Restorative Justice: Assessing Optimistic and Pessimistic Accounts

John Braithwaite

The increasing popularity of restorative justice has prompted new and varying applications of its core principles. The authors of The Promise of Restorative Justice highlight the ways that these new ideas now spark innovations both throughout the criminal justice system and in arenas as diverse as business, education, athletics, and the aftermath of ethnic conflict. They offer fresh approaches to addressing the problems of crime and other human conflicts in ways that productively incorporate the values of mutual respect, accountability, and ultimately reconciliation between offenders and victims.

Historically, Justice System refers to the interpretation and application of the law in the name of the community, society and/or state. This, under normal situation, help in conflict resolution, control crime and impose penalties on those who violate laws. In this way, it must be such that the people have absolute trust in and thus approach it with high level of confidence. However, the situation in Nigeria seems to be the contrary as the people seems to have lost hope in it. This paper therefore argues that the failure of the Justice system is as a result of the fact that it lacks the restorative justice system approach especially since the colonial period, as there is a sort of disconnect between the system (society) in which it operates and the system operating it and their orientation. On the other hand, history of Restorative justice in communities where it was operational before colonial rule and its preference for collaborative and inclusive processes, as far as possible, (and outcomes that are mutually agreed upon rather than imposed) shows its effectiveness in repairing broken relationships, addressing the causes and courses of the crime while meeting the needs of victims-offenders and communities. In this way, the offenders (without compulsion) acknowledge and take responsibility for their actions to receive proper punishment, healing and forgiveness. Data obtained from primary and secondary sources were deployed to carry out the study with an analytical and narrative historical method. Findings indicate that contrary to Restorative Justice, the present justice system in operation in most parts of Nigeria (as in other areas that witnessed colonial rule) have failed to address the causes and courses of the crime, build trust and eventually improve damaged relationships. Instead it has often favoured the rich and elites in the society. The paper, therefore, advocate for a return to restorative justice using the Esan traditional example and the lessons for modern societies.

The current Western criminal justice system, with its emphasis on maintaining a balance between offenders&amp;#x27; rights (in their defense) and governmental power (represented by the prosecutorial process) throughout the arrest phase, trial process, and sentencing exercise, has largely ignored restorative justice critiques that have called for a greater emphasis on recognizing emotions, healing injuries, and repairing damage caused by criminal activity. These critiques have come from a number of sources, for example, theological (eg, ...

John Braithwaite Restorative Justice: Assessing Optimistic and Pessimistic Accounts ABSTRACT For informal justice to be restorative justice, it has to be about restoring victims, restoring offenders, and restoring communities as a result of participation of a plurality of stakeholders. This means that victimoffender mediation, healing circles, family group conferences, restorative probation, reparation boards on the Vermont model, whole school antibullying programs, Chinese Bang Jiao programs, and exit conferences following Western business regulatory inspections can at times all be restorative justice. Sets of both optimistic propositions and pessimistic claims can be made about restorative justice by contemplating the global diversity of its practice. Examination of both the optimistic and the pessimistic propositions sheds light on prospects for restorative justice. Regulatory theory (a responsive regulatory pyramid) may be more useful for preventing crime in a normatively acceptable way than existing criminal law jurisprudence and explanatory theory. Evidence-based reform must move toward a more productive checking of restorative justice by liberal legalism, and vice versa. This essay conceives restorative justice as a major development in criminological thinking, notwithstanding its grounding in traditions of justice from the ancient Arab, Greek, and Roman civilizations that accepted a restorative approach even to homicide (Van Ness 1986, pp. 64-68), the restorative approach of the public assemblies (moots) John Braithwaite is a professor in the Research School of Social Sciences, Australian National University. My thanks to Stephen Free, Ellen Foulon, Alison Pilger, and Chris Treadwell for their diligent research assistance for this essay. Helpful comments on earlier drafts were provided by many people including Kathy Daly, Carol Heimer, Carol La Prairie, Brenda Morrison, Guy Masters, Christine Parker, Hugh Potter, Kent Roach, Tom Scheff, Lawrence Sherman, Heather Strang, Michael Tonry, and Tom Tyler. © 1999 by The University of Chicago. All rights reserved. 0192 -3 2 34/99/0025 -0005$02 .00 HeinOnline -- 25 Crime & Just. 1 1999 2 John Braithwaite of the Germanic peoples who swept across Europe after the fall of Rome (Berman 1983, pp. 53-56), Indian Hindus as ancient as the Vedic civilization (6000-2000 B.C.) (Beck 1997, p. 77) for whom "he who atones is forgiven" C\Veitekamp 1989), and ancient Buddhist, Taoist, and Confucian traditions that one sees blended today in North Asia (Haley 1996). Contemporary Nobel Peace Prize-winning Buddhists, Aug San Suu Kyi of Burma and the Dalai Lama, are reteaching the West that the more evil the crime, the greater the opportunity for grace to inspire a trans formative will to resist tyranny with compassion. They follow in the footsteps of Hindus such as Mohandas Gandhi and Christians such as Desmond Tutu. In the words of the Dalai Lama: "Learning to forgive is much more useful than merely picking up a stone and throwing it at the object of one's anger, the more so when the provocation is extreme. For it is under the greatest adversity that there exists the greatest potential for doing good, both for oneself and for others" (Eckel 1997, p. 135). Or as Saint Paul put it, "Where sin abounded, grace did much more abound." The implication of this teaching for criminologists is that preventing crime is an impoverished way of conceiving our mission. Crime is an opportunity to prevent greater evils, to confront crime with a grace that transforms human lives to paths of love and giving. If we take restorative justice seriously, it involves a very different way of thinking about traditional notions such as deterrence, rehabilitation, incapacitation, and crime prevention. It also means transformed foundations of criminal jurisprudence and of our notions of freedom, democracy, and community. Restorative justice has been the dominant model of criminal justice throughout most of human history for all the world's peoples. A decisive move away from it came with the Norman Conquest of much of Europe at the end of the Dark Ages (Van Ness 1986, p. 66; Weitekamp 1999). Transforming crime into a matter of fealty to and felony against the king, instead of a wrong done to another person, was a central part of the monarch's program of domination of his people. Interest in restorative justice rekindled in the West from the establishment of an experimental victim-offender reconciliation program in 1974 in Kitchener, Ontario (Peachey 1989). Today, Umbreit (1999) reports that there are at least three hundred of these programs in North America and over five hundred in Europe. The 1990s have seen the New Zealand idea of family group confer- HeinOnline -- 25 Crime & Just. 2 1999 Restorative Justice 3 ences spread to many countries including Australia, Singapore, the United Kingdom, Ireland, South Mrica, the United States, and Canada, adding a new theoretical vitality to restorative justice thinking. Canadian native peoples' notions of healing circles and sentencing circles (James 1993) also acquired considerable influence, as did the Navajo Justice and Healing Ceremony (Yazzie and Zion 1996). Less visible were the rich diversity of African restorative justice institutions such as the Nanante. By the 1990s, these various programs came to be conceptualized as restorative justice. Bazemore and Washington (1995) and Van Ness (1993) credit Albert Eglash (1975) with first articulating restorative justice as a restitutive alternative to retributive and rehabilitative justice. As a result of the popularizing work of North American and British activists such as Howard Zehr (1985, 1990), Mark Umbreit (1985, 1994), Kay Pranis (1996), Daniel Van Ness (1986, 1999), Tony Marshall (1985), and Martin Wright (1982) during the 1980s, and the new impetus after 1989 from New Zealanders and Australians, restorative justice became the emerging social movement for criminal justice reform of the 1990s (Daly and Immarigeon 1998). Since 1995, two organizations, Ted Wachtel's (1997) Real Justice in the United States and John MacDonald's Transformative Justice Australia, have offered training in conferencing to thousands of people worldwide. An evaluation research community also emerged in association with the social movement; this community is dominated by Belgians, Germans, Austrians, and Canadians, though Burt Galaway and Joe Hudson (1975) in Minnesota and Canada were the early and persistent role models of this research community. During the 1980s, there was also considerable restorative justice innovation in the regulation of corporate crime (Rees 1988; Braithwaite 1995a). Clifford Shearing's (1997) historical analysis is more about governmentalities of post-Fordist capitalism than village moots: "Restorative justice seeks to extend the logic that has informed mediation beyond the settlement of business disputes to the resolution of individual conflicts that have traditionally been addressed within a retributive paradigm ... In both a risk-oriented mentality of security [actuarialism] and a restorative conception of justice, violence loses its privileged status as a strategy to be deployed in the ordering of security" (p. 12). Section I of this essay first seeks to conceptualize what restorative justice is against the background of these histories. Sections II-V follow the author's peculiar history of engagement with restorative pro- HeinOnline -- 25 Crime & Just. 3 1999 4 John Braithwaite cesses in business regulation (nursing homes, corporate crime) and in Asia and the Pacific. Section VI summarizes fifteen propositions of an Optimistic Account (Sec. VII) and the thirteen propositions of a Pessimistic Account of restorative justice that form the subsections of Section VIII. The Optimistic Account is that each of a number of theories-about shame and shaming, justice, defiance, self-categorization, and deterrence-might have some partial validity. The Pessimistic Account is that restorative justice processes will often fail or backfire, defeating justice. This dialectic leads to a conclusion about how to hedge restorative justice with deterrence, incapacitation, and liberal rights. 1. What Is Restorative Justice, and Why Is It Beginning to Take Off? Restorative justice is most commonly defined by what it is an alternative to. Juvenile justice, for example, is seen as seesawing back and forth during the past century between a justice and a welfare model, between retribution and rehabilitation. Restorative justice is touted as a long-overdue third model or a new "lens" (Zehr 1990), a way of hopping off the seesaw, of heading more consistently in a new direction while enrolling both liberal politicians who support the welfare model and conservatives who support the justice model. The appeal to liberals is a less punitive justice system. The appeal of restorative justice to conservatives is its strong emphasis on victim empowerment, on empowering families (as in "family group conferences"), on sheeting home responsibilities, and on fiscal savings as a result of the parsimonious use of punishment. When restorative justice is applied to whitecollar crime, probusiness politicians also tend to find the approach more appealing than a retributive approach to business wrongdoing. Every one of these bases of political appeal is subject to horrible perversions, as Section VII (The Pessimistic Account) suggests. In New Zealand, the country with the most developed programmatic commitment to restorative justice, the mainstream conservative and social democratic parties have been joined by Christian profamily parties of the Right in their support for restorative justice. In New Zealand (Maxwell and Morris 1993) and Australia (Moore with Forsythe 1995), the evidence is surprising on how supportive of restorative justice are the police, that traditional ally of law-and-order politicians. The strongest opposition has come from lawyers, including some judges, under the influence of well-known critiques of the justice of informal processing of crime (see Sec. VII). At the same time, in both HeinOnline -- 25 Crime & Just. 4 1999 Restorative Justice 5 New Zealand and Canada, judicial leadership has been at the vanguard, if not the vanguard, of restorative justice reform. In the 1990s, restorative justice became a unifying banner, sweeping up various traditions of justice as "making amends" (Wright 1982), reconciliation (Marshall 1985; Umbreit 1985; Dignan 1992), peacemaking (Pepinsky and Quinney 1991), redress (De Haan 1990), relational justice (Burnside and Baker 1994), trans formative justice (Moore with Forsythe 1995, p. 253; Morris 1995), Real Justice (McDonald et al. 1995), and republican justice (Braithwaite and Pettit 1990). During the same period, similar ideas were also being developed by feminist abolitionists (Meima 1990) and in other feminist analyses that emphasized denunciation of the harm and help for victims as more central than punishment (Lacey 1988, pp. 193 -94; Harris 1991; Braithwaite and Daly 1994; Roach 1999). Feminist thinking about crime has been a dialectic of Portia (an ethic of justice) and Persephone (an ethic of care) (Heidensohn 1986), out of which some feminists want Portia and Persephone each to check the excesses of the other (Masters and Smith 1998) in a manner rather like that discussed in the conclusion to this essay. The most influential text of the restorative tradition has been Nils Christie's "Conflicts as Property" (1977), which defined the problem of criminal justice institutions "stealing conflicts" from those affected. Centuries-earlier philosophies of New Zealand Maori (Pratt 1996), North American Indian (Krawll 1994; Aboriginal Corrections Policy Unit 1997a), Christian (Van Ness 1986), and Japanese/Confucian/ Buddhist (Haley 1996; Masters and Smith 1998) restorative justice have actually been the sources of the deepest influences on the contemporary social movement. Paul McCold (1997) recently convened a Delphi process on behalf of the Working Party on Restorative Justice of the Alliance of NGOs [Nongovernmental Organizations] on Crime Prevention and Criminal Justice to see if these disparate strands of the emerging alternative might settle on a consensual conception of restorative justice. A Delphi process iteratively solicits expert opinion, in this case on the best way to define restorative justice. The consensus was not overwhelming. The most acceptable working definition was offered by Tony Marshall: "Restorative justice is a process whereby all the parties with a stake in a particular offense come together to resolve collectively how to deal with the aftermath of the offense and its implications for the future" (e-mail, Marshall to McCold, 1997). This definition does stake out a shared core meaning of restorative HeinOnline -- 25 Crime & Just. 5 1999 6 John Braithwaite justice. Its main limitation is that it does not tell us who or what is to be restored. It does not define core values of restorative justice, which are about healing rather than hurting, moral learning, community participation and community caring, respectful dialogue, forgiveness, responsibility, apology, and making amends (see Nicholl 1998). I take those who have a "stake in a particular offense" to mean primarily victims, offenders, and affected communities (which include the families of victims and offenders). So restorative justice is about restoring victims, restoring offenders, and restoring communities (Bazemore and Umbreit 1994; Brown and Polk 1996). An answer to the "\Vhat is to be restored?" question is whatever dimensions of restoration matter to the victims, offenders, and communities affected by the crime. Stakeholder deliberation determines what restoration means in a specific context. Some have suggested dimensions of restoration that are found to be recurrently important in restorative justice processes. For example, I have defined the following dimensions of restoration as important from a republican perspective: restoring property loss, restoring injury, restoring a sense of security, restoring dignity, restoring a sense of empowerment, restoring deliberative democracy, restoring harmony based on a feeling that justice has been done, and restoring social support (Braithwaite 1996). All cultures have restorative justice traditions defined in these terms, particularly in their families, schools, and churches, just as they all have retributive traditions. The core belief of the social movement for restorative justice is that all cultures in the circumstances of the modern world will find their restorative traditions a more useful resource than their retributive traditions. Yet all cultures must adapt their restorative traditions in ways that are culturally meaningful to them. Insufficiently retributive societies were often wiped out in the past by more violent cultures. l Punitiveness, however, has less survival value for communities that are more interdependent, a lesson North American states finally learned in their dealings with each other during the nineteenth century and European states in the twentieth. Most of the world continues to live in a zone of violence where international disputes continue to be settled through force of arms; citizens continue to suffer terrible war crimes. However, North America and Western 1 The most famous example was the razing of Carthage by Rome after Hannibal achieved his tactical objectives and left Rome in peace (when he had it at his mercy). HeinOnline -- 25 Crime & Just. 6 1999 Restorative Justice 7 Europe today constitutes a zone of restorative diplomacy, into which the newly economically interdependent states of Asia are being integrated (cf. Goldgeir and McFaul 1992). Within this zone of restorative diplomacy, democracies do not commit war crimes against one another, indeed do not go to war against one another (Doyle 1983, 1986; Maoz and Abdolali 1989; Burley 1992, pp. 394-95). Rather disputes are mostly settled through conciliation, mediation, conferences, and summits. They are dealt with through processes that fit Marshall's definition of restorative justice (see Laue 1991), albeit processes with huge imbalances of power. According to this analysis, there was a profound difference between Versailles, which was a degradation ceremony intended to humiliate a defeated Germany in 1919 and the Marshall Plan after World War II, which was a restorative approach to reintegrate Germany with respect into the international community (Braithwaite 1991, 1999; Offer 1994; Scheff 1994). Retributive emotions are things we all experience and things that are easy to understand from a biological point of view. But, on this view, retribution is in the same category as greed or gluttony; biologically they once helped us to flourish, but today they are corrosive of human health and relationships. There is a contrary view that a more rationalist conception of retribution can be reconciled with restoration, however, and indeed must be if restorative justice is to be a pragmatic program (Van Ness and Strong 1997, pp. 27-28; Daly and Immarigeon 1998). While most of the writing on restorative justice focuses on the comparatively small crimes of juvenile delinquents, in this essay I emphasize its relevance to adult crime as well, including war crimes and crimes at the commanding heights of business power (as in corporate restorative justice) and political power (as in Archbishop Desmond Tutu's Truth and Reconciliation Commission in South Aftica, which he explicitly sees as a restorative justice process). On this view, organizations like Transcend that specialize in peacemaking training for international violence are part of the social movement for restorative justice (www.transcend.org). Reviewers of an earlier version of this essay were wary of the relevance of much of this "noncriminological" material, of the relevance of restorative justice in Africa or Asia to the United States and indeed of my credentials as a restorative justice advocate for undertaking a "dispassionate" criminological review. I am not sure that they are right on three counts. First, regulatory and educative practices and institu- HeinOnline -- 25 Crime & Just. 7 1999 8 John Braithwaite tions might be better objects of research than criminal justice practices, both because there is enormous selectivity in when we react to regulated phenomena as crime and because regulatory and educational theory deliver superior and more general explanations than criminological theory. Second, my working hypothesis is that better theories of U.S. or Australian crime are likely to be stimulated by broadening our horizons to studying the different patterns of regulatory practices and regulatory outcomes observable in radically different societies. Third, my working hypothesis is that superior explanatory theory (ordered propositions about the way the world is) and superior normative theory (ordered propositions about the way the world ought to be) arise from an explicit commitment to integrating explanatory and normative theory. But for this approach to deliver the goods, what must come with it is a serious commitment to research designed to refute both the normative and explanatory claims. I do not defend this approach here, as Christine Parker and I have done so elsewhere (Braithwaite and Parker 1999). Over time science will judge whether such an approach has marshalled theory and evidence around inferior explanations compared with those advanced by criminologists out of a more standard North Adantic mold. Most restorative justice advocates came to the approach through juvenile crime as a result of persistent empirical evidence of the failures of the welfare and justice models. The path that led me and a number of my colleagues who are experts in corporate crime to restorative justice is quite different and instructive. Many young criminologists began to study white-collar crime after Watergate to resurrect Edwin Sutherland's (1983) project. We wanted to document systematically how the crimes of the powerful were unpunished. What we found, in effect, was that the regulation of corporate crime in most countries was rather restorative. The reasons for this were far from ennobling, being about corporate capture combined with high costs of complex corporate crime investigations that states were unwilling to pay. Nevertheless, some of us began to wonder whether we were wrong to see our mission as to make corporate crime enforcement more like street crime enforcement through tougher sanctions. 2 Instead we began to wonder whether street crime enforcement might be more effective if it were more like corporate criminal enforcement. 1 Critics indeed might enjoy the irony that Watergate offender Charles Colson, now of Prison Fellowship Ministries, is a prominent restorative justice advocate. HeinOnline -- 25 Crime & Just. 8 1999 Restorative Justice 9 In my case, engagement with restorative approaches to corporate crime was entangled with my active engagement with social movement politics-particularly the consumer movement, but other social movements as well. In turn, my engagements with regulatory agenciesconcerned with nursing homes, occupational health and safety, antitrust, environment, consumer protection, tax, and affirmative actionwere as much connected to my history as an NGO activist as with a research background in these domains. For all these reasons I write as a reviewer who "belongs" in the social movement for restorative justice. Normatively serious people who engage with a social movement should not be dispassionate about it; they should have a passion for good science to find where its claims are false. In the following sections I describe four examples of restorative justice praxis to open up an understanding of the interface among activism, theoretical innovation, and evaluation: nursing home regulation, Asian community policing, trade practices enforcement, and restorative justice conferences. II. Nursing Home Regulation Valerie Braithwaite, Toni Makkai, Diane Gibson, David Ermann, Anne Jenkins, and I became involved in evaluating nursing home regulation before the Australian federal government took it over from state governments in 1988. Over the next six years we became the government's main consultants in this area. Prior to the change, regulation had consisted of specifying quality of care inputs and prosecuting breaches criminally when enforcement action was required. Since 1988 the move away from the criminal model has been almost total (in our view, it went too far), apart from spectacular cases where multiple deaths from neglect or abuse occurred. In a radical shift from prescriptive regulation, the old rule books were thrown out and replaced with thirty-one outcome standards (compared with over a thousand standards in most U.S. states) settled consensually between the industry and major stakeholders such as consumer groups, unions, and aged care interests. The new regulatory process was dialogic. While a certain amount of time was spent auditing care plans, quality audit reports, and other records, government inspectors spent more time talking to residents and staff about how the quality of care could be improved. This was a shift to a resident-centered process (victim-centered in criminal justice discourse); an evaluation showed that this could work, residents could be empowered dialogically, even in nursing homes with the sickest residents HeinOnline -- 25 Crime & Just. 9 1999 10 John Braithwaite (Braithwaite and Makkai 1993). Performance against each of the thirty-one standards was ultimately discussed at a conference of the inspection team and management to which representatives of owners, staff, residents, and relatives were also invited. Occasionally the elected representatives of the residents' committee would invite someone from an outside advocacy group to attend. These functioned in ways quite similar to the family group conferences for juvenile offenders discussed later. The final evaluation report concluded from a variety of types of data that the new regulatory regime had improved quality of life for Australian nursing home residents and compliance with the law (Braithwaite et al. 1993), notwithstanding the identification of a large number of problems. Action plans agreed at the exit conferences to restore residents' rights to quality of care were overwhelmingly implemented, the most common reason for nonimplementation being coming up with a better plan subsequent to the inspection. More critically to the evaluation of restorative justice, it was also found that inspectors who treated nursing homes with trust (Braithwaite and Makkai 1994), used praise when improvements were achieved (Makkai and Braithwaite 1993a), and had a philosophy of reintegrative shaming (Makkai and Braithwaite 1994), achieved higher compliance with the standards two years later than inspectors who did not. Jenkins (1997) showed that sustaining the self-efficacy of managers for improving quality of care was critical. While defiance (participation in a business subculture of resistance to regulation) did reduce compliance (Makkai and Braithwaite 1991), disengagement was the bigger problem (Braithwaite et al. 1994). Strategies such as praise and avoiding stigmatization were important to sustaining self-efficacy and engagement with continuous improvement. Hence, within a regime that improved regulatory outcomes by shifting from rule-book criminal enforcement to restorative justice, the inspectors who shifted most toward restorative justice improved compliance most (those who used praise and trust more than threat, reintegrative shaming rather than tolerance or stigmatization, those who restored self-efficacy). These results are discussed again when I consider the theories that predict why restorative justice might work better than punitive justice. The biggest attraction of research in this field was that we could measure compliance with the law with far greater reliability (assessed through independent ratings by two inspectors) than can be obtained with traditional individual criminal offenses or in other areas of busi- HeinOnline -- 25 Crime & Just. 10 1999 Restorative Justice 11 ness regulation (Braithwaite and Braithwaite 1995). The limited support in this superior data set for some of the criminological theories that had been influential in our thinking to that point about how to design restorative justice-notably control, differential association, and subcultural theories-shaped our subsequent thinking about restorative justice theory (Makkai and Braithwaite 1991). III. Asian Community Policing After Brent Fisse and I did some limited fieldwork on how Japanese companies and regulators secured compliance with regulatory laws (Braithwaite and Fisse 1985), I became interested in Japanese social control more broadly. The work of many other scholars suggested that it was based rather heavily on dialogue about collective obligation and relationships as opposed to punishment. This seemed true from social control of the largest corporations (which we were studying) down to the regulation of the petty delinquencies of children in schools. As with nursing homes, Guy Masters's (1995, 1997) work shows that Japanese schools use methods of social control very similar to the family group conferences I discuss later. There was plenty of degradation and punitiveness in Japanese policing as well, especially when cases move from local koban policing (Bayley 1976) to policing by detectives and prosecutors (Miyazawa 1992).3 Yet it seemed to me then, and still does, that the restorative elements ofJapanese social control are more influential and sophisticated than in the West. We have much to learn from them (Masters 1997). In an earlier draft of Crime, Shame and Reintegration, I also had a section on Chinese community policing (Braithwaite 1989). I threw it in the bin because Chinese informal justice seemed to involve so much more stigmatization and punitiveness than Japanese justice. Vagg's (1998) Hong Kong data captures well the concerns that beat a path to my wastebasket. Chinese Peoples' Courts, especially as they were projected to us in the cultural revolution, seemed a model of how not to do restorative justice. Yet Hong Lu's (1998) research in Shanghai shows that the most important contemporary restorative justice institution in China, bangjiao meetings (bang means help,jiao means education and admonition) tend to start as rather stigmatizing encounters but to end as reintegrative ones (see also Wong 1996). ) I am indebted to Christopher Murphy for pointing out that on the basis of his considerable observation of Japanese policing, Bayley and Miyazawa may both be right in this way, like the blind Hindus in the legend, feeling different parts of the elephant. HeinOnline -- 25 Crime & Just. 11 1999 12 John Braithwaite Chinese restorative justice, in both its positive and negative aspects, deserves more attention because China has by far the largest and most diverse programs, 155,000 local mediation committees, which accounted for over six million cases, compared to under four million cases that went to court in 1994 (Wong 1998). Many of the mediations were of family or neighborhood disputes that were not necessarily criminal. China also is the home of Confucius (551-479 B.C.), arguably the most influential thinker about restorative justice the world has known. Confucius's quest can be read in part as a search for practices of good government that enable people to understand the effects their actions have on one another and that naturally expose the virtue of the virtuous so that others will follow them. Virtue is inculcated by quiet good example rather than by denunciation. Tsze-kung said, "Has the superior man his hatreds also?" The Master said, "He has his hatreds. He hates those who proclaim the evil of others" (Confucius 1974, p. 143). From the perspective of a European republican philosophy (Braithwaite and Pettit 1990; Pettit 1997), there is much of value to draw on in Confucian thought but also much that might be dangerous. Confucian communitarianism was patriarchal and hierarchic. Perhaps a settled sense of deference was not so dangerous in a stable world where family, village, and a unitary state were the only institutions that mattered. But in a more complex world where there are many levels of government, up to the International Monetary Fund (lMF) and World Trade Organization, many cross-cutting institutions of civil society to which we belong, a world in which we and our parents are geographically mobile, we need strong, independent individuals as well as strong families and communities. Individuation is vital as a practice of socialization if individuals are to be strong enough to resist tyranny as they move from one site of domination to another in a complex world. Moreover, if we do not move away from the notion of society as a holistic unity to the notion of the separation of powers and an important place for the rule of law, a liberal-republican constitutional order, the lesson of this century's history is that we will get tyranny-"political power out of the barrel of a gun." Yet we can read the great sweep of Chinese history as a dialectic of learning and unlearning this lesson. I refer in particular to the great historical struggle between the Legalists and the Confucians, and to the dialectic between both legalism and Confucianism and the dialectic of freedom in Taoism, to the disastrous abandonment of the rule of HeinOnline -- 25 Crime & Just. 12 1999 Restorative Justice 13 law in the Cultural Revolution and the partial return to it since (Gernet 1982; Huang 1988). One reason it was an intellectual mistake to scrap the China section of Crime, Shame and Reintegration is that the study of Chinese history may hold one key to a macrosociology of restorative justice. In the dialectic of Chinese history between the domination of Confucian and Legalist ideas, a high-water mark of Legalist influence was the Ch'in Dynasty. What brought about the fall of the Ch'in Empire in 211 B.C. was not the alienation and hatred of the scholar class, nor the bitter enmity of the surviving remnants of the aristocracy, but the growing popular discontent and mounting outrage over the cruelty of the system of punishments and the intolerable burden of taxes and levies imposed for the massive public works that the emperor commanded. Crime increased as did the number of those condemned, tortured, mutilated, and exiled to labor gangs. As long as the emperor was alive, fear of his powerful and demoniacal personality held the empire together; after his death all the restraints broke, and the empire exploded in rebellion. (Michael 1986, p. 66) Today the movement of the Confucian-Legalist dialectic is in the reverse direction, with the "rule of law" rebounding as a dominant value. Given the continued trampling of human rights and freedoms in China, this may be a hopeful development, yet part of it is a sharp decline of the proportion of criminal cases dealt with by mediation as opposed to criminal trials. What a pity that so few Western intellectuals are engaged with the possibilities for recovering, understanding, and preserving the virtues of Chinese restorative justice while checking its abuses with a liberalizing rule of law. Whatever the rights and wrongs of it, the legalist-restorative contest is more central to the dynamic of Chinese history than to the histories of other nations and therefore more central to the development of a macrosociology of the fluctuating fortunes of restorative justice. IV. Trade Practices Enforcement Between 1985 and 1995, as a part-time commissioner with Australia's national consumer protection and antitrust agency, I attempted with mixed success to persuade my colleagues on the Trade Practices Com- HeinOnline -- 25 Crime & Just. 13 1999 14 John Braithwaite mission to experiment in an Australian way with the restorative principles I saw as underlying Japanese business regulation. Ironically, when the commission decided to run its boldest restorative justice conference, I made the mistake of voting against it, believing the conduct to be so serious that formal criminal charges should be laid. It involved the most widespread and serious consumer protection frauds ever to come before the agency. They implicated a number of insurance companies systematically ripping off consumers through misrepresentations about policies that in some cases were totally useless. The worst abuses occurred in twenty-two remote Aboriginal communities and these were tackled first. Top management from the insurance company visited these communities for days on end at meetings with the victims, the local Aboriginal community council, the regulators, and local officials of the Department of Social Security in cases where useless policy premiums were being deducted from welfare checks. Some of those executives went back to the city deeply ashamed of what their company had done. Back in Canberra, meetings were held with insurance regulators and industry associations and even with the prime minister about followup regulatory reforms. The plurality of participants led to a plurality of remedies from the first agreement with Colonial Mutual Life (CML), who voluntarily compensated two thousand policyholders and also funded an Aboriginal consumer education fund to "harden targets" for future attempts to rip off illiterate people. It conducted an internal investigation to discover failings in the company's compliance program and to identify officers responsible for the crimes. A press conference was then called to reveal the enormity of the problem. No one realized quite how enormous, until a police union realized that its own members were being ripped off through the practices of another company (in this case, there were 300,000 victims and a payout of at least $50 million and perhaps $100 million by the company). As a result of the CML self-investigation, eighty officers or agents of CML were dismissed, including some senior managers and one large corporate agent, Tri-Global. CML also put in place new internal compliance policies. Some procedures relating to welfare checks changed in the Department of Social Security and there were regulatory and selfregulatory changes concerning the licensing of agents and changes to the law (Fisse and Braithwaite 1993, p. 235). This polycentric problemsolving was accomplished without going to court (except with a couple of individuals who refused to cooperate with the restorative justice HeinOnline -- 25 Crime & Just. 14 1999 Restorative Justice 15 process). The disparate array of preventive measures were grounded in the different kinds of theories the rich plurality of players involved in this restorative justice process came up with-theories of education, deterrence, incapacitation, rehabilitation, target hardening, moral hazard, adverse publicity, law, regulation, and opportunity theory. The cynic about restorative justice will say that the Australian insurance cases were unusually sweeping exercises in crime prevention. True, most crime prevention is more banal. Yet this process was so sweeping in its ramifications precisely because it was restorative. What would have happened if we had prosecuted this case criminally? At best the company would have been fined a fraction of what it actually paid out and there would have been a handful of follow-up civil claims by victims. At worst, illiterate Aboriginal witnesses would have been humiliated and discredited by uptown lawyers, the case lost, and no further ones taken. The industry-wide extensiveness of a pattern of practices would never have been uncovered; that was only accomplished by the communitarian engagement of many locally knowledgeable actors. V. Restorative Justice Conferences In Crime, Shame and Reintegration I made reference to the desirability of institutionalizing something like the restorative justice conference for criminal offenders (Braithwaite 1989, pp. 173-74). After reading this, John McDonald of the New South Wales Police came to me and said this had already been done in New Zealand. Terry O'Connell showed me videotaped interviews with people such as Maori chief judge of the New Zealand Juvenile Court, Michael Brown. These revealed that one of the rationales for restorative justice in the Maori tradition was the simultaneous communication of "shame" and "healing." It was a depressing revelation that what I thought was the only limited originality in Crime, Shame and Reintegration had been preceded by several hundreds of years of Polynesian oral tradition, not just in New Zealand. Indeed, I concluded that Maori ways of thinking about whakama or shame were in some important ways an advance on my own thinking. After Crime, Shame and Reintegration became a widely read book, many people from Africa, Melanesia, Asia, and the Americas were in touch with me about restorative justice conferences that were part of their tradition. I learned that the Native American healing circle seeks to institutionalize equality rather than hierarchy and "puts the problem in the center-not the person" (Pranis 1996, p. 46, quoting Ada Melton [1995]). These stories challenged assump- HeinOnline -- 25 Crime & Just. 15 1999 16 John Braithwaite tions I strongly held until the mid-1990s, for example, that traditional Western criminal process was superior at just fact-finding than restorative justice processes. 