The Ritual Elements
The Ritual Elements
Abstract and Keywords
This chapter introduces the key theoretical concepts drawing on contemporary sociological theorists of interaction and emotions, including Erving Goffman and Randall Collins. Key terms are explained, such as interaction ritual, solidarity, emotional entrainment, emotional energy, short- and long-term emotions, and symbolic representations. Using this framework, the author proposes a micro level theory of restorative justice. This chapter also examines the different methodological approaches to studying ritual used in this research. It emphasizes the importance of deploying mixed methods and micro sociological approaches, including in-depth interviews; conversation analysis; the analysis of face and demeanor; the use of dramaturgy and game-playing metaphors to analyse interaction; and quantitative analysis of micro level variations across a sample of conferences.
Rituals in the Justice Process
The research on restorative justice has shown that conferences can be transformative, producing high levels of satisfaction and even a commitment to stop offending. They can also fall flat, with little emotional investment or return for the participants. The theories reviewed in Chapter Two suggested that one reason conferences succeed is because they bring about the ‘good kind’ of shame as opposed to the ‘harmful kind’. That is, the non-hierarchical and participatory nature of the conference instils a sense of legitimacy into the proceedings, leading victims and offenders to believe the criminal justice system is just and fair. While all of this may be true, it is not clear on a micro level how these interactions work to produce the ‘right kind’ of shame, or enable participants to feel they have a say or are empowered. Indeed, numerous scholars have called for an in-depth investigation of the processes and dynamics of conferences (Braithwaite, 2002; Braithwaite, Ahmed, and Braithwaite, 2006; Hayes and Daly, 2003). In order to do that, we need a framework to orient ourselves.
A useful way to think of the processes and dynamics of restorative justice is as a kind of ritual. The concept of justice as a ritual has always been a component of both scholarly writing and the public imagination (Foucault, 1977; Garland, 1993; Maruna, 2011). Rituals are important because of the sense of collective solidarity they engender. In sociology, this line of thinking originates with Durkheim, who demonstrated how, by bringing people together with a common cause, rituals are an important part of community building and help create a sense of togetherness. In his study of religion (Durkheim, 1912), Durkheim suggested that the reason people engage in such rituals is because of the resulting feeling of ‘collective effervescence’, or ecstatic energy, and a feeling of group membership. At the height of such a ritual, participants feel (p.29) swept away, transformed, and at one with the group. As Durkheim described,
It seems to him that he has become a new being…and because his companions feel transformed in the same way at the same moment, and express this feeling by their shouts, movements, and bearings, it is as if he was in reality transported into a special world entirely different from the one in which he ordinarily lives, a special world inhabited by exceptionally intense forces that invade and transform him. (Durkheim, 1912: 220.)
It is this search for collective effervescence that keeps us coming back to participate in rituals like religious ceremonies.
Karstedt (2006) argued that justice rituals are important because of their potential to transform one emotion into another, for example where negative feelings of anger, fear, or humiliation may be transformed through a positive ritual into feelings of solidarity and shared morality. Schechner (1981) suggested that rituals may go further than transforming emotions and can even transform the self. He distinguished between transportative and transformative rituals, specifically in performed rituals (be it theatre, or ‘rites of passage’). Transportative rituals can take us to a different place while we are performing in the ritual, but once it’s over we return to the everyday life we had been living. Transformative rituals, on the other hand, actually change who we are. Schechner’s idea of a performative ritual is apt—dramaturgical metaphors are rife in restorative justice theory and practice. The facilitator follows a ‘script’ to make sure all participants play their ‘role’. Powerful performers are referred to as ‘stars’, while failed conferences are ‘duds’. Restorative justice performances/rituals toe the line between transportative and transformative. As Karstedt pointed out, such rituals aim to transform negative emotions into positive ones. While the other participants/co-performers can be transported and swept up in the collective effervescence, there is the hope that the offender will be transformed. In this way then, the restorative justice ritual can engender more than a collective pleasant emotion of group membership—it can create moral beings.
