The convert,the remorseful and the rescued: Three different processes of desistance from crime

Abstract

One of the key issues in research on criminal desistance is the impossibility of stating with any degree of certainty that an offender's criminal career is in fact over. When no clear demarcation line can be established for the precise moment when criminal activity has ended, researchers instead distinguish between the cessation of criminal behaviour and the process of desistance. A second issue lies in the contradictions inherent in explanatory theories on desistance that focus either on agents or, conversely, on the structures that provoke and support the process of change. An integrative theoretical framework on criminal desistance, influenced by the work of Margaret Archer (1995, 2000, 2002) and showing the interplay between structures and agents, can be found elsewhere (F.-Dufour, Brassard, and Martel, forthcoming). The application of this framework to empirical data collected from 29 Canadian offenders serving conditional sentences reveals the existence of three distinct processes leading to desistance among those we metaphorically call the transformed, the remorseful and the rescued.

Keywords

Critical realism conditional sentence desistance from crime house arrest morphogenetic approach probation officer

Introduction

Every year, some 20,000 Canadians receive conditional sentences (Calverley, 2010). A conditional sentence allows individuals who would normally have been sent to prison for a period of fewer than 2 years to serve their sentences in the community when they are not deemed to be a public safety risk. In most cases, a sentence of this kind includes a period of 24-hour-a-day house arrest and a curfew or other comparable conditions designed to ensure ‘good behaviour’ on the part of the offender. For Roberts (2004), such sentences fall between a prison sentence and probation on a severity continuum.

Few studies have been carried out on the subject of conditional sentences (Roberts, 2004). And none at all seem to exist on desistance among offenders serving conditional sentences. However, one study does show that individuals serving conditional sentences and who benefit from the support of a social network have access to community services, successfully control criminal behaviour through psychosocial counselling and actively cooperate in the follow-up offered by their probation officers have recidivism rates of 4.8%, compared with 15.6% for the remainder of their cohort (F.-Dufour, Brassard and Guay, 2009). This suggests that certain relational and structural factors contribute to desistance among offenders serving conditional sentences. In the desistance literature, this conclusion seems to be shared by a number of researchers (Barry, 2006 Farrall, 2002; Laub & Samson, 2003), all of whom argue that these factors are what begin a person's desistance. At the same time, there are other authors (Giordano, Cernkovich, & Rudolph, 2002; LeBel, Burnett, Maruna, & Bushway, 2008; Maruna, 2001; Paternoster & Bushway, 2009; Shover, 1983) who present an opposing argument, to the effect that such factors come into play only when the agent concerned has already decided to desist from criminal activity. As LeBel and his colleagues (2008) point out, some might think this debate over ‘the chicken and the egg’ is pointless since everybody agrees that both are needed for desistance to happen, but in order to facilitate or even accelerate the process of desistance, it seems crucial to know what works for whom and when. Significant progress has been made in relation to this objective but it is still an important agenda for desistance research to ‘untangle the relationships among individual behaviour and structural […] context’ (Veysey, Martinez, & Christian, 2013, p. 258). It is hoped that this article can contribute to highlight the influence of the different actions and interactions of two levels of factors (agents and social structures) that led conditionally-sentenced offenders in Canada to desist from crime.

Archer's (1995, 2000, 2002, 2003) morphogenetic approach is rooted in a critical realism perspective, where the basic premise is that ‘reality’ exists, but that we have a limited capacity to fully grasp it. The second premise is the impossibility of predicting with certainty that B will follow A. While conditioning influences mean that B will frequently follow A, this is not an absolute. Critical realists also argue that agents are born into pre-existing structures (structures developed by other agents over the course of time) and that this initial involuntary position of agents in the social structure is not neutral. Certain agents are favoured by their position in the structure (thus have access to material resources), while others start out from unfavourable situations.

Typically, people born into a favourable social structure will resist changes to it, while those negatively affected by the structure want changes. Archer calls the conditioning influences common to all those born into given social structures vested interests (Archer, 1995). In her words: ‘Certainly, nothing determines that agents act to promote their vested interests, but costs are involved in not doing so. Since positions on social distribution concern scarcity, then there is a bonus to be lost with ceding a high position and penalties to be shed by not acceding to a low one’ (Archer, 1995, pp. 205–207). Hence agents' involuntary positions in the social structure facilitate or limit their access to certain projects, on the basis of their perception of opportunity costs.

Archer also argues that the interactions between an agent's involuntary position in the social structure, vested interests and opportunity costs will have an impact on that agent's choice of projects. For Archer, ‘the objective distribution of costs and benefits condition both interpretation and action’ (Archer 1995, p. 209). This means that opportunity costs are key determinants in agents' project choices. While we might think, for example, that the child of an offender is objectively less likely to become a judge than the child of a judge, structural causal powers will come into play only if this young person wants to become a judge. Without projects, there can be neither constraints nor enablers. Agents are always free to choose projects they perceive to have few enablers, but they must then expect to have to make a greater investment in time, materials, etc. to achieve their goals.

Personal identity is also a driving force in agents' choices around projects. For Archer (2002), personal identity is a reflection of how individuals see themselves and the world, a mirror of the concerns, attitudes and values most important to them and that make them the unique beings they are (Archer, 2000). Some examples of ‘concerns' could be to ‘change the world’, ‘get rich’ or ‘be in charge’. Personal identity is created by the sum of an individual's ultimate concerns, melded into a whole: ‘We make our lives, at least in part, by deliberating upon the structural context in which we find ourselves, often involuntarily. It is our deliberations which determine what we will make of the constraints and enablers which we confront, what opportunity costs we are prepared to pay, and whether we consider it worthwhile (…) joining others in the pursuit of change or (…) defence of the status quo’ (Archer, 2003, p. 52).

Lastly, any choice of projects is conditioned by an agent's social identities. Everyone has a variety of different social identities (parent, student, wage earner, partner, etc.), all of which provide resources we can draw on to achieve our goals. These identities determine the expectations, rules and standards we must conform to, if we do not wish to put a given social identity at risk. The tendency to conform (arriving on time, carrying out prescribed tasks, etc.) corresponds to the ‘vested interests' of a particular social identity. Moreover, no prescription exists as to ‘how much’ an agent should invest in one social identity or another. We know full well that some people ‘live for their work’ and others ‘work for their living’. It is the agents involved who determine the importance that each of their social identities will play in their lives, based on their ultimate concerns.

While agents choose social identities on the basis of their ultimate concerns, those concerns can likewise change the personal identities of individual agents by providing them with new information on themselves (e.g. ‘I like solving problems'). Furthermore, taking on a social identity alters the agent's position in the social structure, by altering the corresponding opportunity costs. In fact, someone who temporarily takes on the criminal social identity may find that it does not correspond to his ultimate concerns (personal identity). Even so, a prior arrest or arrests could reasonably be expected to constitute a barrier to changing that social identity.