4 Healing circles in the Manitoba First Nation community of Hollow Water began to deal with what many thought of at first as an epidemic of alcohol abuse. As citizens sat in these circles discussing the problems of individual cases, they realized that there was a deeper underlying problem, which was that they lived in a community that was sweeping the sexual abuse of children under the carpet. Through setting up a complex set of healing circles to help one individual victim and offender after another, in the end it had been discovered that a majority of the citizens were at some time in their lives victims of sexual abuse. s Forty-eight adults out of a community of six hundred formally admitted to criminal responsibility for sexually abusing children, forty-six as a result of participating in healing circles, two as a result of being referred to a court of law for failing to do so (Lajeunesse 1993; Ross 1996, pp. 29-48). Ross (1996, p. 36) claims that the healing circles have been a success because there have been only two known cases of reoffending. Tragically, however, there has been no genuinely systematic outcome evaluation of Hollow Water. What is more important than the crime prevention outcome of Hollow Water is its crime detection outcome. When and where has the traditional criminal process succeeded in uncovering anything approaching forty-eight admissions of criminal responsibility for sexual abuse of children in a community of just six hundred? Before reading about Hollow Water, I had always said that the traditional criminal trial process is superior to restorative justice processes for justly getting to the truth of what happened. Restorative justice processes were only likely to be superior to traditional Western criminal process when there was a clear admission of guilt. The significance of Hollow Water is that it throws that position into doubt. New Zealand remained of preeminent importance, however, because it mainstreamed the conferencing innovation into a Western ju• A conversation with Gale Burford about his work on conferencing family violence with Joan Pennell also has me wondering. In a third of their cases, sexual abuse of children came out during conferences. Gale said: "So violence programs that exclude sexual abuse don't really. They just say if they have sexual abuse don't talk about it or you'll be out of the program." 5 La Prairie (1994, p. iii) in a sophisticated study of this problem from a restorative justice perspective in another context found that 46 percent of innercity native people in Canada had experienced child abuse. For an outline of the Hollow Water procedures for dealing with sexual abuse, see Aboriginal Corrections Policy Unit (1997, esp. pp. 221-30). HeinOnline -- 25 Crime & Just. 16 1999 Restorative Justice 17 venile justice system (and into the care and protection of abused and neglected children as well). The importance of New Zealand was not because it adopted Maori restorative philosophies; indeed Pakeha (non-Maori) New Zealand tended to reject much of both the restorative and retributive aspects of Maori philosophy, initially justifying the practice of "family group conferences" in terms of a move from the welfare to the Western justice model. When its innovation became internationally celebrated, New Zealand wisely reinterpreted family group conferences as restorative justice. Indeed, for all of us practice was ahead of theory, and it was well into the 1990s before the North American label "restorative justice" subsumed what had been developing elsewhere for a long time. The way conferences work is very simple. Once wrongdoing is admitted, the offender and his or her family are asked who they would like to have attend a conference as supporters. Similarly, victims are asked to nominate loved ones to attend with them. The conference is a meeting of these two communities of care. First there is a discussion of what was done and what the consequences have been for everyone in the room (the victim's suffering, the stress experienced by the offender's family). Then there is a discussion of what needs to be done to repair those different kinds of harm. A plan of action is agreed and signed by the offender and usually by the victim and the police officer responsible for the case. Asking the offender to confront the consequences of his wrongdoing (and talking them through in the presence of those who have suffered them) is believed by conferencing advocates to have a variety of positive effects in terms of taking responsibility, experiencing remorse, and offering practical help and apology to the victim and the community to right the wrong. Beyond this common core, conferences vary from place to place in how they are run. In Australia, Wagga Wagga was the first conferencing program from 1991 and an important site of early research and development on a culturally pluralized conferencing process suitable for both Western and Australian Aboriginal cases. This research and development is being carried forward by the RISE experiment in Canberra in which thirteen hundred adult and juvenile cases are being randomly assigned to conference versus court by Lawrence Sherman and Heather Strang (Sherman et al. 1998). Drunk driving, property, and violent crimes are covered by the experiment. Only preliminary results are available from the first eleven hundred offenders at the time of writing; RISE is designed to test most of the theories of restorative justice discussed in this essay. HeinOnline -- 25 Crime & Just. 17 1999 18 John Braithwaite VI. Optimistic and Pessimistic Accounts of Restorative Justice The empirical evidence about restorative justice is organized in this essay according to the propositions that flow from an optimistic account of restorative justice and a pessimistic one. The propositions of both accounts are plausible in light of the limited evidence we have at this time. The Optimistic Account: A. Restorative Justice Practices Restore and Satisfy Victims Better than Existing Criminal Justice Practices. B. Restorative Justice Practices Restore and Satisfy Offenders Better than Existing Criminal Justice Practices. C. Restorative Justice Practices Restore and Satisfy Communities Better than Existing Criminal Justice Practices. D. Reintegrative Shaming Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. £. Procedural Justice Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. F. The Theory of Bypassed Shame Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. G. Defiance Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. H. Self-Categorization Theory Predicts that Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. I. Crime Prevention Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices. J. Restorative Justice Practices Deter Crime Better than Practices Grounded in Deterrence Theories. K. Restorative Justice Practices Incapacitate Crime Better than Criminal Justice Practices Grounded in the Theory of Selective Incapacitation. L. Restorative Justice Practices Rehabilitate Crime Better than CriminalJustice Practices Grounded in the Welfare Model. M. RestorativeJustice Practices Are More Cost-Effective than CriminalJustice Practices Grounded in the Economic Analysis of Crime. HeinOnline -- 25 Crime & Just. 18 1999 Restorative Justice 19 N. Restorative Justice Practices Secure Justice Better than Criminal Justice Practices Grounded in "Justice" Or Just Deserts Theories. O. Restorative Justice Practices Can Enrich Freedom and Democracy. The Pessimistic Account: A. Restorative Justice Practices Might Provide No Benefits Whatsoever to Over 90 Percent of Victims. B. Restorative Justice Practices Have No Significant Impact on the Crime Rate. C. Restorative Justice Practices Can Increase Victim Fears of Revictimization. D. Restorative Justice Practices Can Make Victims Little More than Props for Attempts to Rehabilitate Offenders. E. Restorative Justice Practices Can Be a "Shaming Machine" that Worsens the Stigmatization of Offenders. F. Restorative Justice Practices Rely on a Kind of Community that is Culturally Inappropriate to Industrialized Societies. G. Restorative Justice Practices Can Oppress Offenders with a Tyranny of the Majority, Even a Tyranny of the Lynch Mob. H. Restorative Justice Practices Can Widen Nets of Social Control. I. Restorative Justice Practices Fail to Redress Structural Problems Inherent in Liberalism Like Unemployment and Poverty. J. Restorative Justice Practices Can Disadvantage Women, Children, and Oppressed Racial Minorities. K. Restorative Justice Practices Are Prone to Capture by the Dominant Group in the Restorative Process. L. Restorative Justice Processes Can Extend Unaccountable Police Power, Even Compromise the Separation of Powers among Legislative, Executive, and Judicial Branches of Government. M. Restorative Justice Practices Can Trample Rights because of Impoverished Articulation of Procedural Safeguards. VII. An Optimistic Account of Restorative Justice As this essay is about reviewing empirical evidence in a theoretical framework rather than theoretical exegesis, I begin with the three COre propositions of the Optimistic Account of restorative justice. These three (A-C) are the corollaries of propositions D-O of the Optimistic HeinOnline -- 25 Crime & Just. 19 1999 20 John Braithwaite Account. They are that restorative justice does restore victims, offenders, and communities. A. Restorative Justice Practices Restore and Satisfy Victims Better than Existing Criminal Justice Practices A consistent picture emerges from the welter of data reviewed in this section: it is one of comparatively high victim approval of their restorative justice experiences, though lower levels of approval than one finds among other participants in the process. So long as the arrangements are convenient, it is only a small minority of victims who do not want to participate in restorative justice processes. Consistent with this picture, the preliminary RISE data in Canberra show only 3 percent of offenders and 2 percent of community representatives at conferences compared with 14 percent of victims disagreeing with the statement: "The government should use conferences as an alternative to court more often." Most of the data to date are limited to a small range of outcomes; we await the first systematic data on some of the dimensions of restoration discussed in Section I. On the limited range of outcomes explored to date, victims do seem to get more restoration out of restorative justice agreements than court orders and restorative justice agreements seem to be more likely to be delivered than court orders even when the former are not legally enforceable. 1. Operationalizing Victim Restoration. There is a deep problem in evaluating how well restorative justice restores. Empowerment of victims to define the restoration that matters to them is a keystone of a restorative justice philosophy. Three paths can be taken. One is to posit a list of types of restoration that are important to most victims, such as that discussed in Section I. The problem with this is that even with as uncontroversial a dimension of restoration as restoring property, some victims will prefer mercy to insisting on getting their money back; indeed it may be that act of grace which gives them a spiritual restoration that is critical for them. 6 The second path sidesteps a debate on what dimensions of restoration are universal enough to evaluate. Instead, it measures overall satisfaction of victims with restorative justice processes and outcomes, assuming (without evidence) that satisfaction is a proxy for victims getting restoration on the things that are 6 I am reminded of a village in Java where I was told of a boy caught stealing. The outcome of a restorative village meeting was that the offender was given a bag of rice: "We should be ashamed because one from our village should be so poor as to steal. We should be ashamed as a village." HeinOnline -- 25 Crime & Just. 20 1999 Restorative Justice 21 most important for them. This is the path followed in this review, largely because these are the kind of data available at this stage. The third path is the best one, but the most unmanageable in large quantitative evaluations. It is to ask victims to define the kinds of restoration they were seeking and then to report how much restoration they attained in these terms that matter most to them. 2. Victim Participation and Satisfaction. While traditional criminal justice practices are notoriously unsatisfying to victims, it is also true that victims emerge from many restorative justice programs less satisfied than other participants. Clairmont (1994, pp. 16-17) found little victim involvement in four restorative justice programs for aboriginal offenders in Canada. There seems to be a wider pattern of greater satisfaction among aboriginal leaders and offenders than among victims for restorative projects on Canadian aboriginal communities (Obonsawin-Irwin Consulting Inc. 1992a, 1992b; Clairmont 1994; La Prairie 1995). Early British victim-offender mediation programs reported what Dignan (1992) called sham reparation, for example Davis's (1992) reporting of offers rather than actual repair, tokenism, and even dictated letters of apology. In some of these programs, victims were little more than a new kind of prop in welfare programs: the "new deal for victims" came in Britain to be seen as a "new deal for offenders" (Crawford 1996, p. 7). However, Crawford's conclusion that the British restorative justice programs that survived into the 1990s after weathering this storm "have done much to answer their critics" (Crawford 1996, p. 7) seems consistent with the evidence. Dignan (1992) reports 71 percent satisfaction among English corporate victims and 61 percent among individual victims in one of the early adult offender reparation programs. In New Zealand, victims attended only half the conferences conducted during the early years of the program,? and when they did attend they were less satisfied (51 percent satisfaction) with family group conferences than were offenders (84 percent), police (91 percent), and other participants (85 percent) (Maxwell and Morris 1993, pp. 115, 7 The evidence seems to be that this was mainly due to limitations in the program administration that made it difficult for victims to attend, not due to the fact that most victims did not want to attend; only 6 percent did not want to meet their offender (Maxwell and Morris 1996). It is widely believed that victim attendance is much higher in New Zealand today now that attention has been directed to these administrative problems, but no systematic evidence exists. HeinOnline -- 25 Crime & Just. 21 1999 22 John Braithwaite 120). About a quarter of victims reported that they felt worse as a result of attending the family group conference. Australian studies by Daly (1996) and Strang and Sherman (1997) also find a significant minority of victims who feel worse after the conference, upset over something said or victimized by disrespect, though greatly outnumbered by victims who feel healing as a result of the conference. Similarly, Birchall, Namour, and Syme (1992) report 27 percent of victims feeling worse after meeting their offender and 70 percent better in Western Australia's Midland Pilot Reparation Scheme. The Ministry of Justice (1994), Western Australia, reports 95 percent victim satisfaction with their restorative justice conference program (Juvenile Justice Teams). McCold and Wachtel (1998) found 96 percent victim satisfaction with cases randomly assigned to conferences in Bethlehem, Pennsylvania, compared to 79 percent satisfaction when cases were assigned to court and 73 percent satisfaction when the case went to court after being assigned to conference and the conference was declined. Conferenced victims were also somewhat more likely to believe that they experienced fairness (96 percent), that the offender was adequately held accountable for the offense (93 percent), and that their opinion regarding the offense and circumstances were adequately considered in the case (94 percent). Ninety-three percent of victims found the conference helpful, 98 percent that it "allowed me to express my feelings without being victimized," 96 percent believed that the offender had apologized, and 75 percent that the offender was sincere. Ninety-four percent said they would choose a conference if they had to do it over again. The Bethlehem results are complicated by a "decline" group as large as the control group; either offenders or victims could cause the case to be declined. In the Canberra RISE experiment, victim participation is currently 85 percent. Reports on the Wagga Wagga conferencing model in Australia are also more optimistic about victim participation and satisfaction (Moore and O'Connell 1994), reporting 90 percent victim satisfaction and victim participation exceeding 90 percent (Moore and O'Connell 1994). The highest published satisfaction and fairness ratings (both 98 percent) have been reported by the Queensland Department of Justice conferencing program (Palk, Hayes, and Prenzler 1998). Umbreit and Coates's (1992) survey found that 79 percent of victims who cooperated in four U.S. mediation programs were satisfied compared to only 57 percent of those who did not have mediation (for earlier similar findings, see Umbreit 1990a). In a subsequent study, Um- HeinOnline -- 25 Crime & Just. 22 1999 Restorative Justice 23 breit (1999) found at four combined Canadian sites victim procedural satisfaction at 78 percent, and 62 percent at two combined English mediation sites. Victim satisfaction with outcomes was higher still: 90 percent (four u.s. sites), 89 percent (four Canadian sites), and 84 percent (two English sites). However, victim satisfaction was still generally lower across the sites than offender satisfaction. Eighty-three percent of U.S. mediation victims perceived the outcome "fair" (as opposed to being "satisfied") compared to 62 percent of those who went through the normal court process. Umbreit and Coates (1992) also report reduced fear and anxiety among victims following mediation. Victims afraid of being victimized again dropped from 25 percent prior to mediation to 10 percent after. A survey of German institutions involved in model mediation projects found that the rate of voluntary victim participation generally ranged from 81 percent to 92 percent, and never dropped below 70 percent (Kerner, Marks, and Schreckling 1992). 3. Honoring of Obligations to Victims. Haley and Neugebauer's (1992) analysis of restorative justice programs in the United States, Canada, and Great Britain revealed between 64 and 100 percent completion of reparation and compensation agreements. I assume here, of course, that completion of agreements that victims have agreed to is important for victim restoration. Marshall's (1992) study of cases referred to mediation programs in Britain found that over 80 percent of agreements were completed. Galaway (1992) reports that 58 percent of agreements reached through mediation in New Zealand were fully complied with within one year. In a Finnish study, 85 percent of agreements reached through mediation were fully completed (Iivari 1987, 1992). Dignan (1992) reports from England 86 percent participant agreement with mediation outcomes, with 91 percent of agreements honored in full. Trenczek (1990), in a study of pilot victim-offender reconciliation projects in Braunschweig, Cologne, and Reutlingen, West Germany (see also Kuhn 1987), reports a 76 percent full completion rate, and a partial completion rate of 5 percent. Pate's (1990) study of victim-offender reconciliation projects in Alberta, Canada, found a rate of noncompletion of agreements of between 5 and 10 percent, and less than 1 percent in the case of a Calgary program. Wundersitz and Hetzel (1996, p. 133) found 86 percent full compliance with conference agreements in South Australia, with another 3 percent waived for near compliance. Fry (1997, p. 5) reported 100 percent completion of agreements in a pilot of twenty-six Northern Territory police-coordi- HeinOnline -- 25 Crime & Just. 23 1999 24 John Braithwaite nated juvenile conferences, and Waters (1993, p. 9) reported 91 percent payment of compensation in Wagga Wagga conferences. In another Wagga-style program, McCold and Wachtel (1998, p. 4) report 94 percent compliance with the terms of conference agreements. Umbreit and Coates (1992) compared 81 percent completion of restitution obligations settled through mediation to 58 percent completion of court-ordered restitution in their multisite study. Ervin and Schneider (1990), in a random assignment evaluation of six U.S. restitution programs, found 89 percent completion of restitution compared with 75 percent completion of traditional programs. Most of these restitution programs, however, were not restorative in the sense of involving meetings of victims and offenders. 4. Symbolic Reparation. One reason that the level of satisfaction of victims is surprisingly high in processes that so often give them so little material reparation is that they get symbolic reparation, which is more important to them (Retzinger and Scheff 1996). Apology is at the heart of this: preliminary results from the RISE experiment in Canberra show that 74 percent of victims whose cases were randomly assigned to a conference got an apology compared to 11 percent in cases randomly assigned to court (Strang and Sherman 1997). Sixty percent of victims felt "quite" or "very" angry before the Canberra conferences, 30 percent afterward. Obversely, the proportion of victims feeling sympathetic to the offender almost doubled by the end of the conference (Strang and Sherman 1997). I discuss below a large body of research evidence showing that victims are not as punitive as the rather atypical victims whose bitter calls for brutal punishment get most media coverage. Both the Strang and Sherman (1997) and Umbreit (1992, p. 443) studies report victim fear of revictimization and victim upset about the crime as having declined following the restorative justice process. In Goodes's (1995) study of juvenile family group conferences in South Australia, where victim attendance ranges from 75 to 80 percent (Wundersitz and Hetzel 1996), the most common reason victims gave for attending their conference was to try to help the offender, followed by the desire to express feelings, make statements to the offender, or ask questions such as "Why me?" (what Retzinger and Scheff (1996) call symbolic reparation) followed by "curiosity and a desire to 'have a look'," followed by "responsibility as citizens to attend." The desire to ensure that the penalty was appropriate and the desire for material reparation rated behind all of the above motivations to attend. The response rate in the Goodes (1995) study was poor and there may be a HeinOnline -- 25 Crime & Just. 24 1999 Restorative Justice 25 strong social desirability bias in these victim reports; yet that may be precisely because the context of conference attendance is one that nurtures responsible citizenship cognitions by victims. Eighty-eight percent of Goodes's victims agreed with the conference outcome, 90 percent found it helpful to them, and 90 percent said they would attend again if they were a victim again (Goo des 1995). With all these quantitative findings, one can lose sight of what most moves restorative justice advocates who have seen restorative processes work well. I am not a spiritual enough person to capture it in words: it is about grace, shalom (which Van Ness [1986, p. 125] characterizes as "peace as the result of doing justice"). Trish Stewart (1993, p. 49) gets near its evocation when she reports one victim who said in the closing round of a conference: "Today I have observed and taken part in justice administered with love." Psychologists are developing improved ways of measuring spiritualityself-transcendence, meaning in life beyond one's self. So in future it will be possible to undertake systematic research on self-reported spirituality and conferences to see whether results are obtained analogous to Reed's (1986, 1987, 1992) findings that greater healing occurred among terminally ill individuals whose psychosocial response was imbued with a spiritual dimension. For the moment, we must accept an East-West divide in the way participants think about spiritual leadership in conferences. Maori, North American, and Australian Aboriginal peoples tend to think it important to have elders with special gifts of spirituality, what Maori call Mana, attend restorative justice processes (Tauri and Morris 1997, pp. 149-50). This is the Confucian view as well. These traditions are critical of the ethos Western advocates such as myself have brought to conferences, which has not seen it as important to have elders with Mana at conferences. Two years ago in Indonesia I was told of restorative justice rituals in western Sumatra that were jointly conducted by a religious leader and a scholar-the person in the community seen as having the greatest spiritual riches and the person seen as having the greatest riches of learning. My inclination then was to recoil from the elitism of this and insist that many (if not most) citizens have the resources (given a little help with training) to facilitate processes of healing. While I still believe this, I now think it might be a mistake to seek to persuade Asians to democratize their restorative justice practices. There may be merit in special efforts to recruit exemplars of virtue, grace, Mana, to participate. Increasingly, I am tempted to so interpret HeinOnline -- 25 Crime & Just. 25 1999 26 John Braithwaite our experience with RISE in recrUItmg community representatives with grace to participate in drunk-driving conferences where there is no victim. B. Restorative Justice Practices Restore and Satisfy Offenders BetWo than Existing Criminal Justice Practices This section concludes that offender satisfaction with both corporate and traditional individual restorative justice programs has been extremely high. The evidence of offenders being restored in the sense of desisting from criminal conduct is encouraging with victim-offender mediation, conferencing, restorative business regulatory programs, and whole-school anti bullying programs, though not peer mediation programs for bullying. No study has shown restorative justice to make things worse. However, only some of these studies adequately control for important variables and only two randomly assigned cases to restorative versus punitive justice. The business regulatory studies are instructive in suggesting that restorative justice works best when it is backed up by punitive justice in those (quite common) individual cases where restorative justice fails, and that trying restorative justice first increases perceived justice. 1. Fai172ess and Satisfaction for Offenders. As I show in Section VIIE, offenders are more likely to respond positively to criminal justice processing when they perceive it as just. Moore with Forsythe's (1995, p. 248) ethnographic work concludes that most offenders, like victims, experienced quite profound "procedural, material, and psychological justice" in restorative justice conferences. Umbreit (1992) reports from his cross-site study in the United States an 89 percent perception of fairness on the part of offenders with victim-offender mediation programs, compared to 78 percent perceived fairness in unmediated cases. Across four Canadian studies, Umbreit (1999) reports 80 percent offender perception of fairness of victim-offender mediation and 89 percent at two combined English sites. The Ministry of Justice (1994), Western Australia, reports 95 percent offender satisfaction with their restorative justice conference program (Juvenile Justice Teams). McCold and Wachtel (1998, pp. 59-61) report 97 percent satisfaction with "the way your case was handled" and 97 percent satisfaction with fairness in the Bethlehem police conferencing program, a better result than in the four comparisons with Bethlehem cases that went to court. Coats and Gehm (1985, 1989) found 83 percent offender satisfaction with the victim-offender reconciliation experience based on a study of HeinOnline -- 25 Crime & Just. 26 1999 Restorative Justice 27 programs in Indiana and Ohio. Smith, Blagg, and Derricourt (1985), in a limited survey of the initial years of a South Yorkshire mediation project, found that ten out of thirteen offenders were satisfied with the mediation experience and felt that the scheme had helped alter their behavior. Dignan (1990), on the basis of a random sample of offenders (n = 50) involved in victim-offender mediations in Kettering, Northamptonshire, found 96 percent were either satisfied or very satisfied wi th the process. The strongest published result was again on 113 juvenile offenders in the Queensland Department of Justice conferencing program where 98 percent thought their conference fair and 99 percent were satisfied with the agreement (Palk, Hayes, and Prenzler 1998). 2. Reduced Reoffending as Offender Restoration. Pate (1990), Nugent and Paddock (1995), and Wynne (1996) all report a decline in recidivism among mediation cases (as does Neimeyer and Shichor [1995], which is cited in Umbreit [1999]). Umbreit, Coates, and Kalanj (1994) found 18 percent recidivism across four victim-offender mediation sites (n = 160) and 27 percent (n = 160) for comparable nonmediation cases at those sites, a difference that fell short of statistical significance. Similarly, Marshall and Merry (1990, p. 196) report for an even smaller sample that offending declined for victim-offender mediation cases, especially when there was an actual meeting (as opposed to indirect shuttle diplomacy by a mediator), while offending went up for controls. However, the differences were not statistically significant. Schneider (1986, 1990) in an experimental evaluation of six U.S. restitution programs found a significant reduction in recidivism across the six programs. This result is widely cited by restorative justice advocates as evidence for the efficacy of restorative justice. However, all but one of these programs seem to have involved mandated restitution to victims without any mediation or restorative justice deliberation by victims and offenders. The one program that seems to meet the definition of restorative justice here, the Washington, D.C., program, did produce significantly lower rates of reoffending for cases randomly assigned to victim-offender mediation and restitution compared to cases assigned to regular probation. This test is reported in Schneider (1986), but for mysterious reasons Schneider (1990) reports only the nonsignificant differences between before and after offending rates for the control and experimental groups separately, rather than the significant difference between the experimental and control group (which is the relevant comparison). HeinOnline -- 25 Crime & Just. 27 1999 28 John Braithwaite There is no published evidence on the impact of New Zealand family group conferences on recidivism. Maxwell and Morris (1996, p. 106) conclude that the recidivism of their conference cases is "certainly no worse" than recidivism of like offenders prior to the conferencing reforms, but these are rather speculative data. The story is similar with Wagga. Forsythe (1995) shows a 20 percent reoffending rate for cases going to conference compared to a 48 percent rate for juvenile court cases. This is a big effect; most of it is likely a social selection effect of tougher cases going to court, as there is no matching, no controls, though it is hard to account for the entire association in these terms given the pattern of the data (see Forsythe 1995, pp. 245-46). Another big effect with the same social selection worry was obtained with only the first sixty-three cases to go through family group conferences in Singapore. The conference reoffending rate was 2 percent compared to 30 percent over the same period for offenders who went to court (Chan 1996; Hsien 1996). The most determined attempt to tackle social selection problems through randomization, McCold and Wachtel's (1998) experimental evaluation of Bethlehem, Pennsylvania's Wagga-style police conferencing program, ultimately fell victim to another kind of selection effect. For property cases, there was a tendency for conferenced cases to have higher recidivism than court cases, but the difference was not statistically significant. For violence cases, conferenced offenders had a significantly lower reoffending rate than offenders who went to court. However, this result was not statistically valid because the violent offenders with the highest reoffending rate were those who were randomly assigned to conference but who actually ended up going to court because either the offender or the victim refused to cooperate in the conference. In other words, the experiment failed to deliver an adequate test of the effect of conferences on recidivism both on grounds of statistical power and because of unsatisfactory assurance that the assigned treatment was delivered. One conferencing program that has dealt convincingly with the social selection problem is in the Canadian coal-mining town of Sparwood, British Columbia, a Royal Canadian Mounted Police program. For almost three years from the commencement of the program in 1995 until late 1997, no young offender from Sparwood went to court. 8 8 I am indebted to Glen Purdy, a Sparwood lawyer in private practice, for these data. The data until early 1997 are also available at www.titanlink.com. HeinOnline -- 25 Crime & Just. 28 1999 Restorative Justice 29 All were cautioned or conferenced. Three youths who had been conferenced on at least two previous occasions went to court in late 1997. No cases have been to court during 1998 up until the time of writing (October 1998). In the year prior to the program (1994) sixty-four youth went to court. Over the ensuing three years and nine months, this net was narrowed to eighty-eight conferences and three court cases. This was probably not just a net-narrowing effect, however. It may also have been a real reduction in offending. According to police records, compared to the 1994 youth offending rate, the 1995 rate was down 26 percent; the 1996 rate, 67 percent. Reoffending rates for conference cases were 8 percent in 1995, 3 percent in 1996, 10 percent in 1997, and 0 percent for the first nine months of 1998, compared to a national rate of 40 percent per annum for court cases (which is similar in towns surrounding Sparwood). Reoffending rates for Sparwood court cases prior to 1995 have not been collected. \Vhile social selection bias is convincingly dealt with here by the universality of the switch to restorative justice for the first three years, eighty-eight conferences is only a modest basis for inference. Burford and Pennell's (1998) study of a restorative conference-based approach to family violence in Newfoundland found a marked reduction in both child abuse/neglect and abuse of mothers/wives after the intervention. A halving of abuse/neglect incidents was found for thirty-two families in the year after the conference compared to the year before, while incidents increased markedly for thirty-one control families. Pennell and Burford's (1997) research is also a model of sophisticated process development and process evaluation and of methodological triangulation. \Vhile sixty-three families might seem modest for quantitative purposes, it is actually a statistically persuasive study in demonstrating that this was an intervention that reduced family violence. This is because within each case a before-and-after pattern is tested against thirty-one different types of events (e.g., abuse of child, child abuses mother, attempted suicide, father keeps income from mother) where events can be relevant to more than one member of the family. Given this pattern matching of families x events x individual family members, it understates the statistical power of the design to say it is based on only sixty-three cases. Burford and Pennell (1998, p. 253) also report reduced drinking problems after conferences, something I doubt is happening after Canberra conferences. The Newfoundland conferences were less successful in cases where young people were abusing their mothers, a matter worthy of further investigation. HeinOnline -- 25 Crime & Just. 29 1999 30 John Braithwaite Restorative antibullying programs in schools, generally referred to as "whole-school" approaches (Rigby 1996) that combine community deliberation among students, teachers, and parents about how to prevent bullying with mediation of specific cases, have been systematically evaluated with positive results (Farrington 1993; Pepler et al. 1993; Pitts and Smith 1995; Rigby 1996), the most impressive being a program in Norway where a 50 percent reduction in bullying has been reported (Olweus 1993). Gentry and Benenson's (1993) data further suggest that skills for mediating playground disputes learned and practiced by children in school may transfer to the home setting, resulting in reduced conflict, particularly with siblings. The restorative approaches to bullying in Japanese schools, which Masters's (1997) qualitative work found to be a success, can also be read as even more radically "whole school" than the Norwegian innovations. However, Gottfredson's (1997) and Brewer et al.'s (1995) reviews of school peer mediation programs, which simply train children to resolve disputes when conflicts arise among students, showed nonsignificant or weak effects on observable behavior such as fighting. Only one of four studies with quasi-experimental or true experimental designs found peer mediation to be associated with a decrease in aggressive behavior. Lam's (1989) review of fourteen evaluations of peer mediation programs with mostly weak methods found no programs that made violence worse. It appears a whole-school approach is needed that tackles not just individual incidents but that links incidents to a change program for the culture of the school, in particular to how seriously members of the school community take rules about bullying. Put another way, the school must not only resolve the bullying incident; it must use it as a resource to affirm the disapproval of bullying in the culture of the school. With the possible exception of McCold and Wachtel's (1998) statistically nonsignificant increase in property reoffending, a point Walgrave (1993, p. 4) made in 1993 remains true today: however widely one defines restorative justice programs, even including his own group's work on court-ordered community service as restorative justice (Walgrave 1999), after more than thirty studies discussed above, "no research shows an increase in recidivism." However, statistical power, randomization, and control have mostly been weak to very weak in this research. 3. Reduced Reoffending in Corpo1"ate Restorative Justice Programs. In Section I, I recounted how corporate crime researchers like myself be- HeinOnline -- 25 Crime & Just. 30 1999 Restorative Justice 31 gan to wonder if the more restorative approach to corporate criminal law might actually be more effective than the punitive approach to street crime. What made us wonder this? When we observed inspectors moving around factories (as in Hawkins's [1984] study of British pollution inspectors) we noticed how talk often got the job done. The occupational health and safety inspector could talk with the workers and managers responsible for a safety problem and they would fix it. No punishment, not even threats of punishment. A restorative justice reading of regulatory inspection was also consistent with the quantitative picture. The probability that any given occupational health and safety violation will be detected has always been slight and the average penalty for OSHA violations in the post-Watergate United States was $37 (Kelman 1984). So the economically rational firm did not have to worry about OSHA enforcement: when interviewed they would say it was a trivial cost of doing business. Yet there was quantitative evidence that workplace injuries fell after OSHA inspections or when inspection levels increased (Scholz and Gray 1990). The evidence of the impact of Mine Safety and Health Administration inspections in the United States was even stronger that it saved lives and prevented injuries (Lewis-Beck and Alford 1980; Perry 1981a, 1981b; Boden 1983; Braithwaite 1985, pp. 77-84). Boden's data showed a 25 percent increase in inspections was associated with a 720 percent reduction in fatalities on a pooled cross-sectional analysis of 535 mines with controls for geological, technological, and managerial factors; these were inspections at a time when the average penalty for a successful citation was $173 (Braithwaite 1985, p. 3). They were inspections that ended with an "exit conference" that I observed to be often quite restorative. Boden (1983) and the Mine Enforcement and Safety Administration (1977) found no association between the level of penalties and safety improvement, however. This was just the opposite to the picture we were getting from the literature on law enforcement and street crime. On the streets, the picture was of tough enforcement, more police, and more jails failing to make a difference. In coal mines we saw weak enforcement (no imprisonment) but convincing evidence that more inspectors reduced offending (Braithwaite 1985). That book was called To Punish or Persuade: Enforcement of Coal Mine Safety (Braithwaite 1985) and it concluded that while persuasion works better than punishment, credible punishment is needed as well to back up persuasion when it fails. Writing the book was a somewhat emo- HeinOnline -- 25 Crime & Just. 31 1999 32 John Braithwaite tional conversion to restorative justice for me as I came to it as a kind of victim's supporter, a boy from a coal-mining town who wanted to write an angry book for friends killed in the mines. My research also found strong empirical evidence that persuasion works better when workers and unions (representing the victims of the crime) are involved in deliberative regulatory processes. 