• 1 Introduction: what have we learned from restorative justice?

  The popularity and appetite for restorative justice waxes and wanes with different political, cultural and budgetary climates. While various forms of the practice have grown and shrunk alongside, or in opposition to, domestic criminal justice institutions for a number of decades, it is perennially viewed as an ‘up and coming’ idea – one which has the potential to transform criminal justice (Johnstone, 2013).
  Criminologists tend to paint a fairly grim picture of criminal justice policy and practice (Zedner, 2002). The story of the decline of the welfare state, the politicisation of criminal justice and the rise of the culture of control is well known (Garland, 2001; Reiner, 2016). Restorative justice has always sat somewhat uncomfortably alongside these developments, though it is possible that its concurrent rise has offered a ‘humanising cosmetic’ for increasingly harsh and bureaucratic practices at work in the larger criminal justice system (Kamenka & Tay, 1975, cited in Bottoms, 2003: 102; see also Maglione, 2019). Even so, various forms of restorative justice have enjoyed renewed support in Western democratic countries. To take England and Wales as an example, in 2013 the Ministry of Justice committed funding to local Police and Crime Commissioners to deliver restorative justice. The government also revised the Code of Practice for Victims to include provisions for restorative justice. Finally, the Crime and Courts Act 2013 gave statutory footing for restorative justice, allowing judges to defer sentence until restorative justice took place, if both victim and offender wanted it.
  The most recent Restorative Justice Action Plan was published by the Ministry of Justice in 2014, covering the period to 2018. In it, they explain their vision for ‘good quality, victim-focused restorative justice to be available at all stages of the criminal justice system in England and Wales’ (2014: 22). This includes ensuring all victims have equal access to restorative justice at all stages of the criminal justice system, the public has an awareness and understanding of restorative justice and that good quality restorative justice is consistently delivered. While it is unclear how such promises can be realised (Wright, 2015), for the moment, we might say that restorative justice is approaching mainstream status, at least more than it has at any other point. One reason for this is the solid evidence base from a range of different empirical projects documenting the positive benefits of restorative justice for both victims and offenders (Rossner, 2017; Shapland, Robinson & Sorsby, 2011; Sherman et al., 2015).
  At the same time, there is concern that the expectations for restorative justice are too high. While restorative justice can take many forms, the model most commonly researched, and the one that reports the most promising empirical findings is the ‘conferencing model’, where victims, offender and a wide range of stakeholders meet face to face to discuss the impacts and consequences of an offense (Rossner, 2017). This model is resource intensive, and (contrary to much practice and doctrine) potentially most suitable for more serious crimes and with adult offenders. Restorative justice practices that have developed to deal with the lower end of the offending spectrum, such as ‘street-level’ restorative encounters by the police, have been subject to less empirical scrutiny (Marder, 2018; Shapland, Crawford, Gray & Burn, 2017). While it does not seem likely that the conferencing model will become widespread, it is not clear that other forms of restorative justice can achieve appropriate constraining principles and standards (Braithwaite, 2002). A further worry about mainstreaming is that the progressive aims of restorative justice could be twisted to serve non-progressive ends, such as coercive treatment, a lack of procedural safeguards and net-widening (Levrant, Cullen, Fulton & Wozniak, 1999).
  In the text that follows I offer an expansive vision of what restorative justice can achieve as a part of the criminal justice system. In particular, I draw on lessons learned over the past few decades of research and practice of restorative justice conferencing in different settings and in different parts of the world. By now, we have had enough experience, both of success and failures in restorative justice, to look backwards at what we have learned and forward towards how these lessons can help move restorative practices and principles more towards the centre of criminal justice (see also Doak, 2011; Hoyle & Rosenblatt, 2016).
  In particular this paper will focus on three elements of restorative justice that might be immediately relevant to rethinking criminal justice. The first is an understanding of the ritual dynamics of the encounter. The second is the role that negative emotions, such as anger, can play in such rituals. The final component considers the transformative power of apology – forgiveness. This paper will focus on each of these three elements of restorative justice – drawing on insights from three decades of practice and research to offer a refinement of the relationship between anger, apology, forgiveness and restorative justice. I will conclude by suggesting that the ‘mainstreaming’ of restorative justice may not mean a future where all victims and offenders take part in a restorative justice conference, or even in ‘street-level’ restorative justice. Rather, I encourage us to think creatively about how we can design effective justice rituals that incorporate these emotions and symbolic exchanges.

• 2 On ritual and solidarity

  A significant area of research within restorative justice focuses on the role of ritual and collective emotion (Braithwaite & Mugford, 1994; Rossner, 2013). Such an interaction between victim, offender, family, friends and other stakeholders can be viewed as a ritual of reconciliation, reintegration and/or redemption. This line of inquiry builds on the long tradition of ritual theorists in sociology and anthropology (Collins, 2004; Maruna, 2011), premised on the idea that we develop a sense of who we are, our values and morals, our statuses, our networks, even our sense of citizenship, through the rituals that we participate in (Rossner & Meher, 2014). A ritual, simply defined as a focused interaction that contains symbolic elements, can be as formal as a marriage or a church service, or as informal as the polite exchanges with strangers on the street (Goffman, 1967). When a ritual is successful, it can charge us up with a sense of solidarity, shared emotion, and belonging (for instance, a royal wedding, a political rally, or an intense conversation with close friends). When a ritual fails, it can leave us feeling deflated or let down (a dinner party that sours or a lecture where one’s students stare back blankly). Rituals can also be stratified or asymmetrical, filling some people with positive feelings of belonging and membership and others with negative ones of status degradation and shame. A classic example of this is the well-known conception of criminal justice as a status-reducing degradation ritual (Carlen, 1976; Garfinkel, 1956; Maruna, 2011). Notably, rituals take work, participants continually adjust and adapt to the changing contours of the interaction (Goffman, 1967, 1974).
  Collins (2004) has offered a refinement of this perspective, articulating a theory of precisely how the micro dynamics of social interaction take on ritual forms. In short, an interaction ritual is a social encounter with the following distinct features: (1) people are physically together in a room and aware of each other’s bodily presence; (2) there are delineated boundaries between who is participating in the interaction and who is an outsider; (3) participants have a shared focus of attention and a (4) shared mood. Each of these elements feeds into each other to create a rhythmic coordination and synchronisation in participants’ conversation, bodily movements, and emotions. When people become ‘caught up in the rhythm and mood of the talk’ (Collins, 2004: 48), a sense of collective effervescence emerges along with feelings of social solidarity and shared emotion (Durkheim (1995 [1912]). This sense of shared emotion comes to represent a shared morality, influencing one’s longer-term emotional states, what Collins terms emotional energy. Emotional energy can include feelings of confidence, elation, initiative and a recommitment to the group’s standards of morality. Collins draws on a wide range of cases to offer empirical support for this perspective, ultimately arguing for the micro level, ritual foundation of larger normative frameworks.
  This framework can provide an explanatory theory of the restorative justice mechanism (Daly, 2016).¹xNote that this is different from a normative theory of how restorative justice should work. See Braithwaite and Pettit (2000). Restorative justice encounters are likely to be successful interaction rituals because they are likely to feature high intensity of emotions and are structured to include all the ingredients the theory posits. Key elements of the ritual are the inclusion of lay people and the privileging of (sometimes complex) narratives (Rossner, 2017). Victims and offenders come together, often sitting in a circle, to ‘tell their story’. Facilitators might guide the discussion from expressions of fear, anxiety and anger to expressions of reconciliation and solidarity. When successful, this can be observed through the articulation of apology and forgiveness, and symbolic integration through handshakes, eye contact and hugs (Rossner, 2011). In theory, the micro-level production of solidarity and shared emotion are what drives success in restorative justice (measured by victim and offender reports, and by measures of future offending). However, it is crucial to recognise that this is an ideal-type description of a successful interaction ritual. Restorative justice interactions can also fail to achieve these outcomes, leaving participants flat, deflated or angry. More often, though, much of the practice can be rather more mundane (Daly, 2002; Rossner & Bruce, 2018).
  In different empirical projects examining restorative justice conferences in the United Kingdom and Australia, I have tried to show the creative ways that practitioners plan, stage and choreograph these interactions in a way that brings about a shared mood, focus and rhythm (Rossner, 2013; Rossner & Bruce, 2018).²xThe dramaturgical metaphors are intentionally used here. Participants perform a role, but it is often one that is central to their identities and experiences (see Dignan et al., 2007). There is, however, worry that restorative justice interactions can be merely performative, more theatre than real. This has been demonstrated, for instance, by Rosenblatt (2015) in a case study of how ‘community participation’ in restorative justice can be rather perfunctory and passive. Community participants play a role, but their performance is rote and empty (see also Rossner & Bruce, 2016). While I acknowledge this danger, I would say this is an example of a failed, or empty, ritual. A dramaturgical language is vital to restorative justice, as performed rituals are precisely the places where actors (and their emotions) are transformed (Karstedt, 2006; Schechner, 1981). I have also suggested that successful rituals can have long-term benefits. For example, quantitative analysis of conferences dynamics suggests that conferences that were ritually successful – where participants indicated a sense of solidarity and shared emotion (measured by eye contact, touching and hugging, turn taking in conversation and other signs of solidarity) – lead to a lower frequency of offending compared to low-solidarity conferences, controlling for a range of demographic factors and prior offending behaviour (Rossner, 2013). This research contributes to developing an empirical agenda to lend credence to the idea that powerful rituals shape our beliefs and behaviours.³xResearch in this vein focuses on the conferencing model of restorative justice, where a wide range of stakeholders are invited to participate. While the conferencing model might easily be conceived as a solidarity creating (or alienating) ritual, it is also useful to think about the ritual elements of more diverse forms of restorative justice, such as restorative policing.
  Framing restorative justice as a particular form of ritual practice provides a useful way of thinking about the dynamics of conference interactions. In the next section, I will examine some of the research around symbolic reparation, or the exchange of apology and forgiveness. This ‘core sequence’ is often positioned as a unique element of the restorative justice encounter. Following that I will offer a refinement of the core sequence, focusing on (1) the ability of the ritual to incorporate strongly felt negative emotions such as anger and (2) a re-examination of the apology-forgiveness exchange. In the final section, I will offer some ideas for how these particular elements can be ‘mainstreamed’ into broader criminal justice practices.