In addition, agents connect with others through their social identities. Critical realists describe relations of this kind as being ‘institutionalized’, because they must also ‘respect standard rules and conventions' (Vandenberghe, 2007, p. 505). This also means that agents' institutional relations are tied to ‘vested interests', given that the agents are not entirely free to interpret what is expected of them. One example might be expectations that fathers (social identity) show affection to their children, ensure their safety and are involved in their education (institutional relations). In point of fact, the importance any father attaches to each of these expectations will vary depending on his own personality. This means that there exists a dynamic interdependence between personal and social identities. And since each person's ultimate concerns change over time, some social identities and institutional relations are relinquished if they no longer correspond to a particular personal identity, while others become increasingly important in that person's life.

In consonance with Margaret Archer's theoretical framework,¹ a number of concepts have been drawn on to describe the desistance process(es) adopted by offenders who abandon crime: their involuntary position within the social structure, ultimate concerns, enablers and constraints, opportunity costs changes occurring inside the dynamic interface between personal and social identities. We believe this constitutes another valuable aspect of this study as, over the 15 years since publication of the first volume of Archer's quadrilogy, the application of her theoretical framework has been minimal at best, due to its complexity (Horrocks, 2009; Kivinen & Piiroinen, 2006). In the present case, this framework has enabled us to differentiate among the ways three distinct processes of desistance from criminal behaviour are initiated and maintained.

Data collection and sampling strategy

The data were collected using a semi-structured interview format, as this enabled us to take into account the perceptions and experiences of respondents and also make some comparisons among the answers obtained (Savoie-Zajc, 2009). It also allowed us to explore aspects of respondents' answers that had not been anticipated (Charmaz, 2003). In accordance with the recommendations of Miles and Huberman (2007), data analysis was carried out in parallel with the data collection. As a result, participants were selected gradually, as different elements with implications for the nature of the emerging desistance process became increasingly evident. Initially, the sole criteria were that of being an adult male² who had received a conditional sentence of at least 1 year between 2001 and 2009, and had not subsequently reoffended. Then, by degrees, comparisons and contrasts were made between respondents, based on age, marital status, primary source of income, nature of the crime committed, criminal record, etc. Data collection continued until theoretical saturation occurred (Pires, 1997).

Sample

A total of 29 participants were recruited. Considering the possible age–crime curve effect, participants of different ages were recruited: 13 were aged from 18 to 35, 12 from 36 to 55 and 4 others were over 56. A high proportion of respondents (22) had received sentences of 18 months or more, 7 had been given sentences of between 12 and 16 months and 2 had been sentenced to more than 24 months.³ Respondents were also diversified in regard to the crime that led to their criminal sentence: 6 had committed violent crimes, 6 had committed crimes against property, 7 had committed crimes of a sexual nature, 7 had committed drug-related crimes and 2 had been found guilty of gangsterism.⁴ Within the entire group of respondents, 13 (44, 8%) had prior criminal records and the majority of them (75% of those with prior records) had served prison terms. Given the importance for this study of the length of time since the cessation of criminal behaviour, a comparison of respondents was also made on the basis of this characteristic. Four (4) respondents had not engaged in criminal behaviour for 8–9 years, 13 had not for 6–7 years, 7 had not reoffended in the 4–5 preceding years and 5 others had not in the previous 2–3 years. At the time of the interview, 10 respondents were living with a partner, 11 shared or had custody of children and the majority of respondents (21) were wage earners or received insurance benefits. These characteristics are illustrated in Table 1.

Table 1.

Sociodemographic and Sociojuridicial Characteristics of Participants.

								At the time of the interview
	Age at sentence	Crime that led to conditional sentence	Prior record	Committed crimes for more than 1 year	Length of sentence (months)	Length of cessation (years)	Estimated age at desistance	Living with partner	Sharing/ having custody of children	Having a job or insurance benefits
Transformed (N = 9)
John	33	Violence	Yes	Yes	24	2	35	No	No	No
Nathan	23	Drugs	Yes	Yes	12	4	27	No	Yes	Yes
Patrick	23	property	Yes	Yes	24	5	28	No	No	Yes
Francis	37	Gang	Yes	Yes	18	5	41	No	No	No
Walter	46	Gang	No	Yes	18	6	50	Yes	Yes	Yes
David	22	?	Yes	Yes	18	6	24	No	No	Yes
Xavier	35	Property	Yes	Yes	36	7	40	Yes	Yes	Yes
Ivan	53	Property	Yes	Yes	18	7	56	No	Yes	Yes
Ulysse	21	Drugs	Yes	Yes	27	8	25	Yes	No	Yes
Negative cases, Transformed (N = 2)
Kevin	32	Drugs	Yes	Yes	14	3	n/a	No	Yes	Yes
Yvon	56	Violence	No	Yes	18	6	n/a	Yes	No	No
Remorseful (N = 8)
Roger	42	Property	No	Yes	18	4	44	Yes	No	Yes
Victor	45	Drugs	No	Yes	12	5	47	Yes	Yes	Yes
Lucien	33	Violence	No	No	18	6	35	Yes	Yes	Yes
Mathias	25	Drugs	No	Yes	18	6	28	No	No	Yes
Oscar	32	Property	No	Yes	18	7	34	No	No	Yes
Jean-Claude	42	Sex	No	Yes	12	7	44	No	No	Yes
Denis	41	Violence	No	No	24	9	42	Yes	Yes	Yes
George	41	Drugs	Yes	Yes	24	9	47	Yes	Yes	Yes
Negative cases, Remorseful (N = 3)
Claude	33	Drugs	Yes	Yes	16	2	n/a	No	Yes	No
Edouard	35	Property	No	Yes	18	5	n/a	No	No	Yes
Sam	40	Sex	Yes	Yes	24	6	n/a	No	Yes	Yes
Rescued (N = 5)
Thomas	40	Sex	No	No	18	2	42	No	No	No
Albert	60	Sex	No	Yes	24	6	63	No	No	No
Benoit	55	Sex	No	Yes	24	7	59	No	No	Yes
Alphonse	38	Sex	No	Yes	15	7	40	No	No	Yes
Charles	70	Sex	No	Yes	24	8	70	Yes	No	No
Negative case Rescued (N = 2)
Baptiste	63	Violence	No	Yes	24	2	n/a	No	No	Yes
Hubert	34	Violence	Yes	Yes	15	4	n/a	No	No	No

Data analysis strategy and validity criteria

Data were analysed using the general inductive approach (Thomas, 2006) with QDA Miner software (version 4.0.8). Data coding was conducted in three phases. Open coding was carried out for the raw data represented by the material contained in the detailed written transcripts of the recorded interviews (induction). This was followed by coding based on ‘sensitizing categories' from the theoretical framework: the person's involuntary position in the social structure, opportunity costs, vested interests, etc. (deduction). In the third phase of the analysis, we compared emerging and sensitizing categories so as to establish broad categories enabling us to better understand desistance. To ensure the reliability of the analysis, one interview was reassessed for intercoder reliability using the ‘code overlap’ criterion; the result showed a 95.9% ratio. Once patterns and trends were identified in the data, Patton's (1990) negative case sampling technique was used to illustrate what happens when a desistance process does not have the expected results. These negative cases are therefore used to test the limits of each process and as such, contribute to the integrity of the analysis (Patton, 1999).