9 Nearly all serious mine safety accidents can be prevented if only the law is obeyed (Braithwaite 1985, pp. 20-24, 75-77); the great historical lesson of the coal industry is that the way to accomplish this is through a rich dialogue among victims and offenders on why the law is important, a dialogue given a deeper meaning after each fatality is investigated. The shift from punitive to restorative justice in that industry has been considerable and the results considerable. During the first fifty years of mine safety enforcement in Britain (until World War I) there were a number of years when a thousand miners lost their lives in the pits. Fatalities decreased from 1,484 in 1866 to forty-four in 1982-83, after which the British industry collapsed. In the years immediately prior to World War I, the average number of annual criminal prosecutions for coal mine safety offenses in the United Kingdom was 1,309. In both 1980 and 1981, there were none (Braithwaite 1985, p. 4). The qualitative research doing ride-alongs with mine safety inspectors in several countries resolved the puzzle for me. Persuasion worked much of the time; workers' participation in a dialogue about their own security worked. However, the data also suggested that persuasion worked best in the contexts where it was backed by the possibility of punishment. In the United Kingdom during the 1970s, fifty pits were selected each year for a special safety campaign; these pits showed a consistently greater improvement in accident rates than did other British pits (Collinson 1978, p. 77). I found the safety leaders in the industry were companies that not only thoroughly involved everyone concerned after a serious accident to reach consensual agreement on what must be done to prevent recurrence, they also did this after "near accidents" (Braithwaite 1985, p. 67) and they discussed safety audit results with workers 9 For example DeMichiei et al.'s (1982, p. i) comparison of mines with exceptionally high injury rates with matched mines with exceptionally low injury rates found that at the low-injury mines: "Open lines of communication permit management and labor to jointly reconcile problems affecting safety and health; Representatives of labor become actively involved in issues concerning safety, health and production; and Management and labor identify and accept their joint responsibility for correcting unsafe conditions and practices." HeinOnline -- 25 Crime & Just. 32 1999 Restorative Justice 33 even when there was no near-accident. In a remarkable foreshadowing of what we now believe to be reasons for the effectiveness of wholeschool approaches to bullying and family group conferences, Davis and Stahl's (1967, p. 26) study of twelve companies who had been winners of the industry's two safety awards, found one recurring initiative was a "safety letter to families of workers enlisting family support in promoting safe work habits." That is, safety leaders engaged a community of care beyond the workplace in building a safety culture. In To Punish or Persuade, I shocked myself by concluding that after mine disasters, including the terrible one in my home town that had motivated me to write the book, so long as there had been an open public dialogue among all those affected, the families of the miners cared for, and a credible plan to prevent recurrence put in place, criminal punishment served little purpose. The process of the public enquiry and helping the families of the miners for whom they were responsible seemed such a potent general deterrent that a criminal trial could be gratuitous and might corrupt the restorative justice process that I found in so many of the thirty-nine disaster investigations I studied. Joe Rees (1988, 1994) is the scholar who has done most to work through the promise of what he calls communitarian regulation, which we might read as restorative regulatory justice. First Rees (1988) studied the "Cooperative Compliance Program" of the Occupational Safety and Health Administration between 1979 and 1984. OSHA essentially empowered labor-management safety committees at seven Californian sites to take over the law enforcement role, to solve the underlying problems revealed by breaches of the law. Satisfaction of workers, management, and government participants was high because they believed it "worked." It seemed to. Accident rates ranged from one-third lower to one-fifth as low as the Californian rate for comparable projects of the same companies, as the rate in the same project before the cooperative compliance program compared with after (Rees 1988, pp. 2-3). Rees's next study of communitarian regulation was of U.S. nuclear regulation after Three Mile Island. The industry realized that it had to transform the nature of its regulation and self-regulation from a rule book, hardware orientation to one oriented to people, corporate cultures, and software. The industry's CEOs set up the Institute of Nuclear Power Operations (INPO) to achieve these ends. Peers from other nuclear power plants would take three weeks off their own jobs to join an INPO review team that engaged the inspected facility in a HeinOnline -- 25 Crime & Just. 33 1999 34 John Braithwaite dialogue about how they could improve. Safety performance ratings were also issued by the review team; comparative ratings of all the firms in the industry were displayed and discussed at meetings of all the CEOs in the industry and at separate meetings of safety officers. Rees (1994) sees these as reintegrative shaming sessions. Here is an excerpt from a videotape of a meeting of the safety officers: It's not particularly easy to come up here and talk about an event at a plant in which you have a lot of pride, a lot of pride in the performance, in the operators ... It's also tough going through the agonizing thinking of what it is you want to say. How do you want to confess? How do you want to couch it in a way that, even though you did something wrong, you're still okay? You get a chance to talk to Ken Strahm and Terry Sullivan [INPO Vice Presidents] and you go over what your plans are, and they tell you, "No, Fred, you've got to really bare your soul.." .. It's a painful thing to do. (Rees 1994, p. 107) What was the effect of the shift in the center of gravity of the regulatory regime from a Nuclear Regulatory Commission driven by political sensitivities to be tough and prescriptive to INPO's communitarian regulation (focused on a dialogue about how to achieve outcomes rather than rule book enforcement)? Rees (1994, pp. 183-86) shows considerable improvement across a range of indicators of the safety performance of the U.S. nuclear power industry since INPO. Improvement has continued since the completion of Rees's study. For example, more recent World Association of Nuclear Operators data show scrams (automatic emergency shutdowns) per unit declined in the United States from over seven per unit in 1980 to one by 1993 . . In Section II, I showed that shifting nursing home regulation from rule-book enforcement to restorative justice improved regulatory outcomes and that the inspectors who shifted most toward restorative justice improved compliance most (those who used praise and trust more than threat, reintegrative shaming rather than tolerance or stigmatization, those who restored self-efficacy). These results are discussed again when I consider the theories that predict why restorative justice might work better than punitive justice. For the moment, I simply note that communitarian regulation has had considerable documented success in restoring coal mining firms, nuclear power plants, and nursing homes to a more responsible approach to compliance with the law. HeinOnline -- 25 Crime & Just. 34 1999 Restorative Justice 35 Equally, writers such as Gunningham (1995) and Fiona Haines (1997) have shown that there are serious limits to communitarian regulation-rapacious big firms and incompetent little ones who will not or cannot respond responsibly. Deterrence and incapacitation are needed, and needed in larger measure than these regimes currently provide, when restorative justice fails (see also Ayres and Braithwaite 1992; Gunningham and Grabosky 1998). Carol Heimer pointed out in comments on a draft of this essay that "if high-level white collar workers are more likely to get restorative justice, it may be because their corporate colleagues and other members of the society believe that their contributions are not easily replaced, so that offenders must be salvaged" (see Heimer and Staffen 1995). This is right, I suspect, and a reason why justice is most likely to be restorative in the hands of communities of care who can see the value of salvaging the offender and the victim. C. Restorative Justice Practices Restore and Satisfy Communities Better than Existing Criminal Justice Practices In every place where a refonn debate has occurred about the introduction of family group conferences, two community concerns have been paramount: while victims might be forgiving in New Zealand, giving free reign to victim anger "here" will tear at our community; while families may be strong elsewhere, "here" our worst offenders are alienated and alone; their families are so dysfunctional and uncaring that they will not participate meaningfully. But as Morris et al. (1996, p. 223) conclude from perspectives on this question summarized from a number of jurisdictions: "Concerns about not being able to locate extended family or family supporters, to engage families or to effectively involve so-called 'dysfunctional' families, about families fonning a coalition to conceal abuse and about families' failing to honour agreements do not prove to have been well-founded in any of the jurisdictions reported in this book." In his discussion of the Hollow Water experience of using healing circles to deal with rampant sexual abuse of children in a Canadian First Nations community, Ross (1996, p. 150) emphasizes the centrality of restoring communities for restoring individuals: "If you are dealing with people whose relationships have been built on power and abuse, you must actually show them, then give them the experience of, relationships based on respect ... [so] ... the healing process must involve a healthy group of people, as opposed to single therapists. A HeinOnline -- 25 Crime & Just. 35 1999 36 John Braithwaite single therapist cannot, by definition, do more than talk about healthy relationships." The most sophisticated implementation of this ideal that has been well-evaluated is Burford and Pennell's (1998) Family Group Decision Making Project to confront violence and child neglect in families. Beyond the positive effects on the direct objective of reducing violence, the evaluation found a posttest increase in family support, concrete (e.g., babysitting) and emotional, and enhanced family unity, even in circumstances where some conference plans involved separation of parents from their children. The philosophy of this program was to look for strengths in families that were in very deep trouble and build on them. Members of the community beyond the offender and the victim who attend restorative justice processes tend, like offenders, victims and the police, to come away with high levels of satisfaction. In Pennell and Burford's (1995) family group conferences for family violence, 94 percent of family members were "satisfied with the way it was run"; 92 percent felt they were "able to say what was important," and 92 percent "agreed with the plan decided on." Clairmont (1994, p. 28) also reports that among native peoples in Canada, the restorative justice initiatives he reviewed have "proven to be popular with offenders ... and to have broad, general support within communities." The Ministry of Justice (1994), Western Australia, reports 93 percent parental satisfaction, 84 percent police satisfaction, and 67 percent judicial satisfaction, plus (and crucially) satisfaction of Aboriginal organizations with their restorative justice conference program (Juvenile Justice Teams). In Singapore, 95 percent of family members who attended family group conferences said that they benefited personally from the experience (Hsien 1996). For the Bethlehem police conferencing experiment, more parents of offenders were satisfied (97 percent) and likely to believe that justice had been fair (97 percent) than in cases that went to court (McCold and Wachtel, 1998, pp. 65-72). A study by Schneider (1990) found that completing restitution and community service was associated with enhanced commitment to community and feelings of citizenship (and reduced recidivism). While the evidence is overwhelming that where communities show strong social support, criminality is less (Cullen 1994; Chamlin and Cochran 1997), it would be optimistic to expect that restorative justice could ever have sufficient impacts in restoring microcommunities to cause a shift in the macro impact of community on the crime rate (cf. Brown and Polk 1996). But building the micro community of a school or restoring social HeinOnline -- 25 Crime & Just. 36 1999 Restorative Justice 37 bonds in a family can have important implications for crime in that school or that family. Moreover, the restoring of microcommunity has a value of its own, independent of the size of the impact on crime. In the last section I showed how whole-school approaches to bullying can halve bullying in schools. There is a more important point of deliberative programs to give all the citizens of the school community an opportunity to be involved in deciding how to make their school safer and more caring. It is that they make their schools more decent places to be in while one is being educated. There is Australian evidence suggesting that restorative sexual harassment programs in workplaces may reduce sexual harassment (Parker 1998). Again the more important value of these programs than the improved compliance with the law may be about more general improvements in the respect with which women are treated in workplaces as a result of the deliberation and social support integral to such programs when they are effective. I have known restorative justice conferences where supporters of a boy offender and a girl victim of a sexual assault agreed to work together to confront a culture of exploitative masculinity in an Australian school that unjustly characterized the girl as "getting what she asked for" (Braithwaite and Daly 1994). Conversely, I have seen conferences that have missed the opportunity to confront homophobic cultures in schools revealed by graffiti humiliating allegedly gay men and boys (Retzinger and Scheff 1996). After one early New Zealand conference concerning breaking into and damaging the restaurant of a refugee Cambodian, the offender agreed to watch a video of The Killing Fields and "pass the word on the street" that the Cambodian restaurateur was struggling to survive and should not be harassed. A small victory for civil community life perhaps, but a large one for that Cambodian man. One of the most stirring conferences I know of occurred in an outback town after four Aboriginal children manifested their antagonism toward the middle-class matriarchs of the town by ransacking the Country Women's Association Hall. The conference was so moving because it brought the Aboriginal and the white women together, shocked and upset by what the children had done, to talk to each other about why the women no longer spoke to one another across the racial divide in the way they had in earlier times. Did there have to be such an incivility as this to discover the loss of their shared communal life? Those black and white women and children rebuilt that communal life as they restored the devastated Country Women's Association Hall, working together, respectfully once more (for more details on this case, see the Real Justice website http://www.realjustice.org/). HeinOnline -- 25 Crime & Just. 37 1999 38 John Braithwaite One might summarize that the evidence of restorative justice restoring communities is of very small accomplishments of microcommunity building and of modest numbers of community members going away overwhelmingly satisfied with the justice in which they have participated. Maori critics of Pakeha restorative justice such as Moana Jackson (1987) and Juan Tauri (1998) point out that it falls far short of restoring Maori community control over justice. Neocolonial controls from Pakeha courts remain on top of restorative justice in Maori communities. This critique seems undeniable; nowhere in the world has restorative justice enabled major steps toward restoring precolonial forms of community among colonized peoples; nowhere have the courts of the colonial power given up their power to trump the decisions of the indigenous justice fora. At the same time, there is a feminist critique of this indigenous critique of community restoration. I return later to at least one case where male indigenous elders in Canada used control over community justice as a resource in the oppression of women complaining of rape by dominant men. In this case the community was torn asunder to the point of a number of women leaving it. With all the attention given to the microcommunity building of rOutine restorative justice conferences, we must not lose sight of historically rare moments of restorative justice that reframe macrocommunity. I refer, for example, to the release of IRA terrorists from prison so that they could participate in the IRA meetings of 1998 that voted for the renunciation of violent struggle. I refer to much mOre partially successful examples, such as the Camp David mediations of President Carter with the leaders of Egypt and Israel (more partially successful because it excluded the Palestinians themselves), and to mOre completely successful local peacemaking such as that of the Kulka Women's Club in the Highlands of New Guinea (Rumsey 1999). D. Reintegrative Shaming Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices rna te whakama e patu! "Leave him alone, he saying) IS punished by shame." (Maori Crime, Shame and Reintegration (Braithwaite 1989) gives an account of why restorative justice processes ought to prevent crime mOre effec- HeinOnline -- 25 Crime & Just. 38 1999 Restorative Justice 39 tively than retributive practices. The core claims are: tolerance of crime makes things worse; stigmatization, or disrespectful, outcasting shaming of crime, makes crime worse still; while reintegrative shaming, disapproval within a continuum of respect for the offender, disapproval terminated by rituals of forgiveness, prevents crime. In developing the theory, I was much influenced by the restorative nature of various Asian policing and educational practices, by what I saw as the effectiveness of restorative regulatory processes for dealing with corporate crime both in Asia and the West, and by the restorative nature of socialization in Western families that succeed in raising lawabiding children. That child development literature is not reviewed again here. Essentially, what it shows is that both laissez-faire parenting that fails to confront and disapprove of children's misconduct and punitively authoritarian parenting both produce a lot of delinquents; delinquency is less likely when parents confront wrongdoing with moral reasoning (Braithwaite 1989). One implication for restorative justice advocates of this substantial body of empirical evidence is that the justice system will do better when it facilitates moral reasoning by families over what to do about a crime as an alternative to punishment by the state. Restorative justice conferences work by inviting victims and supporters (usually family supporters) of the victim to meet with the offender and the people who care most about the offender and most enjoy the offender's respect (usually including both the nuclear and extended family, but not limited to them). This group discusses the consequences of the crime, drawing out the feelings of those who have been harmed. Then they discuss how that harm might be repaired and any steps that should be taken to prevent reoffending. Attendance of over forty people can occur, but average attendance (beyond the offender) reported is six in New Zealand (Robertson 1996), six in Victoria (Ban 1996), five in Bethlehem (McCold and Wachtel 1998, p. 30), eight in Canberra (unpublished RISE data), and twenty-three in Manitoba (Longclaws, Galaway, and BarkwellI996). Wachtel (1997, p. 73) reports a five-hour conference in Pennsylvania with an attendance of seventy-five. In terms of reintegrative shaming theory, the discussion of the consequences of the crime for victims (or consequences for the offender's family) structures shame into the conference; the support of those who enjoy the strongest relationships of love or respect with the offender structures reintegration into the ritual. It is not the shame of police or HeinOnline -- 25 Crime & Just. 39 1999 40 John Braithwaite judges or newspapers that is most able to get through to us; it is shame in the eyes of those we respect and trust. These are not new ideas. They have existed for hundreds of years in Maori philosophies of justice. Maori thought about whanau conferences repeatedly use the words shame (whakama) and healing in equivalent ways to my use of shaming and reintegration. In Maori thinking, it is the shame of letting one's extended family down that is a particularly important type of shame to discuss. The advantage of this sort of shame over the individual guilt/shame one is expected to experience as one stands alone in the dock of Western justice is that it is readily transcended when family members extend forgiveness to the offender. Evidence from 548 adult and juvenile cases randomly assigned to court versus conference in Canberra, Australia, is that offenders both report and are observed to encounter more reintegrative shaming in conferences than in court, that conference offenders experience more remorse and more forgiveness than court offenders, and are more likely to report that they have learnt from the process that there are people who care about them (Sherman and Strang 1997 a). Eighty percent of conference offenders compared to 40 percent of court offenders said after the process that they felt they had repaid their debt to victims and to society. Another rwo years of data collection are required in this study before we know whether reoffending was less in the cases where reintegrative shaming was experienced. Harris and Burton's (1997) work at least shows that reliable observational measurement of reintegrative shaming is possible: ratings of how much reintegrative shaming occurred in forty-five conferences and court cases by independent raters agreed berween 67 percent and 93 percent of the time. Makkai and Braithwaite's (1994) test of the theory in the domain of compliance of Australian nursing homes with quality of care standards has the attraction of test-retest reliabilities of the measure of compliance with the law berween .93 and .96, obtained by having an independent inspector check compliance. Makkai and Braithwaite found that homes checked by inspectors with a reintegrative shaming philosophy experienced improved compliance with the law in a follow-up inspection rwo years later. Nursing homes inspected by stigmatizing inspectors suffered an equivalent drop in compliance rwo years later, while homes checked by tolerant and understanding inspectors suffered an intermediate fall in compliance. Lu (1998) has produced a different kind of encouraging data on the validity of the theory of reintegrative HeinOnline -- 25 Crime & Just. 40 1999 Restorative Justice shaming in the very different context of community justice Shanghai. 41 III E. Procedural Justice Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices The idea of reintegrative shaming is that disapproval is communicated within a continuum of respect for the offender. A key way to show respect is to be fair, to listen, to empower others with process control, to refrain from bias on the grounds of age, sex, or race. More broadly, procedural justice communicates respect (Lind and Tyler 1988; Tyler 1990). Conferences do not have all the procedural safeguards of court cases; yet the Optimistic Account predicts offenders and victims will find them fairer. Why? Conferences are structurally fairer because of who participates and who controls the discourse. Criminal trials invite along those who can inflict maximum damage on the other side; conferences invite those who can offer maximum support to their own side, be it the victim side or the offender side. In other words those present are expected to be fair and therefore tend to want to be fair. They tend not to see their job as doing better at blackening the character of the other than the other does at blackening theirs. Citizens are empowered with process control, rather than placed under the control of lawyers. In the study of nursing home regulation discussed above, Makkai and Braithwaite (1996) found that of the various facets of procedural justice, perceived process control on the part of citizens is the one that predicts subsequent compliance with the law. Other research suggests other dimensions of procedural justice may be important, however. For example, in the Milwaukee domestic violence experiment (Bridgeforth 1990, p. 76), "arrestees who said (in lockup) that police had not taken the time to listen to their side of the story were 36 percent more likely to be reported for assaulting the same victim over the next 6 months than those who said the police had listened to them" (Sherman 1993, p. 463; see also Paternoster et al. 1997). More broadly, in Why People Obey the Law, Tyler (1990) found that citizens were more likely to comply with the law when they saw themselves as treated fairly by the criminal justice system. Sherman (1993) has reviewed further more recent supportive evidence on this question. The key questions are whether citizens feel they are treated more fairly in restorative justice processes than in courts and whether they are more likely to understand what is going on. The answer seems HeinOnline -- 25 Crime & Just. 41 1999 42 John Braithwaite clearly to be yes. Early results from the Canberra conferencing experiment show that offenders are more likely to understand what is going on in conferences than in court cases, felt more empowered to express their views, had more time to do so, were more likely to feel that their rights were respected, to feel that they could correct errors of fact, and to feel that they were treated with respect, and were less likely to feel in conferences that they were disadvantaged due to "age, income, sex, race, or some other reason" (Sherman and Barnes 1997; Sherman et a1. 1998). Without the randomized comparison with court, a number of other studies have shown absolutely high levels of citizen satisfaction with the fairness of restorative justice processes (Sec. VIIA-C). Given that there is now strong evidence that restorative justice processes are perceived to be fairer by those involved and strong evidence that perceived procedural justice improves compliance with the law, the Optimistic prediction follows that restorative justice processes will improve compliance with the law. F. The Theory of Bypassed Shame Predicts that Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Scholars working in the affect theory tradition of Sylvan Tomkins (1962), most notably Donald Nathanson (1992) and David Moore (with Forsythe 1995), have a theoretical perspective on why restorative justice should reduce crime based more on the nature of shame as an affect than on shaming, reintegration, and stigmatization as practices. According to this perspective, shame can be a destructive emotion because it can lead one to attack others, attack self, avoid, or withdraw (Nathanson's [1992] compass of shame). All of these are responses that can promote crime. A profound deficiency of Braithwaite's (1989) theory is that it is just a theory of shaming, with the emotion of shame left undertheorized. From this perspective, therefore, a process is needed that enables offenders to deal with the shame that almost inevitably arises at some level when a serious criminal offense has occurred. Denial, for example being "ashamed to be ashamed," in Scheff's words, is not an adaptive response. Shame is a normal emotion that healthy humans must experience; it is as vital to motivating us to preserve social bonds essential to our flourishing as is fear to motivating us to flee danger. Indeed Scheff (1990, 1994), Retzinger (1991), and Scheff and Retzinger (1991) finger bypassed shame as the culprit in the shame-rage spirals that characterize our worst violence domestically and internationally. HeinOnline -- 25 Crime & Just. 42 1999 Restorative Justice 43 The evidence these authors offer for the promotion of anger through bypassed shame is voluminous but of a quite different sort from the more quantitative evidence adduced under the other propositions in this section of the essay. It consists primarily of collections of clinical case notes (preeminently.Lewis's 1971 research) and microanalyses of conversations (preeminently Retzinger's 1991 marital quarrels). Yet the thrust of this work is also supported by Tangney's (1995) review of quantitative studies on the relationship between shame and psychopathology: Guilt about specific behaviors, "uncomplicated by feelings of shame about the self," is healthy. The problem is "chronic self-blame and an excessive rumination over some objectionable behavior" (Tangney 1995, p. 1141). Scheff and Retzinger take this further, suggesting that shame is more likely to be uncomplicated when consequences that are shameful are confronted and emotional repair work is done for those damaged. Shame will become complicated, chronic, more likely to descend into rage if it is not fully confronted. If there is nagging shame under the surface, it is no permanent solution to lash out at others with anger that blames them. Then the shame and rage will feed on each another in a shame-rage spiral. Consistent with this analysis, Ahmed (1999) has shown in a study of bullying among twelve hundred Canberra schoolchildren that bullies deal with shame through bypassing it, victims acknowledge and internalize shame so that they suffer persistent shame, while children who avoid both bullying and being victimized by bullies have the ability to acknowledge and discharge shame so that shame does not become a threat to the self. According to Retzinger and Scheff's work, if we want a world with less violence and less dominating abuse of others, we need to take seriously rituals that encourage approval of caring behavior so that citizens will acquire pride in being caring and nondominating. With dominating behavior, we need rituals of disapproval and acknowledged shame of the dominating behavior, rituals that avert disapproval-unacknowledged shame sequences. Retzinger and Scheff (1996) see restorative justice conferences as having the potential (a potential far from always realized) to institutionalize pride and acknowledged shame that heals damaged social bonds. Circles in this formulation are ceremonies of constructive conflict. When hurt is communicated, shame acknowledged by the person who caused it, respect shown for the victim's reasons for communicating the hurt, and respect reciprocated by the victim, constructive conflict has occurred between victim and offender. It may be that in the "abused spouse syndrome," for example, shame is HeinOnline -- 25 Crime & Just. 43 1999 44 John Braithwaite bypassed and destructive, as a relationship iterates through a cycle of abuse, manipulative contrition, peace, perceived provocation, and renewed abuse (see Retzinger 1991). Crime wounds, justice heals; but only if justice is relational (Burnside and Baker 1994). Moore with Forsythe (1995, p. 265) emphasize that restorative justice should not, in the words of Gypsy Rose Lee, accentuate the positive and eliminate the negative; rather it should accentuate the positive and confront the negative. Tomkins (1962) adduces four principles for constructive management of affect: "(1) That positive affect should be maximized. (2) That negative affect should be minimized. (3) That affect inhibition should be minimized. (4) That power to maximize positive affect, to minimize negative affect, and to minimize affect inhibition should be maximized." Nathanson (1998, p. 86) links this model to an hypothesized capacity of restorative justice processes to build community, where community is conceived as people linked by scripts for systems of affect modulation. Community is built by: "(1) Mutualization of and group action to enhance or maximize positive affect; (2) Mutualization of and group action to diminish or minimize negative affect; (3) Communities thrive best when all affect is expressed so these first two goals may be accomplished; (4) Mechanisms that increase the power to accomplish these goals favor the maintenance of community, whereas mechanisms that decrease the power to express and modulate affect threaten the community." In the most constructive conflicts, shame will be acknowledged by apology (reciprocated by forgiveness) (Tavuchis 1991). Maxwell and Morris (1996) found in New Zealand family group conferences that the minority of offenders who failed to apologize during conferences were three times more likely to reoffend than those who had apologized. Interpreting any direction of causality here is admittedly difficult. Moore (1994, p. 6) observes that in courtroom justice shame is not acknowledged because it is "hidden behind impersonal rhetoric about technical culpability." Both Moore with Forsythe (1995) and Retzinger and Scheff (1996) have applied their methods to the observation of restorative justice conferences, observing the above mechanisms to be in play and to be crucial to shaping whether conferences succeed or fail in dealing with conflicts in ways that they predict will prevent crime. For Retzinger and Scheff (1996), conferences have the ostensible purpose of material reparation; but underlying the verbal and visible process of reaching HeinOnline -- 25 Crime & Just. 44 1999 Restorative Justice 45 agreement about material reparation is a more nonverbal, less visible process of symbolic reparation. It is the latter that really matters according to their theoretical framework, so the emphasis in the early restorative justice literature on how much material reparation is actually paid becomes quite misguided. The evidence now seems strong that bypassed shame contributes to violence; Sherman and Barnes's (1997) and Sherman et al.'s (1998, pp. 127-29) admittedly preliminary evidence suggests that in conferences offenders may accept and discharge shame more than when they go through court cases. If both propositions are correct, conferences might do more to reduce crime than court cases. G. D.efiance Theory Predicts that Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices "Disrespect begets disrespect," claims Howard Zehr (1995), and few things communicate disrespect as effectively as the criminal exploitation of another human being. Sherman (1993) has woven the propositions from Subsections D-F about procedural justice, the social bonds that render shaming reintegrative and bypassed shame into an integrated theory of defiance. It has three propositions: 1. Sanctions provoke future defiance of the law (persistence, more frequent or more serious violations) to the extent that offenders experience sanctioning conduct as illegitimate, that offenders have weak bonds to the sanctioning agent and community, and that offenders deny their shame and become proud of their isolation from the sanctioning community. 2. Sanctions produce future deterrence of law-breaking (desistance, less frequent or less serious violations) to the extent that offenders experience sanctioning conduct as legitimate, that offenders have strong bonds to the sanctioning agent and community, and that offenders accept their shame and remain proud of solidarity with the community. 3. Sanctions become irrelevant to future law breaking (no effect) to the extent that the factors encouraging defiance or deterrence are fairly evenly counterbalanced. (Sherman 1993, pp. 448-49) Sherman hypothesizes that restorative justice processes are more likely to meet the conditions of proposition 2 than traditional punitive processes. The evidence to date supports this. We have already seen that restorative processes are accorded high legitimacy by citizens, that they are better designed to empower those with strong bonds with the of- HeinOnline -- 25 Crime & Just. 45 1999 46 John Braithwaite fender, and that they outperform court in inducing the acknowledgement and discharging of shame for wrongdoing. While Sherman (1993) reviews some suggestive evidence that law breaking might vary under the conditions that are hypothesized to vary defiance, a systematic test of defiance theory remains to be undertaken. Results from the RISE experiment are still very preliminary here, only laying the foundations for the test of this theory. One published result encouraging to defiance theory, however, is that while 26 percent of drunk drivers randomly assigned to court felt bitter and angry after court, only 7 percent of offenders felt bitter and angry after a conference (Sherman and Strang 1997 b). Hagan and McCarthy (1997, pp. 191-97) have tested Sherman's defiance theory against the prediction that children who have been humiliated, treated unfairly, and had bonds severed by virtue of being victims of sexual abuse or physical violence (with bruising or bleeding) will have their criminal behavior amplified by traditional criminal justice processing more than offenders who have not been abused. Their data, collected among homeless children in Toronto and Vancouver, supported the defiance theory prediction. H. Self-Categorization Theory Predicts that Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices Self-categorization theory (Turner et al. 1987) explains the conditions under which a social self-concept or social identity becomes salient through individuals categorizing their self as having a similar identity to that shared by various social groups. These emergent identities shape what we are and how we act. I act the way I do because I am an Australian, male, a criminologist, a consumer advocate, a republican, and so on. According to self-categorization theory, it is group identities that matter more than group interaction. I do not have to spend time going to Australian Republican Movement meetings for my identity as a republican to affect how I act. The notion of group influence is therefore different in emphasis from that proposed in the theory of reintegrative shaming, which emphasizes interdependence. Like most criminological theories, Braithwaite's (1989) is sloppily theorized on this question, slipping back and forth between interaction-based and identity-based accounts of how criminal subcultures influence action. This is true of Sutherland's (1983) theory as well, headlined as a theory of differential association, it actually defines differential association cognitively rather than inter- HeinOnline -- 25 Crime & Just. 46 1999 Restorative Justice 47 actively: "An excess of definitions favorable to violations of law over definitions unfavorable to violation of law" (Sutherland and Cressey 1978, p. 81). Albert Cohen's (1955) subcultural theory is more incipiently a selfcategorization theory than other classic criminological theories. For Cohen, children who fail in the status system of the school have a status problem. They can solve that status problem by identifying with other groups that invert the values of the school. If the school values being "square," there is attraction to being "cool," feeling membership in a cool group. If the school values control of aggression, then there is attractiveness in a group that values free expression of aggression. While there is evidence that children experience Cohen's reaction formation (Koh 1997), there is more evidence in more contexts for Matza's (1964) view that delinquents drift between law-supportive and law-neutralizing identities, though some studies do not find a lot of drift away from law-supportive identities among delinquents (Box 1981, pp. 107-8; Ball 1983; Thurman 1984; Agnew and Peters 1986; Anderson 1999; Koh 1997). Sykes and Matza (1957) have suggested five techniques of neutralization that make drift possible: denial of victim ("We weren't hurting anyone"); denial of injury ("They can afford it"); condemnation of the condemners ("They're crooks themselves"); denial of responsibility ("I was drunk"); and appeal to higher loyalties ("I had to stick by my mates"). Restorative justice conferences may prevent crime by facilitating a drift back to law-supportive identities from law-neutralizing ones. How might they accomplish this? At a victim-offender mediation or conference when the victim is present, it is hard to sustain denial of victim and denial of injury. In contrast, these techniques of neutralization are fostered by criminal justice institutions that sustain separations of victims and offenders. Admittedly, victims often do not convince the offender in a conference that they were hurt in a way they could ill afford. Yet when this occurs, victim supporters will often move offenders through the communicative power, the authenticity, that comes from their love of the victim. An upset daughter explaining how frightened her mother now is in her own house can have a more powerful impact on the offender than direct expressions of concern by the victim. Condemnation of the condemners is also more difficult to sustain when one's condemners engage in a respectful dialogue about why the criminal behavior of concern to them is harmful. Katz, Glass, and Co- HeinOnline -- 25 Crime & Just. 47 1999 48 John Braithwaite hen's (1973) research shows that outgroup derogation is the preferred way of handling shame when the victim is a member of an outgroup. Conferences and healing circles are designed to make the condemners members of an in-group rather than an outgroup by two moves: inviting participants from all the in-groups that matter most to offenders; encouraging victims and victim supporters to be respectful, even forgiving, of them as a person thus rendering their outgroup location more ambiguous. One of the advantages of the presence of victim supporters is that if the victim is irrevocably a member of an outgroup, the consequences of the crime might be effectively communicated by a victim supporter who happens to be a member of an in-group. The evidence is that the transience of in- and outgroup categorizations is contextually responsive to variables like politeness and respectfulness, the very modes of interaction restorative justice processes seek to nurture (Turner et al. 1987, pp. 55-56). From a self-categorization perspective, an advantage of Chinese social structuring is the relative lack of clear boundaries in defining an in-group, for example in the elastic definition of Chia or family, depending on the problems at issue (Bond and Wang Sung-Hsing 1983, p. 68). Denial of responsibility