for forgiveness sub- reflection of a more fundamental issue, namely the con- jects them to the victims’ (or community’s) will and ception of the relationship between the affected parties in control. By granting forgiveness, victims (and/or the terms of self and identity. Specifically, we assume the community) can assert a moral superiority, and their endorsement of a retributive versus restorative justice ‘‘magnanimity emphasizes [the offender’s] inferiority’’ notion depends on whether or not offender and victim are (Heider, 1958, p. 269). Having said this, empirical evi- regarded as sharing membership in the relevant community dence suggests that narcissistic individuals who greatly or inclusive group (Wenzel and Thielmann 2006). The value self-respect and have a strong sense of entitlement argument is, firstly, based on a more detailed analysis of 123 Law Hum Behav (2008) 32:375–389 383 the two underlying psychological motivations of status/ the endorsement of a ‘‘just desert notion of justice’’ power versus value restoration. (retributive justice) was a better predictor of sanctioning Status and power are inherently relative or competitive. decisions when respondents indicated a relatively low level A person/group has high status in relation to another, of identification in terms of the relevant social category lower-status party; and it has power insofar as it has power that includes the offender and themselves (national iden- over somebody else. When a transgression is interpreted as tity). In contrast, the endorsement of a ‘‘value reaffirmation the offender’s illegitimate appropriation of status and notion of justice’’ (restorative justice) predicted sanction- power, victims or communities have framed their rela- ing decisions only when respondents indicated a strong tionship to the offender essentially as an antagonistic, level of identification with the relevant inclusive category. negatively interdependent one (see Sherif 1966). If, These findings support our reasoning: restorative justice through their actions, the offender is perceived to have seemed more relevant to participants’ responses when they gained status or power over the victim, the offender’s gain identified strongly with the category that included the in status/power is the victim’s loss. This implies that vic- offender, whereas retributive justice was psychologically tims distinguish psychologically between themselves and less relevant under these conditions. the offender, as between self and non-self (Turner 1981). If Hence, we argue a retributive notion of justice is more they were to regard themselves and offenders as members likely salient when respondent and offender lack a common of the same group, a competition over status and power identity, whereas a restorative notion of justice is more would be less likely, as they would focus more on what likely salient when they share a common identity. This is they have in common than on what separates them. Indeed, mainly so because when the affected parties lack a shared as members of the same group they could even have power identity (i.e., they see each other as different individuals through powerful other in-group members (Turner 2005). and construe the context as interpersonal, or they see each Conversely, a transgression is more likely to be interpreted other as members of different groups and construe the as a threat to one’s status or power when one conceptual- context as intergroup) they are more likely to perceive each izes oneself as different from the offender. Assuming that other as negatively interdependent in terms of status and retributive justice is primarily motivated by status/power power, and react predominantly to the status/power concerns, this justice notion should therefore be more implications of a transgression. In contrast, when they salient when the involved parties are perceived to have share a social identity (i.e., they construe the context as different identities. intragroup) they are more likely to perceive each other as In contrast, values are valid insofar as they are shared positively interdependent in terms of the validation of within a relevant social category (Haslam et al. 1996; shared values, and they react more out of a concern for Turner 1987). A transgression should only be interpreted as values and the restoration of identity. a violation of shared values when one expects the offender The importance of social identity can be restated in two to hold and live by the same values. If an offense is further, though related, ways. First, at the beginning of this interpreted as inconsistent with one’s values, then this article, we referred to a common understanding of the should only lead to uncertainty and, thus, the motivation to restorative justice model as aiming at the healing of com- seek consensus, when one expects the offender to agree on munities and mending of social relationships. This those same values (McGarty et al. 1993). Put differently, immediately raises the question of what this community is one should only want to seek consensus and agreement and what those relationships