Processes of desistance from crime

Our study identified three distinct criminal desistance processes. To differentiate among them, we use three metaphors to illustrate the changes they produce: (1) the transformed become law-abiding citizens, (2) the remorseful return to the ‘right path’ and (3) the rescued are saved by good Samaritans.

The transformed

The transformed (n = 9) began illegal activities early in their lives, between the ages of 9 and 16. Most came from disadvantaged backgrounds and respondents said that they knew they would be the last to benefit from any ‘material resources' (territory and wealth) in their environment. Crime seemed a good way to obtain them:

Crime isn't the solution for everyone. Not an open door for all and sundry. Often just for guys with specific backgrounds … Definitely not an option for anyone at all. (David).

In this respect, the transformed resemble subjects described in the desistance literature (Barry, 2006; Farrall, 2002; Giordano et al., 2002; Laub & Sampson, 2003; Maruna, 2001; Shover, 1996) and also came from underprivileged backgrounds. But the transformed do not perceive their involuntary unfavourable position in the social structure as being the one determining factor that drove them to a ‘life of crime’. They also mention escaping from the informal social control⁵ of their families and being able to establish institutional relationships with procriminal peers:

There never were any rules in our house. Even when I was 9 or 10, I'd come home at 11 o’clock or midnight. When I was 13 or 14, I could disappear for a week and they [the parents] never asked me about it. (Xavier).

It would however be unwise to think that the young teenagers who choose to take on a criminal social identity are making an informed choice. As noted by Shover (1996) and Paternoster and Bushway (2009), they often carry ‘romanticized’ images of criminals in their heads and may even have dreamed of becoming ‘the greatest criminal’ of all time:

I’ve always had a fascination with crime. When I was young, I devoured stories about criminals like Jacques Mesrine and Richard Blass, and I'd say to myself: “Well, boys, you ain't seen nothin’ yet!” (Xavier).

Once they have taken on this social identity, it is given a place of pride in their personal identity, as can be seen in the words they use to refer to themselves (crooks, criminals, bad guys). And they invest considerable energy in becoming ‘good criminals' (Maruna, 2001; Shover, 1996). Nathan effectively illustrates how this social identity reinforces the personal identity (by reflecting the person's ultimate goal of becoming a good criminal):

You know, being a crook, people don't realize it, but it's tough work. Yep, you work hard. It's just not on what other people might call […] the right path.

Contrary to Shapland and Bottoms's (2011) respondents, these offenders do not share conformist values. Instead, they are more similar to Burnett's (2004) ‘hedonist’ and ‘earners' respondents: they have personal identities so imbued with their criminal social identity; they see themselves as being completely different from people in the non-criminal world:

I know I’m different. I don't have the law-abiding citizen mentality, but I can live with that [...]. Society is a totally unknown quantity for me. You know, for me, it's ordinary people who don't act normal. (Walter)

In assuming the criminal social identity, the transformed also inherits its vested interests or its standards and expectations for loyalty, respect and recognition. This leads the transformed to create a personal identity in opposition to the unfavourable social structure they were involuntarily born into. They dismiss norms, penalties and rules defined ‘outside’ the criminal world as being totally meaningless. As Uggen, Manza, and Behrens (2004) found, and Bracken, Deane, and Morrissette (2009) later also showed, offenders' sense of occupying a lower social status than other citizens is expressed as mistrust and bitterness towards the community. As a result, the transformed see no point in developing institutional ties with anyone ‘associated with law-abiding citizens' as the latter are seen as being ‘liars and hypocrites'. Xavier's comments capture this feeling about upright citizens among the transformed:

For me, those people act just like brainless sheep, following each other blindly. And they’re all the same. The only ones who don't act the same are the crooks. At least there's some respect in a crook's world.

For a relatively long period of time, the transformed, who may also have other social identities (most often that of father, husband or son) will maintain a fairly separate line between the world of ‘crooks' and that of ‘citizens'. As long as the boundary between those worlds remains unbreached, the transformed appear to be fairly satisfied with their lives. Their illegal activities bring them substantial income and they can also keep company with like-minded people. For a certain stretch of time, they are convinced they are living the ‘high life’, in that they are benefitting from the vested interests associated with their criminal social identity.

After a while, they come to realize that this identity has its disadvantages, the most serious one, according to our interviewees, being a kind of unconcern about what the future holds for them that permeates their lives. Here is David's description of the transition from a ‘groovy’ social identity to one that could end in death:

To begin with, it's just a game. You’re young; you don't really understand the full implications. Then comes a time when you start to think about the future. And you realize there are really only so many possible things this kind of life leads to… that you could easily die at any time, and miss out on a ton of stuff.

Recognizing the limits of their criminal social identity does not necessarily mean giving it up. While respondents admitted that they felt increasingly uncomfortable in that social identity, they could not imagine how to change it. David sums it up in these words:

I was going through what you might call my own ethical crisis. But I was just paralyzed. It was like I was stuck in a black hole.

Even though almost all the men in this group had served prison sentences over the course of their criminal careers, the descriptions they give of their experiences in prison are far from suggesting they felt they had ‘wasted’ time there. In fact, the transformed describe being left with relatively ‘pleasant’ memories of their time in prison. John is one such example:

I went to prison three times. And each time, I really got involved. I was the leader for a whole cell block of the F3 [a wing of the detention centre]. We were even paid salaries. My role was to organize ping pong tournaments. I had it easy there, I really did.

They learned from their time in prison to be less worried about prison terms and in fact to trivialize the experience. Xavier commented:

You get out and you tell yourself: [expletive deleted], it wasn't that bad. I’ve got more money than when I went in. I’ve learned stuff. I know more people. So thanks, guys!

Based on this kind of data – and contrary to what Cusson and Pinsonneault (1986) and Paternoster and Bushway (2009) suggest – we can deduce that the likelihood of going to prison or simply having become ‘fed up’ with a life of crime are not what lead people to desist from crime. All the men in the transformed group commented that they had become really ‘fed up’, with this way of life, but none of them could see how they could escape that trap. They may have come to recognize the limits of their criminal social identity, but they also quickly learned that their criminal records make it well-nigh impossible to get into even the lowest paying jobs, as Patrick notes:

I applied at X for a job as a forklift operator. They turned me down because I had a criminal record. The job paid 12 bucks an hour. That's when I understood… I wasn't asking for a heap, just a job at 12 bucks an hour, but I had a record.