is tested at a conference. The presence of supporters who know and care for an offender risk that a denial of responsibility like "I was drunk" might lead to a discussion of his responsibility for recurrent drunkenness that has induced irresponsible behavior in the past. Obversely, criminal trials only test those denials of responsibility legally relevant to mitigating guilt. Even for that legally relevant subset of the psychologically relevant denials, they are tested in ways that are least likely to be persuasive to the offender-by attacking his credibility as a person in the eyes of a judge or jury. The restorative conference supports him as a person while questioning the usefulness of his denials to him as a person and to clearing things up for those who have been hurt. The restorative process, by showing a path to redemption, provides an alternative to denial. This contrasts with the two paths the court proffers-guilt and punishment or innocence and impunity-a choice that makes denial an attractive posture. Criminal offenders are criminal offenders partly because they are good at denial. When a shaft of shame is projected across the room from victim to offender, the offender may have a shield that deflects the shame, only to find the deflected shame spears through the heart of his mother who quietly sobs beside him. What I have observed in many conferences is that it may then be mother's or father's or sister's HeinOnline -- 25 Crime & Just. 48 1999 Restorative Justice 49 shame that gets behind his shield of denial. This only happens, of course, when he loves one of these intimates. Appeals to higher loyalties like loyalties to one's mates is the technique of neutralization of greatest interest from a self-categorization perspective. Emler and Reicher's (1995) interviews with delinquents reveal that they are simultaneously concerned about having a reputation for whatever it is their delinquent group values (say toughness) while being concerned about maintaining a different reputation with their families. Their delinquents worked hard at keeping families unaware of the different values and conduct they manifest in the delinquent group. Delinquents' parents rarely met their peers. Delinquents were more likely than nondelinquents to keep peers and parents apart (Emler and Reicher 1995, p. 204). Koh (1997, p. 376) found that incarcerated Singaporean delinquents endorsed neutralization techniques to a lesser extent when their family identity was salient and when confrontation with authority was seen to be public rather than private. Goffman (1956) is the preeminent theorist of what he calls strategies for matching audience segregation to role segregation. In the nineteenth century village, all our roles were played out for the same audience. The condition of modernity, however, is of a proliferation of group identities-mother, criminologist, golfer, Christian, cat breeder-but where those groups are scattered across global space. Most of us are actually not more alone in the modern city; but our togetherness is not unified with place (Braithwaite 1993a). This means, as Benson (1989) shows empirically, that the white-collar criminal in the contemporary world is peculiarly vulnerable to shame if only his business activities might be revealed to his church group. Restorative justice conferences are designed to do just this-to bring together the audiences the criminal would most want to be segregated. This design can and does backfire. On rare occasions, we have had restorative justice conferences in Australia where a delinquent gang, or two rival gangs on the victim and offender sides, have dominated the conference numerically and persuasively (in neutralizing shame). On many occasions, we have observed adult restorative justice conferences for drunk driving where the offender's drinking group has dominated the conference with denials of victim, of injury, of responsibility (Mugford and Inkpen 1995). Overall, my observation from sitting through more than a hundred conferences of different types is that such cases are in the minority. Why? One reason is Matza (1964) was right that drift toward and away HeinOnline -- 25 Crime & Just. 49 1999 50 John Braithwaite from rejection of the law's moral bind is more common than outright rejection of moral commitment to the law. For example, while parents of serious delinquents are more likely to have been delinquents themselves (Wilson and Herrnstein 1985, pp. 95-103), they are not Fagins. Criminal parents almost always disapprove of their children's delinquency (West and Farrington 1973, p. 116). Even when we put together a conference dominated by multiproblem families concerning a violent offense, we find empirically that very few of the utterances are approving of violence. One reason for this is that philosophers in the Aristotelian tradition of truth-finding through undominated dialogue, like Habermas (1996), are right that the closer we get to conditions of undominated speech, the more overwhelmingly it will turn out to be the case that evils such as violence will be near-universally condemned. That is, there is a moral fact of the matter that gratuitous violence is wrong and undominated dialogue will converge on consensus about contextual judgments of the wrongness of specific violent acts. A nice moral feature of restorative justice from this perspective is that restorative justice might only work with crimes that ought to be crimes. If a group of citizens cannot agree in an undominated conference that an act of obscenity is wrong, then the obscenity should not be a crime; and the conference will fail in controlling obscenity. But the fact of the matter is that most criminal offenses brought to justice in democratic societies are more like the violence case than the obscenity case: they are unambiguously wrong to most citizens attending a conference. Put another way, when a victim comes to a conference with a broken nose, denial of victim and denial of injury are likely to be revealed as bad arguments. From a Habermasian perspective, techniques of neutralization for violence can only be sustained by avoiding undominated dialogue about their justice. Restorative justice breaks through that avoidance. The social psychological research literature supports the interpretation that self-interested egotistical neutralizations are vulnerable to group dialogue: "In situations without strong social bonds [courtrooms?], people are egotistical. Once a group identity is created, however, people are increasingly responsive to group-centered motives" (Tyler and Dawes 1993, p. 102). The challenge for circles is to forge a common group identity in the face of the other identities that divide them; they are a group committed to achieving restoration. Of course, circles are never free of domination, so the degree of truth of the Habermasian analysis is contingent. However, some of the HeinOnline -- 25 Crime & Just. 50 1999 Restorative Justice 51 ineradicable dominations of social life systematically conduce to lawabiding in-groups having more power in the long haul than law-neutralizing ones, at least with juveniles. It is well documented that delinquency declines beyond a certain age; one reason is that collective support for delinquency declines from about age sixteen (Emler and Reicher 1995). However much delinquent peer groups dominate a young person, she is not unaware that these peers are not going to be around forever; she knows that when they go off the scene, family will still be there lending money, caring, giving emotional support. At least she knows this in those cases where the conference facilitator has succeeded in getting to the conference communities of care (including non family ones) beyond the delinquent peer group who will stick by the offender in the long haul. Where the offender is so dominated by a delinquent peer group that the longer term nature of family bonds does not trump this domination, a restorative justice strategy still has time on its side. Empirically, the peer group is more likely to disintegrate between ages thirteen and twenty than the family. Very few of the gang members in Esbensen and Huizinga's (1993) Denver survey reported being in a delinquent gang for more than one year. Many members indicated that they would like not to be members and expected to leave the gang in future. If we just hang in with one unsuccessful conference after another in which delinquent peers dominate family, eventually the balance will shift in the other direction. Restorative justice rewards the patient. As Siti Hamidah of the Association of Muslim Professionals said of Singaporean conferences: "Many want to change but don't know how, so it's a time to make concrete plans, like returning to school or finding a job" (Hsien 1996). It is often the case in the short term that peer influences dominate family influence because though the delinquent group "is characterized by a lack of intimacy or affection, there is a strong sense of belongingness" (Koh 1997, p. 201). Yet where that belongingness is grounded in its provision of an alternative status system to the status system of a school that fails them, removal of the original cause by dropping out of school may undermine a belongingness so grounded. Indeed, there is evidence of reduced delinquency following school drop-out (Elliott and Voss 1974). An unattractive way of applying the lessons of self-categorization theory to restorative justice would be to exclude delinquent peers from the conference, or to exclude drinking mates in the difficult case of HeinOnline -- 25 Crime & Just. 51 1999 52 John Braithwaite the shameless Aussie drunk driver. There is little point persuading a delinquent during an hour stacked with the law abiding when she will spend the next thousand hours in a world surrounded by the law violating. Better to confront the whole delinquent group or the whole drinking group with the indefensibility of their techniques of neutralization. Better to win the conscience of the delinquent in the presence of his delinquent peers than to win a Pyrrhic victory in their absence. VVhat one must guard against, however, is allowing a law-neutralizing group to dominate a conference. VVhere the law-neutralizing group is strong, a lot of work is needed to balance them with a plurality of lawabiding citizens who also enjoy the respect and trust of the offender (Mugford and Inkpen 1995). Ross (1996, p. 182) finds special virtue in the participation of healed victims and healed victimizers of sexual abuse who can cut through the (often shared) neutralizations that they had to cut through in confronting their own abuse: In Hollow Water, ex-offenders are not shunned forever, but seen as important resources for getting under the skin of other offenders and disturbing the webs of lies that have sustained them. Better than anyone, they understand the patterns, the pressures and the ways to hide. As they tell their personal stories in the circle, they talk about the lies that once protected them and how it felt to face the truth about the pain they caused. It is done gently but inflexibly, sending signals to offenders that their behaviour has roots that can be understood, but that there are no such things as excuses. (Ross 1996, p. 183) Indeed, at Hollow Water, before they met their own victim in a healing circle, sexual abusers met other offenders and other offenders' victims, who would simply tell their stories as a stage in a process toward breaking down the tough-guy identity that pervaded the dominating relationship with their own victim. Note what an interesting strategy this is from a defiance theory perspective as well. Averting defiance is about getting offenders to put their caring identity rather than their defiant self in play. I can summarize by suggesting that self-categorization theory might be read to make the following predictions about restorative justice: 1. Restorative justice prevents crime when (a) justice rituals are structured so that condemners are harder to condemn because they are members of an in-group, (b) if condemners are irrevocably members HeinOnline -- 25 Crime & Just. 52 1999 Restorative Justice 53 of an outgroup, condemnation still influences intermediaries who are in-group members present at the conference (who can pass that influence on to the offender), (c) discussion of consequences reveals that denial is a coping strategy that blocks in-group acceptance, (d) justice rituals break down the segregation of law-abiding and law-neutralizing in-groups in circumstances where the law-abiding groups will (i) have more persuasive arguments to the extent that speech is undominated and (ii) be more dominant to the extent that speech is dominated. 2. Restorative justice will more often achieve conditions a-d than traditional trials because trial lawyers have a trained competence at exaggerating evil, condemning condemners, denying victim, denying injury, and denying responsibility, at blackening grey and whitening brown, in short in consolidating offenders and victims into opposed out-groups. 1. Crime Prevention Theory Predicts That Restorative Justice Practices Reduce Crime More than Existing Criminal Justice Practices Lon Fuller (1964, p. 33) suggests that only two types of problems are suited to full judicial-legal process: yes-no questions such as "Did she do it?" and more-less questions such as "How much should be paid?" Polanyi (1951, pp. 174-84) distinguishes polycentric problems from these. They require reconciliation of complex interacting consequences of multidimensional phenomena. Polycentric problems are not well suited to the judicial model. Because most crime problems beyond the determination of guilt are polycentric, courts are rather ineffective at preventing crime. In response to the recognition that courts cannot be expected to be competent at crime prevention, crime prevention has expanded as a largely police-facilitated alternative to expending criminal justice resources on dragging cases through the courts. From a restorative justice perspective, an uncoupling of crime prevention from case processing amounts to lost opportunity in two ways. First, every police officer knows that the best time to persuade a householder to invest in security is after a burglary; every business regulator knows the best time to persuade a company to invest in a corporate compliance system is after something goes wrong and someone gets into trouble. They also know that they do not have the resources to get around and persuade all households and all businesses to invest in security or compliance systems. Given that the police or the regulator must make contact with victims and offenders when an offense is cleared, it is a suboptimal HeinOnline -- 25 Crime & Just. 53 1999 54 John Braithwaite use of resources not to seize that opportunity for crime prevention. Moreover, it brings finite crime-prevention resources to bear at the moment when motivation for implementing demanding preventive measures is at its peak and at its peak for good reason: one study has shown prospects of another burglary four times as high as in houses that had not been burgled before (Bridgeman and Hobbs 1997, p. 2). Hence, a project in Huddersfield that focused resources such as temporary alarms on prior victims reduced domestic burglary by 24 percent; in a Rockdale project by 72 percent (Bridgeman and Hobbs 1997, p. 3). Focusing crime prevention on existing cases of victimization (Pease 1998) also mainstreams crime prevention to where the resources are-street-level enforcement-rather than leaving crime prevention ghettoized in specialist areas. This of course is not to deny that there will always be circumstances where crime prevention is best deployed before any offense occurs. Restorative justice resolves the tension between the incapacity of the court for polycentric problem-solving and the imperatives for mainstreaming crime prevention into case management. It also resolves the most fundamental tension between crime prevention theory and practice. The theory says "involve the community"; the practice says "citizens don't turn up to neighborhood watch meetings except in highly organized communities that don't need them." I don't go to neighborhood watch meetings, even though I think I should. But if the kid next door gets into trouble, if my secretary is a victim, and they ask me to attend to support them, I attend. I am touched by the invitation, that they have chosen me as one whose support they value in a time of stress. Corruption and capture are worries with problem-oriented policing that leaves discretion totally with law enforcement agencies to decide the preventive measures required. This is especially true with business regulation-be it police regulating prostitution or drug markets or antitrust agencies regulating competition policy. Ayres and Braithwaite (1992, chap. 3) have shown game-theoretically and in terms of republican theory how transforming the crime prevention game from a bipartite game between state and business into a tripartite state-businesscommunity game prevents corruption and capture. "Community" is the ingredient needed to prevent the crimes that arise from crime prevention; and restorative justice may deliver community to deliberative forums better than any strategy yet attempted. At the same time, abuse of police powers in mainstream processes of arrest is rendered account- HeinOnline -- 25 Crime & Just. 54 1999 Restorative Justice 55 able to community when a mother complains during a conference that the police used unnecessary force on her son. I have observed mothers do this in conferences (because they are polycentric) but not in courtrooms (because they are not). IO Crime prevention is a preeminently important area of criminal justice practice and evaluation research, but a theoretical backwater. In some respects this is a good thing because one should want prevention practitioners not to be theoretically committed, to be interpretively flexible, searching to read situations from the different angles illuminated by multiple theories. Plural understandings of a crime problem stimulate a disparate range of action possibilities that can be integrated into a hedged, mutually reinforcing package of preventive policies (Braithwaite 1993b). Plural understandings are best generated out of a dialogue between crime prevention professionals, such as police, and community members with disparate perspectives from their direct experience with the problem phenomenon. In the discussion of the CML case in Section IV, a disparate array of preventive measures was discovered grounded in the different kinds of theories the rich plurality of players involved in this restorative justice process came up with-theories of education, deterrence, incapacitation, rehabilitation, target hardening, moral hazard, adverse publicity, law, regulation, and opportunity theory. Restorative justice rituals can be a lever for triggering prevention of the most systemic and difficult-to-solve crimes in contemporary societies, like sexual abuse in families (Hollow Water), like the crimes of finance capital (CML). We should take seriously the possibility of family group conferences with leaders of Colombian cocaine cartels. How do we know they are beyond shame? How do we know that they would not like to retire at seventy instead of fear violent usurpation by a rival. Even common thieves retire because they find managing a criminal identity takes its toll: "You get tired. You get tired trying to be a tough guy all the time. People always expecting this and that" (Shover 1996, p. 137). How do we know that organized crime bosses might not find very attractive an agreement that allowed them to pass on some of their wealth to set up legitimate businesses for their children so they did not need to bequeath to them the life they had led (see Rensselaer 10 There is another reason. Mothers do not complain in court against the police for the same reason their sons do not-because legal aid lawyers in Australia are fairly systematic in warning clients that complaining about the police is likely to backfire in a way that leads to a longer sentence. HeinOnline -- 25 Crime & Just. 55 1999 56 John Braithwaite 1992)? How do we know that they do not actually hate killing other human beings in order to survive themselves? An incipient and only very partially successful model here is the Raskol gang surrenders and gang retreats in Papua New Guinea that have involved surrenders of up to four hundred alleged gang members (Dinnen 1996). In summary, restorative justice can remove crime prevention from its marginal status in the criminal justice system, mainstreaming it into the enforcement process. It can deliver the motivation and widespread community participation crime prevention needs to work and to protect itself against corruption and capture by organized interests (including the crime prevention industry itself). It can sometimes deliver the political clout to crime prevention that it needs to tackle systemic problems systemically. J. Restorative Justice Practices Deter Crime Better than Criminal Justice Practices Grounded in Deterrence Theory Bentham would be disappointed at the current state of the evidence on how well deterrence works (Sherman et al. 1997). I do not review here the vast literature on the limited effectiveness of criminal punishments as deterrents. In another essay (Braithwaite 1997), I have reviewed some of the reasons why deterrence does not work as well as it ought. Deterrence is shown to fail as a policy not so much because it is irrelevant (though it is for many) but because the gains from contexts where it works are cancelled by the losses from contexts where it backfires. Evidence surrounding Brehm and Brehm's (1981) theory of psychological reactance is particularly instructive. The theory of reactance asserts that intentions to control are reacted to as attempts to limit our freedom, which lead us to reassert that freedom by acting contrary to the direction of control. Reactance is found to be greatest for those who care most about the freedom. This insight motivates a fundamental reframing of deterrence theory. Because deterrence works well (without reactance) for people who care little about the freedom being regulated, what we need to do is search for such people who are in a position to prevent the crime. In Braithwaite (1997) it is argued that for most crimes there are many actors with the power to prevent it. The victimization of a child by a fourth-grade bully can be prevented by the intervention of every child in the playground in grade five or above who observes it. This may be why whole-school approaches to bullying work, while peer me- HeinOnline -- 25 Crime & Just. 56 1999 Restorative Justice 57 diation programs that target only the bully do not (see Sec. VIlE). The sanctioning that counts is not that directed at the bully, but the softer sanctions of disapproval directed at those who fail to intervene to prevent bullying before it gets out of hand. I I When Canberra drunk-driving conferences work best, loved ones, drinking mates, and friends from work become key players in suggesting preventive agreements that draw on the capacity of many hands to prevent. Drinking mates may sign a designated driver agreement. Bar staff at the drinker's pub may undertake to call a taxi when the offender has had too much and make him take it. Uncle Harry may undertake to ensure that the car is always left in the garage on Friday and Saturday nights. Even with an offense as seemingly solitary as drunk driving, often there are many with preventive capabilities who can be rendered responsible for mobilizing those capabilities through a restorative justice dialogue. While reactance may be strong with the young male drink-driver who is a "petrol head," proud of his capacity to hold his drink, there may be no reactance from any of the other targets at a restorative justice conference. Indeed, when there is a collective reaction of nonreactance, we observe this to calm the anger of a young offender. Common garden varieties of juvenile crime are even more collective, proffering more soft targets, than drunk driving (Zimring 1981). Again, it was my empirical work with Fisse on corporate crime that led to the conclusion that the way to deter crime was not to seek to deter the criminal who benefits most from the crime, but to look for a softer target who has preventive capabilities. The paradigm-transforming moment in our praxis with this insight was the Solomons Carpet case (Fisse and Braithwaite 1993). Solomons had committed a false advertising offense. There were problems of proof and the penalty likely to be imposed by the courts was light. At the Trade Practices Commission we conferenced it without success. Involving even the CEO in successive conferences did not work; he was a hard target for deterrence, calling our bluff to take the case to court. In a final attempt, when we involved Mr. Solomon, the chairman of the board, he turned out to be a soft target who was ashamed that his company was flouting its legal obligations. He sacked the CEO and put in place a II It is common for other children to be involved in "holding" the victim for the bully or preventing him from getting away (Rigby 1996, p. 151). Victims themselves have preventive capacities that research evidence shows can be developed to protect them from bullying (Rigby 1996, p. 226). HeinOnline -- 25 Crime & Just. 57 1999 58 John Braithwaite remarkable program of compensation for consumers and industry-wide preventive (self-regulatory) measures. So we learned that a good regulatory strategy was to conference, conference again, and conference again with ever wider circles of executives with preventive capabilities until we found the soft target. Move up the organization until we found the soft target who could be moved by reason or deterred by fear of a personal sense of shame. We have applied this strategy in nursing-home regulation as well. Dialogue proceeds for about an hour among the stakeholders at the end of an inspection on the positive things that have been accomplished, what the problems are, and who will take responsibility for what needs to be done. In this process, most participants turn out to be soft targets, wanting to put their responsible self forward, volunteering action plans to put right what has been found wrong. This is why it succeeds in improving compliance with the law (Braithwaite et al. 1993). At the same time, it is clear from our data that there are cases where dialogic regulation fails-where the hardest of targets are in charge, dominating, and intimidating softer targets who work under them. Empirical experience gives good reasons for assuming that even the worst of corporate malefactors has a public-regarding self that can be appealed to, a self-categorization as "responsible businessman," for example (Ayres and Braithwaite 1992). However, when trust is tried and found to be misplaced, there is a need to escalate to deterrence as a regulatory strategy. When deterrence fails-because of reactance, or simply because noncompliance is caused by managerial incompetence rather than rational calculation of costs and benefits-then there is a need to move higher up an enforcement pyramid to an incapacitative strategy. Incapacitation can mean withdrawing or suspending a license for a nursing home that has proved impregnable to both persuasion and deterrence. Hence, there are increasingly solid empirical grounds for suspecting that we can often reduce crime by replacing narrow, formal, and strongly punitive responsibility with broad, informal, weak sanctionsby making the many dialogically responsible instead of the few criminally responsible. By dialogically responsible I mean responsible for participating in a dialogue, listening, being open to accountability for failings and to suggestions for remedying those failings. The theory I have advanced (Braithwaite 1997) is that this is more likely when there are many actors with causative or preventative capability with respect HeinOnline -- 25 Crime & Just. 58 1999 Restorative Justice 59 to that abuse. Where we can engage all of those actors in moral reasoning and problem-solving dialogue, the more of them there are, the more likely one or more will be a soft target. When just one player with causative responsibility or with a powerful preventative capability turns, empirical experience shows that many other actors who had hitherto been ruthlessly exploitative suddenly find a public-regarding self that becomes surprisingly engaged with a constructive process of righting the wrong. The implication of the analysis in this section of the essay is that punishing crooks is a less efficient deterrence strategy than opening up discussion with a wide range of actors with preventive capabilities, some of whom might be motivated by a raised eyebrow to change their behavior in ways that prevent reoffending. It is to keep expanding the number of players involved in a restorative justice process until we find someone who surprises us by being influenced through the dialogue to mobilize some unforeseen preventive capability. The hypothesis is that creative restorative processes have enormous potential to surprise us as Mr. Solomon did. You do not give up after a first conference because no one turns up who can deliver that surprise. You keep convening new conferences with new carers, new stakeholders, new resource people until someone walks through the door who can pull one of the levers to prevent a criminality that is almost always "overdetermined" (Lewis 1986). Again, restorative justice rewards the patient. Parker, in commenting on a draft of this essay, pointed out that I think this because of my view (some would say naive view) of human beings as social beings that are almost always enmeshed in multiple communities: "The Braithwaite argument is that there almost always are many with that capacity [to prevent] because we all live in a community wherein many individuals can pull strings of informal control and evoke bonds of responsibility" (see also Parker 19990). The argument draws sustenance from empirical findings such as those of Pennell and Burford (1996, p. 218) in their Canadian study of family violence conferences: the conferences "generated a sense of shame across the extended family for not having acted in the past to safeguard its relatives as well as a sense of shared identity because often the problems which their relatives experienced were common in their own lives." It also draws sustenance from that other Canadian experience at Hollow Water. How can we understand the accomplishment of no fewer than forty-eight child abusers brought to justice in such a small HeinOnline -- 25 Crime & Just. 59 1999 60 John Braithwaite Canadian community? Without the restorative process, could we have expected Western punitive justice to have convicted even three or four, or any? Probably not. It was the restorative process that flushed out those with knowledge of the evil. The fact is that for any kind of crime, communities know about and are concerned about countless crimes of which the police are ignorant. Karstedt-Henke and Crasmoller (1991) showed in Germany that for every juvenile crime the police detect, parents detect at least four, teachers detect about two, and peers detect more than five. Given the stronger evidence for an effect of certainty of punishment on crime than an effect of severity of punishment (Braithwaite 1997, n. 47), "soft" restorative justice for forty might just accomplish more general deterrence than tough incarcerative justice for four. With respect to knowledge, restorative justice is a virtuous circle, retributive justice a vicious circle. When the community knows about many crimes and reacts to them restoratively, the benefits of restoration motivate others to speak up, increasing community knowledge of crimes they will want to do something about. When the police know about few crimes and respond punitively, the collateral costs of punishment silence citizens into minding their own business, reducing reporting of crime. Again, in a world where certainty of sanctions matters more than severity of sanction and where informal sanctions deter more than formal ones, the corollary is that virtuous circles of restorative justice deter more than vicious circles of punitive justice. So the process implication of our analysis is dialogic regulation of social life of the sort we get in a family group conference or a restorative exit conference such as we see with nursing home or nuclear safety (Rees 1994) inspection. There is a structural implication as well, which is developed in Braithwaite (1997): more robust separations of powers within and between the private and public sectors. The number of third-party enforcement targets is greater to the extent that we have richer, more plural, separations of power in a polity. Dialogue among a wider range of citizens beyond the offender himself means that ripples of general deterrence spread out more widely. When many different types of subcriminal responsibility are known to be at risk of exposure to people we care about in restorative justice conferences, we are all deterred in our many roles. This is why Australian nursing home regulation has worked reasonably well. Whether we are the responsible nurse, the aide, the chaplain, the gardener, or the man who visits the lady in the next bed, if no one raises the alarm HeinOnline -- 25 Crime & Just. 60 1999 Restorative Justice 61 about a resident who is being abused, we know that our inaction might be disapproved in a conference. Restorative justice in other words is not just about specific deterrence of the offender: it also widens the scope of general deterrence (albeit a more benign general deterrence). The benign nature of this general deterrence will be seen by most critics as the greatest weakness of restorative justice. The crunch is that restorative justice sets free many whom deterrence or desert theories say should go to jail-like the insurance executives from CML. While it is clear that offenders and others who attend restorative justice processes do not view them as a soft option but rather as a difficult and demanding experience (Umbreit and Coates 1992; Sherman and Strang 1997 c; Schiff 1998), of course the agreements reached are softer than prison. A final qualification about general deterrence arises from the assumption that restorative justice will often fail and fail again and again until deterrent justice must be tried in an attempt to protect the community. Since, for the reasons outlined in Braithwaite (1997), deterrence will also often fail, we will sometimes need to escalate our response to incapacitation. Figure 1 represents this articulation of restorative justice to deterrence and incapacitation. The idea of the pyramid, which is justified in detail in Ayres and Braithwaite (1992, chap. 2), is that we start with the restorative strategy at the base of the ASSUMPTION Incompetent or Irrational Actor Rational Actor Virtuous Actor FIG. RESTORATIVE JUSTICE I.-Toward an integration of restorative, deterrent, and incapacitative justice HeinOnline -- 25 Crime & Just. 61 1999 62 John Braithwaite pyramid. The possibility of escalation channels regulatory aCtIVIty down to the base of the pyramid. This model transcends the limitations of passive deterrence in criminology by learning from the shift to active from passive deterrence in international relations theory. With passive deterrence, one simply calculates the probability of compliance on the basis of the expected size and risk of punishment. Active deterrence, in contrast, is dynamic, open to escalating threats in response to moves by the other player, as well as to graduated reduction in tension strategies. The pyramid dynamically meets the challenge that unless the threat of punishment lingers in the background, there will be a class of ruthless criminals who will exploit the opportunity of restorative justice with a deceitful pretence of cooperation. Where restorative justice for a first or second offense is backed up by a passive deterrent tariff for a third, the rational actor will cheat for one or two free throws. If enforcement is the product of restorative justice negotiation, Langbein and Kerwin (1985) show game-theoretically that rational actors will avoid immediate compliance. Langbein and Kerwin's model is only true, however, if deceit, holding back on compliance, does not cause an escalation of penalties. In practice it does; deterrence is active rather than passive, which is why Langbein and Kerwin's prediction is false as a description of most regulatory activity (see, e.g., Bardach and Kagan 1982; Braithwaite 1985). The reality of active deterrence as a strategy that works, at least in the business regulatory domain, where it has been more systematically studied and theorized than with common crime, commends Fisse's (1983) suggestion of giving it a jurisprudentially principled foundation through implementing "reactive fault" as the core criterion of criminal fault. In its most radical version, this would mean in a case of assault, the alleged assailant would go into a circle not on the basis of an admission of criminal guilt, but on the basis of admitting responsibility for the actus reus of an assault ("I was the one who punched her"). Functionally, New Zealand law already accomplishes this result by putting cases into family group conferences not on the basis of an admission of criminal guilt, but on the basis of formally "declining to deny" criminal allegations. Whether the mental element required for crime was present would be decided reactively, on the basis of the constructiveness and restorativeness of his reaction to the problem caused by his act (Braithwaite 1998b). If the reaction were restorative, the risk of criminal liability would be removed; only civil liability would re- HeinOnline -- 25 Crime & Just. 62 1999 Restorative Justice 63 main. However, if reactive criminal fault were found by a court to be present, that would be insufficient for a conviction; the mental element for the crime would also have to be demonstrated before or during its commission. 