• 3 The scarcity of symbolic reparation: the ‘core sequence’ of the restorative justice ritual

  It has long been understood that a key element of a restorative justice encounter is the moment of symbolic reparation (Retzinger & Scheff, 1996; Strang, 2002). This is the ‘magic’ exchange of remorse-apology-forgiveness. In the ideal-type restorative justice conference the offender comes to see the harmfulness and the wrongness of his or her action and then meaningfully expresses genuine remorse. This is followed by feelings and expressions of forgiveness by the victim, or the offender’s family, or someone else. This sequence may not follow this exact temporal order or take place sequentially or consecutively, but these exchanges are needed for this ‘core sequence’ to occur. Retzinger and Scheff have suggested that it is ‘the vital element that differentiates conferences from all other forms of crime control’ (1996: 317).
  When Retzinger and Scheff (1996) first identified this core sequence, based on their observations of a handful of conferences in Australia – they noted that the successful achievement of this sequence was vital, but rare. Indeed one of Daly’s (2002) great insights was to point out how very rare this sequence actually is. In interviews with restorative justice practitioners in New South Wales (Rossner, Bruce & Meher, 2013), we asked practitioners to give us a concrete example of what makes a good restorative justice conference. Even though the practitioners were spread across a large geographic area, working in different courts with different populations, again and again we heard the same few stories – of dramatic instances of symbolic reparation, victims who go on to mentor offenders, broken families beginning to heal. Daly calls these ‘mythical true stories’ (2002). They are true in that they actually happened. But they take on mythical status because they are a departure from the more mundane interactions that make up many restorative justice encounters.⁴xThere are plenty of good reasons for the same few stories to circulate and for practitioners to focus on mythical true stories. These stories can affirm restorative values, maintain morale in a challenging work environment and provide a focus for best practice. I am not critiquing the focus on these stories, but pointing out that the core sequence, as envisioned by Retzinger and Scheff (1996), is relatively rare.
  In an in-depth analysis of 34 restorative justice conferences, Jasmine Bruce and I identified eight cases that demonstrated the core sequence of apology-remorse-forgiveness, at least in its ‘mythical’ form (Rossner & Bruce, 2018). More often we saw more subtle forms of reparation, apologies not followed by the expression of forgiveness but instead by a nod of the head or a wave of the hand. Many victims agreed that the offender should not make material reparation, but rather could make amends by addressing his or her own ‘criminogenic’ problems, such as alcohol dependency or other addiction (see also Strang, 2002; Halsey et al., 2015).
  It is regularly reported in the literature that apologies are a common element of restorative justice encounters. Eighty-six per cent of Australian victims and 96 per cent of British victims who participated in a restorative justice conference received one, compared to 19 per cent and 7 per cent respectively of victims who went to court (Strang et al., 2013; see also Dhami, 2012). This is not surprising, as the structure of the courtroom interaction makes apologies less likely. As Shapland et al. note, ‘Having the victim present, particularly in a face-to-face meeting in which everyone speaks, seems definitely to be a major element in the offender working through embarrassment and nervousness … to remorse, expressed in apology’ (2011: 130). A closer look at apology, however, suggests a somewhat conflicting picture. In Daly’s research on restorative justice for young people in South Australia (2003), 40 per cent of offenders apologised spontaneously, apologies were drawn out by other group members in 28 per cent of cases, and in 30 per cent there was no apology at all. Of those who apologised, 27 per cent thought it would help their case (though this was not necessarily their motivation for doing so). At the same time, while 62 per cent of offenders said in interview that they were genuinely sorry, only 25 per cent of victims thought the main reason that the offender apologised was because he was sorry. This suggests that while apology is common, it may not always meet the dramatic performative criteria articulated by Retzinger and Scheff, and may not always be ‘genuine.’ On the other hand, offenders who do apologise are much less likely to reoffend compared to offenders who do not (Hayes & Daly, 2003; Rossner, 2013).
  The relationship between apology and forgiveness further confuses this story. For example, most victims (91 per cent) in Shapland et al.’s (2007) research indicated that they accepted the apology that was offered to them. However, only 45 per cent thought that the offender was sincere when he apologised and 21 per cent said they were not. A similar result is found by Strang (2002): 77 per cent of victims indicated the apology was sincere and 23 per cent did not. Hayes (2006) has also found in Australia that the core sequence of apology and forgiveness was achieved in less than half of the time. In most of this research, expressions of forgiveness were much less common than apologies. In Dhami’s research on mediation in the United Kingdom (2012), only 18 per cent of victims offered forgiveness, though they accepted the apology 91 per cent of the time during a mediation.
  The empirical research in this area suggests a kind of lopsided sequence. Apologies are common, but they are not always seen as genuine, and forgiveness is less often expressed.
  I suggest that that there are two things missing from this account of symbolic reparation. The first is that an understanding of forgiveness in restorative justice needs to include an analysis of the role that anger plays (since, after all, forgiveness is often defined as the cessation of anger or resentment). The second is that we need to be more creative and flexible about what forgiveness actually looks (and feels) like in restorative justice interactions. In what follows, I will provide a brief discussion of the role anger plays in restorative justice rituals. I will then re-examine the relationship between apology and forgiveness in the core sequence.