So respondents considering adopting a new work-based social identity soon find that their record has negatively affected their opportunity costs. They can only reassume their criminal social identity, though this time not by choice, so they no longer feel the earlier balance with their personal identity. This imbalance between their criminal social identity and their nascent identity of ‘good citizens' is crippling and makes the transformed feel they have nothing else to lose. They describe this as a time when they were probably the most dangerous to themselves and to others. As David puts it, when you are not even slightly personally concerned about whether you live or die, you put others at risk:

I was in total despair. Nothing matters. If you die tomorrow, so what? Or in ten years' time? That's just ten more years. That's what's scary. When people start to think like that, things can get out of control.

The transformed tend to see the moment of their arrest (before they receive their conditional sentence) as an ‘opportunity’ to reclaim their lives. Their descriptions of their experience show they clearly believe that it was their choice to incriminate themselves. Without their cooperation the system would not have had enough evidence to convict them. In an odd reversal of fate, this experience was what allowed them to get their lives back.

I signed everything they put in front of me [the charges]. Eyes closed. Just told myself, come on, enough's enough! (Walter)

If ‘cooperating’ to assist in one's own arrest sounds surprising, it should be remembered that most of the transformed are not worried about doing jail time. They see it as an opportunity to ‘take a break’, make new contacts and plan future crimes. None of them enjoy the idea of losing their freedom, but some do see it as preferable to continuing their ‘criminal careers'. So it comes as no surprise that the majority of offenders sentenced to prison say they want to ‘go straight’ (Burnett, 2004). And although the transformed have no fears about going to prison, the intensity of their reaction to receiving conditional sentences is so strong because they realize the huge challenge it is going to be, as David explains:

What to do with my life? What's happening to me? What the heck am I gonna do? Zero ideas. Pure agony. No reason, but you panic. Not to minimize what started all this [the conditional sentence], but it was like I had struck rock bottom, and had nowhere to go.

They gradually come to realize that the conditional sentence will in itself give them the distance they need to think about choosing new avenues for action that will reflect their new ultimate concerns. And since they take pride in ‘doing their time’ properly, they see real changes happening when they respect their conditional sentence conditions, as Xavier's remarks show:

You know, I never cheated on my sentence. Always did my time. To begin with, you’re like fixated on the same old things [crimes to commit], then you start to think about stuff you should do and you tell yourself: “Tomorrow, I need to fix this or that, do something or other, remember to mow the yard, etc.”. Silly little things. The kind of silly things citizen people think about […]. When you’ve done your time, are you gonna go back to the kind of stuff you did before? Not damn likely!!! If you’ve got a brain in your head, you say, “No way, José!”

As this excerpt shows, the transformed who become ‘prisoners' in their own homes increasingly focus on other things than crime: becoming closer to their children, developing new skills, questioning the importance of money, or of power, in their lives. The conditions attached to their conditional sentences help them to think in other ways about their ultimate concerns (the basis of their personal identity) and to gradually distance themselves from their former criminal activities. With time, they tire of being obliged to stay at home. But the only conditions that allow offenders with conditional sentences to leave their homes during their period of house arrest are a return to school, finding a job or being in a therapy programme. As a result, all nine transformed took advantage of this chance to leave their homes.⁶

Initially, they make these choices simply to escape the ‘prison’ of their own home. Even so, it is not entirely by chance that they identify particular sectors of the job market to find work in or train for. The employment options the social structures lead them to choose are in environments that are easier for people with criminal records to get into. The jobs they obtain either directly or after training should not be seen simply as ‘good things happening to bad people’ (Laub, Nagin, & Sampson, 1998, p. 227). On the contrary, actions taken by the transformed clearly illustrate the extent to which they have reflected on these issues.

In the construction business, people are … let's just say they’re not exactly like ones in an office (amused laugh) … they’re more open, you see. Everybody knew [that he had a record]; it just wasn't a problem. It didn't change how … they acted toward me, you know. (Nathan)

Our results also indicate that, contrary to the findings of Giordano et al. (2002) and Paternoster and Bushway (2009), offenders actually take advantage of these opportunities before they are ready to change. It is in fact when they take on the social identities of students or employees that they find that their institutional relations have improved. In the example below, David explains that, during his conditional sentence, he went back to school and also began to work part-time in a laundromat. And that ‘in the course of things', ‘good people’ crossed his path. More precisely, he came to see that his institutional relationship with his mother was on the mend when she came to believe that he had really given up his life of crime. By degrees, David established new institutional relationships with prosocial adults. They provided guidance and helped him ‘close the doorway’ on his criminal life. And there came the day when he realized he was no longer an offender:

I was back in school, and working. And in the course of things, I grew closer to certain people. My mother started to believe in me again. I didn't want to spoil that! The doorway [to crime] was now closed. With good people around you, you make sure that doorway is locked tight. It all happens naturally and things just fall into place. And people change. OK, I’m not a saint (he laughs) but neither am I a crook now.

Although the transformed gradually assimilate into a ‘culture of good citizenship’, as a result of employability programmes, the jobs they take on or the psychosocial counselling they receive, we would be mistaken to assume that these opportunities lead them to immediately put their criminal social identity behind them. The transition from this latter identity to a more highly regarded one (father, student, employee) can be a long process. It took them an average of between 2 and 4 years to feel convinced they had really and truly ceased their criminal activities. Nathan describes the long, doubt-filled path that eventually led him to clearly identify the ‘right choice’ between his new life as a non-offender and his previous criminal social identity:

The first two or three years, I just kept my head low [avoided criminal activity] and I'd say to myself, I dunno… But now, it's different. I can clearly see the alternatives, and I know I want to stay outside and maintain a family life, not live as a criminal. But you gotta get a taste of this other life [non-criminal] before you can choose… really.

The transformed, negative cases

When the desistance process is not concluded, some people are not ‘transformed’ yet: they are still hesitant between their new life as a non-offender and their previous criminal social identity. We found two respondents who fit this profile and, at the time of the interview, it was difficult to evaluate if they will go back to crime or not. This ambivalence is probably what explains the ‘zigzag’ pattern of criminal careers (Laub & Sampson, 2003; Piquero, 2004):

I’ve always wanted to be a criminal. So…how0004865814523972? How0004865814523972? You know, I’ve tried for the last two years to stay clean. I still try. It's not so bad…but… [Interviewer: Do you think you might commit a crime again?]. Probably, yes. There is no point denying it. But you know, I would do only minor things; nothing serious…Well, I don't know…really… (Kevin)

As illustrated by Kevin's words, the process of transforming personal identity isn't easy and that's probably why the respondents corresponding to this profile take more time than others to reach the desistance point.