are. Do victim and offender with the offender when consensus and agreement can see themselves as part of the same community, and do they indeed reaffirm the validity of the values. If there is no see each other as engaged in some social relationship expectation to agree on the same values, then agreement worthy to maintain? Does the offender want to be part of does little to validate one’s values. In line with self-cate- the wider community as the victim sees it? Does the victim gorization theory (Turner 1987), the expectation to agree is want the offender to be part of their community? The way based on the view that one shares a relevant social identity. the two parties define their social identity in relation to Likewise, the motivation to reduce disagreement and each other should provide answers to these questions, and it achieve consensus is based on a sense of shared identity. should indicate whether the parties can and want to affirm Assuming that restorative justice is primarily motivated by their identities through rebuilding a consensus about value concerns, this justice notion should therefore be more identity-defining values. Second, to see their values rec- salient when the parties share a relevant social identity. ognized, victims might want the offenders to accept Research by Wenzel and Thielmann (2006) provides responsibility and acknowledge the harm and wrong done. initial support for these predictions. In two correlational When victims do not think they share a group identity with studies involving offenses with non-specific victims (tax the offender, they might be more inclined to resort to a evasion and social security fraud), there was evidence that coercive process of trying to impose their views on the 123 384 Law Hum Behav (2008) 32:375–389 offender by means of punishment and intimidation, However, it is generally acknowledged that people can believing that only this process will have an effect on identify with groups at different levels of inclusiveness outgroup members (Haslam 2004; Reynolds and Platow (Gaertner and Dovidio 2000; Turner 1987). Thus, while 2003). In contrast, victims who perceive to share a relevant victim and offender may identify as members of different social identity with offenders are more likely to think they groups at one level (e.g., in terms of their ethnic back- are open to reason and argument, to processes of persua- ground), they may also identify with a more abstract group sion and influence, which seem essential to the dialogical to which they both belong (e.g., their nation). This means morality of restorative justice. the fact that, in intergroup contexts, offenders and victims We acknowledge that the role of social identity could be regard themselves as members of different groups does not more complicated due to the fact again that status/power preclude the possibility that they also see each other as relations and group values are not completely independent sharing membership in another, more abstract group. from each other. Concerns about status/power could also Research has shown that when members of different groups occur in intragroup contexts (where parties refer to a identify with a higher-order, inclusive group, their notions common identity) when a member’s status or standing is of entitlement are based on the perceived values of that defined with reference to consensual, identity-defining shared group (rather than on sub-group interests; Wenzel value dimensions (e.g., intergroup attraction: Hogg 1993; 2002), and acceptance and legitimacy of authority deci- leadership: Hogg and van Knippenberg 2003; Turner and sions are based on relational concerns about fairness and Haslam 2001; distributive justice; Wenzel 2004). Likewise, respect (rather than on the favorability of the decisions for value concerns might also be possible in interpersonal or their sub-group; Huo 2003; Huo et al. 1996). Similarly, the intergroup contexts (where the parties do not refer to a intergroup quality of a transgression does not rule out that common identity) when individuals or groups derive a the affected parties adopt a restorative notion of justice. sense of pride or status from their identity-defining values, Indeed, Wohl and Branscombe (2005) showed that mem- specifically when these are favorably compared to others’ bers were more willing to forgive an outgroup that values. These interdependencies between status/power and committed atrocities against their own group when they value processes are interesting and challenging, but they do regarded both groups as part of a more inclusive group. not diminish our main argument that a sense of a common While restorative justice is thus a possibility even in identity underlies the concern for presumably shared values intergroup contexts due to people’s capacity to identify and thus a restorative justice notion, whereas status/power simultaneously with a more abstract social category, there concerns come to the fore when the affected parties lack a may still be arguments that responses could here be less sense of common identity. restorative and more punitive compared