Based on this, we can conclude that not only do rituals create and reaffirm a sense of group membership, they are actually the source of morality and public order. For example, Durkheim argued that public rituals in response to a crime or misbehaviour serve to bind a community together, both through the formal spectacle of punishment and the informal rituals of people getting together to gossip (p.30) about the offender and the punishment imposed. Through the enactment and re-enactment of justice rituals, we come to a shared moral standard of right and wrong. This has important implications for reintegrative shaming theory. As Braithwaite argued, reintegrative shaming works to stop reoffending not just because we want to avoid future instances of shame. Rather, it is through such shaming that we develop a ‘learned conscience’ and a new set of morals (Braithwaite, 1989: 37). The Durkheimian tradition in sociology would suggest that this happens because reintegrative shaming can be a community-building, transformative ritual.
Goffman (1967) extended this insight to show that solidarity- creating rituals do not only occur in formal rituals (such as religious ceremonies) but appear in all facets of everyday life. He shifted the focus of attention away from the individual, and what they are feeling, to the encounter as the unit of analysis, examining every interaction for its ritual elements. This way, we come to see how our sense of who we are and what we mean to other people is constantly negotiated through interaction rituals, be it a chat with an acquaintance whom we meet on the street, a paper we give at an academic conference, or an intimate conversation with a loved one.
Goffman is particularly adept at showing what happens when rituals fail. Interaction rituals can break down for a number of reasons: a participant can fail to respond appropriately, or may offend other participants, become distracted, misunderstand what is happening, or lose face in some other way. When this happens, they feel the opposite of solidarity—perhaps awkward, uninterested, or even repelled by the experience. However, neither Durkheim nor Goffman specified the conditions under which rituals are likely to succeed and fail. Partly in response to this, Collins (2004) has developed a theory of ‘interaction ritual chains’ which provides a set of principles for predicting when an interaction will work and make us feel good, or fail and make us feel bad. Over time, these repeated chains of ritualistic social interactions raise or lower feelings of long-term social bonding, motivation towards social goals, and commitment to group symbols and shared morality.
Collins conceived of social interaction as a series of rituals that build on each other to provide an individual with varying levels of positive or negative emotional energy. He identified four main ingredients for a successful ritual: group assembly, a barrier to outsiders, mutual focus, and a shared mood. (p.31)
A key feature of this model is whether a threshold or take-off point is reached, after which these elements feed back upon each other and lead the group to develop rhythmic coordination and synchronization in their conversation, bodily movements, and emotions. When participants become ‘caught up in the rhythm and mood of the talk’ (Collins, 2004: 48), the collective effervescence Durkheim predicted in formal religious ceremonies can occur in any social interaction. The outcomes of successful interaction rituals, therefore, are group solidarity and emotional energy, a development or evocation of symbols that represent the social relationship, and the emergence of group standards of morality. These outcomes will be explored below.
Collins’s theory lends itself well to understanding the process of restorative justice. A powerful justice ritual that brings about feelings of group membership, solidarity, and a shared morality can certainly be considered a successful reintegration ceremony (Braithwaite and Mugford, 1994; Maruna, 2011).
The language of ritual has long been used to describe restorative justice encounters. In an early treatise on restorative approaches to justice, Zehr (1990) uses Pfhol’s (1981) idea of a ‘ritual or reordering’ to describe restorative justice events. Such rituals can reset a wrong and bring an individual to the place he or she was before the commission of an offence. With a nod to Durkheim, Zehr noted that perhaps churches are the best places to organize such rituals (Zehr, 1990: 209). Retzinger and Scheff (1996) have also pointed out how restorative justice works as a powerful ritual to repair broken social bonds (as have Braithwaite, 2000; Karstedt, 2006; and Maruna, 2011).
Maruna (2011) has noted how contemporary criminal justice processes fail at providing successful reintegration rituals. Our (p.32) society has created elaborate spectacles of punishment, but nothing to reintroduce an offender back into society. He argued that such status-degradation ceremonies in criminal justice, without adequate status-elevation ceremonies on the other end, have negative consequences and that status-elevating ‘redemptive rituals’ are needed to symbolize to an offender that he is part of a moral community. Restorative justice programmes are examples of such a ritual. Indeed, Braithwaite and Mugford (1994) specified the conditions under which a restorative justice encounter can be a successful reintegration ritual as including the decoupling of the offence from the perpetrator, by ensuring the presence of supporters, empowering victims, and encouraging respectful interactions.