The remorseful

The attribute distinguishing the remorseful (n = 8) from the transformed is that the former started out with a favourable involuntary position in the social structure. They were better educated, and functioned as ‘good citizens' before beginning their criminal activities. Their criminal behaviour began when they were around 30 years old, making them ‘late-blooming offenders' (Krohn, Gibson, & Thornberry, 2013, p. 186). While the majority of the remorseful have committed crimes over a relatively long period of between 1 and 10 years, members of this group did not think of themselves as offenders:

I never felt like I was a criminal. Sure, I knew I was doing something illegal, I just didn't feel like a criminal. (George)

These respondents in fact use their prosocial social identity to hide their illegal activities. George, for one, used his social identity as a restaurant owner to carry out drug trafficking activities for a criminal organization. Normally, no one, apart from their accomplices and their victims, is aware that remorseful are engaged in criminal activities. And contrary to the transformed, they do not see their arrest as a moment of deliverance. In fact, it is quite the opposite, because their arrests provoke judgements and vilification from those around them.⁷ The very people who never wanted to wear the social label of ‘offenders' suddenly find themselves branded in exactly this way by their peers and their community. Roger, for example, had the impression that everyone else was stamping him with a label (Becker, 1963) he himself had refused to wear:

Well, yeah, but I’m not a criminal! And I had the impression the whole world was against me.

Roger's experience suggests we should perhaps re-examine certain aspects of Archer's morphogenetic approach, since some social identities are not always chosen on the basis of the ultimate concerns of personal identity. Archer argues that it is the ‘fracture’ of personal identity⁸ which explains this phenomenon. A fracture occurs when people lose contact with their ultimate concerns, become incapable of analysing the situation they are in, and so are unable to set any order of priority for projects they want to carry out. People living with fractured personal identities cannot ‘understand why they feel trapped, do not know how to orient their strategic actions and are incapable of measuring the consequences of their actions. It is extremely likely that it will only be when they are confronted by the actual consequences of their actions that they will be really become aware of the situation’ (Archer, 2003, p. 300). The predominant characteristic in people with fractured personal identities is their conviction that everything that happens to them is beyond their control: (1) they have an almost purely emotional reaction to what has happened to them, instead of rationally deciding on a strategy to get them out of the situation they are in; (2) they are no longer able to define what their ultimate concerns really are, so cannot decide what projects they actually want to carry out and (3) they display attitudes of resigned passivity (Archer, 2003, p. 305). Following the death of his mother, Roger experienced a fracturing of his personal identity and tried to explain it in terms of his criminal activities:

I wasn't really unhappy, but then I wasn't happy either. I just felt empty, totally empty, it's hard to explain … And I never really understood why I did that stuff [his crimes]. I'd always thought it was perhaps just to spice up my life. I could never quite nail it down, but I tried, man, did I try. I just can't explain (he laughs). In a way, there was just nothing important left in my life. And crime was like a sort of detour that presented itself in my life, and that I should never have taken, ya know… and I never really knew why I did it.

The fracture of personal identity is not a permanent state of mind. Typically, the remorseful are also characterized by the significant effort they make to ‘repair’ their personal identity. The first positive factor contributing to restoring their identity is the fact that they are open and receptive to the forms of assistance proposed by their probation officers:

He [probation officer] really gave me the help I needed. ‘Cause, when a wall in your house is rotting, ya gotta know what's causing it. Whatever crime someone has committed, ya gotta know why. And I don't think that person can find out on their own. They need assistance, outside assistance. That really helped me, ya know. It was tough going, but looking back on it [probation monitoring], I’m glad it happened. (Jean-Claude)

The second positive factor that contributed to restoring their identity is the fact that the remorseful opt to undergo psychosocial therapy, a choice they usually make while their pre-sentence report is being prepared. This could explain why over half of them (5) were required, as a mandatory condition of their conditional sentence, to receive psychosocial treatment. Every single respondent saw this as a positive experience. Lucien described himself as a ‘sponge’, because of the extent to which he committed to his therapy, precisely because he wanted to restore his identity (or, as he put it, to ‘rebuild his fundamental self’)⁹:

When I got my conditional sentence… it was so obviously linked to my recovery…that made the conditional sentence easier to take…and helped my recovery, too… So here I am today. I…let's say…I rebuilt my fundamental self. It was just…awesome, that therapy.

As this excerpt shows, the remorseful, unlike the transformed, do not change their personal identity, they rediscover it. Similarly, if they change their social identities, it is in qualitative terms. They might describe themselves as being more committed to their intimate partner or their children. If they have to leave their jobs (because of their crime), they will often look for new positions similar to the ones they had before their arrest. The remorseful become the people they were before their crimes, without the fracture of their personal identity. Denis comments that they do not change; they just now know what they want to do with their lives:

Comes a moment when you’re forced to think about things. Look at your life. What you might call….getting back on track…putting the focus somewhere else… I wouldn't say I’m a changed person but my focus is now my family, and my kids.

By rediscovering their personal identity, the remorseful are able to redefine what have become new ultimate concerns. Roger, who cannot even understand why he got into crime, has regained his balance. All he wants is a quiet life and he describes his ultimate concerns in these terms:

I want some peace and calm in my life. To live quietly and enjoy life, before I die. […] I just want a place where I can live, eat my meals, and go about life quietly. That's really it, I need peace and quiet. I’m doing OK. I’m working, like my job and I do what I have to do…When I can't work anymore, I'd like to be a shoe-shine man, a couple of hours or so a day. Ya meet different people, ya chat with them…not to get rich, just to enjoy life.

References to remorseful offenders are few and far between in desistance literature. Most studies on the subject focus on individuals who had previously ‘engaged in persistent patterns of criminal offending’ (Maruna, 2001, p. 26). Accordingly, they usually have histories of multiple arrests. Yet if, as Farrall (2004) shows, only 4% of crimes result in penal sanctions, then large numbers of people commit crimes without being arrested. This can be seen as a strong argument to include a majority of respondents who had been committing crimes for more than a year at the time of the sentence (some were engaged in criminal activities for more than a decade) in our analyses, since they embark upon a fairly different process of desistance. In fact, the identification of this previously uncovered process could be viewed as one of the most significant contributions of our research.

The remorseful, negative cases

The remorseful desistance process is characterized by the ‘restoration’ of respondent's personal identity mostly through psychosocial therapy. It was possible to locate three interviewees who share all other characteristics of the remorseful but did not go through therapy and are still struggling with their ‘fractured identity’.

I use to have a good life you know. But now…I came to this interview because I thought, maybe you could help me? I need to see a psychologist. I need to be put back on the map. I am always on my guard. I don't know if I will do something wrong again. That's why I don't do anything anymore. It's all empty around me. So… how am I supposed to rehabilitate myself? I just wish I had my life back. (Sam)

At the time of the interview, it was difficult to foresee if those individuals would be able to repair their fractured identities on their own and keep away from crime.

