Hook for change or shaky peg? Imprisonment, narratives and desistance

Given the twin challenges of the fiscal costs of high levels of imprisonment and of high reconviction rates post-release, the question of whether imprisonment can (and does) contribute to processes of desistance is a crucial one. This article discusses the repeated, and controversial, research finding that some prisoners see themselves as transformed by imprisonment. It argues, with reference to the narratives of Scottish prisoners, that this is not owing to intentional rehabilitative input, but driven by the need to craft a positive future. The article explores under what circumstances a transformation narrative is most likely to emerge and whether such narratives have the power to make desistance more likely.


Introduction
In the context of the on-going fiscal crisis and the related squeeze on public expenditure, the significant economic costs of imprisonment have driven politicians with diverse ideologies to discover or rediscover the merits of penal reductionism. For example, in the USA, some of the unlikeliest jurisdictions (for example, Texas and Georgia) have implemented decarceration strategies under the influence of a 'Right on Crime' movement led by conservative republicans offended by the 'big state' wastefulness of penal expenditure (Dagan and Teles, 2014). Among the UK jurisdictions, in Scotland and Northern Ireland, the current Ministers of Justice (Kenny MacAskill and David Ford), coming from more liberal political traditions and inspired by more than fiscal imperatives, have accepted the case for reducing reliance on imprisonment and enhancing the role of community sanctions. Importantly, both ministers point to high post-release reconviction rates as evidence of imprisonment's failure to support desistance (see, for example, Scottish Prisons Commission, 2008).
Despite very similar arguments surfacing during Ken Clarke's tenure at the UK Ministry of Justice, the current UK Government's Minister for Justice, Chris Grayling, has taken a different tack. In line with his vision of 'transforming rehabilitation' through market disciplines, his attempt to reduce the economic costs of prison centres on driving down its unit costs (i.e. the cost per prison place), partly by increasing the 'austerity' of conditions, rather than seeking to reduce the number of prison places. For Grayling, the question of reducing reconviction after imprisonment becomes more a matter of effective resettlement supervision and less one of appropriate sentencing or of the quality of the custodial experience itself. He demands less expensive punishment and more effective rehabilitation.
Although these political responses differ in important ways, they share in common a desire to meet the twin challenges of prison costs and reconviction rates and thus, in different ways, they raise and respond to questions about the utility of imprisonment as an investment of scarce public resources. Evidently, the relationship between imprisonment and desistance from crime is crucial in this respect. Although sentences of imprisonment may be imposed for reasons other than reducing reoffending (for example, for general deterrence, denunciation, retribution or public protection by incapacitation), since almost all prisoners get out, the nature of their post-release conduct and prospects matters. In this respect, the evidence seems bleak; the UK Ministry of Justice's own data shows that reoffending by former prisoners costs £7-10b per annum in England and Wales (Ministry of Justice 2010).
Yet, research evidence on the relationships between imprisonment and desistance is more equivocal than one might expect. In studies on criminal careers, imprisonment and on desistance, there is a recurring finding that some prisoners credit their imprisonment with transforming them as people, and making their desistance from crime more likely (see, for example, Aresti et al., 2010;Ashkar and Kenny, 2008;Barry, 2006;Comfort, 2008;Farrall and Calverley, 2006;Giordano et al., 2002). Indeed, this finding is pervasive enough for Farrall and Calverley (2006) to argue that the transformative power of prison is a dangerous myth perpetuated by criminologists, who should instead emphasise the obvious and deleterious effect of imprisonment on desistance. However, a finding that so many criminologists and prisoners have reported, if usually only in passing, merits greater attention. Comfort (2008), in one of the few works that tries to analyse the perception of imprisonment as transformative in greater detail, interviewed 10 couples, in which the men had recently been released from US prisons. Despite a lack of useful programmes or other interventions on the part of the prison regime, some of the men felt they had been rehabilitated. She writes: for betterment to virtually all inmates, paradoxically leading them to cast an environment barren of social welfare services as fertile ground for redemption. (Comfort, 2008: 272) Comfort calls the way in which her respondents felt rehabilitated in the absence of rehabilitative interventions 'as if rehabilitation' and argues that their sense of being rehabilitated was the product of their need to 'retroactively make sense of periods of incarceration, and in so doing salvage aspects of that experience that can be recast in a positive light ' (2008: 259). In Australia, Ashkar and Kenny (2008) similarly found that the 16 young offenders they interviewed saw themselves as having changed for the better through their incarceration, despite adverse conditions and a lack of rehabilitative activities, because they needed to find meaning in their sentence. This perception of rehabilitation existed in conjunction with their view that they had been deterred by the adverse experience of imprisonment.
Building on these and other studies, and on original research with long-term prisoners in Scotland, this article's central purpose is to examine why and how some people come to see their imprisonment as transformative and whether and how this is likely to have a positive effect on their subsequent desistance. We examine how, under certain conditions, imprisonment can prompt narratives of transformation, but narratives that are not closely related to intentionally rehabilitative aspects of prison regimes. Instead, we will argue that these narratives are a strategy adopted by some men, under certain circumstances, in order to make sense of their imprisonment and their desired future. We explore under what circumstances a transformation narrative is most likely to emerge and what other types of stories are used to make sense of the present and the future. Finally, we consider whether those who see themselves as transformed through imprisonment are more likely to desist than those who tell other types of stories; and whether transformation narratives seem likely to survive the 'second sentence' that follows release. The key contribution of our article lies in exposing both how and when imprisonment provokes transformation narratives and the limitations of its transformative potential. In sum, we suggest imprisonment is less a 'hook for change' than a 'shaky peg' on which fragile transformation narratives sometimes hang.