A successful interaction ritual has long-term and short-term outcomes. Short-term outcomes include positive feelings of group solidarity and a momentary rise in emotional energy. This solidarity creates symbols of group membership which remind participants of these positive feelings and, theoretically, extend the high emotional energy to future interaction rituals.
Solidarity, shared morality, and short-term emotional energy
A successful interaction ritual is marked by a number of ritual outcomes, including displays of solidarity, a sense of shared morality, and a feeling of emotional energy. Solidarity is externalized when people synchronize their body movements and take part in conversational turn-taking (Collins, 2004: 66). In high solidarity interactions conversation will flow more easily, there will be fewer embarrassing silences (see Goffman, 1956), and people will make eye contact and touch each other more often than in lower solidarity interactions. This contributes to a general feeling of group membership marked by a shared morality in which participants are in agreement on what is right and good, and gain positive feelings from this mutual agreement.
In a restorative justice conference one can observe a rise in solidarity between the offender, victim, and supporters as the conference progresses. This may be externalized by a lessening of gaps in conversation, increased eye contact between participants, a synchronization (p.33) of sounds such as group crying or laughing, and touching—hugs, handshakes, or pats on the shoulder.
Solidarity in a successful interaction ritual is accompanied by a momentary burst of emotional energy or ‘buzz’ that arises from the social interaction. In fact, Collins uses restorative justice as an example of how an interaction can be turned into an emotional energy-creating ritual (Collins, 2004: 111). While this emotional energy can be seen and measured during an interaction ritual, it is most powerful as a long-term emotion—this will be discussed further in Chapter Seven. In a restorative justice conference, these externalizations of emotional energy can be observed in the body posture and facial expressions of individuals. For example, offenders who sit hunched over, downcast, and disengaged at the beginning of the conference may be sitting upright and smiling by the end. Emotional energy states can also be evaluated through interviews about a person’s emotions and attitudes.
A successful interaction ritual creates strong collective symbols that act as a reminder of group membership and solidarity. These tide individuals over between interaction rituals. Without them, the consequences of the interaction may just be fleeting.
Can restorative justice conferences provide meaningful symbols? A useful way of thinking about this is by making the distinction offered by Retzinger and Scheff (1996) between material and symbolic reparation. Material reparations include an undertaking agreed upon by participants that repairs the harm—often through compensation, restitution, or community service. This negotiation happens towards the end of a conference, when participants conventionally formulate and sign an outcome agreement or a plan for action. This agreement may note if an apology was made and accepted but primarily lists instrumental steps the offender agrees to take. Copies are made for each participant to take home with them. These agreements, while focusing on material reparation, can potentially serve as a symbol of the interaction that has taken place, although, in reality, they often feel ‘tacked on’ to the end of the conference, well after the emotional peak or displays of solidarity.
In Retzinger and Scheff’s conception, symbolic reparation is the real heart of the conference and ‘it is the vital element that differentiates conferences from all other forms of crime control’ (Retzinger and Scheff, 1996: 317). (p.34) The core sequence of symbolic reparation involves an expression of shame or remorse by the offender, and some sign or gesture of forgiveness by the victim (or at least the hint that forgiveness may be possible somewhere down the line). This type of reparation is more effective at repairing broken bonds between offender and victim, and also comes to symbolize the potential for repairing bonds broken between an offender and their family, support system, and larger community. Retzinger and Scheff argued that material reparation is not likely to be considered effective, or meaningful, without symbolic reparation. Similarly, in Bennett (2008) the symbolic meaning of an ‘apology ritual’ in criminal justice is the foundation of his theory of punishment. He argued that we want to make an offender ‘act as she would were she genuinely sorry for her offence’ (Bennett, 2008: 146). Perhaps this exchange of remorse and forgiveness between offender and victim, even if only fleeting and lasting a few seconds, can act as a powerful symbol of the social bond.