The rescued

Five respondents fall into this group. Like the remorseful group, they also began criminal activity as adults, the difference being their involuntary unfavourable position in the social structure. All of them are from economically underprivileged backgrounds and, in addition, have poor institutional relationships with family members, as Albert's description shows:

I did have a father and a mother, but we could never talk. There I was in a big family, people all round, but… I was all alone.

The rescued group had limited opportunities for positive social roles in their youth and as adults either had jobs involving little peer contact (working as a dishwasher, custodian, etc.) or were on welfare. Most never had an intimate partner or even close friends. Benoit's case illustrates how an absence of institutional relationships deprives the person of social identities (partner, friend, etc.):

I had no social life, no friends, no gang of pals. No close relationships, no sex life either.

In short, while the main element characterizing the remorseful group is the fracture of their personal identity, the rescued group is marked by a relative absence of social identities. Archer (2000) finds that this situation exists because the involuntary position of agents in the social structure determines the range of social identities they have access to. The more unfavourable their position, the more limited the range is. As a result, it is harder for people in unfavourable positions in the social structure to identify social identities they see as being in harmony with their personal identity, and this means they have a greater likelihood of abandoning the less gratifying identities they may have chosen. Moreover, the opportunity cost of reorienting their lives increases every time they give up on those social identities , so there comes a time when the people concerned find themselves unable to take on new social identities and simply float along, adrift,¹⁰ because the ‘absence of personal identity and the accumulation of circumstances make it harder to form one’ (Archer, 2000, p. 247). It is the beginning of the downward spiral.

The people who are ‘adrift’ find their lack of social identities absolutely intolerable; most respondents in the group said they had entertained ideas of suicide:

You say to yourself, “What can I do? What else can I find? Can I even get a job? Or have a normal life? Is my life worth anything?” And the answer was no, there was nothing. So I decided I would just die. The only way I could find some relief was to die. (Thomas)

This probably explains why the majority of respondents in this group took advantage of their arrests to express a need for help. Benoit's example is clear:

Right up front, I said to my lawyer, “Do you know anyone who can help me? I realize it's not just by chance that I’m here. I need help.”

Since most respondents in this group were sex offenders, it was mandatory for them to undertake psychosocial treatment. And for all of them, the psychosocial follow-up required by the court became the main (often the sole) instrument through which they became able to desist from crime. In the absence of facilitating institutional relations or satisfying social identities, they committed themselves totally to the therapy they were participating in and, in the process, developed extremely strong ties with the people working with them:

They really helped me. I just loved those people. And to top it off, they seemed to like me […] Yep; I got incredibly attached to those therapy people. (Thomas)

For these respondents, not only did the therapeutic process constitute an opportunity to establish institutional relations that had been lacking in their lives but they were also ‘rescued’, in that the psychosocial support they received also made it possible for them to identify at least one satisfying social identity they could assume. Albert, for example, was advised to consider looking at the senior citizens' movement and has remained actively involved with them ever since:

I’m involved in a senior citizens' group. And they organize stuff. I started out slowly, but now there's more all the time. I’m even asked to go to other activities. I feel appreciated – and you make friends as you go. It's fun, and there are so many activities. This has all helped me so much, so much.

Finding satisfying social identities for themselves allows them to discover their own importance and worth and identify what they consider as most important in their lives. And in identifying the priorities that create their new personal identity, they find there is no room left for crime. In Albert's words, ‘There's just so much to do now, I don't even have time to think of that’.

However, even when they find satisfying social identities, the rescued group remains on shaky ground at several levels, including their economic status. The jobs they manage to find are usually at minimum wage. In fact, most members of the group receive a form of income support known in Canada as ‘social assistance’, a situation that places significant limits on the projects they would like to engage in. As Alphonse explains:

I would have liked to go back to school, but I wasn't eligible under the provincial loans and bursaries system. I was a social assistance recipient… How could anyone on $200 to $250 a month pay rent, school fees and books, plus buy food to survive?

The rescued group also remains on shaky ground in terms of institutional relations. This could be due to the public reaction to this category of offenders. Although their new social identities bring them into contact with a variety of other people, they are not necessarily in a position to develop close relationships with them. Alphonse does not have any friends he can confide in:

Sure, I see people when we go bowling or play chess; we even do things together from time to time. [Interviewer: Do you have a friend you can confide in?] Well, no-one really. I confide in myself. I deal with my problems all by myself.

The rescued group is therefore left alone to manage its process of desistance. Even so, they do not seem to see this as a problem. Albert describes how, during his treatment, he had the impression that he was mapping out the path to desistance he would then be able to take. After that, he simply had to follow the path:

Once the path has been prepared, it's all a lot easier. You clear the land, cut down the trees, dig up the stumps – and take advantage of all the help you can get. Suddenly your dirt road has become a highway and you’re ready to roll. You’re drifting along on a cloud. As AA says: “take it one day at a time”. One day at a time. Time flies past, so fast. When you always look on the bright side of life, things just fall into place.

Even if the sampling process to track possible ‘rescued- non sex offenders' was extended to 9 months, none were found. A possible explanation lies in the nature of the conditional sentences. Only sex offenders go through a 2-year mandatory community-based therapy. As noted by other researchers (Kruttschnitt, Uggen, & Shelton, 2000) this therapy plays a pivotal role in their desistance process. In fact, one might think that people sharing ‘rescued’ characteristics (begin criminal activity as adults; come from economically underprivileged background; have poor institutional relation with family members and have limited opportunities for positive social roles in their lives) would also benefit from a therapeutic help. In fact, this idea is supported by the two negative cases found for this process.

The rescued, negative cases

The ‘negative cases' of the rescued process are characterized by the fact that they are ‘adrift’ (Archer, 2000). They are in a position that makes it impossible for them to know what tomorrow is made of.

I have my “dark side of the moon”. I pray every day because, you know, I have been incarcerated a lot in the last years and I never know where I’ll end up. I never know where I am going. Sometimes I live in the street. I never know what could happen. I just never know. (Hubert)

In short, these respondents are still stuck in their rotating downward spiral and no one knows where or when it will stop. Therefore, they cannot be considered desisters yet.

Conclusion

The main contribution of this article, besides its original theoretical approach, is to highlight the process of desistance of three subgroups of offenders that are not typically studied in desistance literature. First, our ‘transformed’ group started offending as juveniles, but their mean estimated age at desistance is 36 years. In regard to Laub and Sampson (2003, p. 104) trajectory data, they relate more closely to the ‘moderate rate chronic’ offenders considered as the most entrenched in their pattern of prior criminality, which, in turn, would explain the very criminal set of attitudes found in this group prior to desistance. Since they represent roughly 20% of offenders, only a minority of studies on desistance has focused on these respondents. Second, our unusual focus on respondents who did not have prior criminal records let us uncover the desistance processes of two other subgroups of offenders, namely the ‘remorseful’ and the ‘rescued’, who, for the majority (10 of 13), had been committing crimes for more than 1 year before their arrest and therefore could truly be regarded as ‘real desisters'.