to interpersonal or intragroup contexts. First, while people can identify simulta- neously with groups at different levels of inclusiveness (i.e., Retributive and Restorative Justice in Intergroup two differently inclusive groups can both be central to one’s Contexts self), there is a certain ‘‘functional antagonism’’ between two differently inclusive self-categorizations as to their psycho- The preceding discussion suggests a rather pessimistic logical salience in a given situation (Turner 1987). An view for transgressions in intergroup contexts (if we con- attribute that defines the difference between two groups in a sider a restorative justice option a good thing), but this is given context is less likely to be seen as a similarity that worth analyzing in more detail. In intergroup contexts, defines their common identity in terms of a more abstract where the offender is regarded as a member of a salient group, and vice versa—except when the two groups are con- outgroup, it would seem as if, by definition, the affected sidered similar at a higher level because of their differences, parties lack a common identity. Hence, they would be more such as their complementarities (Haslam 2004). For example, likely to endorse retributive justice, revenge, and punish- when victims define their ingroup as holding relevant values ment (rather than restorative justice). Indeed, there is different from the offender group, this is incompatible with the evidence that members tend to endorse stronger punish- view that they, as members of the inclusive category, sub- ment toward outgroup offenders than ingroup offenders scribe to the same values. Hence, value restoration would be (Graham et al. 1997; Kerr et al. 1995; Sommers and Ells- less of a concern and restorative justice less likely endorsed as worth, 2000; but see Feather 1999; Vidmar 2002). This is a justice notion. This would be the case unless victim and consistent with the argument in the literature that aggres- offender groups are defined as holding different values but sive conflicts tend to be more severe or have greater both subscribe, for example, to tolerance as the value defining escalation potential in intergroup than interpersonal con- their more inclusive, common identity. As tolerance means texts (e.g., Meier and Hinsz 2004; Mummendey and Otten acceptance or appreciation of difference, there is no functional 1993). antagonism between self-categorizations at both levels 123 Law Hum Behav (2008) 32:375–389 385 (rather, there needs to be subgroup difference for tolerance to rule-breaking. Whereas most previous research has focused have any meaning). Restorative justice could here still be on a retributive, or just desert, notion of justice and peo- endorsed, with the aim of restoring the shared value of toler- ple’s desire to punish offenders in order to restore justice, ance toward value differences. our approach takes up the challenge that the recently Second, when the parties affected by a transgression advancing philosophy and practices of restorative justice identify simultaneously with their sublevel ingroup and a mount to such a narrow view. While a variety of practices more abstract social category inclusive of the offender tend to be subsumed under the banner of restorative justice, group, their subgroup’s consensus might imbue the violated our psychological conceptualization identifies as its defin- value with sufficient validity for members to claim that their ing element a striving for consensus (on harm, ingroup’s values are representative and normative of the responsibility, and values) as necessary to restore justice. inclusive group generally (ingroup projection; Mummendey The seeking of consensus implies a dialogical morality, and Wenzel 1999; Wenzel et al. 2003). In a way, the ingroup where all the affected parties have a voice in an open does not consider itself as positively interdependent with the dialogue that is geared toward reaffirming what are con- outgroup for the validation of shared values. Rather, the sidered shared and identity-defining values of their ingroup considers its values as valid for the inclusive cate- community. gory and, thus, the outgroup as deviant or subversive. The community referred to in restorative justice is Indeed, due to its deviance the offender group might be understood in the widest possible sense and can also be a considered to deserve greater punishment than if there were ‘‘community’’ of two individuals as, for example, in a close no sense of shared inclusive identity. Consistent with this relationship. It thus refers to a psychological group or argument, Wenzel and Thielmann (2006) found that a value category that the victim and offender perceive themselves restoration notion of justice was positively related to calls to be members of in a given social context. In our view, the for punishment only when the norms and values of the salience and definition of a shared group identity are cen- inclusive category were considered clear and consensual. In tral to people’s endorsement of a restorative justice notion. contrast, the value restoration