• 4 Anger can focus ritual dynamics

  The debate over whether restorative justice is a punitive response to wrongdoing or whether it can satisfy the goals of a retributive justice system is hotly contested (Daly, 2012; Duff, 2003; Walgrave, 2008). It is clear, however, that the first generation of restorative justice activists and thinkers explicitly viewed the underlying philosophy as one that rejects both retributivism and punitive impulses more broadly (Zehr, 2015). Anger is arguably a core element of a retributivist justification for state punishment (Lacey & Pickard, 2013). An offense is seen as a moral wrong that makes us, represented by the state, angry. We come to see the offender as deserving of punishment. We express this anger through the criminal procedure, through the rituals of courts and through censure and punishment.⁵xThe preceding description is of a retributive theory of state punishment. There is also evidence that on an individual level, people who feel more anger about crime are also more likely to harbor punitive beliefs (Johnson, 2009). See also Vidmar (2002) for a discussion of retribution and anger.
  Restorative justice theorists, like many liberal scholars working in law, philosophy, theology and sociology have rejected this view, arguing that anger is not the way to achieve justice, both on the interpersonal and the political level. A particularly elegant recent example comes from Nussbaum (2016). She argues that anger is both a morally inappropriate and instrumentally ineffective response to wrongdoing, in both political and interpersonal realms.
  To briefly summarise a complex set of ideas: Nussbaum points out that anger is morally problematic in that it is too closely connected to a desire for payback or revenge. She argues that this way of thinking is a mistake: there is too much ‘magical thinking’ involved in proportional payback – retribution and suffering cannot undo one’s own suffering. Furthermore, to desire the degradation of another is a moral flaw. Second, she argues that anger does not serve a useful or instrumental purpose. It is alienating towards your opponents, it can make you feel worse and it does not help you achieve your desired ends. While she acknowledges that anger is commonly felt, she suggests that it should quickly transition into more generous sentiments. In a ‘sane and not excessively anxious status-focused person, anger’s idea of retribution or payback is a brief dream or cloud, soon dispelled by saner thoughts on personal and social welfare’ (2016: 30-31).
  We can see echoes of these ideas in the critiques of criminal justice made by many in the restorative justice community. Nussbaum spends a significant portion of her book discussing strategies for justice that do not involve anger, most notably those pursued by Mandela, Gandhi and King. These are the same heroes of the restorative justice world. Their example, she says, ‘will help us see why the idea of “noble anger” is a false guide … and why a generous, even overgenerous, frame of mind is both more appropriate and more effective’ (2016: 212).
  Restorative justice has been positioned as a way to escape the retributivist anger of the state.⁶xIndeed, Nussbaum discusses a variation of restorative justice in these terms (2016: 200-202). But there is a tension here: restorative justice encounters are full of angry emotions (and at times retributive urges). How do we make sense of this? Nussbaum argues that anger is inappropriate both in the political and interpersonal realm. But, as the political philosopher Amia Srinivasan has pointed out (2016), this is a simplified reading of the lessons from history. In a review of Nussbaum she asks:

  Is it not somewhat historically naïve to think the embrace of King’s programme of racial harmony had nothing to do with the angry politics of Malcom X? Or that Mandela and Gandhi would have been successful without the anger of their followers? Would there have been the anticolonial struggles in North Africa or the Arab Spring without anger? And where would the labour, feminist, LGBTQ, and disability-rights movement be – would they even be at all? (Srinivasan, 2016).

  Anger has utility. Audre Lorde describes women’s anger as ‘a liberating and strengthening act of clarification’ that is ‘loaded with information and energy … when we turn from anger we turn from insight’ (1981: 127; see also Ahmed, 2013). Anger can help us, it can motivate us, it can provide us with agency. Even if it does not have utility in certain instances, it may still be an appropriate response to injustice. In this way, anger can have both intrinsic and instrumental values, and may not always be self-defeating (Srinivasan, 2018).
  Bringing this back to a discussion of restorative justice, early thinkers in this area have pointed out how anger is only ‘bad’ when it is not ‘properly channelled’ (Zehr, 2015). This lack of proper channelling can help explain why in contemporary criminal justice practices the concept of retribution has become ever more abstracted, leading to public demands for punishment that stereotype and degrade (Dzur & Werthereimer, 2002). As Zehr suggests:

  Victims need opportunities to express and validate their emotions: their anger, their fear, their pain. Even though such feelings may be hard for us to hear and may not square with what we would wish them to feel, these feelings are a natural, human response to the violation of crime. Anger, in fact, needs to be recognised as a common stage of suffering and one that cannot simply be short circuited (2015: 33).

  But what does it mean to ‘properly channel’ anger in restorative justice? Drawing on the ideas of Srinivasan and Lorde, anger can provide that mutual focus that seems to be the key to pulling off a successful ritual. Nussbaum, while dismissing anger as morally inappropriate and ineffective, suggests that when it is felt and expressed, there ought to come a ‘transition’ moment where the angry person moves away from a backward-looking desire for payback to a forward-looking focus on the future. Such a focus has been a feature of conferences that I have observed and interviews that I have conducted with practitioners and participants. The expression of anger can act as a turning point. One example comes from my field notes about a conference in London (from Rossner, 2013: 94):

  This is a robbery case where the victim’s wife suffered an attack of stress-related eczema prior to the conference. The conference started off quite badly when she refused to shake hands with the offender, stared at him with open hostility, and interrupted both the offender and her husband with comments that were ‘full of anger’. There was a change, however, when it came to be her turn to speak uninterrupted. As the facilitator described it, ‘All at once the anger and disdain that had been present in all my conversations with her prior to the conference was channelled to constructive words’. She clearly and emotionally described why she was so angry and the effect the robbery had on her and her family. She said, ‘Before I met you, I wanted to kill you’. And she slammed her fist into her hand. The offender moved back. She slowly and in order told of all the effects. The effects on her son, her family, her work, her eczema. The facilitator told me ‘I don’t think he expected such an outpouring, or for it to be so descriptive. She went into it.’ According to the facilitator, you could see her story ‘sink into’ the offender.

  Similarly, the following is an excerpt from an interview with an offender after his conference for a burglary. He describes the early stages of his conference as ‘awkward’ and then explains what precipitated a change:

  R: At first they [the victims] didn’t understand because I wasn’t talking. So they kinda got a bit angry – they didn’t know why they were bothering. I just sat there. I wanted to have words, but I just couldn’t say anything.
  I: So you think they were angry at you toward the beginning?
  R: Yeah, ‘cos it was like I wasn’t paying attention, or even caring. Then my brother had to stand up and say something … but at the end of it I told them what I was really thinking. It was good.