In regard to Margaret Archer's morphogenic approach, the data collected for this study seem to indicate that processes of desistance are modulated by the involuntary position of offenders within the social structure, their ultimate concerns, the enablers and constraints they face in the structure, the opportunity costs they are willing to invest to change and of course, the complex interplay between their initial personal identity and social identities. Our analysis based on these concepts, suggest that the ‘transformed’ process starts by taking an opportunity in the structure to ‘escape’ house arrest. They take the ‘less costly’ opportunity, since they choose an environment less stigmatizing for (ex)offenders. By developing the basic skills they need for job entry and retention, they also develop or revitalize a range of social identities for themselves before they can feel confident enough to ‘join the ranks of law-abiding citizens'. In the long run, these new social identities can help to mitigate the anger and frustration created by their involuntary unfavourable position in the social structure. Lastly, they need to redefine their own particular ultimate concerns with the express goal of abandoning their former criminal social identity. Data on the remorseful show that they need to feel they have the support of their psychosocial therapists and probation officers to be able to repair their ‘fractured identity’ in order to recreate the social identities they had ‘before’ their arrest. These situations also call for significant support at the institutional relations level if offenders are to successfully re-establish those social identities. In the case of the rescued, the data show that they also need to receive support from their therapists and their therapy group, to help them pinpoint at least one social identity they would like to assume. Respondents themselves stressed the importance of finding at least one opportunity to act on this social identity within the social structure, if they are going to confirm their own personal identity, hence putting an end to being ‘adrift’.

Of course, these results should be interpreted with some caution. Even if theoretical saturation (Guest, Bunce, & Johnson, 2006) were obtained for every process identified and negative cases were analysed to reinforce the analysis, the small sample size in every group prevents generalization. In addition, ‘many researchers view retrospective reports with scepticism’ (Hardt & Rutter, 2004, p. 260), mainly because of the vagaries of memory. Despite this pitfall, research comparing the validity of retrospective recall when compared with other sources of information (official reports, parent official reports, sibling, etc.) consider it ‘sufficiently valid (in spite of substantial measurement error)’ (Hardt & Rutter, 2004, p. 270) especially for between-group case-control comparison. Nevertheless, since the average time from cessation of offending to interviews is about 6 years, we recommend that future research improve upon this study by using a longitudinal method.

Despite these caveats, the findings provide some support for the importance of structural opportunities in all process of desistance identified through the morphogenic approach. It seems that, in large part, the decisions that began the process of desistance were instrumental in nature (to escape from the constraints of house arrest, out of ‘consideration’ for a partner or parents, or to rebuild a fractured identity). It is only when newly discovered or rediscovered social identities lead to a transformation of (ex)offenders ultimate concerns (to lead a quiet life, have a good family life, find fulfilment in work) that respondents come to see the idea of desisting from criminal activity as a realistic goal and one they can achieve.¹¹ Clearly, once they are able to benefit from the vested interests associated with their new social identities, and as the related opportunity costs drop, they are in a better position to abandon their previous criminal social identity.

However they are described, as ‘opportunities', ‘hooks for change’ (Giordano et al., 2002), or ‘social capital’ (Farrall, 2002), the data collected support previous studies that showed that there can be no desistance from crime without the contribution of the social structure. This prior influence of structural opportunities has a particular relevance in the real world. We suggest that interventions should be adapted to each process of desistance. For instance, the transformed seem reluctant to individual counselling at the beginning of their process since they still ‘reject’ citizen values and way of life. Work-related interventions could be the best way to incrementally foresee their process of desistance. The opposite is true for the remorseful and the rescued since they express great need for counselling at the beginning of their process. It is only when they are able to repair the fractured identity or to fix their personal identity that they seem to embark on their process of desistance. We hope these conclusions will enrich the debate on ‘the chicken or the egg’ of desistance and will encourage more robust analysis to further disentangle the role of structures and agents in the process(es) of desistance.

Footnotes

Acknowledgement

Isabelle F.-Dufour would like to thank Dr Stephen Farrall, Sheffield University, UK for his precious feedbacks on this manuscript.

Funding

This work was supported by Vanier Canada Graduate Scholarship.

Notes

References

Archer

M. S.

(1995) Realist social theory: The morphogenetic approach, Cambridge, England: Cambridge University Press.

Archer

M. S.

(2000) Being human: The problem of agency, Cambridge, England: Cambridge University Press.

Archer

M. S.

(2002) Realism and the problem of agency. Alethia 5(1): 11–20.

Archer

M. S

. Structure, agency and the internal conversation. 2003Cambridge: Cambridge University Press.

Barry

(2006) Youth offending in transition: The search of social recognition, Abington, MA: Routledge.

Becker

H. S.

(1963) Outsiders: Studies in the sociology of deviance, New York, NY: Free Press.

Bracken

D. C.

Deane

Morrissette

(2009) Desistance and social marginalization: The case of Canadian Aboriginal offenders. Theoretical Criminology 13(1): 61–78.

Burnett

(2004) To reoffend or not to reoffend? The ambivalence of convicted property offenders. In: Maruna

Immarigeon

(eds) After crime and punishment: Pathways to offender reintegration, Cullompton, England: Willan, pp. 152–180.

Burnett, R., & Maruna, S. (2004). So ‘prison works', does it? The criminal careers of 130 men release from prison under Home Secretary Micheal Howard. Howard Journal of Criminal Justice, 43, 390–404.

10.

Calverley

(2010) Les services correctionnels pour adultes au Canada, 2008–2009 [Adult correctional services in Canada, 2008–2009]. Juristat Vol. 30, 32p.. (3, publication 85-002-x au catalogue de Statistique Canada).

11.

Charmaz

(2003) Qualitative interviewing and grounded theory analysis. In: Holstein

J. A.

Gubrium

J. F.

(eds) Inside interviewing: New lenses, new concerns, Thousand Oaks, CA: Sage, pp. 311–330.

12.

Cusson

Pinsonneault

(1986) The decision to give up crime. In: Cornish

D. B.

Clarke

R. V.

(eds) The reasoning criminal: Rational choice perspectives on offending, New York, NY: Springer-Verlag, pp. 72–82.

13.

Farrall

(2002) Rethinking what works with offenders: Probation, social context and desistance from crime, Cullompton, England: Willan.

14.