notion of justice was posi- To the extent that the affected parties believe they share an tively related to alternative, more restorative forms of identity in terms of a common group, they will expect to punishment when the norms and values of the inclusive share the same values. When these values are violated by a category were considered to be diverse, in flux, and in transgression, victims will be motivated to reaffirm the constant need of renewed consensus. values through social consensus and, correspondingly, To sum up, intergroup transgressions involve parties endorse a restorative justice notion. Conversely, when the with different salient identities who thus tend to see each parties lack shared identification with a relevant inclusive other as negatively interdependent with regard to their group, they will more likely see each other as negatively status and power; intergroup transgressions will thus more interdependent in terms of status and power. Victims will likely elicit notions of retributive justice. Restorative jus- tend to interpret a transgression as an illegitimate appro- tice is a possible response to the extent that the affected priation of status and power and will seek to undo their parties simultaneously define themselves in terms of a humiliation and disempowerment through the imposition superordinate category that includes them all. However, of punishment on the offender and subscribe to retributive due to functional antagonism between self-categorizations justice. at different levels of inclusiveness, a sense of a relevant This is the basic argument we have advanced in this shared identity may (under certain circumstances) be article, together with a discussion of some theoretical reduced as long as subgroup identities remain salient. complexities surrounding it. We want to conclude by Moreover, when the affected parties simultaneously iden- pointing to some omissions and implications. First, the tify at subgroup and inclusive levels, group members may analysis put forward in this article focused on rather cog- project their subgroup’s values onto the inclusive category nitive appraisals of the meaning of transgressions. and regard the offender group as deviant with respect to However, it is clear that injustice is an affective experi- those values. Even when the involved groups refer to a ence, involving a variety of emotions (e.g., see Feather shared inclusive identity, intergroup transgressions can 2006; Kristjánsson 2005; Mikula et al. 1998; Weiner thus lead to more punitive responses. 2006). These different emotions could be differentially related to, or predictive of, the endorsement of retributive and restorative justice notions, but such an analysis is Conclusions beyond the scope of this article. Second, we have considered retributive and restorative In this article, we have outlined a social-psychological justice foremost in terms of their subjective appropriate- approach to lay people’s notions of justice after ness or morality. However, victims and observers may also 123 386 Law Hum Behav (2008) 32:375–389 have expectations as to whether a certain form of justice favor retributive justice and punishment (e.g., see Carroll can be realized or is likely to be effective in a given situ- et al. 1987; Feather 1996; Gollwitzer 2004). ation, and such expectations could also affect the adoption More specifically, right-wing authoritarianism (Alte- of a certain justice notion. For example, when they see (or meyer 1988) can be interpreted as a cluster of conservative anticipate) that an offender does not seriously engage in a values (e.g., conformity, tradition, security; Feather 1996) restorative process and does not intend to accept respon- which, in part, define who has or should have status and sibility for the offence, they may endorse a more retributive power in society. When a transgression violates and notion instead (Gromet and Darley 2006). Indeed, victims apparently questions those prescribed status and power and observers (as it has been suggested for regulators; relations (e.g., legitimate authority), retributive justice is Ayres and Braithwaite 1992) may generally want to start likely to be more strongly endorsed (see Feather 1998, with a cooperative restorative approach, but when this fails 2002; Vidmar 2000). Indeed, individual value preferences turn to a more retributive one. are another variable likely to affect how people evaluate Third, our theoretical analysis focused on the prediction and respond to transgressions and victimization (Feather of rather abstract, lay-philosophical notions of justice after 1996, 1999). Consistent with our approach, Braithwaite rule-breaking. From these we need to distinguish concrete (2000) found in the context of school bullying that parents’ actions that people can take, such as punishment, revenge, endorsement of security values (e.g., protection of interests apology and forgiveness, which, although typically fea- and status; V. Braithwaite 1998) was positively related to turing to a different degree in retributive and restorative their support for punitive strategies