  In both these cases, the expression of anger acts as a turning point that helps to focus the rhythm and the emotional tone of the conference. In the first example, the dramatic description of anger and trauma shocks and draws in the participants. In the second example, due to anxiety or fear, initially the offender felt unable to speak in the conference. It continued in a disjointed manner with everyone getting increasingly upset. Only when his brother stood up and confronted him angrily did a transitional moment occur. He pointed to this exchange as the moment that allowed him to participate and this contributed to what he judged to be a successful conference.
  One might ask how does the expression of anger lead to a mutual focus and rhythm? Anger is not the only emotion to help bring such a focus. Other emotions might include anxiety, fear, shame or embarrassment. What these emotions have in common is that (1) they are strongly embodied and (2) they are disruptive and confrontational in a way that requires repair work by other participants. The expression of strong emotions can be startling, and can break the everyday rules of polite social interaction (Goffman, 1956). When participants are in an intimate setting such as a restorative justice conference, such emotions need attending to. This, for example, is why facilitators let ‘embarrassing silences’ drag on in the early stages of a conference. They know that participants will work together to ‘fix’ such disruptions, eventually leading them to follow turn-taking rules and engage with each other. When anger is expressed in a restorative justice setting, it may be the type of communicative act that jars an offender into an understanding of harm done and generates feelings of remorse. This is an ideal-type, mythical version of what anger does. More prosaically, strong emotions such as anger require repair work – participants attend to the angry person by engaging with them, asking them more questions, drawing out their narrative, and offering additional viewpoints. This is precisely how mutual focus and rhythm develop in an interaction ritual.
  Thinking of anger in these terms is consistent with larger concepts within the sociology of emotion and social psychology of anger. For instance, Katz (1999) has pointed out how anger arises in social interaction, has a specific phenomenology, and lends itself to narrative elements. It is a strongly embodied emotion, and when expressed in social interaction can provide a powerful focal point. Similarly, restorative justice may provide a way around ‘brute retributivism’ in that it provides a space for an offender to ‘get the message’ about a victim’s anger, fear or pain (Funk, McGeer & Gollwitzer, 2014; McGeer & Pettit, 2015). In other words, the focused anger expressed in a restorative justice dialogue is preferable to the abstracted anger of state retribution.
  This is not to say that the expression of anger is required or is an unmitigated good in restorative justice. There are numerous examples where anger disrupts ritual dynamics. People can be overwhelmed with anger that they cannot move past. Or the expression of anger may be seen by others as disproportionate to the harm caused and therefore illegitimate (Rossner & Bruce, 2018). While there is no simple calculus determining the ‘right’ amount of anger, proportionality is assessed and negotiated within each interaction. Anger is a useful emotion only in that it can help focus the attention of all participants, shaping the rhythmic contours of the ritual.
  To summarise, anger may provide a useful focus that helps to create an effective ritual. This is not a guarantee, and anger and other strongly felt emotions need to be managed to prevent a degradation ritual. In the next section, I will offer a closer look at the elements of symbolic reparation – apology and forgiveness. In the final section, I will shift focus to discuss how these interpersonal dynamics of restorative justice ritual can be reimagined in a broader criminal justice context.

• 5 Apology and restoration

  I argue above that anger can provide a ritual focus that can lead to this ‘vital but rare’ symbolic reparation – which is meant to be what makes restorative justice restorative. This apology-forgiveness transaction is often seen as the ‘restorative element’ of a conference. It is relatively unclear what restoration actually means in restorative justice, and many people have suggested different things. I will rely on Bottoms’ definition, where restoration is defined as ‘a restoration of prior social relationships in a community, with an understood structural and normative framework’ (2003: 93).
  How does apology lead to restoration? Tavuchis (1991: vii) notes that apology is a ‘delicate and precarious transaction’ that works a kind of magic (see also Goffman, 1971). ‘No matter how sincere or effective, [an apology] does not and cannot undo what has been done. And yet, in a mysterious way and according to its own logic, this is precisely what it manages to do’ (Tavuchis, 1991: 5). Both Tavuchis and Bottoms (2003) suggest that apology can provide restoration, at least in situations where the victim and offender share the same ‘understood structural and normative framework’. The so-called magic of apology is that it is a social mechanism that can help to restore breaches of this shared normative order. According to Bottoms, ‘The key to its success lies squarely in the normative and relational realms’ (2003: 97).
  This is a satisfying account of apology, but it leaves one wondering how this process works when there is not a ‘shared normative framework?’ Can we rely on there actually being a relationship to restore? A common critique of restorative justice is that the relationships it purports to repair may never have existed in the first place. In other words, in contemporary restorative justice practices, is it useful to value restoration? If not, then do apology and expressions of remorse become less important? Before addressing these questions, in the next section I will further discuss the meaning and role of forgiveness.

• 6 Forgiveness and transformation

  The second half of the core sequence is forgiveness. Forgiveness is widely seen as willingness to overcome resentment or anger, again reminding us that anger and other negative emotions must be a part of a ritual in order for forgiveness to occur. There is a significant amount of social-psychological research promoting the benefits of forgiveness, for both the forgiver and forgivee (Lacey & Picard, 2015). Among other things, forgiveness can promote reparative behaviour and de-escalation, increasing well-being and reducing a desire for revenge by victims of injustice. This helps explain the reduced desire for revenge among victims who participate in restorative justice (Sherman & Strang, 2011).
  While we may believe that forgiveness is beneficial and important,⁷xThough certainly not something that we can compel (See Braithwaite, 2002). the evidence suggests that it does not occur as often as apology does in restorative justice conferences. Braithwaite (2016) makes two important points about forgiveness in restorative justice. The first is that forgiveness exists along a continuum, and will rarely take the form of a dramatic demonstration. More common in restorative justice is an encounter where people reach a practical agreement about how to ‘offer some repair for the harm and close the matter’ (2016: 79). Drawing on the work of Chapman and Chapman (2016), Braithwaite argues that while these encounters fall short of explicit forgiveness, they are a success in that victims can say ‘that’s good enough for me’. He notes, ‘quite often victims hope never to see the offender again, but if they did meet them on the street at least it could be civil. Thin rather than thick civility’. (2016: 80). The second point that Braithwaite makes is that such ‘thin civility’ is part of the civilising impact of restorative justice.
  Armour and Umbreit (2018) conceptualise forgiveness as a kind of shift in energy. In their empirical work on victims’ narratives, they point to the subtle shifts in victims’ orientations as they prepare for, undertake, and reflect on restorative dialogue. This way of thinking about forgiveness is consistent with the ‘thin’ forgiveness articulated by Chapman and Chapman. When one tries to pinpoint this shift, it seems to be when the orientation of the ritual moves from a backward-looking exercise (expressions of anger and remorse) to a forward-looking one. Seen this way, forgiveness is an emergent property of the ritual, similar to the solidarity and emotional energy as discussed by Collins. It takes on symbolic dimensions, coming to represent the shared mood, solidarity and forward-looking orientation of participants.
  As described earlier, there is an imbalance in the core emotional sequence – apologies are common, forgiveness less so. But taking into account the subtle energy shifts and the thin civility of restorative justice, a different picture emerges. Shapland (2016) has noted that in her study very few victims explicitly articulated forgiveness. Instead, they were more likely to use implicit externalisations, such as acknowledging the offender’s remorse, nodding at them, shaking their hand, or even at times offering a hug. I report similar findings – participants and practitioners told me about the importance of turning points and the subtle cues that represent symbolic reparation and forgiveness, such as smiling, touching, laughing, or making small talk over tea after the formal end to the conference (Rossner, 2013: see also Strang, 2002). In-depth analysis of victims narratives, such as conducted by Umbreit and Armour (2018), reveals this subtle energy of forgiveness.
  Bringing together these ideas about apology and forgiveness, we can begin to rethink the core sequence. The standard account is that an apology, when genuine (Retzinger & Scheff, 1996) and given to someone who inhabits the same shared normative framework (Bottoms, 2003), can lead to forgiveness. This can be restorative in that it can restore offender and victim to their prior social relations. The problems with this are that (1) apology is not always genuine and (2) a shared normative framework and prior social relations cannot be assumed.
  Forgiveness is more ambitious than this. On the micro level, it can be seen as a transition, a shift in energy towards a future-focused orientation. In some ways it may be unfortunate that restorative justice is called ‘restorative,’ as such a backward-looking connotation can be limiting. While apologies can restore, forgiveness can transform. This is consistent with social-psychological research that suggests that most people seem to desire transformative, forward-looking outcomes from criminal justice (McGeer & Funk, 2017). Furthermore, there is no need for a shared normative framework for forgiveness to bring positive benefits. Forgiveness, both the thick and thin variety, can be a mechanism for transformation independent of any prior social relations.
  I want to pause here to consider the related issues of ‘genuine’ remorse and forgiveness. The core sequence of symbolic reparation suggests that it is imperative for an offender to feel ‘genuine’ remorse and for the victim to express ‘genuine’ forgiveness. However, much of the empirical research suggests that such sincerity occurs less often than one might like (Choi & Severson, 2009; Martin & Zappavigna, 2018). For example, Strang (2002) found that while 39 per cent of victims said they felt forgiving after their conference, researchers observed forgiveness in around two-thirds of the cases, suggesting that victims were performing something that they may not have genuinely felt. In such an intimate and emotional setting, offenders and victims might feel enormous pressure to apologise and forgive. Even if these feelings and expressions are genuine, they may be short-lived once participants return to the everyday settings of their lives. Conceivably, this does not reduce the moral power of the apology-forgiveness exchange. Dzur and Wethereimer (2002) have suggested that the performative action of both apology and forgiveness carries moral weight independent of the feelings attached to it. Drawing on Austin (1981), they suggest that enacting forgiveness may serve to generate the relevant feelings. And even if it does not, it still commits the victim and offender to certain ways of interacting in the future. Bennett picks up this line of thinking with apology, arguing that when requiring offenders to ritually apologise, ‘the apology will be valid regardless of whether it is sincere or blatantly insincere’ (2006: 132).⁸xBennett’s use of the term ritual seems to suggest something empty or rote. While this is a common understanding of ritual in everyday vernacular, the tradition of ritual theory in Sociology and Anthropology would suggest that even so-called ‘empty’ rituals carry significant symbolic and moral weight. Again, this is because of the moral value of the performative act (see also Celermajer, 2013). While I do think, based on the empirical evidence that restorative justice encounters are more likely than other forms of criminal justice encounters to engender ‘genuine’ apology and forgiveness of some form, it is worth noting that even in the instances where it does not there may still be value in such an exchange.
  To summarise, the mechanism that seems to make restorative justice a potentially powerful criminal justice encounter relies on its successful ritual dynamics. The key elements of this ritual are the expression of anger (and other strongly embodied emotions) and the value of symbolic reparation, commonly articulated through the exchange of apology and forgiveness. After taking a close look into this exchange, I suggest a more creative understanding of symbolic reparation. It involves worrying less about apology and its restorative power and more on rethinking forgiveness, or similar forward-looking transformative energies. In the next section, I will offer some ideas for how these insights can be integrated into broader criminal justice practices.