This may be problematic, however, when apology is virtually built into the script. Once the participants have expressed how they have been affected by the crime, the facilitator will often turn to the offender and ask if there is anything they would like to say and they are practically compelled to offer an apology at this point. While there are many instances of spontaneous apology and expressions of forgiveness in conferences, relying on a formal apology at this point may not create an effective apology ritual, or be filled with the symbolic meaning described by Retzinger and Scheff. While the strength of these symbols is an empirical issue that can be measured, perhaps the memory of the emotional peak in the interaction is sufficient to ensure short- and long-term benefits.
In interaction rituals the group symbols and feelings of solidarity created lead to a short-term, transient emotion or buzz. Such instances of solidarity and emotional energy also have the potential to be translated into a long-term emotional state that, theoretically, the participants can add to their stock of symbols and emotional energy and take into future solidarity-creating interactions as part of an ‘interaction ritual chain’ (Collins, 2004).
Having experienced the emotional energy from a successful interaction ritual, individuals will need to participate in other (p.35) interaction rituals in order to maintain, or increase, their emotional energy. This is because emotional energy decays over time and the benefits of the initial interaction decrease. So, like a battery, the individual will need to engage in more solidarity-producing interactions in order to be ‘recharged’. Once the initial interaction ritual ends, the individual enters a kind of interactional market (Collins, 1993) where they will endeavour to reinvest their current stocks of emotional energy in future interaction rituals. The more they invest, the bigger the long-term payoff. In this process, people become emotional energy seekers, always moving towards the highest emotional energy payoffs they can find relative to their current resources.
A common criticism with shame and procedural justice theories of restorative justice is that they do not provide a realistic account of the long-term effects of a conference. The restorative justice conference is usually a single intervention. What good could one instance of reintegrative or acknowledged shame do in the life of an offender? How could it affect someone whose life may be characterized by stigmatizing interactions and where shame, anger, and violence are a part of everyday life? As with most theory and research on emotion in which the focus is on short-term or transient emotions, little attention has been paid to long-term emotional states (Collins, 1990). Long-term emotional energy, however, may theoretically be the most important issue in setting out a social theory of human behaviour. In restorative justice, long-term emotional energy may prove to be the key that keeps people from reoffending.
For an offender with a low stock of emotional energy prior to a restorative justice conference—especially if they are a repeat offender who has already experienced rituals of stigmatic shame in the courts, prisons, or probation and parole—they may desire emotional energy but lack the necessary resources to invest in positive interaction rituals. Alternatively, they may invest in an interaction ritual that provides some emotional energy, such as the rush incurred by committing a crime (Katz, 1988), but this will soon end when they are compelled to participate in a series of status-reducing interactions such as attending court or receiving a prison sentence.
Emotional energy does not have to be created from previous stocks of emotional energy, it can also be created ‘out of thin air’ by a particularly intense interaction ritual. According to Collins, interaction ritual is a mechanism of change: ‘as long as there are potential occasions for ritual mobilization, there is the possibility for sudden and abrupt periods of change’ (Collins, 2004: 43). In (p.36) the case of restorative justice, a powerful interaction may tip the emotional energy scales in the offender’s favour, and encourage them to seek out interactions that will top up their emotional energy.
While an actual restorative justice conference may not be repeated, the emotional energy gained from the single interaction may be the key to motivating a person to engage in further positive interactions that lead to a reduction in offending. Seen from this perspective, it is not reintegrative shame, acknowledged shame, or any other particular emotion at work that makes restorative justice successful. Instead, it may be that a taste of emotional energy leaves an individual hungry for more positive interactions, thus motivating them to engage in pro-social behaviour.
Power and Status Rituals
Power and status are further aspects of interaction rituals that influence each other and lead to stratified rituals. Broadly speaking, power rituals in an interaction determine who is an ‘order giver’ and who is an ‘order taker’ while status rituals show the extent to which an individual is part of a group. People who have power and status in an interaction are more likely to have more positive long-term benefits in the form of emotional energy.