12 But it would be the reactive fault that would be the more important determinant of penalty than the concurrent fault. In practice, criminal justice systems vary enormously in the reactiveness versus proactiveness of their criminal law in action: Japan being unusually strong on reactive fault, the United States on proactive or causal fault (Haley 1996; Braithwaite 1998b). According to this analysis, this helps Japan enjoy lower crime rates than the United States in a way that has a profound jurisprudential justification. Encouraging findings on deterrence are emerging as the surprising positive result of the RISE experiment on restorative conferencing in Canberra. Preliminary data reveal a modest "Sword of Damocles" effect, something actually revealed in previous criminological research (Sherman 1992). For example, Dunford's (1990) study suggests that a warrant for arrest may deter domestic violence better than either actual arrest or nonarrest. To date, offenders randomly assigned to conferences in Canberra are coming out somewhat more fearful that they will be rearrested if they offend again, more fearful of family and friends finding out about rearrest, more fearful of a future conference, more fearful of at least one other consequence of a court case than those assigned to court (Sherman and Strang 1997 b; Sherman et al. 1998). In this variety of ways, conferences may sharpen our perceptions of how bad the punitive consequences would be if we were caught again. This is a somewhat unusual result because what much criminological research shows is that actual experience of the justice system reduces its terrors. For example, tax audits can have counterproductive effects by teaching many of those who are audited that they can cheat on their tax without going to jail and teaching them "how to avoid being caught when they evade taxes" (Kinsey 1986, p. 416). What the preliminary RISE data suggest is that changes at the margin to send increasing numbers of offenders to conferences may simultaneously increase the deterrent power of both conferences and court. From the deterrence perspective against which I am measuring restorative justice in this section, this is good news. The problem is that if deterrent threats cause defiance and reactance, restorative justice may 12 Brent Fisse takes the more radical view that if criminal liability is about punishing conduct known to be harmful and if failure to respond responsibly is harmful, then such reactive fault can be sufficient to establish criminal liability. HeinOnline -- 25 Crime & Just. 63 1999 64 John Braithwaite be compromised by what sits above it in a dynamic pyramidal strategy of deterrence and incapacitation. For Ayres and Braithwaite (1992), this is the greatest challenge facing responsive regulatory institutions. The challenge is to have the Sword of Damocles always threatening in the background but never threatened in the foreground. The criminal justice system must have an image of invincibility at the same time as it has an image of mercy and forgiveness. Police have a lot to learn here from the wisdom of business regulatory inspectors, such as Hawkins's British pollution inspectors: "Negotiating tactics are organized to display the enforcement process as inexorable, as an unremitting progress, in the absence of compliance, towards an unpleasant end" (Hawkins 1984, p. 153). Here is a New York nursing home inspector's account of how their surprisingly restorative regulatory system keeps cooperation in the foreground while coercion looms in the background: You can maintain the same demeanor when confronted with tension and stress, when the facility gets aggressive and unpleasant [in one case this involved putting a gun on the table]. You can be friendly if they don't correct. You just pass it on. You never have to be anything but assured and friendly. The enforcement system will take on the battle ... The team leader just tells them [the nursing home] what the repercussions are if you don't correct. You just let the system take over. That's all you have to do. A good team leader is confident, friendly, and explains consequences. She never uses a standover approach. (Braithwaite 1994, p. 30) Part of the trick of deterrence that is always threatening but never threatened (Ayres and Braithwaite 1992, pp. 44-53) is to enculturate trust in regulatory interactions (Braithwaite and Makkai 1994), while institutionalizing distrust through an enforcement system (Braithwaite 1998a). It is to surprise the very worst of people by treating them as trustworthy; because if we can persuade them to put their best self forward (in the presence of people whose respect they crave) we will regularly be surprised to find that the most socially responsible of their many selves is restorative. Restorative justice is not about picking good apples for reconciliation and bad apples for deterrence; it is about treating everyone as a good apple as the preferred first approach. This implies that to be effective restorative justice requires considerable nuance in administration, yet a nuance most human beings have HeinOnline -- 25 Crime & Just. 64 1999 Restorative Justice 65 at their disposal. Just as they know from experience with life that it is better to discuss consequences, allowing the offenders to discover their own shame, than to say "shame on you," they also know that direct threat engenders defiance in a wayan image of invincibility does not. The prediction here, that will be tested in the RISE experiment, is that conferences will fail if they are either "shaming machines" (Retzinger and Scheff 1996) or threat machines. The widespread understanding of this wisdom in the community is reflected in the fact that the majority of parents of children in societies such as the United States succeed in raising nondelinquent children because they do have an "authoritative" rather than an "authoritarian" parenting style (Baumrind 1973, 1978). Durkheim (1961, p. 10) understood it as well when he said: "Punishment does not give discipline its authority, but it prevents discipline from losing its authority." K. Restorative Justice Practices Incapacitate Crime Better than Criminal Justice Practices Grounded in the Theory of Selective Incapacitation Incapacitation means removing an offender's capacity to reoffend; there are many ways to do this beyond incarceration, execution, and cutting off the hands of pickpockets. A useful feature of restorative justice is that it empowers communities of care to be creative about how to incapacitate. The empirical evidence on selective incapacitation in criminology is almost exclusively limited to a consideration of selecting the most dangerous criminals for incarceration. That evidence suggests that we are not very good at getting the selection right (Gottfredson and Gottfredson 1994; MacKenzie 1997, p. 9). Failures to incapacitate those who commit serious further offenses tend to be well publicized. Less well publicized is the likely more serious problem of false positives whose criminal career might have ended had we not thrust them into daily interaction with criminals in a prison where they learn new skills in the illegitimate labor market or suffer demeaning experiences that engender defiance, shame, and rage. Through using incarceration much more selectively, restorative justice should be able to avert a lot of damage that makes our crime problems worse. That is mere speculation, however, as there is no empirical evidence to support such a hope. At the same time, the pyramidal theory of restorative justice outlined in the last section means that there is a willingness to resort to incapacitation when both restorative justice and deterrence repeatedly fail to protect the community from a serious risk. HeinOnline -- 25 Crime & Just. 65 1999 66 John Braithwaite However, imprisonment is not the principal method of incapacitation to which restorative justice would want to resort. Again there is much that criminology can learn from business regulation here. VVhen a company continually creates a serious risk to the community, a common alternative to putting the company in jail is to put the jailer into the company. An example was the "resident inspector" program run by the Mine Safety and Health Administration in the United States for repeat offending high-accident mines. The presence of the inspector in these mines stopped certain unsafe practices from being contemplated, substantially reducing deaths and injuries to well below the national average in mines that had been the least safe in the country (Braithwaite 1985, p. 83). Similar resident inspector programs have been applied in the nuclear industry, the nursing home industry, and more recently in relation to the environmental compliance problems of Consolidated Edison in New York. Braithwaite and Daly (1994, p. 200) have outlined how successive restorative justice conferences might escalate incapacitative response for domestic violence: for example, there could be escalation from weekly reporting by all family members of any violent incidents to the man's aunt or brother-in-law (conference 1), to a relative or other supporter of the woman moving into the household (conference 2), to the man moving to a friend's household (conference 3). That essay also makes much of flipping the incapacitation targetincapacitating the male offender by assuring the female victim of the resources and guaranteed shelter to walk out, leaving the offender alone in a house without a victim and therefore without a capacity to victimize. In cases such as the Aboriginal insurance scandals discussed earlier, agents who make fraudulent claims can be incapacitated by licensing schemes that deny them a license for this kind of work. Doctors, lawyers, and company directors can be delicensed through either positive or negative licensing schemes. Drunk drivers can also be deprived of a license to drive, a form of incapacitation that works badly in Australia, where drunks driving without licenses is pandemic. More social and less legal assurances of incapacitation may sometimes have more promise and restorative justice conferences can deliver these. Drinking mates can sign undertakings that they will prevent him from driving after drinking and will make him comply with a designated driver agreement. Uncle Harry can incapacitate him from drinking and driving on Friday and Saturday HeinOnline -- 25 Crime & Just. 66 1999 Restorative Justice 67 nights when he goes out with the boys by taking ownership of the car and its keys on those nights. Such incapacitation can be escalated by a conference in response to noncompliance by agreement up front that the consequence of failure to hand over the car at these times is that Uncle Harry will take permanent possession of the car for a year. The theory of restorative justice here is that Uncle Harrys have a more plural range of incapacitative keys they can turn than a prison guard who can turn just one key. Uncle Harry can respond dynamically when his incapacitative ideas backfire. But they are less likely to backfire when the offender voluntarily commits to them. As we have seen, unenforceable restorative justice agreements enjoy higher compliance than enforceable court agreements (see Sec. VIL4). Beyond the greater commitment we all have to undertakings we choose ourselves, the further reasons for superior compliance are that the Uncle Harrys of this world come up with ideas more attuned to the reality of the offender's circumstances than can a judge, and are better monitors of their implementation than police officers because one Uncle Harry might have more contacts with the offender in a month than all the police in the city during a year. Intimates, in short, can incapacitate more intensively, more creatively, more sensitively, more consensually, and in a more dynamically responsive way than the criminal justice system. At this stage, the Optimistic Account of Incapacitation lacks systematic support, but does map a promising new research agenda for the possibilities of restorative justice. L. Restorative Justice Practices Rehabilitate Crime Better than Criminal Justice Practices Grounded in the Welfare Model In Section VIIB I showed that there is some evidence that, while limited, all suggests that restorative justice processes may prevent reoffending better than traditional criminal justice processing. The qualitative literature on restorative justice is certainly littered with case studies of offenders who have been rehabilitated as a result of the deliberation at conferences. What is clear from the criminological literature is that when rehabilitation of criminal behavior does occur, it is at the hands of families more than any other institution. Obversely, family dysfunction correlates as consistently with delinquency as any variable. Hence, the Maori critiques of the Western justice system that led to the restorative justice reforms of 1989 have a strong empirical foundation: Western justice weakens families because it takes away their responsibility HeinOnline -- 25 Crime & Just. 67 1999 68 John Braithwaite for dealing with crime and preventing recurrence. Weaken family responsibility, especially for cultures with deeply embedded traditions of family responsibility, and you destroy the fabric of crime control (Hassail 1996). Some Australian Aboriginal peoples articulate a similar critique-in a culture where the father of an adolescent's future wife has the primary role in social control, a justice system that wrenches young offenders away from any influence by that person or other relevant elders will destroy, has destroyed, the basis for social control. One reason why restorative justice ought to do better at rehabilitation than rehabilitative justice is that it does not have rehabilitation as its aim. Rehabilitation is like spontaneity as an objective: when you try to be spontaneous you are not very spontaneous. When the criminal justice system is seen as setting out to change people, even by offering rewards, that engenders reactance, though reactance to reward does not seem as great as to punishment (Brehm and Brehm 1981, p. 229). The practical focus on the consequences of the crime and the needs this creates for victims and the community, more act focused and less focused on the offender as a person, more victim focused and less offender focused, means that the process is less stigmatizing and more dignified for the offender. It is hard for the communication of disapproval to be respectful when the focus is on the twisted psyche of the offender or his defective conscience. By definition, stigmatic labeling is not averted when words such as sociopath are bandied around with the family. However good the diagnosis, however good the rehabilitation program it commends, the very fact that it comes out of a program designed to deliver a diagnosis and a treatment renders the process stigmatic. This means that the crime-reduction effects of the rehabilitative program have to be very strong before they can outweigh the crimeinstigating effects of the stigmatization. Any program where social workers, psychologists, or psychiatrists come in to do things to or for people risks stigmatization by the very fact of professionalized doing or helping. Retributivist critics of rehabilitation are right when they say rehabilitation strips the offender of dignity in this way (Murphy and Hampton 1989); they are wrong to suggest that punishment confers dignity; a space that gives the offender an opportunity to choose to put things right is what restores dignity. It is such a choice to put things right that most nurtures a continuing commitment to keep things right, that nurtures rehabilitation. HeinOnline -- 25 Crime & Just. 68 1999 Restorative Justice 69 Of course offenders are often desperately in need of a drug rehabilitation program, face-to-face counseling, job training, remedial education, all manner of rehabilitative programs. If that need is desperate, citizens should speak up for the need in a restorative justice program and the offenders should see committing to it as part of putting things right. This empowerment of the offender, together with their community of care, to choose from rehabilitative programs offered by health and welfare professionals in the state, private, and voluntary sectors is different from state monopolies of social work and health care provision in the traditional welfare model. I must confess to seeing these as empty ideals in all the restorative justice programs of which I have experience. I have seen many drunkdriving conferences where the offender is a tottering alcoholic, but where no one in the community of care raises the need for a drug treatment program, sometimes because most supporters are also excessive drinkers. In New Zealand, the rhetoric of citizens being empowered to choose rehabilitation programs without having them forced down their throat by the state is impressive; yet this occurs in a context where the retrenchment of the once exemplary New Zealand welfare state by successive conservative governments means there are no programs left to choose (cf. Maxwell and Morris 1996). Australia is almost as bad in this respect. It therefore seems highly doubtful that restorative justice conferences are having major rehabilitative effects at this time. They may, however, be averting some of the dis empowerment of traditional "corrections," the stigmatization of rehabilitation oriented to changing pathology. Two of the things we know from the vast literature on the effectiveness of programs for the rehabilitation of criminals are that voluntarily chosen programs outperform enforced rehabilitation, and that programs that strengthen community support for the offender outperform those that wrench offenders out of communities of care into the hands of professionals who offer individual treatment (Cullen 1994). In sum, what should make restorative justice more effective at rehabilitation than rehabilitative justice has historically been are its empowering, communitarian, dignifying, and victim-centered characteristics. Delivering on this potential is unlikely to be demonstrated at the moment when the welfare state is being dismantled, even as it requires the dismantling of welfarist justice monopolized by state correctional professionals. HeinOnline -- 25 Crime & Just. 69 1999 70 John Braithwaite M. Restorative Justice Practices Are More Cost Effective than Criminal Justice Practices Grounded in the Economic Analysis of Crime The economic analysis of law (e.g., Posner 1977) provides a more theoretically sophisticated, though transparently false explanatory structure than the other utilitarian analyses in deterrence, rehabilitation, and incapacitation theory. It makes false predictions because it is myopic. Its models assume that rational choosing of costs and benefits provides a total explanation of compliance when emotions and twisted cognition's play havoc with the reality (see Braithwaite 1997). Makkai and Braithwaite (1993b) found that actual costs of compliance with nursing home laws explained only 19 percent of the variance in the subjectively expected costs that should inform rational choices. While there is a powerful effect of expected cost of compliance on compliance, this is not a monotonically increasing effect. There is a turning point in the relationship explained by the behavior of "disengagers." Their behavior is not to be understood in terms of rational game-playing but in terms of dropping out of the enforcement game. Regulatory disengagers, rather like many heroin addicts, are in the regulatory system but not of it and certainly not economically calculative about it. All this means that the underspecification in economic analyses of law is of a fatal sort. It is not that the models are basically right and can be improved by tinkering that includes more of the excluded variables. When a variable like reactance can reverse the direction of a deterrence coefficient, when disengagers come into play in a way that turns an increasing relationship into a decreasing one, advice on the optimal level of deterrence will be not only wrong, but very wrong. Moreover, these influences mean it will be wrong in a way that assumes increasing deterrence will deliver more economic benefits than it ever in fact does (to the extent that defiance and disengagement neutralize or reverse deterrence). Finally, even if an empirically correct economic analysis of the optimal level of penalties were discovered, its implementation would lead us into a deterrence trap that would create economic chaos in respect of some of our most serious crimes (Braithwaite 1997). For example, if the probability of detection for insider trading is one in a hundred and the expected returns are a million dollars, fines for insider trading will have to exceed $100 million to deter the average insider trader, and be much higher to deter the biggest sharks. Coffee's (1981) deterrence trap is that fines of this magnitude will cause bankruptcies, punishing innocent workers who are retrenched. It is difficult therefore to imagine the construction of a purely eco- HeinOnline -- 25 Crime & Just. 70 1999 Restorative Justice 71 nomic model of crime that will not set deterrence at a counterproductively high level. The responsive theory of regulatory deterrence in Ayres and Braithwaite (1992) certainly draws heavily on economic analysis, but it uses the economic analysis as an element in the design of a dynamic model that moves from restorative justice when experience proves it a failure and then moves from deterrent justice when experience proves that a failure. That is, there should be no reliance on a statically optimal level of deterrence. We do not want to rely on that because, for the reasons adduced, it will always be wrong. A dynamic model based on a regulatory pyramid where restorative justice is privileged at the base of the pyramid is more likely to get it right, albeit clumsily. It iterates through one failed strategy after another until contextual deliberation declares one effective. It should also be cheaper because it averts maximally expensive options like imprisonment and courts staffed by highly paid judges, prosecutors, and other professionals as it privileges the efforts of volunteers from the citizenry. on the costs of restorative justice compared to Systematic ･ｶｩ､ｾｮ｣＠ punitive justice is scarce, though Peter Reuter has a study underway as his contribution to the RISE experiment in Canberra. Claims are regularly made about multimillion dollar savings in New Zealand, particularly as a result of closure of juvenile institutions. While the number of residential places has dropped by almost two-thirds since 1989 (Maxwell and Morris 1996), this seems plausible, but no published studies exist of the magnitude of the claimed savings. It certainly is true that nations such as Germany, Austria, New Zealand, and China, which are vigorously committed to restorative programs for juveniles, pay for extremely modest numbers of institutional beds per capita compared to nations like the United States, the United Kingdom, and Australia. However, to be maximally effective in the terms of the last subsection, restorative justice requires a more credible investment in the welfare state and this does not come cheap. Of course, the benefits of a decent welfare state should not be measured primarily in terms of crime prevention. The most thorough study is of Scottish mediation of disputes largely among neighbors, family, and friends (Knapp and Netten in Warner 1992, pp. 105-37). Theoretically, these were supposed to be cases that otherwise would have been prosecuted. In the comparison group, nineteen of the forty-four cases were not prosecuted. Across the two programs, when prosecutions did occur, the average prosecution case costs HeinOnline -- 25 Crime & Just. 71 1999 72 John Braithwaite were just over £200, compared with about £300 for mediation and reparation cases, leading Knapp and Netten to conclude that for comparatively simple matters that would not lead to either a not guilty plea or imprisonment, mediation and reparation was rather more expensive than prosecution. N. Restorative Justice Practices Secure Justice Better than Criminal Justice Practices Grounded in ''Justice'' or ''Just Deserts" Theories Allowing offenders to buy their way out of prison with monetary and nonmonetary compensation to victims unacceptably confounds the private goals of mediation and the public goals of criminal law (Brown 1994, p. 1253). For "just deserts" theorists, it is unjust that offenders get unequal treatment depending on whether they have a merciful or a punitive victim, a poor one who needs compensation or a rich one who does not, a victim who will cooperate in the diversion from court or one who will not. Some restorative justice advocates turn this around by saying that it is morally wrong to privilege equality of treatment for offenders over "equality of justice [which] means equal treatment of victims" (Barnett 1981, p. 259). Or equal justice might mean equality of opportunity for victims with known offenders to pursue the forms of restoration most important to them in the way of their choosing (see generally Roach 1999). Because equality for victims and equality for offenders are utterly irreconcilable, the more practical justice agenda is to guarantee victims a minimum level of care and to guarantee offenders against punishment beyond a maximum limit. The normative theory of restorative justice illuminates a practical path to those guarantees. The fundamental problem restorative justice advocates have with the justice model has been most eloquently captured by Martin Wright (1992, p. 525): "Balancing the harm done by the offender with further harm inflicted on the offender only adds to the total amount of harm in the world." As with previous sections of this essay, the analysis of the justice of restorative justice compared with the so-called justice or just deserts model is influenced by a consideration of white-collar crime that is so often lacking in the work of desert theorists. There is now, as we have seen, a good deal of evidence that citizens are more likely to feel that restorative processes are just and respect HeinOnline -- 25 Crime & Just. 72 1999 Restorative Justice 73 their rights after they have experienced them than are citizens who have experienced the justice of courts (see Subsecs. A-C above). Desert theorists have to respond to this by saying that citizens in a democracy do not understand what justice entails, do not understand what their own interests in justice should cause them to want. While this response may be largely false, given the worries we ought properly to have about tyrannies of the majority, we must recognize it can sometimes be true. The systematic evidence we have from judicial oversight of New Zealand conferences is not consistent with any widespread tyranny of the majority. Maxwell and Morris (1993) report that 81 percent of family group conference plans were approved without modification by courts, with the overwhelming majority of changes (in 17 percent of cases) being to make orders at a higher level rather than at a lower level (a lower level being what one would expect to see if there were a tyranny of the majority to be checked). Almost identical results have been obtained in the Restorative Resolutions project for adult offenders in Manitoba (83 percent judicial ratification of plans, with five times as much modification by addition of requirements as modification by deletion) (Bonta, Rooney, and Wallace-Capretta 1998, p. 16). There is no consensus within the social movement for restorative justice on what should count as unjust outcomes. Most advocates want it to be a more modest philosophy than to aspire to settle this question. Rather, restorative justice should settle for the procedural requirement that the parties talk until they feel that harmony has been restored on the basis of a discussion of all the injustices they see as relevant to the case. Within that dialogue about justice, Braithwaite and Pettit (1990) and Braithwaite and Parker (1999) have made the case for a republican conception of justice. Most restorative justice advocates do not know or care what is involved in a republican rationale for restorative justice, let alone subscribe to it. The most popular philosophical foundations among advocates for the justice of restorative justice are spiritual (e.g., Van Ness 1986). Yet civic republicanism is one secular philosophical foundation that has a critique of the just deserts model enjoying some support among restorative justice theorists. One of the virtues of republican theory is that its justification of restorative justice does not depend on all those involved in restorative justice processes subscribing to a republican conception of justice, in the way that just deserts does depend on a consistent commitment of judges and juries to the conception of justice in its justice model. HeinOnline -- 25 Crime & Just. 73 1999 74 John Braithwaite In considering the Pessimistic Account of restorative justice, Section VIII of this essay assesses a number of concerns about the injustice of restorative justice, both procedurally and in terms of outcomes. I conclude that the propositions of the Pessimistic Account about the injustices of restorative justice are true insofar as we can judge from the current state of the evidence. The truth of that critique, I want to contend, is consistent with the conclusion in this section that restorative justice is more just than just deserts. Following Campbell (1988, pp. 3-4), Parker (1999b, pp. 45-47) makes a Rawlsian distinction between the concept and the conception of justice in her republic of justice. Parker's concept of justice is: "Those arrangements by which people can (successfully) make claims against individuals and institutions in order to advance shared ideals of social and political life" (Parker 1999, p. 46). A concept of justice thus conceived as means, formal and informal, by which people seek to secure social and individual relations they think are right will yield different views (conceptions) of rightness. Parker's (1999) republican conception of substantive justice is of freedom as nondomination (following Skinner 1984; Pettit 1993, 1997). The just society then institutionalizes processes of disputing that will maximize freedom as nondomination. So Parker (1999, p. 49) integrates concept and conception in a definition of justice that I will adapt only slightly here: Justice is "that set of arrangements that allow people to make claims against other individuals and institutions in order to secure freedom against the possibility of domination." Freedom as nondomination is the same republican conception of freedom as a citizenship status that Braithwaite and Pettit (1990) called dominion. Freedom as nondomination is contrasted with freedom as noninterference, which is at the core of the liberal tradition. Republicans from Rome to Montesquieu, Madison, and Jefferson wanted more than liberty in the impoverished individualistic sense favored by the liberals who came to dominate Western political discourse through the nineteenth century. Resilient liberty required community assurance against domination through the guarantees of a rule of law, a separation of powers, uncoerced deliberation in governance, welfare policies that guarantee protection from the dominations of poverty, and norms of civic virtue. It required liberty, equality, and fraternity/sorority. Braithwaite and Pettit (1990) have sought to rework what they consider all the key normative questions in the criminal justice system in accordance with the maximization of freedom as nondomination. They HeinOnline -- 25 Crime & Just. 74 1999 Restorative Justice 75 also compare a full retributivist position with a full republican position and conclude that a full just-deserts policy would increase injustice while a republican policy would reduce it. This results from certain facts about complex modern societies. These are mostly facts about the distribution of power, which prevent punishment from being imposed on those most deserving of it. A policy of attempting punishment of all those who deserve it (and who can be caught) has the effect of increasing injustice, worsening tendencies to punish most where desert is least. This is because of a tendency for the law to be "the most powerful where the least needed, a sprinkler system that turns off when the fire gets too hot" (Geertz 1983, p. 217). Braithwaite and Pettit (1990, chap. 9) argue that a number of bureaucratic realities about criminal justice systems conduce to the theorem that where desert is greatest, punishment will be least. One is the problem of system capacity (Nagin 1978; Pontell 1978). Braithwaite and Pettit rely on this literature to show that those locations in time and space where crime is greatest, and those types of crime where offending is most widespread and serious, are precisely where the criminal justice system resorts to leniency in order to keep cases flowing and avert system overload. But bureaucratic pressures are not the main reason for the truth of the theorem. Structural realities of power are more important. Braithwaite and Pettit (1990) argue that in the terms of just deserts theory, there are more white-collar criminals deserving severe punishment in any society than blue-collar criminals deserving severe punishment. Attempts to give deserved punishment to all who are guilty, however, successfully impose desert on blue-collar offenders while being systematically unsuccessful with white-collar offenders. The white-collar crime enforcement system in every country operates on comparatively restorative principles. Braithwaite and Pettit (1990) argue this is sociologically inevitable as well as desirable. Retributive corporate crime prevention would fail because of deterrence traps, formidable defiance, and the superior capacity of the powerful to deploy rational countermeasures against deterrence-like the appointment of "vice presidents responsible for going to jail" (Braithwaite 1984). The best path to equal justice for equal wrongs is therefore to move blue-collar criminal enforcement down the same restorative path that white-collar enforcement has long followed. The injustice of the justice model arises from its reactive quality in a world where equal reactions produce unequal results. Parker (1999) works through the proactive reforms required for republican access to HeinOnline -- 25 Crime & Just. 75 1999 76 John Braithwaite justice. She suggests, for example, that all organizations above a certain size have access to justice plans that, through consultation with stakeholders, identify the various types of injustices (to consumers, workers, minorities, creditors, and so on) that are common consequences of its activities; set up restorative justice fora to correct these injustices when they arise; and deploy preventive law measures to ensure compliance with the law and remove blockages to access to justice. Performance indicators would be required under these plans to demonstrate improved access to justice this year compared to last year (continuous improvement). The results of independent audits against these performance indicators would be made public. Responsively regulated access to restorative justice plans in the large organization sector then frees up more finite legal aid resources for injustices inflicted in small organizations like families and by individuals. Parker's imagined world of access to restorative justice is one where most victims of the most serious crimes (organizational crimes) that currently get no justice are given access to corporate restorative justice. It is a world of profoundly greater justice than the "justice" model imagines. Of course justice imagined is not justice accomplished. It can be said, however, that the justice model skew of our present system, a skew toward just deserts for the poor and impunity for the powerful, accomplishes profound injustice. Moves in the direction of restorative justice for poor offenders and restorative justice for more victims of corporate offenders are the practical moves toward an amelioration of that injustice. o. Restorative Justice Practices Can Enrich Freedom and Democracy Christie's (1977) claim is that the king's justice stole conflicts from citizens; it was a significant accomplishment in the progressive consolidation of the domination of monarchs over their people in Europe from the eleventh to the nineteenth centuries. For much of Europe, justice was centralized under state control and local restorative justice was substantially extinguished by 1200 (Weitekamp 1999). Yet restorative justice as mainstream disputing between and within clans was not extinguished by the English in Scotland until well into the nineteenth century; it was never extinguished by the Dutch in Indonesia, where Adat (local) criminal laws work in parallel with a dominant Dutch criminal law of the Indonesian state. Elsewhere, Cree, Navajo, and Maori restorative justice survived, though barely. Even in England and France, the greatest imperial extinguishers of restorative justice, re- HeinOnline -- 25 Crime & Just. 76 1999 Restorative Justice 77 storative justice practices remain profoundly influential in civil society, in schools, for example. The globalized centrality of the prison and professional police forces in the statist revolution's new justice model actually came quite late. \Vhile the story of our criminal law is a story of imperial oppression to extinguish restorative justice, its major victories are historically recent enough across most of the globe for there to be substantial residues of more democratic modes of doing justice available to be revitalized. Recent empirical experience in places like New Zealand is that the flames of restorative justice can be rekindled surprisingly quickly because citizens find that they like restorative justice and popular demand for it spreads. Control over punishment systems (combined with discretion to issue royal or presidential pardons) strengthened the power and legitimacy of rulers (see, e.g., Foucault 1977; Garland 1985, 1990). The new democratic rulers of the past two centuries continued to see their control of the secret police as vital to combating organized threats to their monopoly on the legal use of violence, the control of the regular police as vital to their control of disorganized threats. Yet abuse of that power (executing someone popular and innocent; the Guilford four) proved at times such a threat to their legitimacy that rulers were forced by political opponents to institutionalize certain principles of fairness and consistency into the state system. That process started with Magna Carta. These are accomplishments of liberalism that are worth preserving in a civic republican justice system. At the same time, the pretence that the state punishes crime in a consistent, politically evenhanded way, so vital to the legitimation of statist criminal justice, is seen by citizens as a pretence. One law for the rich, another for the poor. That is the reality that seems transparent to citizens in totalitarian and democratic states alike. This is another dimension of the democratic appeal of shifting the control over mercy from the monarch back to the people affected, while using state law to constrain excesses of community through defining maximum punishments, rights, and procedural requirements. The theme of how to set up a just interaction between the peoples' justice and the law's justice is one to which I return in the conclusion of the essay. There is more to the democratic virtue of restorative justice than returning conflicts to the citizens from whom they have been stolen. Western democratic institutions were planted in the shallow soil of societies where disputing had been taken over by the king. Disputing HeinOnline -- 25 Crime & Just. 77 1999 78 John Braithwaite over daily injustices is where we learn to become democratic citizens. And the learning is more profound when those daily injustices reveal deeply structured patterns of injustice. Engagement with them is de Tocqueville's apprenticeship of liberty. In Benjamin Barber's terms, democratic disputing is educative, central to learning to be free: While we root our fragile freedom in the myth that we are born free, we are in truth born dependent. For we are born fragile, born needy, born ignorant, born unformed, born weak, born foolish, born unimaginative-born in chains ... Our dependency is both physical-we need each other and cannot survive alone-and psychological; our identity is forged through a dialectical relationship with others. We are inescapably embedded in families, tribes, and communities. As a consequence, we must learn to be free. That is to say, we must be taught liberty. We are born small, defenseless, unthinking children. We must be taught to be thinking, competent, legal persons and citizens. We are born belonging to others; we have to learn how to sculpt our individuality from common clay. (Barber 1992, pp. 4-5) I remember in 1991, in the early days of restorative justice con ferencing in Australia, suggesting to Sergeant Terry O'Connell that it was a mistake to allow young children to attend and participate in conferences. Sometimes it is, but basically empirical experience has proved me wrong. In conferences, children are learning to be democratic citizens. The adults are mostly wise enough to make allowance for the unsophistication of much of what they say and to support them, help them establish the relevance of their point of view. Often it is the very unsophistication of the child's legitimate perspective that is so moving: "I've listened to what you've said about [my big brother]. It's not true. He is always kind to me; he helps me when I don't know what to do. I don't know any boy who is kinder than my brother." We might hope for the town we love (I do) that the thousands of children who have now experienced participatory antibullying programs in our schools, the thousands of adults who have experienced restorative justice conferences in our police stations or community halls, will learn how to do justice restoratively and apply those lessons in the families, clubs, and workplaces where they face their sharpest conflicts. Most especially we might hope conferences are educating the police for democracy. Experience is the best educator, more so the more nuanced the skills required. We hope that citizens are learning HeinOnline -- 25 Crime & Just. 78 1999 Restorative Justice 79 in conferences how to deliberate respectfully in the face of the greatest of the provocations of daily life. If they can learn to deliberate wisely and respectfully in the most provocative contexts, then they are citizens well educated for democracy. My observation is that the citizens of my town are learning, however disappointed I become at the slowness of the learning and at the many setbacks restorative justice has suffered in Canberra. The hope is that the seeds of our democratic institutions will be planted in slightly deeper soil in the next century as a result. VIII. A Pessimistic Account of Restorative Justice My disposition is transparently optimistic about restorative justice. Partly this manifests a bias, a personality that suffers pathological optimism. But it also represents a considered belief that the criminal justice system needed a new and positive vision, that criminologists had become depressingly nihilistic in the 1970s and 1980s. The optimistic bias that gives pessimists something better to shoot at can yet be the kind of optimism that we see among the best natural scientists-the medical researcher whose very optimism about a new theory of disease motivates extraordinary rigor in putting in place randomized controlled trials to refute it. But that is not enough. The scientific optimist is also required to develop and test ideas about the side effects and the contraindications of her new drug. The adverse side effects and contraindications of restorative justice are numerous. Many have already been introduced in the course of qualifying the Optimistic Account. In this section, I rework the concerns into a systematic Pessimistic Account with thirteen propositions. A. Restorative Justice Practices Might Provide No Benefits Whatever to over 90 Percent of Victims Most victims of crime are victims of white-collar crimes without ever coming to realize this. They pay higher prices every day for products whose prices have been fixed by criminal price-fixing conspiracies. Even for offenses like burglary, where the victim is acutely aware of victimization, in every country in the world it is only a small minority of cases that are cleared by arrest. Even for offenses like domestic violence, where the victim knows she has been victimized and by whom, reports to the police followed by admissions of guilt are extremely rare. Of course, if restorative justice does reduce the crime rate, many people who would otherwise have been victimized get a benefit. But re- HeinOnline -- 25 Crime & Just. 79 1999 80 John Braithwaite storative justice has nothing to offer the overwhelming majority of citizens who are acrually victimized by crime. The documented volume of unapprehended white-collar crime and domestic violence alone (Braithwaite and Pettit 1990, chap. 9) makes it easy to demonstrate that it would be foolishly optimistic to believe that the criminal justice system could do something for the known victims of known criminals for even 1 percent of our crime. While there are limits on what the state can do to heal when there is no known criminal, these limits are less for organizations in civil society. The women's shelter movement can help to heal survivors who will not lay a complaint; circles can and do heal victims in the absence of offenders. There can and should be a level of state funding for victim support groups that allows them to provide professional, material, and emotional support at least to all victims of violence who request it. ° B. Restorative Justice Practices Have No Significant Impact on the Crime Rate Because more than 90 percent of victimizations will always be untouched by state restorative justice processes, preventive effects of restorative justice interventions would have to be massive to register any measurable impact on the overall crime rate. It would take a much bigger conferencing program than exists anywhere in the world to conference one percent of all (detected and undetected) criminal offenders. In the unlikely circumstance that conferences halved their reoffending, that would reduce the crime rate by half a percent. Yet no one thinks that the effects of an eighty-minute conference on days, months, and years of competing influences will be massive. Many of the pessimists reasonably say that even if restorative justice theory is right (which they doubt), the impact of transitory restorative justice interventions are sure to be so small as to be detectable only on a massive (unaffordable) sample. The theory