for dealing with bul- practices, may in fact have multiple functions and mean- lying, whereas the endorsement of harmony values (e.g., ings. For example, we briefly discussed the possibility that concern for equality, mutual respect, peace; V. Braithwaite, punishment can help restore values as well as status/power; 1998) was positively related to their support for dialogic that an apology and forgiveness can communicate status/ strategies for dealing with bullying. power in addition to contributing to a value consensus. Finally, although we deliberately developed a theory Future research should specifically focus on the shades of that abstracts from particular contexts, we must not lose meaning of these concrete acts (e.g., Eaton et al. 2006; sight of its implications specifically for the legal system in Frantz and Bennigson 2005). which restorative justice originally emerged and continues Fourth, we do not conceptualize the endorsement of to grow as an alternative paradigm. We need to consider retributive and restorative justice as individual difference how our approach can inform the design of a fair, satis- variables that are stable across incidents and situations, as factory, and effective legal system. Above all, we must test punishment goals are usually conceptualized and measured, our theoretical predictions in the legal field and develop for instance (e.g., De Keijser et al. 2002; but see Orth 2003). empirical approaches that are effective despite obvious Instead, our approach emphasizes that the endorsement of a practical constraints. However, as indicated earlier, what certain justice notion depends on situational factors, on sit- has been a constraint in research testing the relative uationally relevant and salient self-definitions and effectiveness of restorative justice, namely the self-selec- corresponding interpretations of the specific incident. tion of participants (Latimer et al. 2005), can be considered However, individual differences could of course affect the a research opportunity in the realm of our theory. Given the meaning we give to the situation and to ourselves. For option, when and why are people willing to participate in example, individuals with greater power motivation restorative programs, and when and why are they not? (McClelland 1987), when victimized, could be particularly Does the psychological meaning they attribute to the concerned about losing power or being seen as powerless transgression, and/or their perceived identity relation with and weak. They might be more likely to endorse retributive the other party, affect that decision? Next to survey justice and attempt to reassert their power by punishing and methodologies, more powerful experimental designs could diminishing the offender. Likewise, individuals with a nar- here, for example, involve an intervention directed at cissistic personality (Raskin and Terry 1988), characterized eliciting a sense of shared identity prior to participants’ by a sense of superiority and inflated self-esteem, might be decision to take the restorative option or not. Similarly, it more likely to interpret a transgression against them as a could be investigated whether the theoretically proposed personal insult and choose retribution to reassert their factors and processes affect perceptions of justice and positive sense of self (Baumeister et al. 2000; Bushman and satisfaction of those who actually participated in restorative Baumeister 1998, 2002). Similarly, an authoritarian per- justice; parallel studies could be conducted in the con- sonality has been described as ‘‘an excessive concern with ventional court system. power and status’’ (Myers 2005, p. 345) and, as a conse- To conclude, the present framework promises to provide quence, could predispose people to interpret an offence for a fruitful research agenda. It suggests a social-psycho- primarily in terms of its status and power implications and to logical conceptualization of restorative justice and, thus, 123 Law Hum Behav (2008) 32:375–389 387 helps to open up this largely neglected justice notion to Braithwaite, V. (2000). Values and restorative justice in schools. In H. psychological inquiry. Eventually, it should lead to insights Strang & J. Braithwaite (Eds.), Restorative justice: Philosophy to practice (pp. 121–144). Aldershot: Ashgate. about the potential and limits of restorative justice, about Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A the conditions under which restorative justice is considered theory of freedom and control. New York: Academic Press. appropriate, or which need to be established first before it Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, is considered appropriate. Such insights might help to narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of promote ideas and practices of restorative justice in crim- Personality and Social Psychology, 75, 219–229. inal justice as well as social regulation, individual-level Bushman, B. J., & Baumeister, R. F. (2002). Does self-love or self- counselling, and group-level peace-keeping. hate lead to violence? Journal of Research in