• 7 Institutionalising restorative justice

  In this article, I have explored three particular strengths of the conferencing process – its unique ritual elements, its ability to turn anger into a useful emotion and its ability to transform relationships through sometimes subtle forms of forgiveness. I conclude by returning to the idea of mainstreaming. While restorative justice privileges the relationship between the offender and victim, the criminal justice system is focused on the relationship between the offender and the state. An abolitionist restorative strategy might be to refocus the entire system away from offender-state relations towards offender-victim relations. Mainstreaming in this sense would involve upscaling something like the conferencing model to all offenders and victims. However, there are a number of instances where the focus on the offender-victim relationship is not possible or not appropriate. Victims may be dead, or unwilling to participate, or unwilling to forgive. Or the crime may be ‘victimless’ in the legal sense. More pragmatically, the conferencing model is resource intensive and not likely to be widely offered. As yet, it is unclear how other forms of restorative justice, such as ‘street-level’ restorative policing, can meet these criteria.⁹xThey might, they just have not been properly evaluated as such. Given these constraints, in this final section I ask whether the offender-state relationship can be reimagined to include the elements of restorative justice discussed above.
  In order to conduct such an exercise, I will shift the focus from the offender-victim relationship of restorative justice to the offender-state relationship of criminal law and criminal justice. In some ways this is straightforward – ritual and symbol pervade criminal justice, one can imagine ways to adapt the ritual to promote shared mood and solidarity. In other ways it is not so straightforward – how do we channel ‘state anger’ into a productive, focused ritual? Can the state forgive an offender? I do not pretend to have the answers to these questions. My goal is to offer some programmatic examples that seem to contain some of the same elements of successful restorative justice: focused interaction rituals that promote solidarity; expressions of anger that do not stigmatise or degrade; and creative forms of forgiveness that promote thin civility and a forward-looking orientation to criminal justice.
  We know that criminal justice offers powerful rituals (Maruna, 2011). However, they tend to be asymmetrical degradation rituals, rather than solidarity-building and status elevating ones. The expression of anger is also clearly already a part of criminal justice. But it is the anger of the state expressed through ‘affective blame’ and retribution, not the anger of victims and communities expressed through respectful dialogue. Anger can also be expressed through Victim Impact Statements, but this is also an asymmetric ritual, not a dialogue, and also usually only reserved for very serious crime (see Booth, 2012; Rock, 2010). Examples of criminal justice interactions that take on ritual dynamics, express anger, but do not degrade can be observed in the rise of problem-solving courts where offenders, judges and other criminal justice and service professionals engage in regular supervisory meetings (Snedker, 2016; Winick, 2002). This can be seen as a more general move towards therapeutic jurisprudence, where it is widely recognised that both legal rules and legal actors can be therapeutic agents, and a sensitivity to the dynamics of justice interactions is paramount (Wexler, 1990, 2011; Winick, 2013).¹⁰xThe relationship between restorative justice and therapeutic jurisprudence is explored further in this issue’s ‘Notes from the Field’ by Jo-Anne Wemmers.
  The core sequence of apology-forgiveness in restorative justice is centred on interpersonal dynamics. Can we transpose these interpersonal dynamics to an institutional context, such as criminal courts? Currently apology plays a role in the criminal justice system only in as much as remorse can be used as a mitigating factor at sentencing (for problems with this, see Bandes, 2016a, 2016b). Forgiveness is not a formal part of the system.¹¹xMercy is, but this is different from forgiveness. At least one reason why this is so is that to show mercy you need the power to punish. This suggests that mercy can only happen at the sentencing (or perhaps parole stage). Theoretically, forgiveness can happen anytime. See Lacy and Pickard (2015); Allais (2008). I have argued above that forgiveness in restorative justice conferences involves a shift in energy towards a forward-looking orientation. Can such a shift happen at the institutional level? In other words, can the state forgive an offender?
  Lacey and Pickard (2015) have offered some ideas about ‘institutional forgiveness’. They argue that, at least in an institutional context, forgiveness is not something to be ‘earned’ through remorse or apology. To require this, sets up a transactional exchange that can easily lead to coercion – it can, in their words, ‘smuggle in vengeance and affective blame.’ (See also Nussbaum, 2016).¹²xThey argue, like Nussbaum, that we should do away with anger (or what they call ‘affective blame’) and substitute it with a form of criminal justice that is inspired by clinical models that separate responsibility from blame, followed by a kind of forgiving punishment (Lacey & Pickard, 2013). I would say that we still need some expression of anger or similar negative emotions in order to have a transformative ritual (see also Norrie, 2018 for a related discussion of guilt and love). One could easily imagine a criminal justice interaction that quickly devolves into such a transaction. Asking offenders to acknowledge wrongdoing without excuse or justification can also limit the availability of ‘redemption scripts’ (Lacey & Pickard, 2015; Maruna, 2001). It is possible then that apology is less important in an institutional context than it might be in an interpersonal one (such as a restorative justice conference). When moving from the interpersonal to the institutional, we may want to consider decoupling apology and forgiveness.¹³xConversely one could argue that we disentangle apology and forgiveness, perhaps by requiring the types of sincerity-free apology rituals offered by Bennett (2006). While this may be odious to many restorative justice activists, ‘such a price will need to be paid if the process they recommend is to be integrated into the criminal justice system of an acceptably liberal state’ (2006: 151). A focus on forgiveness then, ‘move[s] us away from backward-looking retaliation and wrath, and towards more humane and forward-looking attitude towards offenders’ (Lacey & Pickard, 2013: 28).
  Interpersonal forgiveness obviously cannot be compelled or mandated. This does not hold true for institutional forgiveness. Lacey and Pickard’s suggestions promote state forgiveness through various forms of what Braithwaite and Chapman and Chapman might call ‘thin civility’.¹⁴xNote here that they are explicitly focusing on how states can forgive offenders. They are not concerned with victim forgiveness. In a later article, they argue that state criminal justice systems cannot meet the needs of individual victims, and as such a ‘dual process model’ that can meet the needs of both parties is needed (Lacey & Pickard, 2019). Given that most research suggests that forgiveness benefits victims, I would similarly endorse such a model. For instance, they propose reforming criminal procedure and evidence law in a way that leaves out any condemnation of character, instead focusing on the damage suffered – similar to the decoupling of the act and actor that is central to reintegrative shaming theory (Braithwaite, 1989). It allows for anger to be expressed, but in a way that respects an offender and makes forgiveness possible. Second, they reimagine a sentencing hearing as a dialogic interaction, where judges, professionals and offenders work together to decide the most appropriate sentence, including reparation and rehabilitation. This is similar to the restorative sentencing circles seen in indigenous communities around the world (Daly & Marchetti, 2012). Finally, they imagine a form of incarceration that does not rupture an offenders’ social bonds, such as therapeutic communities, part time imprisonment, family visits of varying lengths and less isolated prisons. They argue that this is forgiving in that it reminds the offender of both mutually beneficial past relationships, and maintains relationships so that they will stay strong in the future.
  These are only preliminary sketches, and suggest a ‘thin’ variety of civilising practice. Each of these examples shifts the focus towards the future. These suggestions might seem too much a departure from the core business of restorative justice conferencing. They are certainly more ambitious than what is contained in the Restorative Justice Action Plan. But they offer a way to move from the interpersonal to the institutional, and are feasible ways that elements of restorative justice practice can influence the larger machinery of criminal justice.