Methods
The men on whose accounts this article is based were interviewed as part of a research project that examined the meanings of long-term prison sentences (defined in Scotland as all sentences of four years or longer) for those experiencing them. The interviews for this project used a narrative methodology and were carried out in 2009 and 2010. Twelve men were interviewed within one Scottish prison. Recruitment was carried out by prison staff in line with selection criteria determined by the researcher; only prisoners who were either past their earliest parole date (half of the sentence) or coming up to this point were selected, so that they had had substantial experience of life in prison. A considerable number were within months of their release. Nine other men were interviewed in the community, while they were on license. Table 1 gives some basic information about each of the participants, all of whom chose their own pseudonym.
The narrative methodology was chosen to facilitate an in-depth understanding of the men's views of their sentences. A narrative approach allows for participants' views to be analysed and reported in the context of the whole interview, rather than fragmenting them in order to compare them with the views of others (Lieblich et al., 1998). Asking men to tell the story of their sentence made it possible to investigate how they positioned the sentence in their lives, rather than treating it as an isolated occurrence, and also allowed the identity presented by the men in the interview (e.g. active agent vs passive victim) to be taken into consideration (Riessman, 2002). What this study cannot provide is evidence about how common certain narratives are amongst prisoners, so the question of the prevalence of narratives that focus on prison as the site of transformation, while important, is not addressed in this article. Instead, we explore how narrative possibilities created by different circumstances are sketched out.
The stories we tell to and about ourselves need not be stable over time. As Josselson (1995) found when interviewing women several times over 20 years, elements that were important in a story at one point disappeared from view later, only to sometimes reemerge as significant as interviewees' view of themselves and their life story changed. She writes that 'in understanding ourselves, we choose those factors of our experience that lead to the present and render our life story coherent' (Josselson, 1995: 35). Maruna (2001: 85) has noted in relation to desisters that 'the present "good" of the reformed exoffender must be explained somehow through biographical events'. Furthermore, Presser (2008) argues that narratives do not only concern the past and the present, but also the desired future; a future of which the narrative needs to make sense. All envisioned futures are not of equal value. Gergen and Gergen (1997) have noted that progressive narratives, ones in which things get better, allow us to believe in positive change. They argue that a progressive narrative can be self-reinforcing, because the hope generated in telling this type of story provides us with motivation to make the projected positive future a reality. The very end of a sentence may well be an especially significant point at which to tell the story of one's imprisonment. By then the sentence is almost in the past, and a return to society awaits, leaving the future seemingly wide open to narrative possibilities.

Narratives of desistance and transformation
Amongst the men interviewed at the end of their sentence, there were broadly four ways of looking toward the future and thinking about desistance, with some of the men moving between these within the interview. Although some of those with the longest histories of imprisonment and the least favourable prospects upon release related familiar accounts of institutionalisation and hopelessness, the majority of the men desired a future in which they had stopped offending and achieved 'normality' by having a job, a house, a car and a family. However, given their different pasts, these men told three different types of stories to explain their prospective desistance. One emblematic example of each type is discussed below to show what factors played a role in the way prisoners thought about their present and future, before examining transformation stories in more depth.

The non-starters
Several of the men, who were either imprisoned for their first offence, or whose previous offences had happened a long time ago, identified themselves as 'non-starters' (Burnett, 2004): they said their offending had been out of character and were certain they would not offend again in the future. Accordingly, they were able to tell 'stability narratives', with the narrator 'steady in his propensity to act according to moral principles' (Presser, 2008: 62). Like the small number of men who thought they would keep offending, they did not need to explain any change. For example, Devan saw himself as never really having been an offender but as having been led to commit his (first and only) crime through debt he built up when his girlfriend got pregnant and they moved in together. Despite the setback of his imprisonment, he expected to return to his family and his old existence, so that prison was just a glitch in an otherwise favourable life. A further influence on the stories told, one that often coexisted with a 'non-starter' identity, was having significant social and personal resources available upon release. Maruna (2001: 14) has noted that, '[t]he white-collar or corporate deviant can fall back on family savings, a college education, or the support of well-connected friends to aid their transition out of crime'. While none of the respondents quite fit the bill of whitecollar criminals, nonetheless all those who appealed to a 'non-starter' identity had resources to overcome the potential negative impact of imprisonment, such as supportive families, savings, well-paying jobs they could return to and stable accommodation. Devan, for example, had a loving family, who would accommodate him, and job prospects upon his release: I'm either going to work for my father-in-law, at his bathroom store, that, he makes them, ships and supplies bathrooms. Or, I go back to my old job and work as a cook. So, hopefully, there's possibilities there for me to keep myself on the straight and narrow.
Accordingly, desistance was a realistic prospect for Devan and other men like him.