In the restorative justice conference, power dynamics may influence the outcome so the aim in an ideal conference will be to try to circumvent this. As all participants are supposed to be on an equal footing—engaging in democratic deliberation—a conference that is stratified along power dimensions is likely to fail. Similarly, in conferences where people take on the roles of order-giver and order-taker the interaction may be dominated by one participant (either offender or victim). In such cases, there will be low emotional energy (or perhaps unacknowledged, or stigmatizing, shame) on the part of the order-taker.
During a conference there is always the potential for the status of the offender to change. Initially, the victims, as the wronged party, are most likely to have the moral upper hand. By the end of a conference, however, the offender’s status may be elevated as they come to be accepted by the victim and their supporters as a member of the social group. The victim’s acceptance of the offender as a ‘good’ person will raise their status and, consequently, the emotional energy. In an unsuccessful conference, the victim will not accept the offender as a person on the same level as them, further depleting emotional energy.
(p.37) To summarize, when examining restorative justice conferences as interaction rituals one must do so with an eye to the potential status and power imbalances. If we reconceptualize restorative justice conferences as a form of interaction ritual we can theorize how the process serves both restorative and deterrence functions. Through rhythmic coordination and the emotional entrainment that happens over time during the course of conversation, participants will become charged up with feelings of solidarity, group membership, and emotional energy by the conclusion of a conference. Ideally this will create some kind of collective symbol and positive feelings that will follow individuals into future interactions, helping them to engage in further emotional energy-building rituals.
For offenders, this kind of interaction may provide them with just enough emotional energy to motivate them to stop (or at least reduce) offending. All of these propositions can be empirically tested and in Chapters Four to Seven of this book I will analyse elements of this theory through a combination of qualitative and quantitative methods.
Are Good Rituals Driving the Success of Restorative Justice?
While shame and procedural justice theories do a good job of identifying specific emotions at work in restorative justice, they do not specify how emotions and behaviour can be transformed through such a process. Ritual theories, specifically Collins’s theory of interaction ritual chains (Collins, 2004), can fill this gap. The reason ‘emotional’ conferences are ‘good’ conferences is that the emotions are used to create a strong interaction ritual which results in feelings of solidarity and emotional energy. This is generally felt by all present, as seen when the facilitator returns from ‘good’ conferences on a high, eager to share the experiences with colleagues. As Chapter Five will show, facilitators employ a number of strategies to create the necessary ingredients for a successful interaction ritual and emotions truly are at the heart of their success. Expressions of strong emotions in the conference (be it anger, fear, or shame) often act as a turning point that allows the participants to develop a shared rhythm to their interaction. Facilitators are trained to be attuned to these as the careful way emotions are incited and expressed allows for a shared mood, rhythm, and balance of power that results in expressions of solidarity and shared morality.
(p.38) A number of questions arise from this discussion of emotions and ritual. What does an interaction ritual actually look like? How can we observe and analyse one? Can we develop an empirical model for interaction ritual in restorative justice that tests the conditions under which a restorative justice conference can lead to solidarity, feelings of group membership, and a symbolic representation of these emotions? Can this model predict the likelihood of future offending? In Chapter Four I explore the ritual dynamics of a single conference in-depth and in Chapter Five examine how facilitators prepare for a successful conference and use strategies to ensure that the ingredients are present for a successful ritual. I discuss what short-term outcomes, like solidarity, actually look and feel like and then present a statistical model of interaction ritual in restorative justice using quantitative data collected by systematic observations of conferences in Chapter Six. Finally, in Chapter Seven I will examine how successful interaction rituals are related to offending behaviour after a conference.
Methodological Approaches to Studying Ritual
The research uses a three-pronged approach to studying the micro dynamics of restorative justice conferences. This involves a combination of qualitative interviews; detailed observations of discourse, face and demeanour; and the quantitative analysis of systematically observed conferences. I focus on displays of emotions, emotional turning points, and the emergence of rhythm and solidarity between participants as the key to successful short-and long-term outcomes, demonstrating the impact of emotionally successful conferences on recidivism.