is therefore useless because its benefits (if true) could never be demonstrated by economically feasible scientific research. Consider the Canberra drunk-driving restorative justice experiment being conducted by Sherman and Strang. For every officially recorded drunk-driving offense that comes into the experiment, the offenders are reporting eighteen other undetected drunk drives during that year. Add to that all the undetected drunk drives of those who are never caught and it is clear that the 450 RISE drunk-driving conferences HeinOnline -- 25 Crime & Just. 80 1999 Restorative Justice 81 touch only a tip of the iceberg. However it is the really serious repeat offenders who are most likely eventually to be caught. So at least the restorative justice process might eventually get a shot at a good proportion of the worst offenders (one would hope so in Canberra with as many as fifteen hundred drunk-driving convictions every year in a city of just 310,000). Furthermore, if the Optimistic Account's analysis of the general deterrent and crime prevention (e.g., culture changing) superiority of restorative justice over punitive justice is right, then there might be a measurable impact on the overall crime rate. Even if not, the economic and protective value to the community of reducing reoffending for that minority of offenders who do find their way into the criminal justice system should not be discounted. But this is a much more modest claim than the claim that restorative justice can significantly reduce the crime rate. It may be made more modest by the fact that restorative justice processes cause crime as well as prevent them. Offenders learn the identity of their victim as a result of meeting them in a restorative justice process, or learn some other fact of their lifestyle that makes it easier to revictimize them. Offenders who experience heightened rather than reduced anger over what the victim says may revictimize, for example, through intimidating speech or threats during the conference. These concerns will be explored systematically in the RISE experiment. C. Restorative Justice Practices Can Increase Victim Fears of Revictimization The studies reported in Section VIL4 clearly establish that this can happen. However, they also establish that reduction of victim fears of revictimization appears to be about twice as common. While victims are mostly surprised to learn how shy, ashamed, and inadequate offenders are, some offenders are formidable and scary. Such cases can destabilize restorative justice programs in the media. Our worst case in Canberra involved an offender who threatened a woman with a syringe filled with blood. The conference was not well run and feelings between offender and victim deteriorated. Subsequently, the victim found a syringe left on the dashboard of her car, which she took to be a threat from the offender (though this allegation was never proved). The case was covered by a local television station. Out of two thousand Canberra conferences (some with no victims, some with twenty), this is the only case of escalated victim fear that hit the media. But one HeinOnline -- 25 Crime & Just. 81 1999 82 John Braithwaite can be enough. Restorative justice programs need to offer much more comprehensive support to the victims who face such traumas. D. Restorative Justice Practices Can Make Victims Little More than Props for Attempts to Rehabilitate Offenders This concern became acute with a number of British mediation programs during the 1980s where it was common for the offender and victim not to meet face-to-face, but rather for the mediator to be a go-between. Where no meeting occurs, Retzinger and Scheff's (1996) symbolic reparation, which we have seen is more important to most victims than material reparation, is more difficult. In these circumstances we can expect the dissatisfaction of victims to focus on the limits of the material reparation they get, "projects which claim to provide reparation for victims actually operating to maximise the potential for diversion of children from prosecution" (Haines 1999, p. 6). The British concern about victims being no more than props has not been a major issue in the debate in Australia and New Zealand about the pluses and minuses of restorative justice conferences. This is not to deny that victims used as props by a youth lobby who are concerned only to get a kinder deal for young offenders does not emerge as a deficiency in particular cases. Jennifer Brown (1994, p. 1274) is concerned that victim anger may be redirected in ways that may be destructive for victims by mediation ground rules that "forbid blaming and extended discussion of past events" in favor of "a more forward-looking, problem-solving outlook." A connected concern is that approaches by state officials, perhaps particularly if they are police, may create pressure on victims to take part in a restorative justice process when they would rather cut their emotional and material losses. Brown (1994, p. 1266) is probably right that at least for a subset of victims, "the very rhetorical appeal of the program may induce a sense of guilt in a reluctant victim." Indeed, the same point might be made of the moral obligation imposed on victim and offender supporters by restorative justice processes. That is the inevitable fallout of a program that seeks to get things done by nurturing citizenship obligations; it comes with a cost. E. Restorative Justice Practice Can Be a "Shaming Machine" that Worsens the Stigmatization of Offenders The "shaming machine" concern has been well articulated in Retzinger and Scheff's (1996) essay, "Strategy for Community Confer- HeinOnline -- 25 Crime & Just. 82 1999 Restorative Justice 83 ences: Emotions and Social Bonds," written after their observation of a number of Australian conferences, from which they came away concerned about the damaging effects of sarcasm, moral superiority, and moral lecturing in particular: The point about moral indignation that is crucial for conferences is that when it is repetitive and out of control, it is a defensive movement in two steps: denial of one's own shame, followed by projection of blame onto the offender.... For the participants to identify with the offender, they must see themselves as like her rather than unlike her (There but for the grace of God go I). Moral indignation interferes with the identification between participants that is necessary if the conference is to generate symbolic reparation. In our judgement, uncontrolled repetitive moral indignation is the most important impediment to symbolic reparation and reintegration. But on the other hand, to the extent that it is rechannelled, it can be instrumental in triggering the core sequence of reparation ... Intentional shaming in the form of sustained moral indignation or in any other guise brings a gratuitous element into the conference, the piling of shame on top of the automatic shaming that is built into the format. This format is an automatic shaming machine ... in a format that is already heavy with shame, even small amounts of overt shaming are very likely to push the offender into a defensive stance, to the point that she will be unable to even feel, much less express, genuine shame and remorse. (P. 13) Restorative justice processes are "already heavy with shame" as a result of the simple process of victims and their supporters talking about the consequences of the crime. In effect, that is all one needs. Umbreit (1994, p. 4) makes a similar point on victim defensiveness: "For individual victims, use of such terms as 'forgiveness' and 'reconciliation' is highly judgmental and preachy, suggesting a devaluing of the legitimate anger and rage the victims may be feeling at that point." Braithwaite and Mugford (1994) think that the best protection against the vices of moral lecturing and sarcasm is to do a good job of inviting a large number of caring supporters for both the victim and the offender, a point also discussed by Retzinger and Scheff (1996). If these invitees really do care about the offender, they will counter moral lecturing with tributes to the sense of responsibility and other virtues of the offender. Then, even if the sort of connection with the moral HeinOnline -- 25 Crime & Just. 83 1999 84 John Braithwaite lecturer that would allow productively reparative communication is severed, the bond with the other participant who comes to her defense is strengthened in the same sequence. For Braithwaite and Mugford (1994) this is the genius in the design of a Maori conference, a Cree healing circle, or Japanese school discipline, that is absent in the design of dyadic Western victim-offender mediationY Of course, training of facilitators to intervene against moral lecturing and ask for respectful discussion of consequences and solutions is also a remedy. Training of citizens through learning how to do restorative justice in school disputes is even more important: reason is more likely to prevail in democratic deliberation when citizens are educated to reasonableness (Barber 1992). Over the next few years there will be a flood of research coming out of the RISE experiment on what predicts the degeneration of conferences into defensive self-righteousness and their elevation into the symbolic reparation Retzinger and Scheff want. F. Restorative Justice Practices Rely on a Kind of CO'l1l11lltnity That Is Culturally Inappropriate to Industrialized Societies The most common assertion of critics of restorative justice, even in the face of thriving programs in large multicultural cities like Auckland, Minneapolis, Adelaide, and Singapore, is that it might work well in rural contexts but not in the metropolises of industrialized societies. The theory outlined in Section VII really makes a different kind of prediction, however: In our cities, where neighborhood social support is least, where the loss from the statist takeover of disputing is most damaging, the gains that can be secured from restorative justice reform are greatest. When a police officer with a restorative justice ethos arrests a youth in a tightly knit rural community who lives in a loving family, who enjoys social support from a caring school and church, that police officer is not likely to do much better or worse by the child than a police officer who does not have a restorative justice ethos. Whatever the police do, the child's support network will probably sort out the problem so that serious offending does Il As Mark Umbreit has pointed out to me, much victim-offender mediation is not dyadic. Other participants are often involved. He also rightly points out that dyadic encounters have their advantages too. Some things might be said one-on-one that could never be drawn out in front of the wider group. In short, some of the most successful conferences may adjourn for dyadic mediations; some of the most successful mediations may expand into conferences. HeinOnline -- 25 Crime & Just. 84 1999 Restorative Justice 85 not occur. But when a police officer with a restorative justice ethos arrests a homeless child in the metropolis like Sam, who hates parents who abused him, who has dropped out of school and is seemingly alone in the world, it is there that the restorative police officer can make a difference that will render him more effective in preventing crime than the retributive police officer. (Braithwaite 19980, pp. 18-19)14 Hagan and McCarthy's (1997, p. 163) research shows that homeless youth in Toronto and Vancouver were far from alone. A majority speak of their "street families" who look out for them: "you really learn what friendship is .... If I need them, they're there for me." In other words, part of our stigmatization of the homeless is to view them as somehow asocial, noncommunal. Certainly the restorative justice movement could be more conscious of helping other peoples to recover their own restorative traditions rather than showing them our own. I have suggested (Braithwaite 1996) the need for culturally specific investigation of how to save and revive the restorative justice practices that remain in all societies. Thence the following two elements for a research agenda: helping indigenous community justice to learn from the virtues of liberal statism-procedural fairness, rights, protecting the vulnerable from domination; and helping liberal state justice to learn from indigenous community justice-learning the restorative community alternatives to individualism (Braithwaite 1996). The design of restorative justice institutions can be rather minimalist. A conference, for example, can be defined by a strategy for who is invited and a small number of procedural rules about advising the defendant of a right to leave and take their chances in court, speaking in turn, and so on. Even "speaking in turn" may be too Eurocentric to be a minimal requirement as in some cultures it shows polite engagement to finish another person's sentence or to speak at the same time. Perhaps the ideal is undominated speech. The ideal is certainly not to be culturally prescriptive: to allow participants to begin and end 14 Put another way, the kind of theory I favor does not have a structurally or culturally detenninate explanation of crime: low Japanese crime rates are not about village culture or Japanese culture; to understand them we look to Japanese regulatory practices and Japanese provision of opportunities for human development. Changing regulatory practices and development opportunities can explain why Japanese crime rates fell from World War II in a way that it is hard for changing Japanese culture or the demise of village life to explain (Masters 1997). HeinOnline -- 25 Crime & Just. 85 1999 86 John Braithwaite a conference with a prayer if that is their wish, to include noisy babies if they wish or exclude them if they wish, to allow Samoan offenders to kneel at the feet of victims and First Nations Canadians to wash the feet of victims (Griffiths and Hamilton 1996), to communicate by a storytelling that may appeal to less formally educated members of a community (Young 1995) more than by a deductive reasoning that appeals to certain dominant men, or to lawyers. The sad fact is that this ideal is often not realized in restorative justice processes. Cunneen (1997), Findlay (1998), and Blagg (1998), for example, are right to point out that the interest of Australian Aboriginal people in participating in restorative justice alternatives was often assumed rather than discovered by reformers through an empowering dialogue with Aboriginal people (and other silenced minorities). Sometimes, we even inflict Maori process on young Maori who say they don't believe in "too much shit about the Maori way" (Maxwell and Morris 1993, p. 126). For all these failings, the design of restorative justice processes is for participant ownership and adaptation whereas the design of the Western criminal trial is for consistency-to be determinedly uniculturalone people, one law. The complex challenge for restorative justice is to improve the match between aspirations of design and reality of accomplishment. G. Restorative Justice Practices Can Oppress Offenders with a Tyranny of the Majority, Even a Tyranny of the Lynch Mob Empowering indigenous justice in many parts of the world can and does at times empower communities to kill offenders and more commonly to punish them corporally. Police in outback Australia are not coy to confess to criminologists that they allow the latter to happen; they would never let themselves be seen to allow the former. Liberal justice regimes that turn a blind eye to violent indigenous justice succumb to a dangerous kind of cultural relativism. It is one thing to accept the legitimacy of traditional forms of social control in a unicultural traditional society. In a multicultural society where all people learn to count on the state for protection of their rights, without state oversight of respect for fundamental human rights there is no way of being sure that those punished really are members of the traditional society, or even if they are, that they are not cultural dissidents who wish to call on the protections afforded to all citizens by the state regardless of race. Without state oversight, there is no way of assuring that the rights of a victim from a different cultural group than the offender will HeinOnline -- 25 Crime & Just. 86 1999 Restorative Justice 87 be protected. Moreover, as Ross (1996, p. 234) points out: "In many communities, the overnight withdrawal of the Western justice system would not be followed by the immediate substitution of effective aboriginal approaches, but by significant violence." Some Australian outback police, black and white, show considerable wisdom in communicating the message that traditional justice processes are encouraged to run their course so long as they do not cross certain lines. "If you do that, blackfella law will be pushed aside by whitefella law." Put another way, what such police do is encourage Aboriginal restorative traditions, but when they want to exercise their retributive traditions with any vigor, require them to put their case to a court of law. No citizens can feel secure in their rights when in some contexts the state is willing to sacrifice them to the tyranny of the lynch mob. This is not to say that courts have generally been less tyrannous than the mob. On the contrary, public executions stopped because the mob booed and pelted executioners as they carried out the horrors ordered by the courts (Hay 1975, pp. 67-68; Foucault 1977, pp. 61-67) and because juries refused to convict for minor offenses that would lead to the gallows (Trevelyan 1978, p. 348). Courts around the world still order executions (in private) in numbers that surely exceed those imposed by popular justice tribunals. Today, no popular justice fora impose other sanctions as barbarous as imprisonment, which Graeme Newman (1983) points out in practice can be considerably more barbarous than corporal punishment. Certainly, contemporary liberal courts have some upper limits on the barbarism they can indulge. However, it seems empirically wrong, both as a matter of attitude and practice, that courts are less punitive than victims and restorative justice fora. In practice, if courts in New Zealand were less punitive than family group conferences, they would be cutting conference agreements for community work and other sanctions on a regular basis, but we have seen that increments are much more common than cuts (Maxwell and Morris 1993). The Clotworthy case before the Court of Appeal of New Zealand, for example, has been challenging to the principles of restorative justice. 15 Mr. Clotworthy inflicted six stab wounds, which collapsed a lung and diaphragm of an attempted robbery victim. Justice Thorburn of J; The Queen v. Patrick Dale Clotworthy, Auckland District Court T. 971545, Court of Appeal of New Zealand, CA 114/98. HeinOnline -- 25 Crime & Just. 87 1999 88 John Braithwaite the Auckland District Court imposed a two-year prison sentence, which was suspended, a compensation order of $15,000 to fund cosmetic surgery for an "embarrassing scar," and two hundred hours of community work. These had been agreed at a restorative conference organized by Justice Alternatives. The judge found a basis for restorative justice in New Zealand law and placed weight on the wish of the victim for financial support for the cosmetic surgery and emotional support to end through forgiveness "a festering agenda of vengeance or retribution in his heart against the prisoner." The Court of Appeal allowed the victim to address it, whereupon the victim "reiterated his previous stance, emphasizing his wish to obtain funds for the necessary cosmetic surgery and his view that imprisonment would achieve nothing either for Mr. Clotworthy or for himself" (p. 12). The victory for restorative justice was that "substantial weight" was given by the court to the victim's belief that expiation had been agreed; their honors accepted that restorative justice had an important place in New Zealand sentencing law. The defeat was that greater weight was given to the empirical supposition that a custodial sentence would help "deter others from such serious offending" (p. 12). The suspending of the twoyear custodial sentence was quashed in favor of a sentence of four years and a $5,000 compensation order (which had already been lodged with the court); the community service and payment of the remaining compensation were also quashed. The victim got neither his act of grace nor the money for the cosmetic surgery. At the level of attitudes, Sessar (1998) has shown that judges and prosecutors in Germany have more punitive attitudes than the general public. In the United States, Gottfredson and Taylor (1987) found the general public to be no more retributive than correctional policy makers. Doob and Roberts's (1983, 1988) research shows that the more information the public has about particular cases, the less punitive they become. Hough and Roberts (1998) use data from the British Crime Survey to suggest that citizens derive their information about punishment primarily from the mass media, underestimate the severity of sentences actually imposed, and approve penalties lower than those actually imposed. Kerner, Marks, and Schreckling (1992) show that while a majority of Cologne victims who had been through victim-offender mediation felt the German justice system was too lenient, only 28 percent felt the treatment of "their" offender was too lenient. When citizens of most Western democracies answer opinion polls, they support capital punishment; when they sit on juries, they are much less supportive. Roberts and Stalans's (1997) literature review suggests that HeinOnline -- 25 Crime & Just. 88 1999 Restorative Justice 89 while in response to opinion polls citizens support tougher sentencing in the abstract, on more specific judgments about appropriate punishment they are not more punitive than the status quo and are very supportive of restitution as an alternative (when it is brought to their attention as an alternative). A number of studies find victims to be considerably less punitive than popular stereotypes would have it (Kigin and Novack 1980; Novack, Galaway, and Hudson 1980; Heinz and Kerstetter 1981; Shapland, Willmore, and Duff 1985; Sessar, Beurskens, and Boers 1986; Weitekamp 1989, pp. 83-84; Sessar 1990; Youth Justice Coalition 1990, pp. 52-54; Umbreit 1990b, 1994, pp. 913). McCold and Wachtel (1998, p. 35) found victims to be less punitive than victim supporters in Bethlehem conferences and much less punitive than offender supporters (though more punitive than the offenders themselves). This evidence is one reason the social movement for restorative justice has generally moved from seeing victims of crime movements as potential sources of resistance (see Scheingold, Pershing, and Olson 1994) to tangible sources of support. All of this serves to keep the problem of the tyranny of the majority in perspective. While it is extremely rare for victims to fly into rages of abuse, it does happen in restorative justice processes: and when it does, justice can be compromised. The remedy has to be an absolute right of the accused to walk out of the restorative justice process and try their chances in a court of law. H. Restorative Justice Practices Can Widen Nets of Social Control Polk (1994, p. 134) and Minor and Morrison (1996), among others, have expressed concern about the net-widening potential of con ferencing and other restorative justice programs. Systematic data to test such concerns are scarce. Maxwell and Morris (1996) do not find evidence of net widening as a result of the New Zealand restorative justice reforms of 1989. This is most particularly true at the most intensive end of intervention, with the number of places in residences for young offenders falling from two hundred to seventy-six and sentences that involve custody declining from an average of 374 a year prior to the juvenile justice reforms to 112 in 1990 (Maxwell and Morris 1996, p. 94). Forsythe (1995) found no net widening in the Wagga Wagga juvenile conferencing program; indeed found a small decrease in juvenile cases being processed by the justice system after the program's introduction. The Sparwood, British Columbia, police conferencing program discussed in Section VIlB was unique in terms of the net-widening hypothesis because it completely abolished court in favor of conferences HeinOnline -- 25 Crime & Just. 89 1999 90 John Braithwaite that were held much less frequently than court cases had been held in the past. Yeats (1997, p. 371) associates the introduction of restorative justice conferences and an expanded cautioning program for juveniles in Western Australia with more than a halving of the number of charges heard in the children's court in the mid-1990s. Dignan (1992, p. 462) found quite modest net widening in the Kettering evaluation of a British adult victim reparation program; while some experienced higher levels of intervention than they might otherwise have experienced, more experienced lower levels of intervention that they would otherwise have encountered. The John Howard Society's Restorative Resolutions project in Manitoba was designed to be confined to adult cases where the Crown had already recommended a prison sentence of six months or more (this being true in 90 percent of the cases) (Bonta, Rooney, and Wallace-Capretta 1998). This approach to restorative probation builds in strong guarantees of net narrowing and has considerable promise for reducing minority overrepresentation in prison. Net widening is most likely to occur with programs the police do not take seriously and which depend on referrals from the police. The police then refer cases they would not normally be bothered doing much about and the restorative justice program is motivated to get more cases by proving to the police that they are a tough option. A reverse of this situation is the New Zealand program. Here police support for conferencing is strong and in any case the police cannot send a juvenile to court without a youth justice coordinator having the chance to opt for a family group conference (with the offender's consent).16 It is worth considering why net widening is believed to occur in critiques of the welfare model of criminal justice, and whether these factors might not be so worrying in a restorative justice program aimed at securing freedom as nondomination (Walgrave 1995). Walgrave makes the point: "Educative and clinical arguments within a legal system imbue subjective and speculative approaches of the clinical practitioner with the coercive power of justice" (p. 230). With restorative justice, educative and clinical services are options selected by the joint deliberation of offenders and their community of care, who enjoy an absolute right of veto over them. Welfare professionals actually suffer a diminution in coercive state backing of their discretion to intervene 16 The New South Wales Young Offenders Act 1997 empowers a "specialist youth officer" to opt for a conference against the advice of the police and precludes police commencement of proceedings before giving the specialist youth officer an opportunity to consider the conference option. HeinOnline -- 25 Crime & Just. 90 1999 Restorative Justice 91 under the restorative justice model. As Walgrave (1995, p. 233) puts it, the restorative model is a move from "the state of power" and "the welfare state" to "the empowering state." On the present limited evidence, restorative justice more often narrows than widens nets of formal state control; but it does tend to widen nets of community control. Whether the nets that are widened are state or community nets, an assumption that net widening is a bad thing seems wrong. From a republican normative perspective, net widening that increases freedom as non domination is a good thing (Braithwaite and Pettit 1990). From this republican perspective, it is important to widen nets of social control over white-collar crime and domestic violence. Even with juveniles, we need to widen nets of community control over school bullying because this so demonstrably is oppressive of the freedom of victims, because restorative justice programs aimed at reducing bullying work (Olweus 1993), and because bullying is connected to other problems-victims being more susceptible to suicide than nonvictims and offenders being more supportive of adult wife abuse than children who are not bullies (Rigby, Whish, and Black 1994). If it is true that restorative justice narrows nets of judicial control and widens nets of community control, then another kind of critique of restorative justice swings into play. Brown (1994, p. 129) quotes Barbara Babcock who in turn is referring to an argument by Thurman Arnold: "The trial is an important occasion for dramatic enactment, the symbolic representation of the community's most deeply held values." Restorative justice advocates see this as the romantic vision of the law of trial lawyers (and Hollywood). The guilty plea cases that are the bread and butter of both criminal courts and restorative justice programs are devoid of courtroom drama; on average, Canberra RISE cases that go to court are over in about ten minutes and very few people are present to observe the "drama." Yes, then, the Pessimistic Account is right that restorative justice programs can widen nets of community control and even nets of state control for some types of crime. The question is whether they should. 1. Restorative Justice Practices Fail to Redress Structural Problems Inherent in Liberalism Like Unemployment and Poverty Critics correctly point out that there is little evidence of restorative justice programs conquering not just unemployment and poverty, but family breakdown and other problems that underlie offending (Waters HeinOnline -- 25 Crime & Just. 91 1999 92 John Braithwaite 1993). There is little doubt about the validity of this criticism. Even from the perspective of theories that motivate restorative justice, like the theory of reintegrative shaming (see Braithwaite 1995b), the main implications for crime control policy of the theories are not about the redesign of the criminal justice system. From my theoretical perspective, the low-crime society will be one with redemptive schooling (Knight 1985), a jobs compact that guarantees a job or training to all long-term unemployed (Braithwaite and Chappell 1994), vigorous social movement politics (Braithwaite 1995b), robust separations of private and public powers (Braithwaite 1997), a strong welfare state, strong markets, and strong plural communities (which include strong families that constitute independent individuals, and vigorous social movement politics) (Braithwaite 1998b). It will also be one that shames domination and denial of human rights as it promotes restorative justice. In short, it will be a civic republican society with a citizenry that takes seriously the virtues of liberty, equality, and fraternity/sorority. More abstractly, the low-crime society will be one whose citizens enjoy republican freedom as nondomination, with the rich panoply of institutional implications that follow from it (Pettit 1997). For republicans, it is therefore important that restorative justice not undercut other elements vital for progress toward a republican polity. It is hard to see in the literature evidence of it doing this. We do see retributive justice wrench young people away from loving families, give them criminal records that destroy their employability (Hagan 1991); restorative justice does not seem very guilty of this. At its best, it heals families, extends their bonds of care, gets young people who have been expelled back into school, or helps them find jobs. But restorative justice does not normally accomplish these things and it does not, cannot, have any substantial effect on unemployment and educational disadvantage. An entirely different institutional agenda is required for that. One of the things about the social movement for restorative justice is that the people in it are active in other arenas of struggle for justicein the women's movement, aboriginal rights movements, the peace movement, churches, the consumer movement, the environment movement, the development movement, the labor movement. They are not so naive as to believe that restorative justice as an alternative disputing philosophy can be the primary vehicle for securing the transformative changes they seek. Elections need to be won, constitutions rewritten, economies restructured, the World Trade Organization and the IMF reformed. Wearing the hat "criminologist" would make them look silly as they engage in those other struggles. HeinOnline -- 25 Crime & Just. 92 1999 Restorative Justice 93 It would be a mistake to skew any of these political agendas toward a preoccupation with crime control; they are much more important than that (see Crawford 1997). This is not to deny the importance of pointing out that it is harder for restorative justice programs to restore when they are not surrounded by the infrastructure of a decent welfare state and a vigorous civil society. Ken Polk (1994; see also White 1994) is right to emphasize the greater importance of "developmental institutions" of family, school, work, and recreation that confer positive identities. Developmental institutions are more important than disputing institutions (regardless of whether they are restorative or punitive). Shearing's Community Peace Foundation in South Africa is doing exciting work to transform the relationship between (restorative) justice and developmental institutions in circumstances so challenging that it may show us a new path. George Pavlich (1996) has a concern that runs deeper. It is that restorative justice restores individuals as subjects of liberal legality. It becomes "fundamentally implicated in the identity of liberal law" (Pavlich 1996, p. 714), which implies that "in the West popular justice, as it is currently understood, is impossible" (Fitzpatrick 1992, p. 199). Again restorative justice is pretty much guilty as charged here (see Shearing's (1995) appropriation of neoliberal discourses). As argued in Section VIIN and the conclusion, like the analysis in Habermas (1996), my analysis of restorative justice sees much that is worth preserving in liberal legalism. This analysis finds virtue in the institutionally specific kind of interplay between popular justice and the law bequeathed by the liberal tradition described in the conclusion. Liberal legalism is not the right culprit if we want to tackle unemployment, homelessness, educational disadvantage, sexism, racism, and the like. Without it, these problems would likely be even worse. We need to engage with other forms of social movement politics besides the social movement for restorative justice to tackle seriously the economic institutions that are the deeper villains here. At the same time, a republican justice that connects more of our private troubles to public issues will enable the law to be a slightly more useful tool against scourges like racism. J. Restorative Justice Practices Can Disadvantage Women, Children, and Oppressed Racial Minorities Disadvantaging of the powerless can occur through treating them harshly or through silencing them. On the latter, Gabrielle Maxwell (1993, p. 292) actually concludes that restorative justice conferences 'lrp "nlaces where women's voices are heard" (see also Burford and HeinOnline -- 25 Crime & Just. 93 1999 94 John Braithwaite Pennell's [1998] findings). Similarly, Rigby (1996, p. 143) with data from eighty-five hundred students shows that at all ages girls are more interested than boys in talking through bullying problems in school programs. Daly (1996) reports that while a minority (15 percent) of offenders were women in her study of Australian conferences, 54 percent of victims, 58 percent of victim supporters, and 52 percent of offender supporters were women. In Canberra, offenders randomly assigned to conferences are considerably less likely than offenders assigned to court to say that they were disadvantaged in proceedings due to "age, income, sex, race, or some other reason" (Sherman and Barnes 1997). While there was no comparison with court, Joe Hudson's (1998) study of Canadian family group conference participants found 80 percent to be "very satisfied" with the way all conference participants were treated as equals. Morris et a1. (1996, p. 224) conclude from the consideration of the issue across the set of contributions to their volume: "Fears raised by commentators about the disempowerment of women have not been supported by observers and researchers who note their active participation in the process in contrast with their nonparticipation in judicial processes." Our research group's qualitative observations of restorative justice at various sites is that women's voices in restorative justice conferences are often extremely influential; in juvenile conferences if we were to nominate one type of actor who is more likely to be influential in the outcome than any other, it might be the mother of the offender. This may be a good thing from the perspective of empowerment of women in a deliberative democracy. It constitutes a kind of empirical response to feminist critics of republicanism and deliberative democracy who fear the pursuit of communitarian consensus that may be a male consensus (Phillips 1991; Young 1995). At the same time, it opens up another feminist concern about restorative justice. It is that women again bear the burden of all the unpaid caring (Daly 1996). The potential fiscal benefit of conferences that they may be cheaper than courtroom justice is a benefit likely to be carried on women's backs. Hughes (1996) explores this concern through considering Campbell's (1993) analysis that in "Britain's dangerous places," it is women who are the "community builders," while men deal with unemployment by indulging a cult of selfish irresponsibility and brute force. In Britain's highcrime communities: "Crime and coercion are sustained by men. Solidarity and self-help are sustained by women. It is as stark as that" (Campbell 1993, p. 319). No, the data do not suggest it is as stark as HeinOnline -- 25 Crime & Just. 94 1999 Restorative Justice 95 that. Yet there seems little doubt that women do more of the restoring than men in restorative justice processes. The price tag for communitarian empowerment (that most women say they want in all the interview-based research) is a gendered burden of care. A related worry about restorative discourses is that they may domesticate violence. Sara Cobb (1997, p. 414) finds that "the morality of mediation itself" can frame the interpretation of action, subsuming, taming the morality of right and wrong so that "the category 'victim' dissolves" (p. 436). Some restorative justice advocates view it as a good thing for the category victim to dissolve, indeed for the category "crime" to dissolve. There is a divide between some mediation advocates who believe the mediator should be "neutral" and some conferencing advocates who see a conference beginning from an assumption that a crime has been admitted about which the facilitator is not neutral, but disapproving. Rights are not necessarily domesticated to needs (Minow 1990) by a justice that "erases any morality that competes with the morality of mediation." The right kind of interplay between the justice of the law and the restorative justice of the people might secure rights as trumps against the morality of mediation, a topic I return to in the conclusion. An opposite kind of claim about disadvantage sometimes made about restorative justice programs is that they are a benefit granted disproportionately to Caucasians. It is not an allegation I have seen made in New Zealand and Canada where so many of the leading programs have been run by and for First Nation peoples. However, there have been some worrying suggestions of the validity of such a concern in the United States (Gehm 1990; Brown 1994; Schiff 1998) and Australia (Daly 1996; Wundersitz and Hetzel 1996; Blagg 1997). K Restorative Justice Practices Are Prone to Capture by the Dominant Group in the Restorative Process Restorative business regulatory practices are frequently captured by business (Clinard and Yeager 1980, pp. 106-9). As Dingwall, Eekelaar, and Murray (1983) have shown, child protection practice is liable to "family capture." As with business regulation, Dingwall, Eekelaar, and Murray show that a "rule of optimism" prevails among family regulators who have a bigger caseload than they can manage. Sandor (1993) even worries that family group conferences might ignite episodes of physical abuse, so common in the lives of serious juvenile offenders. This is a worry that deserves testing by a major empirical study. Indig- HeinOnline -- 25 Crime & Just. 95 1999 96 John Braithwaite enous justice can empower elders to tyrannize the young of their tribe. Critics have alleged this in a most alarming way in Canada through allegations by women from a reserve that project leaders of a program of indigenous justice administered by a panel of elders "manipulated the justice system to protect family members who had committed violent rapes, had intimidated victims and witnesses into withdrawing charges, had perjured themselves during the trial of the project leader's son (for rape), had slashed tyres of community members who tried to speak out and sent the alleged 'rape gangs' to their homes, and generally had used the project to further their strangle-hold on the community and the justice system." 