• 8 Conclusion

  This paper has offered an analysis of some key elements of restorative justice that set it apart from traditional criminal justice encounters. To be clear, I am not saying that all restorative justice interactions will be successful rituals where anger provides a focus that can lead to symbolic reparation. However, I am saying that (1) these elements of restorative justice can help explain why it has the potential to be a beneficial practice for victims and offenders, and (2) that these elements are more likely to appear in restorative justice encounters compared to other forms of criminal justice. I have suggested that one way of ‘mainstreaming’ restorative justice is not to offer a conference to all victims and offenders, nor to offer a modified version that can be used for street-level encounters. Rather, if we zoom in on the key principles – respectful ritual that allows for anger to be expressed, and a recognition of the transformative, if subtle, energy of forgiveness – then we can be creative in thinking about how such principles are incorporated into every stage of the criminal justice system. This is the larger civilising project that restorative justice seeks to achieve.

References

• Ahmed, S. (2013). The cultural politics of emotion. London: Routledge.

• Allais, L. (2008). Wiping the slate clean: the heart of forgiveness. Philosophy & Public Affairs, 36(1), 33-68.

• Armour, M. & Umbreit, M.S. (2018). Violence, restorative justice, and forgiveness: dyadic forgiveness and energy shifts in restorative justice dialogue. London: Jessica Kingsley Publishers.

• Bandes, S.A. (2016a). Remorse and criminal justice. Emotion Review, 8(1), 14-19. doi: 10.1177/1754073915601222.

• Bandes, S.A. (2016b). Remorse and demeanor in the courtroom: cognitive science and the evaluation of contrition. In J. Hunter, P. Roberts, S.N.M. Young & D. Dixon (eds.), The integrity of the criminal process (pp. 309-326). Oxford: Hart Publishing. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2363326 (last accessed 5 September 2019).

• Booth, T. (2012). ‘Cooling out’ victims of crime: managing victim participation in the sentencing process in a superior sentencing court. Australian & New Zealand Journal of Criminology, 45(2), 214-230. doi: 10.1177/0004865812443680.

• Bottoms, A. (2003). Some sociological reflections on restorative justice. In A. von Hirsch, J.V. Roberts, A.E. Bottoms, K. Roach & M. Schiff (eds.), Restorative justice and criminal justice: competing or reconcilable paradigms (pp. 79-113). Oxford: Hart Publishing.

• Braithwaite, J. (1989). Crime, shame and reintegration. Cambridge: Cambridge University Press.

• Braithwaite, J. (2002). Setting standards for restorative justice. British Journal of Criminology, 42(3), 563-577. doi: 10.1093/bjc/42.3.563.

• Braithwaite, J. (2016). Redeeming the ‘F’ word in restorative justice. Oxford Journal of Law and Religion, 5(1), 79-93. doi: 10.1093/ojlr/rwv049.

• Braithwaite, J. & Mugford, S. (1994). Conditions of successful reintegration ceremonies: dealing with juvenile offenders. British Journal of Criminology, 34(2), 139-171. doi: 10.1093/oxfordjournals.bjc.a048400.

• Braithwaite, J. & Pettit, P. (2000). Republicanism and restorative justice: an explanatory and normative connection. In H. Strang & J. Braithwaite (eds.), Restorative justice: from philosophy to practice (pp. 145-163). Burlington: Ashgate.

• Carlen, P. (1976). The staging of magistrates justice. The British Journal of Criminology, 16(1), 48-55. doi: 10.1093/oxfordjournals.bjc.a046692.

• Celermajer, D. (2013). Mere ritual? Displacing the myth of sincerity in transitional rituals. The International Journal of Transitional Justice, 7(2), 286-305. doi: 10.1093/ijtj/ijt003.

• Collins, R. (2004). Interaction ritual chains. Princeton: Princeton university press.

• Daly, K. (2002). Restorative justice: the real story. Punishment & Society, 4(1), 55-79. doi: 10.1177/14624740222228464.

• Daly, K. (2012). The punishment debate in restorative justice. In J. Simon & R. Sparks (eds.), The handbook of punishment and society (pp. 356-372). London: Sage.

• Daly, K. (2016). What is restorative justice? Fresh answers to a vexed question. Victims & Offenders, 11(1), 9-29. doi: 10.1080/15564886.2015.1107797.

• Daly, K. & Marchetti, E. (2012). Innovative justice processes: restorative justice, indigenous justice, and therapeutic jurisprudence. In M. Marmo, W. De Lint & D. Palmer (eds.), Crime and justice: a guide to criminology (pp. 455-481). Pyrmont: Lawbook Co.

• Dhami, M.K. (2012). Offer and acceptance of apology in victim-offender mediation. Critical Criminology, 20(1), 45-60. doi: 10.1007/s10612-011-9149-5.

• Dignan, J., Atkinson, A., Atkinson, H., Howes, M., Johnstone, J., Robinson, G., Shapland, J. & Sorsby, A. (2007). Staging restorative justice encounters against a criminal justice backdrop: a dramaturgical analysis. Criminology & Criminal Justice, 7(1), 5-32. doi: 10.1177/1748895807072474.

• Doak, J. (2011). Honing the stone: refining restorative justice as a vehicle for emotional redress. Contemporary Justice Review, 14(4), 439-456. doi: 10.1080/10282580.2011.616374.

• Durkheim, E. (1995 [1912]). Elementary forms of religious life. New York: The Free Press.