Personality, 36, 543–545. Acknowledgment This research was supported by a grant from the Carlsmith, K. M., Darley, J. M., & Robinson, P. H. (2002). Why do Australian Research Council (DP0557634). we punish? Deterrence and just deserts as motives for punish- ment. Journal of Personality & Social Psychology, 83, 284–299. Carroll, J. S., Perkowitz, W. T., Lurigio, A. J., & Weaver, F. M. (1987). Sentencing goals, causal attributions, ideology, and References personality. Journal of Personality and Social Psychology, 52, 107–118. Alexander, F., & Staub, H. (1956). The criminal, the judge, and the Christie, N. (1977). Conflicts as property. British Journal of public. Glencoe: Free Press. Criminology, 17, 1–15. Altemeyer, B. (1988). Enemies or freedom: Understanding right-wing Cropanzano, R., & Ambrose, M. L. (2001). Procedural and distrib- authoritarianism. San Francisco: Jossey-Bass. utive justice are more similar than you think: A monistic Ayres, I., & Braithwaite, J. (1992). Responsive regulation: Transcend- perspective and a research agenda. In J. Greenberg & R. ing the deregulation debate. New York: Oxford University Press. Cropanzano (Eds.), Advances in organizational justice (pp. 119– Baumeister, R. F., Bushman, B. J., & Campbell, W. (2000). Self- 151). Stanford: Stanford University Press. esteem, narcissism, and aggression: Does violence result from Darley, J. (2002). Just punishments: Research on retributional justice. low self-esteem or from threatened egotism? Current Directions In M. Ross & D. T. Miller (Eds.), The justice motive in everyday in Psychological Science, 9, 26–29. life (pp. 314–333). New York: Cambridge University Press. Bazemore, G. (1998). Restorative justice and earned redemption: Darley, J. M., Carlsmith, K. M., & Robinson, P. H. (2000). Communities, victims, and offender reintegration. American Incapacitation and just deserts as motives for punishment. Law Behavioral Scientist, 41, 768–813. and Human Behavior, 24, 659–683. Bazemore, G., & Umbreit, M. (2001). A comparison of four Darley, J. M., & Pittman, T. S. (2003). The psychology of restorative conferencing models. In U. S. Department of Justice compensatory and retributive justice. Personality and Social Office of Juvenile Justice and Delinquency Prevention (Ed.), Psychology Review, 7, 324–336. OJJDP Juvenile Justice Bulletin (Vol. February 2001). Wash- De Keijser, J. W., van der Leeden, R., & Jackson, J. L. (2002). From ington, DC: U.S. Department of Justice. moral theory to penal attitudes and back: A theoretically Beven, J. P., Hall, G., Froyland, I., Steels, B., & Goulding, D. (2005). integrated modeling approach. Behavioral Sciences & the Law, Restoration or renovation? Evaluating restorative justice out- 20(4), 317–335. comes. Psychiatry, Psychology and Law, 12, 194–206. Duff, R. A. (2001). Punishment, communication, and community. Bies, R. J., & Tripp, T. M. (1996). Beyond distrust: ‘‘getting even’’ Oxford: Oxford University Press. and the need for revenge. In R. M. Kramer & T. R. Tyler (Eds.), Durkheim, E. (1964). The division of labor in society (trans: Simpson, Trust in organizations: Frontiers of theory and research (pp. G.). Glencoe: Free Press. (Original work published 1902). 246–260). Thousand Oaks: Sage. Eaton, J., Struthers, C., & Santelli, A. G. (2006). The mediating role Braithwaite, J. (1989). Crime, shame, and reintegration. Cambridge: of perceptual validation in the repentance-forgiveness process. Cambridge University Press. Personality and Social Psychology Bulletin, 32, 1389–1401. Braithwaite, J. (1998). Restorative justice. In M. Tonry (Ed.), The Exline, J. J., Baumeister, R. F., Bushman, B. J., Campbell, W., & handbook of crime and punishment (pp. 323–344). Oxford: Finkel, E. J. (2004). Too proud to let go: Narcissistic entitlement Oxford University Press. as a barrier to forgiveness. Journal of Personality and Social Braithwaite, J. (1999). Restorative justice: Assessing optimistic and Psychology, 87, 894–912. pessimistic accounts. In M. Tonry (Ed.), Crime and justice: A Feather, N. T. (1996). Reactions to penalties for an offense in relation review of research (Vol. 25, pp. 1–127). Chicago: University of to authoritarianism, values, perceived responsibility, perceived Chicago Press. seriousness, and deservingness. Journal of Personality and Braithwaite, J. (2002). Restorative justice and responsive regulation. Social Psychology, 71, 571–587. New York: Oxford University Press. Feather, N. T. (1998). Reactions to penalties for offenses committed Braithwaite, J. (2003). Principles of restorative justice. In A. von by the police and public citizens: Testing a social-cognitive Hirsch, J. V. Roberts, A. Bottoms, K. Roach & M. Schiff (Eds.), process model of retributive justice. Journal of Personality and Restorative justice and criminal justice: Competing or reconcil- Social Psychology, 75, 528–544. able paradigms? (pp. 1–20). Oxford: Hart. Feather, N. T. (1999). Values, achievement, and justice: Studies in the Braithwaite, J., & Strang, H. (2001). Introduction: Restorative justice psychology of deservingness. New York: Kluwer/Plenum. and civil society. In