Farrall

(2004) Social capital and offender reintegration: Making probation desistance focused. In: Maruna

Immarigeon

(eds) After crime and punishment: Pathways to offender reintegration, Cullompton, England: Willan, pp. 57–82.

15.

F.-Dufour, I., Brassard, R., & Guay, J.-P. (2009). Le sursis, la récidive et la réinsertion sociale: un équilibre précaire [Conditional sentence, recidivism and social reintegration: A precarious equilibrium]. Canadian Journal of Criminology and Criminal Justice, 51(3), 303–327.

16.

F.-Dufour, I., Brassard, R., & Martel, J. (2013). An integrative approach to apprehend desistance. International Journal of Offender Therapy and Comparative Criminololgy. Retrieved from doi:10.11777/0306624X13509781.

17.

Giordano

P. C.

Cernkovich

S. A.

Rudolph

J. L.

(2002) Gender, crime and desistance: Toward a theory of cognitive transformation. American Journal of Sociology 107(4): 990–1064.

18.

Guest

Bunce

Johnson

(2006) How many interviews are enough? An experiment with data saturation and variability. Field Methods 18: 59–82.

19.

Hardt

Rutter

(2004) Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology and Psychiatry 45(2): 260–273.

20.

Horrocks

(2009) Applying the morphogenetic approach: Outcomes and issues from a case study of information systems development and organisational change in British local government. Journal of Critical Realism 8(1): 35–62.

21.

Hunter

A. J.

(1985) Private, parochial and public social orders: The problem of crime and incivility in urban communities. In: Suttles

G. D.

Zald

M. N.

(eds) The challenge of social control: Citizenship and institution building in modern society, Norwood, NJ: Aldex.

22.

Kivinen

Piiroinen

(2006) On the limits of a realist conception of knowledge: A pragmatist critique of Archerian realism. The Sociological Review 54(2): 224–241.

23.

Krohn

M. D.

Gibson

C. L.

Thornberry

T. P.

(2013) Under the protective bud the bloom awaits: A review of theory and research on adult-onset and late-blooming offenders. In: Gibson

C. L.

Krohn

M. D.

(eds) Handbook of life-course criminology: Emerging trends and directions for the future research, New York, NY: Springer Science+Business Media, pp. 183–200.

24.

Kruttschnitt

Uggen

Shelton

(2000) Predictors of desistance among sex offenders: The interaction of formal and informal social controls. Justice Quarterly 17: 61–87.

25.

Landreville

Lehalle

Charest

(2004) L'emprisonnement avec sursis au Québec: Impact de l'arrêt Proulx et du nouveau cadre de gestion [Conditional sentence in the Province of Quebec: Impact of the Proulx judgment and the new penal regime], Montréal, Canada: Centre de Criminologie Comparée.

26.

Laub

J. H.

Nagin

D. S.

Sampson

R. J.

(1998) Trajectories of change in criminal offending: Good marriages and the desistance process. American Sociological Review 63: 225–238.

27.

Laub

J. H.

Sampson

R. J.

(2003) Shared beginnings, divergent lives. Delinquent boys to age 70, Harvard: Harvard University Press.

28.

LeBel

T. P.

Burnett

Maruna

Bushway

(2008) The “chicken and egg” of subjective and social factors in desistance from crime. European Journal of Criminology 5(2): 131–159.

29.

Maruna

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

30.

Miles

M. B.

Huberman

M. A.

(2007) Qualitative data analysis: An expanded sourcebook, 2nd ed. Thousand Oaks, CA: Sage.

31.

Mulvey

E. P.

Steinberg

Fagan

Cauffman

Piquero

A. R.

Chassin

(2004) Theory and research on desistance from antisocial activity among serious adolescent offenders. Youth Violence and Juvenile Justice 2(3): 213–236.

32.

Paternoster

Bushway

S. D.

(2009) Desistance and the “feared self”: Toward an identity theory of criminal desistance. The Journal of Criminal Law and Criminology 99(4): 1103–1156.

33.

Patton

M. Q.

(1990) Qualitative evaluation and research methods, 2nd ed. Newbury Park, CA: Sage.

34.

Patton

M. Q.

(1999) Enhancing the quality and credibility of qualitative analysis. HSRL Health Services Research 34(5): 1189–1208.

35.

Piquero

A. R.

(2004) Somewhere between persistence and desistance: The intermittency of criminal careers. In: Maruna

Immarigeon

(eds) After crime and punishment: Pathways to offender reintegration, Cullompton, England: Willan, pp. 102–125.

36.

Pires

A. P.

(1997) Échantillonnage et recherche qualitative: Essai théorique et méthodologique [Sampling in qualitative research: A theoretical and methodological essay]. In: Poupart

(ed) La recherche qualitative: Enjeux épistémologiques et méthodologiques [Qualitative research: Epistemological and methodological challenges]., Montréal, Canada: Gaëtan Morin Éditeur, pp. 113–169.

37.

Roberts

J. V.

(2004) The virtual prison. Community custody and the evolution of imprisonment, Cambridge, England: Cambridge University Press.

38.

Savoie-Zajc

(2009) L'entrevue semi-dirigée [The semi-structured interview]. In: Gauthier

(ed) Recherche sociale de la problématique à la collecte de données, Québec, Canada: Presse de l'Université du Québec, pp. 337–360.

39.

Shapland

Bottoms

(2011) Reflections on social values, offending and desistance among young adult recidivists. Punishment & Society 13(3): 256–282.

40.

Shover

(1983) The later stages of ordinary property offender careers. Social Problems 31(2): 208–218.

41.

Shover

(1996) Great pretenders: Pursuits and careers of persistent thieves, Boulder, CO: Westview Press.

42.

Thomas

D. R.

(2006) A general inductive approach for analyzing qualitative evaluation data. American Journal of Evaluation 27(2): 237–246.

43.

Uggen

Manza

Behrens

(2004) “Less than average citizen”: Stigma, role transition and civic reintegration of convicted felons. In: Maruna

Immarigeon

(eds) After crime and punishment: Pathways to offender reintegration, Cullompton, England: Willan, pp. 261–293.

44.

Vandenberghe

(2007) Une ontologie réaliste pour la sociologie: Système, morphogénèse et collectifs [A realist ontology for sociology: System, morphogenesis and collectivities]. Théorie et Méthodes 46(3): 487–542.

45.

Veysey, B. M., Martinez, D. J., & Christian, J. (2013). “Getting out”: A summary of qualitative research on desistance across the life course. In C. L. Gibons, & M. D. Krohn (Eds.), Handbook of Life-Course criminology: Emerging Trends and directions for future research (pp. 233–260). New York, NY: Springer Science+Buisness Media.

46.

Worrall

(1998) Real punishment for real criminals? Community sentences and the gendering of punishment. British Society of Criminology. . Retrieved from http://www.britsoccrim.org/volume1/009.pdf.