• Duff, R.A. (2003). Restoration and retribution. In A. von Hirsch, J.V. Roberts, A.E. Bottoms, K. Roach & M. Schiff (eds.), Restorative justice and criminal justice: competing or reconcilable paradigms (pp. 43-59). Oxford: Hart.

• Funk, F., McGeer, V. & Gollwitzer, M. (2014). Get the message: punishment is satisfying if the transgressor responds to its communicative intent. Personality and Social Psychology Bulletin, 40(8), 986-997. doi: 10.1177/0146167214533130.

• Garfinkel, H. (1956). Conditions of successful degradation ceremonies. American Journal of Sociology, 61(5), 420-424. doi: 10.1086/221800.

• Garland, D. (2001). The culture of control: crime and social order in contemporary society. Chicago: University of Chicago Press.

• Goffman, E. (1956). Embarrassment and social organization. American Journal of Sociology, 62(3), 264-271. doi: 10.1086/222003.

• Goffman, E. (1967). Interaction ritual. New York: Doubleday.

• Goffman, E. (1971). Relations in public: microstudies of the social order. London: Allen Lane.

• Goffman, E. (1974). Frame analysis. Boston: Northeastern University Press.

• Hayes, H. (2006). Apologies and accounts in youth justice conferencing: reinterpreting research outcomes 1. Contemporary Justice Review, 9(4), 369-385. doi: 10.1080/10282580601014292.

• Hayes, H. & Daly, K. (2003). Youth justice conferencing and reoffending. Justice Quarterly, 20(4), 725-764. doi: 10.1080/07418820300095681.

• Halsey, M., Goldsmith, A. & Bamford, D. (2015). Achieving restorative justice: assessing contrition and forgiveness in the adult conference process. Australian & New Zealand Journal of Criminology, 48(4), 483-497. doi: 10.1177/0004865814538433.

• Hoyle, C. & Rosenblatt, F.F. (2016). Looking back to the future: threats to the success of restorative justice in the United Kingdom. Victims & Offenders, 11(1), 30-49. doi: 10.1080/15564886.2015.1095830.

• Johnson, D. (2009). Anger about crime and support for punitive criminal justice policies. Punishment & Society, 11(1), 51-66. doi: 10.1177/1462474508098132.

• Johnstone, G. (2013). Restorative justice: ideas, values, debates. London: Routledge.

• Karstedt, S. (2006). Emotions, crime, and justice: exploring durkheimian themes. In M. Deflem (ed.), Sociological theory and criminological research: views from Europe and the United States (pp. 223-248). Bingley: Emerald Group Publishing Limited.

• Katz, J. (1999). How emotions work. University of Chicago Press.

• Lacey, N. & Pickard, H. (2013). From the consulting room to the court room? Taking the clinical model of responsibility without blame into the legal realm. Oxford Journal of Legal Studies, 33(1), 1-29. doi: 10.1093/ojls/gqs028.

• Lacey, N. & Pickard, H. (2015). To blame or to forgive? Reconciling punishment and forgiveness in criminal justice. Oxford Journal of Legal Studies, 35(4), 665-696. doi: 10.1093/ojls/gqv012.

• Lacey, N. & Pickard, H. (2019). A dual-process approach to criminal law: victims and the clinical model of responsibility without blame. Journal of Political Philosophy, 27(2), 229-251. doi: 10.1111/jopp.12160.

• Levrant, S., Cullen, F.T., Fulton, B. & Wozniak, J.F. (1999). Reconsidering restorative justice: the corruption of benevolence revisited? Crime & Delinquency, 45(1), 3-27. doi: 10.1177/0011128799045001001.

• Lorde, A. (1981). The uses of anger: women responding to racism. In Sister outsider: essays and speeches (pp. 124-133). Trumansburg: Crossing Press.

• Maglione, G. (2019). The political rationality of restorative justice. Theoretical Criminology, 23(4). doi: 10.1177/1362480618756364.

• Marder, I.D. (2018). Restorative justice and the police: exploring the institutionalisation of restorative justice in two English forces (Doctoral Dissertation). University of Leeds, England.

• Martin, J.R. & Zappavigna, M. (2018). Rites of passion: remorse, apology and forgiveness in youth justice conferencing. Linguistics and the Human Sciences, 12(2-3), 101-121. doi: 10.1558/lhs.36986.

• Maruna, S. (2001). Making good: how ex-convicts reform and rebuild their lives. Washington, DC: American Psychological Association.

• Maruna, S. (2011). Reentry as a rite of passage. Punishment & Society, 13(1), 3-28. doi: 10.1177/1462474510385641.

• McGeer, V. & Funk, F. (2017). Are ‘optimistic’ theories of criminal justice psychologically feasible? The probative case of civic republicanism. Criminal Law and Philosophy, 11(3), 523-544. doi: 10.1007/s11572-015-9381-2.

• McGeer, V. & Pettit, P. (2015). The desirability and feasibility of restorative justice. Restorative Justice: An International Journal, 3(3), 325-341. doi: 10.1080/20504721.2015.1110885.

• Ministry of Justice. (2014). Restorative justice action plan for the criminal justice system for the period to March 2018. Retrieved from https://www.gov.uk/government/collections/restorative-justice-action-plan (last accessed 5 September 2019).

• Norrie, A. (2018). Love and justice: can we flourish without addressing the past? Journal of Critical Realism, 17(1), 17-33. doi: 10.1080/14767430.2018.1426525.

• Nussbaum, M.C. (2016). Anger and forgiveness: resentment, generosity, justice. Oxford: Oxford University Press.

• Reiner, R. (2016). Crime, the mystery of the common-sense concept. New York: John Wiley & Sons.

• Retzinger, S.M. & Scheff, T.J. (1996). Strategy for community conferences: emotions and social bonds. In B. Galaway & J. Hudson (eds.), Restorative justice: international perspectives (pp. 315-336). Monsey: Criminal Justice Press.

• Rosenblatt, F.F. (2015). The role of community in restorative justice. London: Routledge.

• Rossner, M. (2011). Emotions and interaction ritual: a micro analysis of restorative justice. British Journal of Criminology, 51(1), 95-119. doi: 10.1093/bjc/azq075.

• Rossner, M. 2013. Just emotions: rituals of restorative justice. Oxford: Oxford University Press.

• Rossner, M. 2017. Making emotions mainstream: restorative justice in the 21st century. In A. Liebling, S. Maruna & L. McAra (eds.), The Oxford handbook of criminology (pp. 967-1010). Oxford: Oxford University Press.

• Rossner, M. & Bruce, J. (2016). Community participation in restorative justice: rituals, reintegration, and quasi-professionalization. Victims & Offenders, 11(1), 107-125.

• Rossner, M. & Bruce, J. (2018). Trajectories and typologies of pre-sentence restorative justice rituals. Australian & New Zealand Journal of Criminology, 51(4), 502-518. doi: 10.1177/0004865817749263.

• Rossner, M. & Meher, M. (2014). Emotions in ritual theories. In J. Stets & J. Turner (eds.), Handbook of the sociology of emotions (Vol. II, pp. 199-220). Dordrecht: Springer.

• Rossner, M., Bruce, J. & Meher, M. (2013). The process and dynamics of restorative justice: research on forum sentencing. Report to the New South Wales Department of Justice and Attorney General. Sydney: University of Western Sydney. Retrieved from https://researchdirect.westernsydney.edu.au/islandora/object/uws:17926/datastream/PDF/view (last accessed 5 September 2019).

• Schechner, R. (1981). Performers and spectators transported and transformed. The Kenyon Review, 3(4), 83-113.

• Shapland, J. (2016). Forgiveness and restorative justice: is it necessary? Is it helpful? Oxford Journal of Law and Religion, 5(1), 94-112. doi: 10.1093/ojlr/rwv038.