17 In New Zealand, I saw one tragic conference where the state funded the travel of an offender to another community because his whanau (extended family) wanted to separate him from a liaison with a girlfriend it did not want. In pushing for this, the Youth Justice Advocate was not an advocate for the youth, who was heartbroken by this outcome, but was captive of the whanau, which was the repeat player in the use of his legal services. Observational work on juvenile justice conferences quite regularly reports lower levels of offender involvement than involvement by their family members. Maxwell and Morris's (1993, pp. 110-12) interviews found fully 45 percent of young offenders, compared to 20 percent of family members saying they were not involved in making the conference decision. In Canberra and South Australia, Daly (1996) reported 33 percent of offenders not to be engaged with the process. The Maxwell and Morris (1993) data showed family members of the offender having by far the largest influence on the decision, followed by professionals who were present, the young offender, and the victim (not surprising since the victim was absent from a majority of the conferences in this study). Kevin Haines's (1999) critique of conferences as a "room full of adults" who dominate a child is therefore often correct. All such failures are relative, however: the RISE experiment in Canberra shows that young offenders are considerably more likely to believe that they could express their views when they went to a conference than when they went to court (Sherman and Barnes 1997; Sherman et al. 1998, pp. 121-22). The best remedy to this problem is systematic attention in the restorative justice preparatory process to empowerment of the most vul17 This is a quote that I treat as anonymous with respect to person and place. I was able to confirm the same broad story from two other sources. HeinOnline -- 25 Crime & Just. 96 1999 Restorative Justice 97 nerable parties-individual victims and offenders-and systematic disempowerment of the most dominant parties-the police, school authorities, state welfare authorities, sometimes large business corporations. How is this accomplished? The most critical thing is to give the individual offender and the individual victim the one-on-one power in a meeting in advance of the conference to decide who they do and do not want to be there to support them. Unfortunately, the practice is often to empower the parents of young offenders to decide who should be there. They can certainly have a legitimate say; but on the offender side it is only the offender who should make the final decision about who will make her most comfortable, whom she most trusts. To the extent that one is concerned here with imbalances of power berween children and adults, men and women, major corporations and consumers, dyadic victim-offender mediation cements an imbalance. Imbalances are muddied, though hardly removed, by conferences berween rwo communities of care, both of which contain adults and children, men and women, organized interests (like Aboriginal Community Councils in the CML case) and disorganized individuals. Simple rules of procedure can privilege less dominant voices over state voices. The police should never be allowed to give their version of what happened in advance of the offender's version. The New Zealand and South Australian practice of giving attending police a right of veto over the conference agreement is a bad practice from this perspective. The victim and the offender should have a right of veto, and should be formally reminded of it at the start of the conference, but perhaps no agent of the state should have such a right as a conference participant. On this view, if the police feel a victim has dominated an offender by requiring excessive punishment, their approach should be to advise the offender of their right to walk away from the agreement and have the matter heard by a court. Against this view is a nonRISE case we had in Canberra where the police did not veto an agreement, enthusiastically crafted by the victim and the mother of the offender, to have a child wear a T-shirt emblazoned with "I am a Thief." Certainly, formal negative performance indicators for facilitators should include how much they talk, 18 dominating proceedings, and participation in setting the terms of the conference agreement. " Perhaps I am partially wrong here, however, if the more general literature on the effectiveness of problem solving in mediation is a guide. Carnevale and Pruitt's (1992, p. 565) review concludes that when disputants are able to resolve the dispute themselves, mediator intrusiveness gets in the way, but when conflict intensity or hostility are high, interventionist mediation can improve outcomes. HeinOnline -- 25 Crime & Just. 97 1999 98 John Braithwaite With restorative business regulation, Ayres and Braithwaite (1992) have shown game-theoretically the importance of having third parties present during regulatory negotiation to protect against corporate capture. With nursing home regulation, for example, this can mean representatives of the residents' committee, supported by advocacy groups where they want this, relatives, staff and their unions, and outside board members as well as management meeting with the regulators. While there are such measures we can take to counterbalance capture, there can be no doubt the Pessimistic Account is right that the risk of capture by dominant groups is an ineradicable reality of restorative justice Gust as it is of state justice). L. Restorative Justice Processes Can Extend Unaccountable Police Power, Even Compromise the Separation of Powers among Legislative, Executive, and Judicial Branches of Government Critics such as Danny Sandor (1993) and Rob White (1994) are particularly concerned about conferences facilitated by the police, something now happening in Australia, the United Kingdom, the United States, and Canada, and being advocated elsewhere. It is all very well to reply that facilitation is not control, that police should not have a veto power over decisions, but no one can deny that good facilitators have "dramatic dominance" (to use Stephen Mugford's Goffmanesque characterization) even if they exert little direct control. This dramatic dominance ensures among other things that the conference is orderly, that everyone has their turn to speak without interruption, that civility triumphs over abuse. Supporters of police conferencing in police stations, such as the Police-Citizens Consultative Committee in Wagga Wagga, claim this lends "gravitas" to proceedings. These citizens and police also argue that police facilitation, indeed the presence of the police uniform, helps victims feel secure, a critical problem given the evidence that victims often come away from criminal justice processes, including restorative ones, feeling more afraid of their offender (see Sec. VIIA). Interesting and important speculations, but there is no evidence to support the importance of such alleged advantages. Conversely, there is no evidence to support counterclaims that offenders are intimidated by the presence of a police uniform during a conference or by the police station as a venue. It seems unlikely that either is totally false; many young people do distrust the police and many crime victims distrust "do-gooder social workers" and trust the police. What does seem true, by definition, is that hard-line views on either HeinOnline -- 25 Crime & Just. 98 1999 Restorative Justice 99 side of this debate do disempower and disadvantage participants. A rule that police must or must not facilitate conferences has unfortunate consequences when it precludes someone from taking the facilitator role who is the most gifted person who most enjoys the confidence of the disputants. It has perhaps even more unfortunate consequences when it forces police officers to facilitate conferences when they do not believe this is "police work" (see Hoyle and Young 1998). In New Zealand, where police do not facilitate conferences, I saw one conference where a Pakeha official facilitator handed over the effective facilitation of the conference to a police officer because the police officer was Maori and all the participants Maori. A rule that Maori conferences must be held on the traditional Marae might be most unwise if that would terrify a Samoan victim. A rule that it be held on neutral ground is sad if one party is quite stipulative about where she will feel safe about the conference and the other party does not care. In my unsystematic observation on this question, most parties do not seem to care greatly about where conferences are held; but when they do express a strong preference, why not yield to them? The more fundamental question is whether there is something wrong in principle with police facilitating a conference. Does it make the police investigator, prosecutor, judge, and jury? It is only a very partial answer to reply that the investigator is never the police facilitator in such programs. Even if the police have no veto over the conference decision, they can still dominate it and become a de facto judge and jury. At least a lay facilitator who dominates a conference process does not add that domination to institutional domination over the decision to proceed with the matter. At a lower level, this is also a separation of powers argument against facilitation control in the hands of any state agency that already has control over another part of the process-such as control by the courts (as in South Australia) or a juvenile justice or welfare agency. It is not an argument against a prosecutor or judge having a right of veto over the outcome of a conference. Is there therefore any case for control of facilitation by a state agency rather than facilitator recruitment by an institution in civil society, especially when that civilian facilitator has the power to co-opt a police facilitator if that is what the parties want? There are two cases for state control. One is simply politically pragmatic. Restorative justice reform requires enormous energy and political will to struggle against retributivism and vested interests, not least within police forces. My own position has been politically pragmatic in just this way: to ad- HeinOnline -- 25 Crime & Just. 99 1999 100 John Braithwaite mire those with the courage to take on the battle, wherever they pop up institutionally-in civil society, the police, the courts, prosecutors, state welfare agencies. During the research and design era of restorative justice, so long as the activists are sensible and competent as well as committed, let us encourage openness to research results from their innovation. We really have so little data on who is right about these questions of comparative advantage in institutional location. My own suspicion is to think that success is 70 percent driven by attention to getting implementation detail right and only 30 percent by getting the institutional infrastructure right. Gifted people can run wonderful restorative justice programs in an open field with no infrastructure whatsoever. They seem to have done so for millennia. But that suspicion is itself something that must be open to empirical refutation. A second argument for institutional location of conferencing in a police service is about the transformation of police cautioning and police culture more broadly. Even in New Zealand, with the largest juvenile conferencing program anywhere, for every juvenile case dealt with by a conference, more than five are dealt with by a police caution. Rendering police cautioning more restorative is more important than rendering conference or court processing more restorative. From the perspective of the theory of reintegrative shaming, if, as seems to be the case (Braithwaite 1995c), stigmatizing interactions do more to increase crime than reintegrative ones do to reduce it, five stigmatizing police cautions will do more damage than any good from one reintegrative conference. Not just in formal cautioning but in daily interaction on the street, the challenge of transforming police culture from a stigmatizing to a restorative style is important. In a place like Wagga Wagga, the openness to inviting critics from the community, prosecutors, and the like to sit in on the sergeants' committee, where decisions were made to encourage constables to send cases to conference rather than court, constituted an interesting moment in the history of police accountability and culture change. The hope of the leading police reformers in this area, like Terry O'Connell in Australia, has been that police "ownership" of conferencing will imbue police commitment to restorative justice in wider arenas, including their own internal affairs (corruption, sexual harassment). At this stage, however, I doubt if anyone could plausibly demonstrate that any police service has experienced a major change of corporate culture as a result of restorative justice innovation. Perhaps Thames Valley, England, is beginning to approach this situation. The evidence is clear that significant police HeinOnline -- 25 Crime & Just. 100 1999 Restorative Justice 10 1 cultural change has not occurred in Bethlehem, Pennsylvania (McCold and Wachtel 1998, p. 3), though the individual police most exposed to conferences did move toward a more restorative and less crime-control oriented philosophy of policing. But then cultural change is never rapid and always resisted. If the relationship of any restorative justice program with the police is not well managed, disaster is courted. The police are the gatekeepers to the criminal justice system and if they shut the gate to restorative justice, it is nigh impossible to push it open against their resistance. All manner of hybrids are possible. One is for a conferencing unit to be located in police stations, pushing internally to divert cases as they come through the station door, caressing and cajoling police cooperation, but with the facilitators actually being employees of an institution in civil society contracted by the justice minister or other government agency to deliver the service. Some of these facilitators could be respected businesspeople who would have the sophistication to run conferences for complex white-collar crimes, others could have special language skills, others special gifts in gaining the confidence of young people, others could be elders of a local indigenous group. More broadly, from a republican perspective, one wants to see most restorative justice conferencing transacted in civil society without ever going through the police station door-in Aboriginal communities, schools, extended families, churches, sporting clubs, corporations, business associations, and trade unions. Equally, one wants to see those community justice processes subject to state oversight for breaches of citizens' rights and procedural fairness. In such a world, restorative justice would contribute to the building of a republican democracy with a much richer separation of powers. That is not the world we live in yet. For the moment, the restorative justice debate is debilitated byexcessively statist preoccupations to the point where the reforms in place do raise some legitimate worries about impoverishing rather than enhancing the separation of powers in our democracies. M. Restorative Justice Practices Can Trample Rights because of Impoverished Articulation of Procedural Safeguards Robust critiques of the limitations of restorative justice processes in terms of protection of rights have been provided by Warner (1994), Stubbs (1995), Bargen (1996), and Van Ness (1999). There can be little doubt that courts provide superior formal guarantees of procedural fairness than conferences. HeinOnline -- 25 Crime & Just. 101 1999 102 John Braithwaite At the investigatory stage, Warner (1994, p. 142) is concerned: "Will police malpractice be less visible in a system which uses FGCs [family group conferences]? One of the ways in which police investigatory powers are scrutinized is by oversight by the courts. If the police act unlawfully or unfairly in the investigation of a case, the judge or magistrate hearing the case may refuse to admit the evidence so obtained or may criticize the police officer concerned. Allegations of failure to require parental attendance during questioning, of refusal to grant access to a lawyer, of unauthorized searches and excessive force could become hidden in cases dealt with by FGCs." These are good arguments for courts over restorative justice processes in cases where guilt is in dispute. But the main game is how to process that overwhelming majority of cases where there is an open and shut admission of guilt. Here no such advantage of court over conference applies, quite the reverse. As Warner herself points out, a guilty plea "immediately suspends the interests of the court in the treatment of the defendant prior to the court appearance" (Hogg and Brown 1985). In the production line for guilty pleas in the lower courts there is not time for any of that. In restorative justice conferences there is. Mothers in particular do sometimes speak up with critical voices about the way their child has been singled out, has been subject to excessive police force, and the like. Police accountability to the community is enhanced by the conference process. And such deterrence of abuse of police power that comes from the court does not disappear since the police know that if relations break down in the conference, the case may go to court as well. Police therefore have reason to be more rather than less procedurally just with cases on the conference track than with cases on the court track. The preliminary RISE data from Canberra suggest they are. In about 90 percent of cases randomly assigned to a conference, offenders thought the police had been fair to them ("leading up to the conference" and "during"); but they only thought this in 48-78 percent (depending on the comparison) of the cases randomly assigned to court (Sherman et al. 1998). Offenders were also more likely to say they trusted the police after going through a conference with them than after going through a court case with them. At the adjudicatory stage, Warner (1994) is concerned that restorative justice will be used as an inducement to admit guilt. In this, restorative justice is in no different a position than any disposition short HeinOnline -- 25 Crime & Just. 102 1999 Restorative Justice 103 of the prospect of execution or life imprisonment. Its proffering can induce admissions. Systemically though, one would have thought that a shift from a punitive to a restorative justice system would weaken the allure of such inducements. In the preliminary data from the four RISE experiments in Canberra, there is a slight tendency for court offenders to be more likely than conference offenders to agree that, "The police made you confess to something which you did not do in this case." But this difference is only statistically significant in the Juvenile Personal Property experiment (Sherman et al. 1998, pp. 123-24). Warner (1994) is right, however, to point out that guilt is not always black and white. Defendants might not understand self-defense, intoxication, and other defenses that might be available to them. Even so, it remains the case that such matters are more likely to be discussed in a conference lasting about eighty minutes (Canberra data) than in a court case averaging about ten minutes (Canberra data). This may be a simple reason why Canberra offenders who go through a conference are more likely to believe that the proceedings "respected your rights" than offenders who went through court (Sherman and Barnes 1997; Sherman et al. 1998). At the dispositional or sentencing stage, Warner (1994) makes some good points about the care needed to ensure that sentences reflect only offenses the evidence in this case has shown to have been committed and only damage the evidence shows to have been done. We have had conferences in Canberra where victims have made exaggerated claims of the damage they have suffered, in one case many thousands of dollars in excess of what more thorough subsequent investigation proved to be the truth. Warner (1994) and Van Ness (1999) are both concerned about double jeopardy when consensus cannot be reached at a conference and the matter therefore goes to court, though Warner (1994) concedes it is not "true double jeopardy." Indeed it is not. The justice model analogue would seem to be to retrial after a hung jury (which no one would call double jeopardy) rather than retrial after acquittal. Moreover, it is critical that defendants have a right to appeal in court an unconscionable conference agreement they have signed, to have lawyers with them at all stages of restorative justice processes if that is their wish, and that they be proactively advised of these rights. Most restorative justice programs around the world do not legally guarantee the American Bar Association's (1994) guideline that "statements made by victims and offenders and documents and other materi- HeinOnline -- 25 Crime & Just. 103 1999 104 John Braithwaite als produced during the mediation/dialogue process [should be] inadmissible in criminal or civil court proceedings." This is a problem that can and should be remedied by appropriate law reform. Van Ness (1999) has systematically reviewed the performance of restorative justice programs for juveniles against the United Nations Standard Minimum Rules for the Administration of Juvenile Justice ("The Beijing Rules"). Restorative justice programs are certainly found wanting in the review though he concludes that they often tend to outperform traditional court processes on rules such as right to a speedy trial. For example, the New South Wales Young Offenders Act 1997 has the following requirement: "Time Limit Holding Conferences: A conference must, if practical, be held not later than twenty-one days after the referral for the conference is received." While Van Ness's work certainly affirms our hypothesis that restorative justice processes can trample on rights, where rights will be better or worse protected after the introduction of a restorative justice program is a contextual matter. For example, when in South Africa prior to the Mandela Presidency thirty thousand juveniles a year were being sentenced by courts to flogging, who could doubt that the institutionalization of restorative justice conferences might increase respect for children's rights, as Sonnekus and Frank (1997, p. 7) argue: "[Under Apartheid] the most common sentence given was corporal punishment and children often preferred a whipping instead of residential care in a reformatory or school of industry. The time children spent in prison while awaiting trial and placement was not applied toward their sentence, thus a child may have served double and even triple sentences." Nevertheless, our Pessimistic Account is correct that rights can be trampled because of the inferior articulation of procedural safeguards in restorative justice processes compared to courts. The conclusion grapples with how justice might be enhanced in the face of this critique by a creative interplay between restorative fora and traditional Western courts. IX. Conclusion There are good preliminary theoretical and empirical grounds for anticipating that well-designed restorative justice processes will restore victims, offenders, and communities better than existing criminal justice practices. More counterintuitively, a restorative justice system may deter, incapacitate, and rehabilitate more effectively than a punitive system. This will be especially so if restorative justice is embedded in HeinOnline -- 25 Crime & Just. 104 1999 Restorative Justice 105 a responsive regulatory framework that opts for deterrence when restoration repeatedly fails and incapacitation when escalated deterrence fails. We find active deterrence under a dynamic regulatory pyramid to be more powerful than passive deterrence in a sentencing grid; community incapacitation is more variegated and contextually attuned than clumsy carceral incapacitation. In the face of all the discretion that community responsiveness implies, most surprising of all is the conclusion that restorative justice is more just than the justice of the justice model. Empirical evidence of community perceptions of justice under the two models strongly supports this. Normative theory of a republican cast explains why we should get this result. Restorative justice delivers freedom as nondomination in a way just deserts cannot and citizens in democracies have profoundly deep aspirations to freedom and deep distrust of domination. Restorative justice confronts the dilemma that equal justice for offenders is utterly incompatible with equal justice for victims. I have argued that a greater degree of equality for both is delivered by rejecting equality as a goal, guaranteeing victims a minimum level of care, and guaranteeing offenders against punishment beyond a maxImum. For all of this hope about the advantages of restorative justice over the models with which it must compete, restorative justice offers limited prospects of a revolutionary improvement in the circumstances of victims or the control of crime. The primary reason for this is that the most fundamental things we must do to control crime and thereby improve the lot of victims are not reforms to the justice system. They are reforms about liberty, equality, and community in more deeply structural and developmental senses. Even so, just disputing processes have an important role to play in connecting private troubles to public issues. When communities start taking responsibility for the vulnerabilities of their young offenders and start talking about these vulnerabilities at and after conferences, of course they become more engaged with the deeper institutional sources of the problems. 19 When communities begin taking responsibility for family violence, as at Hollow Water, a profoundly institutional debate is triggered (Lajeunesse 1993; Ross 1996; Aboriginal Corrections Policy Unit 1997a, 1997b; Green 1998). When communi19 For the profoundly institutional way the citizens of Wagga did this, see City of Wagga Wagga (1993). HeinOnline -- 25 Crime & Just. 105 1999 106 John Braithwaite ties engage with their victimization by powerful corporations, as with the Aboriginal insurance cases, the imagination of prime ministers can be caught up in the aspirations for restructuring the regulation of finance capital. Restorative justice can trample the rights of offenders and victims, dominate them, lack procedural protections, and give police, families, or welfare professionals too much unaccountable power. Braithwaite and Parker (1998) suggest three civic republican remedies to these problems: contestability under the rule of law, a legal formalism that enables informalism while checking the excesses of informalism; deindividualizing restorative justice, muddying imbalances of individual power by preferring community conferences over individual-on-individual mediation; and vibrant social movement politics that percolates into the deliberation of conferences, defends minorities against tyrannies of the majority, and connects private concerns to campaigns for public transformation. Lawyers who work for advocacy groups-for indigenous peoples, children, women, victims of nursing home abuse-have a special role in the integration of these three strategies. Lawyers are a strategic set of eyes and ears for advocacy groups that use specific legal cases to sound alarms about wider patterns of domination. When appropriate public funding is available for legal advocacy, advocates can monitor lists of conference outcomes and use other means to find cases where they should tap offenders or victims on the shoulder to advise them to appeal the conference agreement because they could get a better outcome in the courts. They thus become a key conduit between rule of law and rule of community deliberation. It is a mistake to see their role as simply one of helping principles of natural justice and respect of rights to filter down into restorative justice. It is also to assist movement in the other direction-to help citizens to percolate up into the justice system their concerns about what should be restored and how. A rich deliberative democracy is one where the rule of law shapes the rule of the people and the concerns of the people reshape the rule of law. Top-down legalism unreconstructed by restorative justice from below is a formula for a justice captured by the professional interests of the legal profession (the tyranny of lawyers). Bottom-up community justice unconstrained by judicial oversight is a formula for the tyranny of the majority. When law and community check and balance each other, according to Braithwaite and Parker (1998), prospects are best HeinOnline -- 25 Crime & Just. 106 1999 Restorative Justice 107 for a rich and plural democracy that maximizes freedom as nondomination. Communitarianism without rights is dangerous. Rights without community are vacuous. Rights will only have meaning as claims the rich can occasionally assert in courts of law unless community disapproval can be mobilized against those who trample the rights of others. Restorative justice can enliven rights as active cultural accomplishments when rights talk cascades down from the law into community justice. None of the problems in the Pessimistic Account is satisfactorily solved. None of the claims in the Optimistic Account is satisfactorily demonstrated. Decades of research and design on restorative justice processes will be needed to explore my suspicion that the propositions of both the Optimistic and Pessimistic Accounts are right. For the moment, we can certainly say that the literature reviewed here does demonstrate both the promise and the perils of restorative justice. It is, however, an immature literature, short on theoretical sophistication, on rigorous or nuanced empirical research, far too dominated by selfserving comparisons of "our kind" of restorative justice program with "your kind" without collecting data (or even having observed "your kind" in action). That disappoints when the panorama of restorative justice programs around the globe is now so dazzling, when we have so much to learn from one another's contextual mistakes and triumphs. REFERENCES Aboriginal Corrections Policy Unit. 1997a. The Four Circles of Hollow Water. Aboriginal Peoples Collection. Ottawa: Solicitor General Canada. - - - . 1997 b. Responding to Sexual Abuse: Developing a Community-Based Sexual Abuse Response Team in Aboriginal Communities. Ottawa: Solicitor General Canada. Agnew, R., and A. A. Peters. 1986. "The Techniques of Neutralization: An Analysis of Predisposing and Situational Factors." Criminal Justice and Behavior 13:81-97. Ahmed, Eliza. 1999. "Shame Management and Bullying." Ph.D dissertation, Australian National University, Department of Psychology. American Bar Association. 1994. "Victim-Offender Mediation/Dialogue Pro- HeinOnline -- 25 Crime & Just. 107 1999 108 John Braithwaite gram Requirements." Resolution adopted by the American Bar Association House of Delegates. Anderson, Elijah. 1999. Code of the Streets. New York: Norton. Ayres, I., and]. Braithwaite. 1992. Responsive Regulation: Transcending the Deregulation Debate. New York: Oxford University Press. Ball, R. A. 1983. "Development of Basic Norm Violation: Neutralization and Self-Concept within a Male Cohort." Criminology 21:75-94. Ban, Paul. 1996. "Implementing and Evaluating Family Group Conferences with Children and Families in Victoria Australia." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Barber, Benjamin R. 1992. An Aristocracy of Everyone: The Politics of Education and Future of America. New York: Oxford University Press. Bardach, E., and R. A. Kagan. 1982. Going by the Book: The Problem of Regulatory Unreasonableness. Philadelphia: Temple University Press. Bargen,]. 1996. "Kids, Cops, Courts, Conferencing and Children's Rightsa Note on Perspectives." Australian Journal of Human Rights 2(2):209-28. Barnett, Randy E. 1981. "Restitution." In Perspectives on Crime Victims, edited by Burt Gallaway and Joe Hudson. St. Louis: Mosby. Baumrind, D. 1973. "The Development ofInstrumental Competence through Socialization." In Minnesota Symposium of Motivation, vol. 7, edited by A. D. Pick. Minneapolis: University of Minnesota Press. - - - . 1978. "Parental Disciplinary Patterns and Social Competence in Children." Youth and Society 9:239-76. Bayley, David H. 1976. Forces of Order: Police Behavior in Japan and the United States. Berkeley: University of California Press. Bazemore, G., and M. Umbreit 1994. Balanced and Restorative Justice: Program Summary: Balanced and Restorative Justice Project. Washington, D.C.: U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention. Bazemore, G., and C. Washington. 1995. "Charting the Future of the Juvenile Justice System: Reinventing Mission and Management." Spectrum (Spring), pp. 51-66. Beck, Guy L. 1997. "Fire in the Atman: Repentance in Hinduism." In Repentance: A Comparative Perspective, edited by Amitai Etzioni and David E. Carney. New York: Rowman & Littlefield. Benson, M. L. 1989. "Emotions and Adjudication: A Study of Status Degradation among White-Collar Criminals." Unpublished manuscript. Knoxville: University of Tennessee, Department of Sociology. Berman, Harold]. 1983. Law and Revolution: The Formation of the Western Legal Tradition. Cambridge, Mass.: Harvard University Press. Birchall, P., S. Namour, and H. Syme. 1992. "Report on the Midland Pilot Reparation Scheme." Unpublished manuscript. Perth: independently prepared. Blagg, Harry. 1997. "A Just Measure of Shame? Aboriginal Youth and Conferencing Australia." British Journal of Criminology 37 :481-50 1. HeinOnline -- 25 Crime & Just. 108 1999 Restorative Justice 109 1998. "Restorative Visions and Restorative Justice Practices: Conferencing, Ceremony and Reconciliation in Australia." Current Issues in Criminal Justice 10:5-14. Boden, Leslie I. 1983. "Government Regulation of Occupational Safety: Underground Coal Mine Accidents, 1973-1975." Unpublished manuscript. Boston: Harvard School of Public Health. Bond, Michael H., and Wang Sung-Hsing. 1983. "China: Aggressive Behavior and the Problem of Maintaining Order and Harmony." In Aggression in Global Perspective, edited by Arnold P. Goldstein and Marshall H. Segall. New York: Pergamon. Bonta, James, Jennifer Rooney, and Suzanne Wallace-Capretta. 1998. Restorative Justice: An Evaluation of the Restorative Resolutions Project. Ottawa: Solicitor General Canada. Box, S. 1981. Deviance, Reality and Society. London: Holt, Rinehart & Winston. Braithwaite, John. 1984. Corporate Crime in the Pha1711aceutical Industry. London and Boston: Routledge & Kegan Paul. ---,. 1985. To Punish 01· Persuade: Enforcement of Coal Mine Safety. Albany, N.Y.: SUNY Press. - - - . 1989. Crime, Shame and Reintegration. Cambridge: Cambridge U niversity Press. - - - . 1991. "Thinking about the Structural Context of International Dispute Resolution." In Whose New World Order: What Role for the United Nations? edited by M. R. Bustelo and P. Alston. Sydney: Federation Press. - - - . 1993a. "Shame and Modernity." British Jou171al of Criminology 33:1-18. - - - . 1993b. "Beyond Positivism: Learning from Contextual Integrated Strategies." Jounzal of Research in Crime and Delinquency 30:383-99. - - - . 1994. "The Nursing Home Industry." In Beyond the Law: Crime in Complex Organizations, edited by Michael Tonry and Albert]. Reiss, Jr. Vol. 18 of Crime and Justice: A Review of Research, edited by M. Tonry. Chicago: University of Chicago Press. - - - . 1995a. "Corporate Crime and Republican Criminological Praxis." In C01porate Crime: Ethics, Law and State, edited by F. Pearce and L. Snider. Toronto: University of Toronto Press. - - - . 1995b. "Inequality and Republican Criminology." In C1·ime and Inequality, edited by ]. Hagan and R. Peterson. Palo Alto, Calif.: Stanford University Press. - - - . 1995c. "Reintegrative Shaming, Republicanism and Policy." In Crime and Public Policy: Putting Theory to Work, edited by Hugh Barlow. Boulder, Colo.: Westview. - - - . 1996. "Restorative Justice and a Better Future." Dalhousie Review 76(1):9-32. - - - . 1997. "On Speaking Softly and Carrying Sticks: Neglected Dimensions of Republican Separation of Powers." University of Toronto Law Jounzal 47:1-57. - - - . 1998a. "Institutionalizing Distrust, Enculturating Trust." In n·ust and Democratic Governance, edited by V. Braithwaite and M. Levi. New York: Russell Sage Foundation. HeinOnline -- 25 Crime & Just. 109 1999 110 John Braithwaite 1998b. "Repentance Rituals." Unpublished manuscript. Canberra: Australian National University. - - - . 1999. "Domination, Quiescence and Crime." Unpublished manuscript. Canberra: Australian National University. Braithwaite, John, and Valerie Braithwaite. 1995. "The Politics of Legalism: Rules versus Standards in Nursing-Home Regulation." Social and Legal Studies 4:307-41. Braithwaite, J., and D. Chappell. 1994. "The Job Compact and Crime." Current Issues in Criminal Justice 5:295-300. Braithwaite, John, and K. Daly. 1994. "Masculinities, Violence and Communitarian Control." In Just Boys Doing Business, edited by T. Newburn and E. Stanko. London and New York: Routledge. Braithwaite, John, and Brent Fisse. 1985. "Varieties of Responsibility and Organizational Crime." Law and Policy 7(3):315-43. Braithwaite, ]., and T. Makkai. 1993. "Can Resident-Centred Inspection of Nursing Homes Work with Very Sick Residents?" Health Policy 24:19-33. - - - . 1994. "Trust and Compliance." Policing and Society 4:1-12. Braithwaite,]., T. Makkai, V. Braithwaite, and D. Gibson. 1993. Raising the Standard: Resident Centred Nursing Home Regulation in Australia. Canberra: Australian Government Publishing Service. Braithwaite, John, and S. Mugford. 1994. "Conditions of Successful Reintegration Ceremonies: Dealing with Juvenile Offenders." British Journal of Criminology 34: 139-71. Braithwaite, John, and Christine Parker. 1999. "Restorative Justice Is Republican Justice." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. Braithwaite, John, and Philip Pettit. 1990. Not Just Deserts: A Republican Theory of Criminal Justice. Oxford: Oxford University Press. Braithwaite, V., J. Braithwaite, D. Gibson, and T. Makkai. 1994. "Regulatory Styles, Motivational Postures and Nursing Home Compliance." Law and Policy 16:363-94. Brehm, Sharon S., and Jack W. Brehm. 1981. Psychological Reactance: A Theory of Freedom and Control. New York: Academic Press. Brewer, D. D., J. D. Hawkins, R. F. Catalano, and H. J. Neckerman. 1995. "Preventing Serious, Violent, and Chronic Juvenile Offending: A Review of Evaluations of Selected Strategies in Childhood, Adolescence, and the Community." In A Sourcebook: Serious, Violent, and Chronic Juvenile Offenders, edited by J. C. Howell, B. Krisberg, J. D. Hawkins, and J. J. Wilson. Thousand Oaks, Calif., and London: Sage. Bridgeforth, Carol A. 1990. "Predicting Domestic Violence from Post-arrest Suspect Interviews." Master's dissertation, University of Maryland, Institute of Criminal Justice and Criminology. Bridgeman, c., and L. Hobbs. 1997. Preventing Repeat Victimisation: The Police Officers' Guide. London: Police Research Group. Brown, Jennifer Gerada. 1994. "The Use of Mediation to Resolve Criminal Cases: A Procedural Critique." Emory Law Journal 43:1247-1309. HeinOnline -- 25 Crime & Just. 110 1999 Restorative Justice III Brown, M., and K. Polk. 1996. "Taking Fear of Crime Seriously: The Tasmanian Approach to Community Crime Prevention." Crime and Delinquency 42:398-420. Burford, G., and]. Pennell. 1998. "Family Group Decision Making Project: Outcome Report, Volume I." St. John's: Newfoundland Memorial University. Burley, A. 1992. "Toward an Age of Liberal Nations." Harvard International Law Journal 33:393-405. Burnside, Jonathan, and Nicola Baker, eds. 1994. Relational Justice: Repairing the Breach. Winchester: Waterside. Campbell, Bea. 1993. Goliath: Britain's Dangerous Places. London: Methuen. Campbell, Tom. 1988. Justice. London: Macmillan. Carnevale, Peter]., and Dean G. Pruitt. 1992. "Negotiation and Mediation." Annual Review of Psychology 43:531-82. Chamlin, M. B., and]. K. Cochran. 1997. "Social Altruism and Crime." Criminology 35 :203-28. Chan, Wai Yin. 1996. "Family Conferences in the Juvenile Justice Process: Survey on the Impact of Family Conferencing on Juvenile Offenders and Their Families." Subordinate Courts Statistics and Planning Unit Research Bulletin, Singapore, February. Christie, Nils. 1977. "Conflicts as Property." British Journal of Criminology 17:1-26. City of Wagga Wagga. 1993. "Wagga Wagga's Communitarian Response to the Juvenile Justice Advisory Council's Green Paper 'Future Directions for Juvenile Justice in New South Wales.''' Wagga Wagga: City of Wagga Wagga. Clairmont, Donald. 1994. "Alternative Justice Issues for Aboriginal Justice." Paper prepared for the Aboriginal Justice Directorate. Ottawa: Department of Justice. Clinard, Marshall, and Peter C. Yeager. 1980. Corporate Crime. New York: Free Press. Coats, R., and]. Gehm. 1985. Victim Meets Offender: An Evaluation of Victim Offender Reconciliation Programs. Valparaiso, Ind.: PACT Institute of Justice. - - - . 1989. "An Empirical Assessment." In Mediation and Criminal Justice, edited by M. Wright and B. Galaway. London: Sage. Cobb, Sara. 1997. "The Domestication of Violence in Mediation." Law and Society Review 31:397-440. Coffee,]. c., Jr. 1981. "No Soul to Damn, No Body to Kick: An Unscandalized Essay on the Problem of Corporate Punishment." Michigan Law Review 79:413-24. Cohen, A. K. 1955. Delinquent Boys: The Culture of the Gang. Glencoe, Ill.: Free Press. Collinson, ]. L. 1978. "Safety: Pleas and Prophylactics." Mining Engineer (July), pp. 73-83. Confucius. 1974. The Philosophy of Confucius, trans. James Legge. New York: Crescent Books. Crawford, Adam. 1996. Victim/Offender Mediation and Reparation in Compara- HeinOnline -- 25 Crime & Just. 111 1999 112 John Braithwaite tive European Cultures: France, England and Wales. Paper presented at the Australian and New Zealand Society of Criminology Conference, Wellington, January/February. - - - . 1997. The Local Goven/ance of Crime: Appeals to Community and Partrzerships. Oxford: Clarendon. Cullen, Francis T. 1994. "Social Support as an Organizing Concept for Criminology: Presidential Address to the Academy of Criminal Justice Sciences." Justice QU01-terly 11 :527-59. Cunneen, C. 1997. "Community Conferencing and the Fiction of Indigenous Control." Australian and New Zealand J01l171al of Criminology 30:292-312. Daly, K. 1996. "Diversionary Conferences in Australia: A Reply to the Optimists and Skeptics." Paper prepared for presentation at the annual meeting of the American Society of Criminology, Chicago, November. Daly, K., and R. Immarigeon. 1998. "The Past, Present, and Future of Restorative Justice: Some Critical Reflections." Contemporary Justice Review 1(1): 21-46. Davis, G. 1992. Making Amends: Mediatio1l and Reparation in Criminal Justice. London: Routledge. Davis, Robert T., and R. W. Stahl. 1967. "Safety Organization and Activities of Award-Winning Companies in the Coal Mining Industry." Bureau of Mines Information circular 8224. Washington, D.C.: Bureau of Mines. De Haan, W. 1990. The Politics of Redress: Crime, Punishment and Penal Abolition. London: Unwin Hyman. DeMichiei, John M., John F. Langton, Kenneth A. Bullock, and Terrance C. Wiles. 