The qualitative data consists of in-depth interviews with police facilitators. These facilitators were employed to work at the Justice Research Consortium, which was conducting a trial of restorative justice for serious adult offenders being prosecuted for robbery or burglary in the Crown Courts of London (see Shapland et al, 2004). Upon entering a guilty plea, but before sentencing, the experimental group of offenders took part in a police-led, face-to-face restorative justice conference with their victim, which was also attended by supporters on both sides—usually close relatives or friends. (p.39) The control group received no treatment outside ordinary prosecution. Due to the serious nature of the offences, most offenders were on remand in custody awaiting sentence. This meant that approximately eighty-five per cent of the conferences were held in prison and the rest were held in police stations and community centres. Conferences lasted about two hours, at which point an outcome agreement was drafted by the facilitator, signed by all those present, and submitted before the judge, who had the discretion to take it into account while sentencing.
In order to gain a deeper understanding of how facilitators distinguish between conferences, I conducted in-depth interviews with facilitators about conferences they chose to be their ‘best’, ‘worst’, and ‘typical’ conferences. Together, we mapped out all of the events leading up to, and during, the conference, examining each situation in detail. All facilitators categorized their best conferences as being the most emotionally intense, and their worst as either emotionally flat or very intense for one party but not the other. The interview guide used to explore these conferences can be found in Appendix 1.
Eight police officers served as facilitators on the research team, with most facilitators having significant experience of running twelve or more conferences. I interviewed each of them on three separate occasions about their most successful, typical, and least successful cases, which resulted in a total of twenty-four situational analyses. This sample size is directly proportionate to the number of facilitators on the original research team, all of whom I interviewed. I systematically coded the interviews based on components of interaction ritual as described by Collins (2004), such as rhythm, emotional entrainment, and solidarity. Additionally, I reviewed all the case logs and observer reports of these conferences. This analysis maps out the dynamics of a successful restorative justice conference by comparing solidarity-creating interactions with less successful rituals. By making these comparisons, I aim to identify the micro level elements of such rituals to build a theory of how emotions and interactions work in restorative justice.
There are two ways of thinking about how to study emotions. One is based on the idea that emotions are an interior experience and a careful interviewer can bring them to the surface. The other way considers emotions to be an external thing, a Durkheimian social fact that can be observed and documented. This is the perspective utilized by contemporary sociologists of emotion such as (p.40) Jack Katz, Thomas Scheff and Suzanne Retzinger, who base their analysis not on direct interviews with participants, but on detailed examination of what people do in an interaction.
Facilitator accounts are the ideal source of data when examining the elements of interaction ritual. Although interviews with conference participants can shed light on their subjective experience of the conference, in the style of Goffman, I am committed to the study of encounters, with the group as the unit of analysis rather than the individual. To prioritize the group-focus with this data, it is appropriate to rely on the facilitators’ accounts of each case, as they are the ones who are trained to be attuned to group-level dynamics. Indeed, as Collins suggests (2004: 97), when studying interaction rituals one should begin with powerful symbols—in this case conferences that especially resonated with facilitators—and work backwards from there. Once we have identified an interaction that strikes us as particularly emblematic, we then,
Reconstruct as best as possible what IRs have surrounded that emblem. Who assembled, in what numbers, with what frequency or schedule? What emotions were expressed, what activities brought a focus of attention, what intensity of collective effervescence was generated? To what degree were individual participants charged with emotional energy; and what did it motivate them to do? What were the barriers to participation; who was divided by the ritual from whom? Who was thereby ranked over whom? (Collins, 2004: 97)
In the context of restorative justice conferences, the facilitator is ideally placed to reveal such elements. Their role requires them to keep track of who made eye contact with whom, who leaned forward to engage, who pulled away to disengage, who was disruptive, and who caused an imbalance. Such dynamics are keenly documented by the facilitator, who must respond to them in the process of guiding the conference along.
Facilitators also have the unique perspective of not being directly involved in the proceedings and can therefore provide accounts of what is happening at the group level. Conference participants may not be very good observers of what happens in an interaction; they are more focused on their own experience and making sense of what it all means, and so may not be attuned to the micro details of emotional expression, body posturing, and tone of others, and how these lead to turning points and a shared rhythm. Facilitators are not infallible, and the accounts they provide can never be fully objective (nor could those of an external observer); however, (p.41) they have the experience and training to report on the micro level dynamics of how the conference unfolds in space and time.