• Shapland, J., Atkinson, A., Atkinson, H., Chapman, B., Dignan, J., Howes, M., Johnstone, J., Robinson, G. & Sorsby, A. (2007). Restorative justice: the views of victims and offenders. Ministry of Justice Research Series, 3(7).

• Shapland, J., Robinson, G. & Sorsby, A. (2011). Restorative justice in practice: evaluating what works for victims and offenders. London: Taylor & Francis.

• Shapland, J., Crawford, A., Gray, E. & Burn, D. (2017). Restorative justice at the level of the police in England: implementing change. Universities of Sheffield and Leed. Retrieved from https://www.sheffield.ac.uk/polopoly_fs/1.743733!/file/DevelopingRestorativePolicing3.pdf (last accessed 5 September 2019).

• Sherman, L. & Strang, H. (2011). Empathy for the devil: the nature and nurture of revenge. In S. Karstedt, I. Loader & H. Strang (eds.), Emotions, crime and justice (pp. 145-168). Oxford: Hart Publishing.

• Sherman, L.W., Trang, H., Barnes, G., Woods, D.J., Bennett, S., Inkpen, N., Newbury-Birch, D., Rossner, M., Angel, C., Mearns, M. & Slothower, M. (2015). Twelve experiments in restorative justice: the Jerry Lee program of randomized trials of restorative justice conferences’. Journal of Experimental Criminology, 11(4), 501-540. doi: 10.1007/s11292-015-9247-6.

• Snedker, K.A. (2016). Unburdening stigma: identity repair through rituals in mental health court. Society and Mental Health, 6(1), 36-55. doi: 10.1177/2156869315598203.

• Srinivasan, A. (2016). Would politics be better off without anger? The nation. Retrieved from https://www.thenation.com/article/a-righteous-fury/ (last accessed 5 September 2019).

• Srinivasan, A. (2018). The aptness of anger. Journal of Political Philosophy, 26(2), 123-144. doi: 10.1111/jopp.12130.

• Strang, H. (2002). Repair or revenge: victims and restorative justice. Oxford: Clarendon Press.

• Strang, H., Sherman, L. W., Mayo-Wilson, E., Woods, D. & Ariel, B. (2013). Restorative justice conferencing (RJC) using face-to-face meetings of offenders and victims: effects on offender recidivism and victim satisfaction. A systematic review. Campbell Systematic Reviews, 9(1), 1-59.

• Tavuchis, N. (1991). Mea culpa: a sociology of apology and reconciliation. Stanford: Stanford University Press.

• Vidmar, N. (2002). Retributive justice: its social context. In M. Ross & D. Miller (eds.), The justice motive in everyday life (pp. 291-313). Cambridge: Cambridge University Press.

• Walgrave, L. (2008). Restorative justice, self interest and responsible citizenship. Cullompton: Willan.

• Wexler, D.B. (1990). Therapeutic jurisprudence: the law as a therapeutic agent. Durham: Carolina Academic Press.

• Wexler, D.B. (2011). From theory to practice and back again in therapeutic jurisprudence: now comes the hard part. Monash University Law Review, 37, 33.

• Winick, B.J. (2002). Therapeutic jurisprudence and problem solving courts. Fordham Urban Law Journal, 30(3), 1055.

• Winick, B.J. (2013). Problem solving courts: therapeutic jurisprudence in practice. In R.L. Wiener & E. Brank (eds.), Problem solving courts (pp. 211-236). New York: Springer.

• Wright, M. (2015). Making it happen or letting it happen. Restorative Justice: An International Journal, 3(1), 119-128. doi: 10.1080/20504721.2015.1049880.

• Zehr, H. (2015). Changing lenses: restorative justice for our times. Harrisonburg: Herald Press.

• Zedner, L. (2002). Dangers of dystopias in penal theory. Oxford Journal of Legal Studies, 22(2), 341-366. doi: 10.1093/ojls/22.2.341.

Noten

• * Contact: meredith.rossner@anu.edu.au.

• 1 Note that this is different from a normative theory of how restorative justice should work. See Braithwaite and Pettit (2000).

• 2 The dramaturgical metaphors are intentionally used here. Participants perform a role, but it is often one that is central to their identities and experiences (see Dignan et al., 2007). There is, however, worry that restorative justice interactions can be merely performative, more theatre than real. This has been demonstrated, for instance, by Rosenblatt (2015) in a case study of how ‘community participation’ in restorative justice can be rather perfunctory and passive. Community participants play a role, but their performance is rote and empty (see also Rossner & Bruce, 2016). While I acknowledge this danger, I would say this is an example of a failed, or empty, ritual. A dramaturgical language is vital to restorative justice, as performed rituals are precisely the places where actors (and their emotions) are transformed (Karstedt, 2006; Schechner, 1981).

• 3 Research in this vein focuses on the conferencing model of restorative justice, where a wide range of stakeholders are invited to participate. While the conferencing model might easily be conceived as a solidarity creating (or alienating) ritual, it is also useful to think about the ritual elements of more diverse forms of restorative justice, such as restorative policing.

• 4 There are plenty of good reasons for the same few stories to circulate and for practitioners to focus on mythical true stories. These stories can affirm restorative values, maintain morale in a challenging work environment and provide a focus for best practice. I am not critiquing the focus on these stories, but pointing out that the core sequence, as envisioned by Retzinger and Scheff (1996), is relatively rare.

• 5 The preceding description is of a retributive theory of state punishment. There is also evidence that on an individual level, people who feel more anger about crime are also more likely to harbor punitive beliefs (Johnson, 2009). See also Vidmar (2002) for a discussion of retribution and anger.

• 6 Indeed, Nussbaum discusses a variation of restorative justice in these terms (2016: 200-202).

• 7 Though certainly not something that we can compel (See Braithwaite, 2002).

• 8 Bennett’s use of the term ritual seems to suggest something empty or rote. While this is a common understanding of ritual in everyday vernacular, the tradition of ritual theory in Sociology and Anthropology would suggest that even so-called ‘empty’ rituals carry significant symbolic and moral weight.

• 9 They might, they just have not been properly evaluated as such.

• 10 The relationship between restorative justice and therapeutic jurisprudence is explored further in this issue’s ‘Notes from the Field’ by Jo-Anne Wemmers.

• 11 Mercy is, but this is different from forgiveness. At least one reason why this is so is that to show mercy you need the power to punish. This suggests that mercy can only happen at the sentencing (or perhaps parole stage). Theoretically, forgiveness can happen anytime. See Lacy and Pickard (2015); Allais (2008).

• 12 They argue, like Nussbaum, that we should do away with anger (or what they call ‘affective blame’) and substitute it with a form of criminal justice that is inspired by clinical models that separate responsibility from blame, followed by a kind of forgiving punishment (Lacey & Pickard, 2013). I would say that we still need some expression of anger or similar negative emotions in order to have a transformative ritual (see also Norrie, 2018 for a related discussion of guilt and love).

• 13 Conversely one could argue that we disentangle apology and forgiveness, perhaps by requiring the types of sincerity-free apology rituals offered by Bennett (2006). While this may be odious to many restorative justice activists, ‘such a price will need to be paid if the process they recommend is to be integrated into the criminal justice system of an acceptably liberal state’ (2006: 151).

• 14 Note here that they are explicitly focusing on how states can forgive offenders. They are not concerned with victim forgiveness. In a later article, they argue that state criminal justice systems cannot meet the needs of individual victims, and as such a ‘dual process model’ that can meet the needs of both parties is needed (Lacey & Pickard, 2019). Given that most research suggests that forgiveness benefits victims, I would similarly endorse such a model.