H. Strang & J. Braithwaite (Eds.), Feather, N. T. (2002). Reactions to supporters and opponents of Restorative justice and civil society (pp. 1–13). Cambridge: uranium mining in relation to status, attitude similarity, and Cambridge University Press. right-wing authoritarianism. Journal of Applied Social Psychol- Braithwaite, V. (1998). The value orientations underlying liberalism- ogy, 32, 1464–1487. conservatism. Personality and Individual Differences, 25, Feather, N. T. (2006). Deservingness and emotion: Applying the 575–589. structural model of deservingness to the analysis of affective 123 388 Law Hum Behav (2008) 32:375–389 reactions to outcomes. European Review of Social Psychology, Marques, J. M., & Paez, D. (1994). The ‘black sheep effect’: Social 17, 38–73. categorization, rejection of ingroup deviates, and perception of Frantz, C. M., & Bennigson, C. (2005). Better late than early: The group variability. European Review of Social Psychology, 5, influence of timing on apology effectiveness. Journal of 37–68. Experimental Social Psychology, 41, 201–207. Marshall, T. F. (1999). Restorative justice: An overview. A report by Gaertner, S. L., & Dovidio, J. F. (2000). Reducing intergroup bias: the Home Office Research Development and Statistics Direc- The common ingroup identity model. Philadelphia: Psychology torate. London: Home Office. Press. McClelland, D. C. (1987). Human motivation. New York: Cambridge Gollwitzer, M. (2004). Do normative transgressions affect punitive University Press. judgments? An empirical test of the psychoanalytic scapegoat McFatter, R. M. (1978). Sentencing strategies and justice: Effects of hypothesis. Personality and Social Psychology Bulletin, 30, punishment philosophy on sentencing decisions. Journal of 1650–1660. Personality and Social Psychology, 36, 1490–1500. Govier, T. (2002). Forgiveness and revenge. London: Routledge. McFatter, R. M. (1982). Purposes of punishment: Effects of utilities Graham, S., Weiner, B., & Zucker, G. S. (1997). An attributional of criminal sanctions on perceived appropriateness. Journal of analysis of punishment goals and public reactions to O. J. Applied Psychology, 67, 255–267. Simpson Personality and Social Psychology Bulletin, 23, McGarty, C., Turner, J. C., Oakes, P. J., & Haslam, M. A. (1993). The 331–346. creation of uncertainty in the influence process: The roles of Gromet, D. M., & Darley, J. (2006). Restoration and retribution: How stimulus information and disagreement with similar others. including retributive components affects the acceptability of European Journal of Social Psychology, 23, 17–38. restorative justice processes. Social Justice Research, 19, 395– Meier, B. P., & Hinsz, V. B. (2004). A comparison of human 432. aggression committed by groups and individuals: An interindi- Haslam, S. A. (2004). Psychology in organizations: The social vidual-intergroup discontinuity. Journal of Experimental Social identity approach (2nd ed.). London: Sage. Psychology, 40, 551–559. Haslam, S. A., McGarty, C., & Turner, J. C. (1996). Salient group Mikula, G. (2003). Testing an attribution-of-blame model. European memberships and persuasion: The role of social identity in the Journal of Social Psychology, 33, 793–811. validation of beliefs. In J. L. Nye & A. M. Brower (Eds.), What’s Mikula, G., Scherer, K., & Athenstaedt, U. (1998). The role of social about social cognition? Research on socially shared injustice in the elicitation of differential emotional reactions. cognition in small groups (pp. 29–56). Thousand Oaks: Sage. Personality and Social Psychology Bulletin, 24, 769–783. Heider, F. (1958). The psychology of interpersonal relations. New Miller, D. T. (2001). Disrespect and the experience of injustice. York: Wiley. Annual Review of Psychology, 52, 527–553. Hogan, R., & Emler, N. P. (1981). Retributive justice. In M. J. Lerner Mummendey, A., & Otten, S. (1993). Aggression: Interaction & S. C. Lerner (Eds.), The justice motive in social behavior between individuals and social groups. In R. B. Felson & J. T. (pp. 125–144). New York: Academic Press. Tedeschi (Eds.), Aggression and violence: Social interactionist Hogg, M. A. (1993). Group cohesiveness: A critical review and some perspectives (pp. 145–167). Washington, DC: American Psy- new directions. European Review of Social Psychology, 4, chological Association. 85–111. Mummendey, A., & Wenzel, M. (1999). Social discrimination and Hogg, M. A., & van Knippenberg, D. (2003). Social identity and tolerance in intergroup relations: Reactions to intergroup differ- leadership processes in groups. In M. P. Zanna (Ed.), Advances ence. Personality and Social Psychology Review, 3, 158–174. in experimental social psychology (Vol. 35, pp. 1–52). San Murphy, J. G., & Hampton, J. (1988). Forgiveness and mercy. Diego: Elsevier Academic Press. Cambridge: Cambridge University Press. Hudson, B. (1998). Restorative justice: The challenge of sexual and Murugesan, V., & Sears, D. (2006, January). The effect of humiliation racial violence. Journal of Law and Society, 25, 237–256. on acceptance for retaliatory aggression. Poster presented at the