1982. Factors Associated with Disabling Injuries in Underground Coal Mines. Washington, D.C.: Mine Safety and Health Administration. Dignan, J. 1990. An Evaluation of an Experimental Adult Reparation Scheme in Kette1-ing, N01-thamptonshire. Sheffield: University of Sheffield, Centre for Criminological and Legal Research. - - - . 1992. "Repairing the Damage: Can Reparation Work in the Service of Diversion?" British Joun/al of Criminology 32 :45 3-72. Dingwall, Robert, J. Eekelaar, and T. Murray. 1983. The Protection of Children: State Intervention and Family Life. Oxford: Blackwell. Dinnen, Sinclair. 1996. "Challenges of Order in a Weak State." Ph.D. dissertation, Australian National University. Doob, A., and J. Roberts. 1983. Sentencing: An Analysis of the Public's View of Sentencing. A Rep01-t in the Department of Justice, Canada. Ottawa: Department of Justice. - - - . 1988. "Public Attitudes towards Sentencing in Canada." In Public Attitudes to Sentencing, edited by N. Walker and M. Hough. Aldershot: Gower. Doyle, M. 1983. "Kant, Liberal Legacies and Foreign Affairs," pts. 1 and 2. Philosophy and Public Affairs 12(3):205-35; 12(4):323-53. - - - . 1986. "Liberalism and World Politics." American Political Science Review 80(4): 1151-69. Dunford, F. 1990. "System-Initiated Warrants for Suspects of Misdemeanour Domestic Assault: A Pilot Study." Justice Quarterly 7:631-53. Durkheim, E. 1961. Moral Education: A Study in the Theory and Application of HeinOnline -- 25 Crime & Just. 112 1999 Restorative Justice 113 the Sociology of Education, trans. E. K. Wilson and H. Schnurer. New York: Free Press. . Eckel, Malcolm David. 1997. "A Buddhist Approach to Repentance." In Repentance: A Comparative Perspective, edited by Amitai Etzioni and David E. Carney. New York: Rowman & Littlefield. Eglash, Albert. 1975. "Beyond Restitution: Creative Restitution." In Restitution in Criminal Justice, edited by J. Hudson and B. Galaway. Lexington, Mass.: Lexington Books. Elliott, D. S., and H. L. Voss. 1974. Delinquency and Dropout. Lexington, Mass.: Lexington Books. Emler, N., and S. Reicher. 1995. Adolescence and Delinquency: The Collective Management of Reputation. Oxford: Blackwell. Ervin, L., and A. Schneider. 1990. "Explaining the Effects of Restitution on Offenders: Results from a National Experiment in Juvenile Courts." In Criminal Justice, Restitution and Reconciliation, edited by B. Galaway and J. Hudson. New York: Willow Tree. Esbensen, Finn-Aage, and David Huizinga. 1993. "Gangs, Drugs and Delinquency in a Survey of Urban Youth." O'iminology 31:565-87. Farrington, David P. 1993. "Understanding and Preventing Bullying." In Crime and Justice: A Review of Research, vol. 17, edited by M. Tonry. Chicago: University of Chicago Press. Findlay, Mark. 1998. "Decolonising Restoration and Justice." Current Issues in Criminal Justice 10:85-88. Fisse, B. 1983. "Reconstructing Corporate Criminal Law: Deterrence, Retribution, Fault, and Sanctions." Southern California Law Review 56:11411246. Fisse, B., and J. Braithwaite. 1993. Corporations, Crime and Accountability. Cambridge: Cambridge University Press. Fitzpatrick, Peter. 1992. "The Impossibility of Popular Justice." Social and Legal Studies 1:199-215. Forsythe, Lubica. 1995. "An Analysis ofJuvenile Apprehension Characteristics and Reapprehension Rates." In A New Approach to Juvenile Justice: An Evaluation of Family Conferencing in Wagga Wagga, edited by David Moore, with Lubica Forsythe and Terry O'Connell. A report to the Criminology Research Council. Wagga Wagga: Charles Sturt University. Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison. London: Pantheon. Fry, Don. 1997. "A Report on Diversionary Conferencing." Alice Springs: Northern Territory Police. Fuller, Lon. 1964. The Morality of Law. New Haven, Conn.: Yale University Press. Galaway, B. 1992. "The New Zealand Experience Implementing the Reparation Sentence." In Restorative Justice on Trial: Pitfalls and Potentials of VictimOffender Mediation-International Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. Galaway, Burt, and Joe Hudson, eds. 1975. Considering the Victim. Springfield, Ill.: Charles C. Thomas. HeinOnline -- 25 Crime & Just. 113 1999 114 John Braithwaite Garland, D. 1985. Punishment and Welfare. Aldershot: Gower. - - - . 1990. Punishment and Modem Society. Oxford: Clarendon. Geertz, Clifford. 1983. Local Knowledge. New York: Basic Books. Gehm, John. 1990. "Mediated Victim-Offender Restitution Agreements: An Exploratory Analysis of Factors Related to Victim Participation." In Criminal Justice, Restitution, and Reconciliation, edited by B. Galaway and J. Hudson. Monsey, N.Y.: Willow Tree. Gentry, Deborah B., and Wayne A. Benenson. 1993. "School-to-Home Transfer of Conflict Management Skills among School-Age Children." Families in Society 74(February):67-73. Gernet, Jacques. 1982. A History of Chinese Civilization. Cambridge: Cambridge University Press. Goffman, E. 1956. "Embarrassment and Social Organization." American Journal of Sociology 62:264-71. Goldgeir, James, and Michael McFaul. 1992. "A Tale of Two Worlds: Core and Periphery in the Post-Cold War Era." International Organization 46:467-92. Goodes, T. 1995. "Victims and Family Conferences: Juvenile Justice in South Australia." Adelaide: Family Conferencing Team. Gottfredson, Denise. 1997. "School-Based Crime Prevention." In Preventing Crime: What Works, What Doesn't, What's Promising: A Report to the United States Congress, edited by Lawrence Sherman, Denise Gottfredson, Doris MacKenzie, John Eck, Peter Reuter, and Shawn Bushway. Washington, D.C.: National Institute of Justice. Gottfredson, S. D., and D. M. Gottfredson. 1994. "Behavioral Prediction and the Problem of Incapacitation." Criminology 32:441-74. Gottfredson, Stephen D., and Ralph B. Taylor. 1987. "Attitudes of Correctional Policymakers and the Public." In America's Correctional Crisis: Prison Populations and Public Policy, edited by Stephen D. Gottfredson and Sean McConville. New York: Greenwood. Green, Ross Gordon. 1998. Justice in Aboriginal Communities. Saskatoon: Purich. Griffiths, Curt Taylor, and Ron Hamilton. 1996. "Sanctioning and Healing: Restorative Justice in Canadian Aboriginal Communities." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Gunningham, Neil. 1995. "Environment, Self-Regulation and the Chemical Industry: Assessing Responsible Care." Law and Policy 17:57-109. Gunningham, Neil, and Peter Grabosky. 1998. Smart Regulation: Designing Environmental Policy. Oxford: Clarendon. Habermas, Jiirgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democmcy. London: Polity. Hagan,]. 1991. "Destiny and Drift: Subcultural Preferences, Status Attainments, and the Risks and Rewards of Youth." American Sociological Review 56:567-82. Hagan, John, and Bill McCarthy. 1997. Mean Streets: Youth Crime and Homelessness. Cambridge: Cambridge University Press. HeinOnline -- 25 Crime & Just. 114 1999 Restorative Justice 115 Haines, Fiona. 1997. Corporate Regulation: Beyond "Punish or Persuade. " Oxford: Clarendon. Haines, Kevin. 1999. "Some Principled Objections to a Restorative Justice Approach to Working with Juvenile Offenders." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. Haley, John. 1996. "Crime Prevention through Restorative Justice: Lessons from Japan." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Haley, John, assisted by A. M. Neugebauer. 1992. "Victim-Offender Mediations: Japanese and American Comparisons." In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-International Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. Harris, M. K. 1991. "Moving into the New Millennium: Toward a Feminist Vision of Justice." In Criminology as Peacemaking, edited by H. E. Pepinsky and R. Quinney. Bloomington: Indiana University Press. Harris, Nathan, and Jamie Burton. 1997. The Reliability of Observed Reintegrative Shaming, Shame, Defiance and Other Key Concepts in Diversionary Conferences. RISE Working papers. Canberra: Australian National University.· Hassall, Ian. 1996. "Origin and Development of Family Group Conferences." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Hawkins, Keith. 1984. Environment and Enforcement: Regulation and the Social Definition of Pollution. Oxford: Clarendon. Hay, D. 1975. "Property, Authority and the Criminal Law." In Albion's Fatal Tree, edited by D. Hay, P. Linebaugh,]. G. Rule, E. P. Thompson, and C. Winslow. London: Allen Lane. Heidensohn, Frances. 1986. "Models of Justice: Portia or Persephone? Some Thoughts on Equality, Fairness and Gender in the Field of Criminal Justice." International Journal of the Sociology of Law 14:287-98. Heimer, Carol A., and Lisa R. Staffen. 1995. "Interdependence and Reintegrative Social Control: Labeling and Reforming 'Inappropriate' Parents in Neonatal Intensive Care Units." American Sociological Review 60:635-54. Heinz, A., and W. Kerstetter. 1981. "Pretrial Settlement Conference: Evaluation of a Reform in Plea Bargaining." In Perspective on C1·ime Victims, edited by B. Galaway and]. Hudson. St. Louis: Mosby. Hogg, Russel, and David Brown. 1985. "Reforming Juvenile Justice: Issues and Prospects." In Juvenile Delinquency in Australia, edited by A. Borowski and J. Murray. Sydney: Methuen Australia. Hough, Michael, and Julian Roberts. 1998. "Attitudes to Punishments: Findings from the British Crime Survey." Home Office Research Study no. 179. London: H. M. Stationery Office. Hoyle, Carolyn, and Richard Young. 1998. "A Survey of Restorative Cautioning with the Thames Valley Police." Oxford: University of Oxford, Centre for Criminological Research. HeinOnline -- 25 Crime & Just. 115 1999 116 John Braithwaite Hsien, Lim Li. 1996. "Family Conferencing Good for Young Delinquents: Report." Straits Times (March 6). Huang, Ray. 1988. China: A Macro Histo1y. Armonk, N.Y.: M. E. Sharpe. Hudson, Joe. 1998. "Conducting the Family Group Conference Process: An Overview." Paper presented at "Conferencing: A New Response to Wrongdoing," Minneapolis, August. Hughes, Gordon. 1996. "Convergent and Divergent Discourses of Communitarian Crime Prevention: A Meditation on Current Developments in Europe, the USA, Australia and New Zealand." Paper presented to the annual conference of the Australian and New Zealand Society of Criminology, January/February, Wellington. Iivari, J. 1987. "Mediation as a Conflict Resolution: Some Topic Issues in Mediation Project in Vantaa." Paper presented to the International Seminar on Mediation, Finland, September. Cited in Crime and Accountability: Victim Offender Mediation in Practice, edited by T. Marshall and S. Merry. 1990. London: Home Office. - - - . 1992. "The Process of Mediation in Finland: A Special Reference to the Question 'How to Get Cases for Mediation.' " In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-Imernational Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. Jackson, Moana. 1987. "The Maori and the Criminal Justice System: A New Perspective-He Whaipaanga Hou." Report for New Zealand Department of Justice. Wellington: Department of Justice, Policy and Research DivisIOn. James, T. M. 1993. "Circle Sentencing." Yellowknife: Supreme Court of the Northwest Territories. Jenkins, Ann L. 1997. "The Role of Managerial Self-Efficacy in Corporate Compliance with Regulatory Standards." Ph.D. dissertation, Australian National University. Karstedt-Henke, Susanne, and Bernhard Crasmoller. 1991. "Risks of Being Detected, Chances of Getting Away." In The Future of the Juvenile Justice System, edited by J. Junger-Tas, L. Boendermaker, and P. van der Laan. Leuven: Acco. Katz, I., D. Glass, and S. Cohen. 1973. "Ambivalence, Guilt and the Scapegoating of Minority Group Victims." Journal of Experimental Social Psychology 9:423-36. Kelman, Steven. 1984. "Enforcement of Occupational Safety and Health Regulations: A Comparison of Swedish and American Practices." In Enforcing Regulation, edited by Keith Hawkins and John M. Thomas. Boston: KluwerNijhoff. Kerner, H., E. Marks, and J. Schreckling. 1992. "Implementation and Acceptance of Victim-Offender Mediation Programs in the Federal Republic of Germany: A Survey of Criminal Justice Institutions." In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-International Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. HeinOnline -- 25 Crime & Just. 116 1999 Restorative Justice 117 Kigin, R., and S. Novack. 1980. "A Rural Restirution Program for Juvenile Offenders and Victims." In Victims, Offenders and Alternative Sanction, edited by J. Hudson and B. Galaway. Lexington, Mass.: Lexington Books. Kinsey, Karyl A. 1986. "Theories and Models of Tax Cheating." CriminalJustice Abstracts (September), pp. 402-25. Knight, T. 1985. "Schools and Delinquency." In Juvenile Delinquency in Australia, edited by A. Borowski and J. M. Murray. Melbourne: Methuen. Koh, Angeline Cheok Eng. 1997. "The Delinquent Peer Group: Social IdenPerspectives." Ph.D. dissertation, Australian tity and ｓ･ｬｦｾｃ｡ｴｧｯｲｩｺｮ＠ National University. Krawll, M. B. 1994. Understanding the Role of Healing in Aboriginal Communities. Ottawa: Solicitor General Canada, Ministry Secretariat. Kuhn, A. 1987. "Koperverletzung als Konflift, Zwischenbericht 1987 zum Project Handschlag." Unpublished paper cited in T. Trenczek, "A Review and Assessment of Victim-Offender Reconciliation Programming in West Germany." In Criminal Justice, Restitution and Reconciliation, edited by B. Galaway and]. Hudson. Monsey, N.Y.: Willow. Lacey, N. 1988. State Punishment: Political Principles and Community Values. London: Routledge. Lajeunesse, T. 1993. Community Holistic Circle Healing: Hollow Water First Nation, Aboriginal Peoples Collection. Ottawa: Solicitor General, Canada, Ministry Secretariat. Lam, J. A. 1989. The Impact of Conflict Resolution Programs on Schools: A Review and Synthesis of the Evidence. Amherst, Mass.: National Association for Mediation in Education. Langbein, L., and C. M. Kerwin. 1985. "Implementation, Negotiation and Compliance in Environmental and Safety Regulation." Journal of Politics 47:854-80. La Prairie, C. 1994. Seen but Not Hem·d: Native People in the Inner City. Report 3, Victimisation and Domestic Violence. Ottawa: Ottawa Department of}ustice. ---.1995. "Altering Course: New Directions in Criminal Justice and Corrections: Sentencing Circles and Family Group Conferences." Australian and New Zealand Journal of Criminology ("Special Issue: Crime, Criminology, and Public Policy," edited by David Dixon) 28(December):78-99. Laue, James H. 1991. "Contributions of the Emerging Field of Conflict Resolution." In Approaches to Peace: An Intellectual Map, edited by W. S. Thompson, K. M. Jensen, Richard N. Smith, and Kimber M. Schraub. Washington, D.C.: U.s. Instirute of Peace. Lewis, David. 1986. "Causation" and "Postscript: Redundant Causation." In Philosophical Papers, vol. 2. Oxford: Oxford University Press. Lewis, Helen B. 1971. Shame and Guilt in Neurosis. New York: International Universities Press. Lewis-Beck, Michael S., and John R. Alford. 1980. "Can Government Regulate Safety: The Coal Mine Example." American Political Science Review 74:745-56. Lind, E. Allan, and Tom R. Tyler. 1988. The Social Psychology of Procedural Justice. New York: Plenum. HeinOnline -- 25 Crime & Just. 117 1999 118 John Braithwaite Longclaws, Lyle, Burt Galaway, and Lawrence Barkwell. 1996. "Piloting Family Group Conferences for Young Aboriginal Offenders in Winnipeg, Canada." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Lu, Hong 1998. "Community Policing-Rhetoric or Reality? The Contemporary Chinese Community-Based Policing System in Shanghai." Ph.D. dissertation, Arizona State University. MacKenzie, D. L. 1997. "Criminal Justice and Crime Prevention." In Preventing Crime: What Works, What Doesn't, What's Promising: A Report to the United States Congress, edited by Lawrence Sherman, Denise Gottfredson, Doris MacKenzie, John Eck, Peter Reuter, and Shawn Bushway. Washington, D.C.: National Institute of Justice. Makkai, T., and J. Braithwaite. 1991. "Criminological Theories and Regulatory Compliance." Criminology 29:191-220. - - - . 1993a. "Praise, Pride and Corporate Compliance." International Journal of the Sociology of Law 21 :73-91. - - - . 1993b. "The Limits of the Economic Analysis of Regulation." Law and Policy 15:271-91. - - - . 1994. "Reintegrative Shaming and Regulatory Compliance." Criminology 32:361-85. - - - . 1996. "Procedural Justice and Regulatory Compliance." Law and Human Behavior 20:83-98. Maoz, Z., and N. Abdolali. 1989. "Regime Types and International Conflict, 1816-1976." Journal of Conflict Resolution 33:3-36. Marshall, T. F. 1985. Alternatives to Criminal Courts. Aldershot: Gower. - - - . 1992. "Restorative Justice on Trial in Britain." In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-International Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. Marshall, T. F., and S. Merry. 1990. Crime and Accountability: Victim Offendel· Mediation in Practice. London: Home Office. Masters, Guy. 1995. "The Family Model of Social Control in Japanese Secondary Schools." Unpublished manuscript. Lancaster: Lancaster University, Department of Applied Social Science. - - - . 1997. Reintegrative Shaming in Theory and Practice. Ph.D. dissertation, Lancaster University. Masters, Guy, and David Smith. 1998. "Portia and Persephone Revisited: Thinking about Feeling in Criminal Justice." Theoretical Criminology 2:528. Matza, D. 1964. Delinquency and Drift. New York: Wiley. Maxwell, Gabrielle M. 1993. "Arrangements for Children after Separation? Problems and Possibilities." In Women's Law Conference Papers: 1993 New Zealand Suffrage Centennial. Wellington: Victoria University of Wellington. Maxwell, Gabrielle M., and Allison Morris. 1993. Family, Victims and Culture: Youth Justice in New Zealand. Wellington: Victoria University of Wellington, New Zealand Social Policy Agency and Institute of Criminology. HeinOnline -- 25 Crime & Just. 118 1999 Restorative Justice 119 1996. "Research on Family Group Conferences with Young Offenders in New Zealand." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. McCold, Paul. 1997. "Restorative Justice Variations on a Theme." Paper presented at the international conference "Restorative Justice for JuvenilesPotentialities, Risks and Problems for Research," Leuven, Belgium, May 12-14. McCold, Paul, and Benjamin Wachtel. 1998. "Restorative Policing Experiment: The Bethlehem Pennsylvania Police Family Group Conferencing Project." Report. Pipersville, Pa.: Community Service Foundation. McDonald, J., D. Moore, T. O'Connell, and M. Thorsborbe. 1995. Real Justice Training Manual: Coordinating Family Group Conferences. Pipersville, Pa.: Pipers Press. Meima, M. 1990. "Sexual Violence, Criminal Law and Abolitionism." In Gender, Sexuality and Social Control, edited by B. Rolston and M. Tomlinson. Bristol: European Group for the Study of Deviance and Social Control. Melton, Ada Pecos. 1995. "Indigenous Justice Systems and Tribal Society." Judicature 79:126. Michael, Franz. 1986. China through the Ages: History of a Civilization. Taipei: SMC. Mine Enforcement and Safety Administration. 1977. A Report on Civil Penalty Effectiveness. Washington, D.C.: Mine Enforcement and Safety Administration. Ministry of Justice, Western Australia. 1994. "Juvenile Justice Teams: A Six Month Evaluation." Perth: Ministry of Justice. Minor, Kevin 1., and]. T. Morrison. 1996. "A Theoretical Study and Critique of Restorative Justice." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Minow, Martha. 1990. Making All the Difference: Inclusion, Exclusion and American Law. Ithaca, N.Y.: Cornell University Press. Miyazawa, Setsuo. 1992. Policing in Japan: A Study on Making Crime. Albany, N.Y.: SUNY Press. Moore, David B. 1994. "Public Anger and Personal Justice: From Retribution to Restoration-and Beyond." Paper to the Silvan S. Tomkins Institute Colloquium "The Experience and Expression of Anger," October, Philadelphia. Moore, David B., with L. Forsythe. 1995. A New Approach to Juvenile Justice: An Evaluation of Family Conferencing in Wagga Wagga. Wagga Wagga: Charles Sturt University. Moore, David B., and Terry O'Connell. 1994. "Family Conferencing in Wagga Wagga: A Communitarian Model ofJustice." In Family Conferencing and Juvenile Justice, edited by Christine Alder and Joy Wundersitz. Canberra: Australian Studies in Law, Crime, and Justice, Australian Institute of Criminology. Morris, Allison, Gabrielle Maxwell, Joe Hudson, and Burt Galaway. 1996. "Concluding Thoughts." In Family Group Conferences: Perspectives on Policy HeinOnline -- 25 Crime & Just. 119 1999 120 John Braithwaite and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Morris, Ruth. 1995. "Not Enough!" Mediation Quarterly 12(3):285-9l. Mugford, S., and N. I. Inkpen. 1995. The Implementation of Shaming Confn'ences as a New Policy Strategy: The Case of Drink Drivel·s. Paper presented at the forty-seventh American Society of Criminology Conference, Boston, November. Murphy, Jeffrie, G., and Jean Hampton. 1989. Forgiveness and Mercy. New York: Cambridge. Nagin, Daniel. 1978. "Crime Rates, Sanction Levels and Constraints on Prison Population." Law and Society Review 12:341-66. Nathanson, Donald. L. 1992. Shame and Pride: Affect, Sex and the Birth of the Self. New York: Norton - - - . 1998. "From Empathy to Community." Paper presented at "Conferencing: A New Response to Wrongdoing," August, Minneapolis. Available on the Real Justice website: www.realjustice.org. Newman, Graeme. 1983. Just and Painful. London: Macmillan. Nicholl, Caroline G. 1998. Implmunting Restorative Justice. Washington, D.C.: U.s. Department of Justice, Office of Community Oriented Policing ServIces. Novack, S., B. Galaway, and J. Hudson. 1980. "Victim and Offender Perceptions of the Fairness of Restitution and Community-Service Sanctions." In Victims, Offenders and Alte17lative Sanctions, edited by]. Hudson and B. Galaway. Lexington, Mass.: Lexington Books. Nugent, W. R., andJ. B. Paddock. 1995. "The Effect of Victim-Offender Mediation on Severity of Reoffense." Mediation Quartn'ly 12:353-67. Obonsawin-Irwin Consulting, Inc. 1992a. "An Evaluation of the Sandy Lake First Nation Justice Project." Ontario: Ministry of the Attorney General. - - - . 1992b. "An Evaluation of the Attawapiskat First Nation Justice Project." Ontario: Ministry of the Attorney General. Offer, A. 1994. "Going to War in 1914: A Matter of Honour?" Paper presented at the Australian War Memorial Conference. Olweus, Dan. 1993. "Annotation: Bullying at School: Basic Facts and Effects of a School Based Intervention Program." Journal of Child Psychology and Psychiatry 35: 1171-90. Palk, Gerard, Hennessey Hayes, and Timothy Prenzler. 1998. "Restorative Justice and Community Conferencing: Summary of Findings from a Pilot Study." Current Issues in Criminal Justice 10:138-55. Parker, Christine. 1998. "Public Rights in Private Government: Corporate Compliance with Sexual Harassment Legislation." Austmlian Journal of Human Rights 5(1): 159-93. - - - . 1999a. "Compliance Professionalism and Regulatory Community: The Australian Trade Practices Regime." Journal of Law and Society 26:21539. - - - . 1999b. Just Lawyers. Oxford: Oxford University Press. Pate, K. 1990. "Victim-Offender Restitution Programs in Canada." In Crimi- HeinOnline -- 25 Crime & Just. 120 1999 Restorative Justice 121 nal Justice, Restitution and Reconciliation, edited by B. Galaway and J. Hudson. New York: Willow Tree. Paternoster, Raymond, Robert Brame, Ronet Bachman, and Lawrence W. Sherman. 1997. "Do Fair Procedures Matter? The Effect of Procedural J ustice on Spouse Assault." Law and Society Review 31: 163-204. Pavlich, George. 1996. "The Power of Community Mediation: Government and Formation of Self-Identity." Law and Society Review 30:707-33. Peachey, D. E. 1989. "The Kitchener Experiment." In Mediation and Criminal Justice: Victims, Offenders and Community, edited by M. Wright and B. Galaway. London: Sage. Pease, Ken. 1998. "Repeat Victimisation: Taking Stock." Crime Detection and Prevention Series Paper no. 90. London: Police Research Group. Pennell, Joan, and Gale Burford. 1995. Family Group Decision Making: New Roles for "Old" Partuers in Resolving Family Violence. Implementation Report, vol!. St. John's: University of Newfoundland, School of Social Work, Family Group Decision Making Project. - - - . 1996. "Attending to Context: Family Group Decision Making in Canada." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. ---.1997. "Family Group Decision Making: After the Conference-Progress in Resolving Violence and Promoting Well-Being." St. John's: University of Newfoundland, School of Social Work, Family Group Decision Making Project. Pepinsky, H. E., and R. Quinney, eds. 1991. Criminology as Peacemaking. Bloomington: Indiana University Press. Pepler, Debra]., Wendy Craig, Suzanne Ziegler, and Alice Charach. 1993. "A School-Based Antibullying Intervention." In Understanding and Managing Bullying, edited by Delwin Tattum. London: Heinemann. Perry, Charles S. 1981a. "Safety Laws and Spending Save Lives: An Analysis of Coal Mine Fatality Rates, 1930-1979." Unpublished manuscript. Lexington: University of Kentucky, Department of Sociology. - - - . 1981b. "Dying to Dig Coal: Fatalities in Deep and Surface Coal Mining in Appalachian States, 1930-1978." Unpublished manuscript. Lexington: University of Kentucky, Department of Sociology. Pettit, P. 1993. "Liberalism and Republicanism." Australian Journal of Political Science 28:162-89. - - - . 1997. Republicanism. Oxford: Clarendon. Phillips, A. 1991. Engendering Democracy. Cambridge: Polity. Pitts, John, and Philip Smith. 1995. "Preventing School Bullying." Police Research Group Crime Detection and Prevention Series Paper no. 63. London: Home Office. Polanyi, Michael. 1951. The Logic ofLiberty. Chicago: University of Chicago Press. Polk, Ken. 1994. "Family Conferencing: Theoretical and Evaluative Questions." In Family Conferencing and Juvenile Justice, edited by Christine Alder and Joy Wundersitz. Australian Studies in Law, Crime, and Justice. Canberra: Australian Institute of Criminology. HeinOnline -- 25 Crime & Just. 121 1999 122 John Braithwaite Pontell, Henry. 1978. "Deterrence: Theory versus Practice." Criminology 16:3-22. Posner, Richard A. 1977. Economic Analysis of Law. 2d ed. Boston and Toronto: Little, Brown. Pranis, K. 1996. "A State Initiative toward Restorative Justice: The Minnesota Experience." In Restorative Justice: International Perspectives, edited by B. Galaway and]. Hudson. Monsey, N.Y.: CriminalJustice Press. Pratt, John. 1996. "Colonization, Power and Silence: A History of Indigenous Justice in New Zealand Society." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: CriminalJustice Press. Reed, P. 1986. "Developmental Resources and Depression in the Elderly." Nursing Research 36:368-74. - - - . 1987. "Spirituality and Well-Being in Terminally III Hospitalized Adults." Research in Nursing and Health 10:335-44. - - - . 1992. "An Emerging Paradigm for the Study of Spirituality in Nursing." Research in Nursing and Health 15:349-57. Rees,Joseph V. 1988. Refornting the Workplace: A Study of Self-Regulation in Occupational Safety. Philadelphia: University of Pennsylvania Press. - - - . 1994. Hostages of Each Other: The Transfo17nation of Nuclear Safety since Three Mile Island. Chicago: University of Chicago Press. Rensselaer, W. L., III. 1992. "Colombia's Cocaine Syndicates." In War on D17lgs: Studies in the Failure of u.s. Narcotics Policy, edited by A. McCoy and A. Block. Boulder, Colo.: Westview. Retzinger, Suzanne. 1991. Violent Emotions. Newbury Park, Calif.: Sage. Retzinger, Suzanne, and Thomas]. Scheff. 1996. "Strategy for Community Conferences: Emotions and Social Bonds." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Rigby, Ken. 1996. Bullying in Schools and What to Do about It. Melbourne: Australian Council for Educational Research. Rigby, Ken, Alison Whish, and Garry Black. 1994. "School Children's Peer Relations and Wife Abuse." Criminology Australia (August), pp. 8-12. Roach, Kent. 1999. Due Process and Victims' Rights: The New Law and Politics of Criminal Justice. Toronto: University of Toronto Press. Roberts, Julian V., and Loretta]. Stalans. 1997. Public Opinion, Crime, and Criminal Justice. Boulder, Colo.: Westview. Robertson, Jeremy. 1996. "Research on Family Group Conferences in Child Welfare in New Zealand." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Ross, Rupert. 1996. Returning to the Teachings: Exploring Aboriginal Justice. London: Penguin. Rumsey, Alan. 1999. "Women as Peacemakers in the New Guinea Highlands: A Case from the Nebilyer Valley, Western Highlands Province?" In Reflections on Violence in Melanesia, edited by Sinclair Dinnen and Allison Ley. Sydney: Federation Press. HeinOnline -- 25 Crime & Just. 122 1999 Restorative Justice 123 Sandor, Danny. 1993. "Juvenile Justice: The Thickening Blue Wedge." Alternative Law Jou17lal 18: 104-8. Scheff, Thomas J. 1990. Microsociology: Discourse, Emotion, and Social Structure. Chicago: University of Chicago Press. - - - . 1994. Bloody Revenge: Emotions, Nationalism and War. Boulder, Colo.: Westview. Scheff, Thomas]., and Suzanne M. Retzinger. 1991. Emotions and Violence: Shame and Rage in Destructive Conflicts. Lexington, Mass.: Lexington Books. Scheingold, Stuart A., Jana Pershing, and Toska Olson. 1994. "Sexual Violence, Victim Advocacy, and Republican Criminology: Washington State's Community Protection Act." Law and Society Review 28(4):501-33. Schiff, Mara F. 1998. "The Impact of Restorative Interventions on Juvenile Offenders." In Restoring Juvenile Justice, edited by Lode Walgrave and Gordon Bazemore. Monsey, N.Y.: Criminal Justice Press. Schneider, A. 1986. "Restitution and Recidivism Rates of Juvenile Offenders: Results from Four Experimental Studies." Criminology 24:533-52. - - - . 1990. Deterrence and Juvenile Crime: Results from a National Policy Experiment. New York: Springer-Verlag. Scholz, John T., and Wayne B. Gray. 1990. "OSHA Enforcement and Workplace Injuries: A Behavioral Approach to Risk Assessment." Journal of Risk and Uncertainty 3:283-305. Sessar, Klaus. 1990. "Tertiary Victimisation: A Case of the Politically Abused Crime Victims." In C17minal Justice, Restitution and Reconciliation, edited by B. Galaway and J. Hudson. Monsey, N.Y.: Willow Tree. - - - . 1998. "Punitive Artitudes of the Public: Reality and Myth." In Restoring Juvenile Justice, edited by Lode Walgrave and Gordon Bazemore. Amsterdam: Kugler. Sessar, K., A. Beuerskens, and K. Boers. 1986. "Wiedergutmachung als Konfliktregelungsparadigma?" Kriminologisches Jou17lal 18:86-105. Shapland, J., J. Willmore, and P. Duff. 1985. Victims in the Criminal Justice System. Cambridge Studies in Criminology. Brookfield, Vt.: Gower. Shearing, Clifford. 1995. "Reinventing Policing: Policing as Governance." In Privatisienmg staatlicher Kontrolle: Befunde, Konzepte, Tendenzen. Interdisziplinare Studien zu Recht und Staat 3:69-88. ---.1997. "Violence and the Changing Face of Governance: Privatization and Its Implications." Cape Town: Community Peace Foundation. Sherman, L. W. 1992. Policing Domestic Violence. New York: Free Press. - - - . 1993. "Defiance, Deterrence and Irrelevance: A Theory of the Criminal Sanction." Journal of Research in Crime and Delinquency 30:445-73. Sherman, L. W., and G. Barnes. 1997. "Restorative Justice and Offenders' Respect for the Law." RISE Working Paper no. 3. Canberra: Australian National University, Research School of Social Sciences, Law Program. Sherman, Lawrence, Denise Gotrfredson, Doris MacKenzie, John Eck, Peter Reuter, and Shawn Bushway. 1997. Preventing Crime: What Works, What Doesn't, What's Promising: A Report to the United States Congress. Washington, D.C.: National Institute of Justice. Sherman, L. W., and H. Strang: 1997 a. "The Right Kind of Shame for Crime HeinOnline -- 25 Crime & Just. 123 1999 124 John Braithwaite Prevention." RISE Working Paper no. 1. Canberra: Australian National University, Research School of Social Sciences, Law Program. - - - . 1997b. "Restorative Justice and Deterring Crime." RISE Working Paper no. 4. Canberra: Australian National University, Research School of Social Sciences, Law Program. - - - . 1997c. "Community Policing and Restorative Justice." Press Release, Australian National University, April 21. Sherman, L. W., H. Strang, G. C. Barnes, J. Braithwaite, N. Inkpen, and M. M. Teh. 1998. "Experiments in Restorative Policing: A Progress Report." Canberra: Australian National University, Research School of Social Sciences, Law Program. Shover, Neal. 1996. Great Pretenders: Pursuits and COl"een of Persistent Thieves. Boulder, Colo.: Westview. Skinner, Q. 1984. "The Idea of Negative Lihrty: Philosophical and Historical Perspectives." In Philosophy in History: Essays on the Historiography of Philosophy, edited by R. Rorty, J. Schneewind, and Q. Skinner. Cambridge: Cambridge University Press. Smith, D., H. Blagg, and N. Derricourt. 1985. "Victim-Offender Mediation Project." Report to the Chief Officers' Group, South Yorkshire Probation Service. Cited in Crime and Accountability: Victim-Offender Mediation in Practice, edited by T. Marshall and S. Merry. 1990. London: Home Office. Sonnekus, Eon, and Cheryl Frank. 1997. "Reconstructing and Developing Juvenile Justice in the New South Africa: Towards Restorative Juvenile Justice." Paper presented to the conference on Restorative Justice for Juveniles, Leuven, Belgium, April. Stewart, Trish. 1993. "The Youth Justice Co-ordinator's Role-a Personal Perspective of the New Legislation in Action." In The Youth Court in New Zealand: A New Model ofJustice, edited by B. J. Brown and F. W. M. McElrea. Auckland: Legal Research Foundation. Strang, H., and L. W. Sherman. 1997. "The Victim's Perspective." Working Paper no. 2. Canberra: Australian National University, Research School of Social Sciences, Law Program. Stubbs, J. 1995. "'Communitarian' Conferencing and Violence against Women: A Cautionary Note." In Wife Assault and the Canadian Criminal Justice System, edited by Mariana Valverde, Linda MacLeod, and Kirsten Johnson. Toronto: University of Toronto, Centre of Criminology. Sutherland, E. H. 1983. White Collar Crime: The Uncut Version. New Haven, Conn.: Yale University Press. Sutherland, E. H., and D. R. Cressey. 1978. Criminology. lOth ed. New York: Lippincott. Sykes, G., and D. Matza. 1957. "Techniques of Netralization: A Theory of Delinquency." American Sociological Review 22:664-70. Tangney, June Price. 1995. "Recent Advances in the Empirical Study of Shame and Guilt." American Behavioral Scientist. 38:1132-45. Tauri, Juan. 1998. "Family Group Conferencing: A Case Study of the Indigenisation of New Zealand's Justice System." Cun"ent Issues in Criminal Justice 10: 168-82. HeinOnline -- 25 Crime & Just. 124 1999 Restorative Justice 125 Tauri, Juan, and Allison Morris. 1997. "Re-forming Justice: The Potential of Maori Processes." Australian and New Zealand Journal of Criminology 30:149-67. Tavuchis, N. 1991. Mea Culpa: A Sociology of Apology and Reconciliation. Stanford, Calif.: Stanford University Press. Thurman, Q. C. 1984. "Deviance and the Neutralization of Moral Commitment: An Empirical Analysis." Deviant Behavior 5:291-304. Tomkins, Sylvan. 1962. Affect/Imagery/Consciousness. New York: Springer. Trenczek, T. 1990. "A Review and Assessment of Victim-Offender Reconciliation Programming in West Germany." In Criminal Justice, Restitution and Reconciliation, edited by B. Galaway and]. Hudson. Monsey, N.Y.: Willow Press. Trevelyan, G. M. 1978. English Social History: A Survey of Six Centuries from Chaucer to Queen Victoria. London: Longman. Turner, John c., with Michael A. Hogg, Penelope]. Oakes, Stephen D. Reicher, and Margaret S. Wetherell. 1987. Rediscovering the Social Group: A Self-Categorization Theory. London: Blackwell. Tyler, Tom. 1990. Why People Obey the Law. New Haven, Conn.: Yale University Press. Tyler, Tom, and Robyn M. Dawes. 1993. "Fairness in Groups: Comparing the Self-Interest and Social Identity Perspectives." In Psychological Perspectives on Justice: Theory and Applications, edited by Barbara A. Mellers and J onathan Baron. Cambridge: Cambridge University Press. Umbreit, Mark. 1985. Crime and Reconciliation: Creative Options for Victims and Offenders. Nashville: Abingdon. - - - . 1990a. "Mediation in the Nineties: Pushing Back the Boundaries." Mediation 6:27 -9. - - - . 1990b. "The Meaning of Fairness to Burglary Victims." In Criminal Justice, Restitution and Reconciliation, edited by B. Galaway and]. Hudson. Monsey, N.Y.: Willow Tree. - - - . 1992. "Mediating Victim-Offender Conflict: From Single-Site to Multi-Site Analysis in the U.S." In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-International Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. - - - . 1994. Victim Meets Offender: The Impact of Restorative Justice and Mediation. Monsey, N.Y.: Criminal Justice Press. - - - . 1999. "Restorative Justice through Juvenile Victim-Offender Mediation." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. Umbreit, M., and R. Coates. 1992. Victim-Offender Mediation: An Analysis of Programs in Four States of the U.S. Minneapolis: Citizens Council Mediation Services. Umbreit, M., with R. Coates and B. Kalanj. 1994. Victim Meets Offender: The Impact of Restorative Justice and Mediation. Monsey, N.Y.: Criminal Justice Press. HeinOnline -- 25 Crime & Just. 125 1999 126 John Braithwaite Vagg, Jon. 1998. "Delinquency and Shame: Data from Hong Kong." British Journal of Criminology 38:247-64. Van Ness, Daniel. 1986. Crime and Its Victims: What We Can Do. Downers Grove, III.: Intervarsity Press. - - - . 1993. "New Wine and Old Wineskins: Four Challenges of Restorative Justice." Criminal Law Forum 4:251-76. - - - . 1999. "Legal Principles and Process." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. Van Ness, Daniel, and Karen Heetderks Strong. 1997. Restoring Justice. Cincinnati: Anderson. Wachtel, Ted. 1997. Real Justice: How We Can Revolutionize Our Response to Wrongdoing. Pipersville, Pa.: Piper's Press. Walgrave, Lode. 1993. "In Search of Limits to the Restorative Justice for Juveniles." Paper presented at the International Congress on Criminology, Budapest, August 23-27. - - - . 1995. "Restorative Justice for Juveniles: Just a Technique or a Fully Fledged Alternative?" Howard Journal 34:228-49. - - - . 1999. "Community Service as a Cornerstone of a Systematic Restorative Response to (Juvenile) Crime." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. Warner, K. 1994. "The Rights of the Offender in Family Conferences." In Family Conferencing and Juvenile Justice: The Way Forward or Misplaced Optimism? edited by C. Alder and J. Wundersitz. Canberra: Australian Institute of Criminology. Warner, S. 1992. Making Amends: Justice for Victims and Offenders. Aldershot: Avebury. Waters, Andrew. 1993. "The Wagga Wagga Effective Cautioning Program: Reintegrative or Degrading?" Bachelor's thesis, University of Melbourne, Department of Criminology. Weitekamp, E. 1989. "Restitution: A New Paradigm of Criminal Justice or a New Way to Widen the System of Social Control?" Ph.D. dissertation, University of Pennsylvania. - - - . 1999. "The History of Restorative Justice." In Restorative Juvenile Justice: Repairing the Harm of Youth Crime, edited by G. Bazemore and L. Walgrave. Monsey, N.Y.: Criminal Justice Press. West, D.]., and D. P. Farrington. 1973. Who becomes Delinquent? Second Report of the Cambridge Study in Delinquent Development. London: Heinemann Educational. White, Rob. 1994. "Shame and Reintegration Strategies: Individuals, State Power and Social Interests." In Family Conferencing and Juvenile Justice, edited by Christine Alder and Joy Wundersitz. Canberra: Australian Institute of Criminology, Australian Studies in Law, Crime, and Justice. Wilson, J. Q., and R. Herrnstein. 1985. Crime and Human Nature. New York: Simon & Schuster. Wong, Dennis. 1996. "Paths to Delinquency: Implications for Juvenile Jus- HeinOnline -- 25 Crime & Just. 126 1999 Restorative Justice 127 tice In Hong Kong and China." Ph.D. dissertation, University of Bristol. ---.1998. "Juvenile Justice in China: The Quest for aJust Restorative Approach." Paper presented at the twelfth International Congress on Criminology, Seoul, August. Wright, M. 1982. Making Good: Prisons, Punishment and Beyond. London: Hutchinson. - - - . 1992. "Victim-Offender Mediation as a Step towards a Restorative System of Justice." In Restorative Justice on Trial: Pitfalls and Potentials of Victim-Offender Mediation-Inte171ational Research Perspectives, edited by H. Messmer and H. U. Otto. Dordrecht and Boston: Kluwer. Wundersitz, Joy, and Sue Hetzel. 1996. "Family Conferencing for Young Offenders: The South Australian Experience." In Family Group Conferences: Perspectives on Policy and Practice, edited by Joe Hudson, Allison Morris, Gabrielle Maxwell, and Burt Galaway. Sydney: Federation Press and Criminal Justice Press. Wynne, Jean. 1996. "Leeds Mediation and Reparation Service: Ten Years Experience with Victim-Offender Mediation." In Restorative Justice: International Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Yazzie, Robert, and James W. Zion. 1996. "Navajo Restorative Justice: The Law of Equality and Justice." In Restorative Justice: I11te171ational Perspectives, edited by Burt Galaway and Joe Hudson. Monsey, N.Y.: Criminal Justice Press. Yeats, Mary Ann. 1997. " 'Three Strikes' and Restorative Justice: Dealing with Young Repeat Burglars in Western Australia." Criminal Law Fornm 8:36985. Young, Iris. 1995. "Communication and the Other: Beyond Deliberative Democracy." In Justice and Identity: Antipodean Practices, edited by Margaret Wilson and Anna Yeatman. Wellington: Bridget Williams Books. Youth Justice Coalition. 1990. Kids in Justice: A Bluep/·int for the 90s. Sydney: Law Foundation of New South Wales. Zehr, Howard. 1985. Retributive Justice, Restorative Justice. New Perspectives on Crime and Justice. Occasional Papers of the MCC Canada Victim Offender Ministries Program and the MCC, U.S. Office of Criminal Justice, vol. 4. Elkhart, Ind.: Mennonite Central Committee; Kitchener, Ontario: Canada Victim Offender Ministries Program. - - - . 1990. Changing Lenses: A New Focus for Criminal Justice. Scottsdale, Pa.: Herald Press. - - - . 1995. "Rethinking Criminal Justice: Restorative Justice." Unpublished manuscript of a conference paper presented in Auckland. Zimring, F. E. 1981. "Kids, Groups and Crime: Some Implications of a WellKnown Secret." J01l171al of Criminal Law and Criminology 72:867-85. HeinOnline -- 25 Crime & Just. 127 1999 HeinOnline -- 25 Crime & Just. 128 1999

RELATED PAPERS

RELATED TOPICS

Log In

Restorative Justice: Assessing Optimistic and Pessimistic Accounts

Restorative Justice: Assessing Optimistic and Pessimistic Accounts

Related Papers

RELATED PAPERS

RELATED TOPICS