In addition to the interviews with all facilitators on the original research team, I reviewed all the police case logs and observer summaries of these particular conferences. As a member of the research team for two years, I participated in facilitator training with the Metropolitan Police; accompanied the facilitators on home visits as they met with victims and offenders; observed many of their conferences; and liaised with the many criminal justice agencies involved in getting this project off the ground. While my own impressions helped to shape the study, the data I present here are based on facilitators’ recollections of conferences and the detailed notes of the conference written up in narrative form by an observer from the Consortium’s research team.
By examining solidarity-creating situations in-depth, and comparing them to less successful interactions, one can begin to map the dynamics of a restorative justice conference. The amount of detail the facilitators recalled was remarkable and the strongest message I took away from my time with them was that organizing a conference is extremely difficult and time-consuming. It requires emotional sensitivity and the ability to strategize and problem solve. These facilitators put great effort into each conference and felt elated when it was a ‘success’ and very let down when it ‘failed’. Because they were so professionally and emotionally invested in the process, they remembered fine details about their conferences long after the event.
To supplement these situational analyses of conferences, I also interviewed six offenders who participated in a restorative justice conference. They are not a representative sample of all restorative justice offenders but were nominated by facilitators based on the criteria that their offences were more serious than most, and they were all sentenced to prison for at least six months. I met with these men between eighteen months and two years after their conference took place. Some had completed their prison sentence and had been released, some were still serving their sentence, and one had served his sentence but was back in prison on another offence. Ten offenders were nominated and I was able to interview six. Influenced by the life course perspective developed by Maruna (2001) and Laub and Sampson (2003), I explore how they talk about their conference long after it took place, and how they do or do not use it to shape a life course narrative. With a larger sample size, this could have (p.42) been a major study on its own. As it stands, the interviews provide useful insight into how some offenders interpret the role restorative justice played in their lives, and connects micro sociological theory to life course studies. This small sample cannot be used to generalize about the long-term impact of restorative justice, but can provide some insight into how offenders talk about their experience of it. These interviews also help evaluate the long-term effects of interaction ritual. As I will discuss, the ritual outcome of emotional energy is a long-term, ongoing process. People are filled with high or low levels of emotional energy based on their ongoing interactional contexts. The best way to assess this is by following participants over time to examine how any emotional energy created in a conference is transmitted through future interactions. In my interviews I attempt to trace their emotional energy states since their conference.
A final source of qualitative data was a video recording of a restorative justice conference held for a robbery in London. This case was originally in the Justice Research Consortium experiment with the Metropolitan Police, but was taken out of the study so that it could be filmed for training and educational purposes. It consisted of a conference between a victim, offender, and their supporters and was facilitated by one of the Justice Research Consortium police facilitators. It is a good example of the types of conferences run by the Metropolitan Police, and I use it to conduct a fine-grained analysis of the ritual elements of such a conference.
The qualitative data provide an in-depth and thick description of the restorative justice process. This can be supplemented by a quantitative analysis of the ritual elements. Can we statistically model the structure of a restorative justice interaction ritual? Can we use this model to examine whether these ritual elements are related to reoffending? This is one more piece of the puzzle, linking together micro and mezzo elements.
The data to explore and test this theory are drawn from the Reintegrative Shaming Experiments (‘RISE’), a restorative justice experiment conducted in Canberra, Australia, between 1995 and 2000. Upon arrest for an eligible offence, and admission of guilt at the police station, offenders were immediately asked to participate in the trial. If they consented, they were randomly assigned into one (p.43) of two groups: the control group (where they were prosecuted as normal) and an experimental group of people diverted from prosecution into a restorative justice conference. The restorative justice conference consisted of a police-led, face-to-face meeting between the offender and the victim, as well as a number of supporters from both the offender’s and victim’s side.1
This study focuses on two of the four experiments conducted in Canberra: juvenile violence and juvenile personal property.2 The Juvenile Violent Crime experiment involved violent offences (assaults and robberies) committed by people below 30 years old (although in practice most were under eighteen). The Juvenile Personal Property experiment consisted of property crimes that involved a personal victim. Offences in this experiment included burglaries, thefts, and criminal damage. A member of the research team observed each conference, and then completed a survey documenting their systematic observations. The experiment was developed in response to Braithwaite’s original 1989 theory (hence the name), so many of the observations focused on measuring reintegrative or stigmatizing shaming, or shame behaviour. Many of the items, however, are also particularly relevant to developing measures for ritual variables. This will be further discussed in Chapter Six.