Huo, Y. J. (2003). Procedural justice and social regulation across 7th annual meeting of the Society for Personality and Social group boundaries: Does subgroup identity undermine relation- Psychology, Palm Springs, CA. ship-based governance? Personality and Social Psychology Myers, D. G. (2005). Social psychology (8th ed.). New York: Bulletin, 29, 336–348. McGraw-Hill. Huo, Y. J., Smith, H. J., Tyler, T. R., & Lind, E. A. (1996). Nisbet, R. E., & Cohen, D. (1996). Culture of honour: The psychology Superordiante identification, subgroup identification, and justice of violence in the South. Boulder: Westview. concerns: Is separatism the problem, is assimilation the answer? Okimoto, T. G. (2007). Outcomes as affirmation of membership Psychological Science, 7, 40–45. value: Monetary compensation as an administrative response to Kahan, D. M. (1998). Social meaning and the economic analysis of procedural injustice: Manuscript submitted for publication. crime. Journal of Legal Studies, 27, 609–622. Okimoto, T. G., & Tyler, T. R. (2007). Is compensation enough? Kerr, N. L., Hymes, R. W., Anderson, A. B., & Weathers, J. E. Relational concerns in responding to unintended harm. Group (1995). Defendant-juror similarity and mock juror judgments. Processes and Intergroup Relations, 10, 399–420. Law and Human Behavior, 19, 545–567. Orth, U. (2003). Punishment goals of crime victims. Law and Human Kristjánsson, K. (2005). Justice and desert-based emotions. Alder- Behavior, 27, 173–186. shot: Ashgate. Pavlich, G. (2001). The force of community. In H. Strang & Latimer, J., Dowden, C., & Muise, D. (2005). The effectiveness of J. Braithwaite (Eds.), Restorative justice and civil society restorative justice practices: A meta-analysis. Prison Journal, (pp. 56–68). Cambridge: Cambridge University Press. 85, 127–144. Peachey, D. E. (1989). The Kitchener experiment. In M. Wright & B. Lerner, M. J. (1991). Integrating societal and psychological rules of Galaway (Eds.), Mediation and criminal justice (pp. 14–26). entitlement. In R. Vermunt & H. Steensma (Eds.), Social justice London: Sage. in human relations (Vol. 1, pp. 13–32). New York: Plenum Raskin, R., & Terry, H. (1988). A principal-components analysis of Press. the narcissistic personality inventory and further evidence of its Lind, E. A., & Tyler, T. R. (1988). The social psychology of construct validity. Journal of Personality and Social Psychology, procedural justice. New York: Plenum Press. 54, 890–902. 123 Law Hum Behav (2008) 32:375–389 389 Retzinger, S. M., & Scheff, T. J. (1996). Strategy for community Tyler, T. R., Boeckmann, R. J., Smith, H. J., & Huo, Y. J. (1997). conferences: Emotions and social bonds. In B. Galaway & J. Social justice in a diverse society. Boulder: Westview. Hudson (Eds.), Restorative justice: International perspectives Tyler, T. R., Degoey, P., & Smith, H. (1996). Understanding why the (pp. 315–336). Monsey: Criminal Justice Press. justice of group procedures matters: A test of the psychological Reynolds, K. J., & Platow, M. J. (2003). On the social psychology of dynamics of the group-value model. Journal of Personality and power and powerlessness: Social power as a symptom of Social Psychology, 70, 913–930. organizational division. In S. A. Haslam, D. Van Knippenberg, Tyler, T. R., & Smith, H. J. (1998). Social justice and social M. Platow & N. Ellemers (Eds.), Social identity at work: movements. In D. G. Gilbert, S. T. Fiske & G. Lindzey (Eds.), Developing theory for organizational practice (pp. 173–188). The handbook of social psychology (4 ed., Vol. 2, pp. 595–629). New York: Psychology Press. Boston: McGraw-Hill. Roberts, J. V., & Stalans, L. J. (2004). Restorative sentencing: Tyler, T. R., & Smith, H. J. (1999). Justice, social identity, and group Exploring the views of the public. Social Justice Research, 17, processes. In T. R. Tyler, R. M. Kramer & O. P. John (Eds.), The 315–334. psychology of the social self (pp. 223–264). Mahwah: Erlbaum. Roche, D. (2003). Accountability in restorative justice. Oxford: Van Prooijen, J.-W. (2004, June/July). Retributive versus restorative Oxford University Press. justice: Is punishing offenders more important than compensat- Roche, D. (2006). Dimensions of restorative justice. Journal of Social ing victims. Paper presented at the 10th biennial meeting of the Issues, 62, 217–238. International Society for Justice Research, Regina, Canada. Schroeder, D. A., Steel, J. A., Woodell, A. J., & Bembenek, A. F. Vidmar, N. (2000). Retribution and revenge. In J. Sanders & V. L. (2003). Justice within social dilemmas. Personality and Social Hamilton (Eds.), Handbook of justice research in law (pp. 31– Psychology Review, 7, 374–387. 63). New York: Kluwer/Plenum. Shaver, K. G. (1985). The attribution of blame: Causality, respon- Vidmar, N. (2002). Retributive justice: Its social context. In M. Ross sibility, and blameworthiness. New York: Springer. & D. T. Miller (Eds.), The justice motive in everyday life (pp. Sherif, M. (1966). Group conflict and cooperation: Their social 291–313). New York: Cambridge University Press. psychology. London: Routledge and Kegan Paul.