The RISE experiment is widely known in the restorative justice community. The various datasets have been subject many times to analysis and re-analysis and continue to reveal new insights into restorative justice.3 In many ways it has set the standard for how restorative justice is researched and evaluated. Portions of the data have been used to compare conferencing versus court in terms of reoffending (Sherman, Strang, and Woods, 2000); victim perspectives (p.44) (Strang, 2002); procedural justice (Barnes, 1999; Tyler et al, 2007), and shame and shaming (Harris, 2001, 2003). I have tried to do something new with this landmark data. While these experiments were designed as randomized controlled trials (the gold standard of policy evaluation), data from the control groups is not utilized in this analysis. In the RISE experiment, the conference cases were diverted entirely from prosecution and the control cases were prosecuted as normal. This usually involved a brief hearing before the judge (often less than ten minutes), where the young offenders usually pleaded guilty while their lawyers negotiated a sentence for them. While fascinating interaction rituals in their own right, these court appearances are qualitatively different from a two hour face-to-face meeting between offender and victim. Court-conference comparisons are useful when testing the effectiveness of restorative justice as a policy; however, as Braithwaite (2002) suggestsed they are not helpful in developing a theory of restorative justice. Rather, I agree with Hayes and Daly (2003) that it is through the study of variations within conferences that an understanding of how they work can develop. Therefore, my analysis focuses on the explicit micro interactions of participants, grounded in interaction ritual theory, and uses the conference data to make this a true test of the variations within restorative justice conferences.4
Using Mixed Methods to Build Theory
Influenced by micro sociologists such as Collins (2004, 2008), Scheff (1990), and Katz (2001) I have been purposely eclectic in my data collection, drawing from a range of sources and conditions. Taken together, this is a multifaceted approach to examining interaction rituals in restorative justice that encompasses an exploration of how rituals unfold dynamically in space and time (Chapter Four); situational analyses of successful and failed rituals (Chapters Five and Six); an empirical test of interaction ritual theory in restorative justice using systematic observations of conferences (Chapter Six); and an investigation of the long-term effects of such rituals, both (p.45) through quantitative modelling of how ritual can impact offending and how offenders talk about the role restorative justice played in their lives (Chapter Seven). These methods, levels of analysis, and theoretical approaches are combined to develop a micro theory of restorative justice.
I kept this multifaceted approach in mind as I conducted my fieldwork and when I was developing my analytic procedures. The themes that I uncovered through my interviews acted as signposts—orientating me and providing focus when I conducted the analysis of the video. They also helped to develop and interpret the statistical constructs and their relationship to each other. In this way, the different types of data speak to each other, providing a full and round account of how emotions and rituals work in restorative justice conferences. Certainly there will be flaws, but this study is the first of its kind to systematically explore and statistically model micro level variations in the restorative justice process.
(2) The other two experiments consisted of adults arrested after failing random alcohol tests at road checkpoints, and juveniles arrested for shoplifting at corporate stores where their offence was detected by store personnel. The offence criteria of these conferences precluded a face-to-face meeting with a personal victim. In order to keep the structures of the conferences consistent for the purposes of this study, only experiments that involved direct contact between offender and victim were included. However, future research may examine interactional dynamics in conferences when there was no victim present.
(3) The data include the systematic observations of conferences and court, multiple waves of surveys of both offenders and victims, and official police records from the Australian Federal Police.
(4) In addition to these theoretical arguments, I conducted a preliminary analysis of the data that included both conferences and court cases. I found that having a conference was the largest predictor of whether the interaction ritual was a success. This suggests that there are distinct differences between conference and court, and it may not be useful to compare them as interaction rituals when developing a theory of restorative justice.