The well-resourced
Subtly different accounts were offered by those who had similar social and personal resources but were not able to portray themselves as 'non-starters'; they did need some explanation for the projected change in their behaviour upon release. Alex, for example, had now served two long-term prison sentences and presented himself as a veteran offender. Despite his entrenched criminal life style, Alex did depict a future of desistance in his narrative. The possibility of desistance was easily explained because of the resources at his disposal. Picturing his future he said: What I'm gonna do is, once I've got all this license and things out of the way, I'm going to relocate to Canada [. . .] So eehm, I'm kind of lucky, I mean [.] when I was a bit younger, I invested some money in some like flats and stuff in Aberdeen and rented them out, then I was lucky enough to get the timing of it right, you know, I sold before the property bubble burst and stuff, so it's not like, I mean, I have, I have resources to go out to.
Alex explained his new motivation to desist through his maturation in prison: I mean, in your late teens, your twenties you have a, you're, [.], you're highly risk tolerant, if not risk reckless. [. . .] I'm 36 now, 37 in April and I can, I can feel now I'm becoming increasingly risk averse.
The way Alex explains his desistance does not rely primarily on a conscious decision on his part to change his ways; instead, factors beyond his control have changed which mean he is now more minded to make the most of his favourable circumstances. Others with resources on the outside did describe some self-determination to go straight, but they did not anticipate many problems in putting this decision in practice. Giordano et al. (2002Giordano et al. ( : 1026 have also commented that those with the most resources do not have to rely on 'agentic moves' in order to desist.

The transformed
This leaves the men whose narratives are the main subject of this article. They told a story of transformation through imprisonment. Their histories meant they were not in a position to appeal to a non-offender identity and they had few obvious resources available upon release, but did aspire to a 'straight' future. For them, the most progressive possible narrative was one in which there had been a change that made future desistance coherent with what had gone before. Precisely because their prison sentence was the only biographical event standing between their pasts and their future, they made it the fulcrum of change; they told stories of personal transformation within prison. For example, Colin had had a troubled past, was expelled from school at 14 for selling acid, and received his first prison sentence at the age of 16, when he was given eight months for stealing cars. Since then, he had been imprisoned repeatedly and had developed a drug addiction. His family was fragmented, and not a reliable source of support. Colin emphasised the distinction between his current, and first, long-term sentence and previous short-term sentences, which had not led to desistance. The main difference, he felt, was his relationship with staff: I think it has got a lot to do with the longer sentence as well, plus the fact that a lot more people have taken an interest in what I am doing and that. I think that's a lot to do with it as well, eh? It's not just like you're left to get on with it.[. . .] Just, as I said, eh, my work party, and just being given the trust. I don't know, as I say, it might have just been me, but I think it has helped, getting a lot of the-getting the drug worker and that. Like people just having faith in me for a change. It is, it's totally changed my outlook.
Colin expressed gratitude for his sentence and saw it as a real turning point in his life. Summing up his view of his imprisonment he said: I am really glad I got it now, eh? I managed to get off all the drugs and that. I have touched nothing since I came into this jail. I have been stable -I am on methadone, eh, I have been stable on that for three years now, and that is the first time I have been clean, so it has really done, really done me good this sentence, eh?
By emphasising the transformative effect of his sentence, Colin was able to look forward to a future of desistance in the absence of any favourable prospects upon release.
Because as I say, I really think things are going to be different this time. Whereas before I have known [.] I have maybe said they're going to be different, but I KNEW. I knew it wasnae likebut this time things ARE different, eh? Just, it's not a case of thinking it, I KNOW they are different. I don't want to be in this life no more.
While Colin attributed his transformation to his interactions with staff and the resources on offer, other prisoners' narratives of transformation-through-imprisonment had different relationships to intentional rehabilitative input by the prison. For example, James, who was from a similarly disadvantaged background and had few resources to draw upon also told what Presser (2008)  Other studies have also found that 'having time to think' is often imbued with rehabilitative power. For example, one of Barry's (2006: 95) respondents is quoted as saying, 'you get the time to think in jail [. . .] I've got more of a conscience now. Jail slows you down'. Similarly, Comfort's (2008) and Ashkar and Kenny's (2008) interviewees attributed their (perceived) rehabilitation to time for reflection. While this used to be a major and intentional feature of penitentiaries, where the incarcerated were kept in almost total isolation (Garland, 1990), now time spent alone in the cell is more often characterised as detracting from the aims of imprisonment. For example, a recent Scottish Inspection Report noted, '[o]pportunities for work and education are insufficient, causing long periods when [. . .] prisoners are 'locked up'. This situation is particularly bad for prisoners [. . .] on remand and at weekends' (HM Inspectorate of Prisons, 2010: 1, emphasis added). While this negative view presumably only applies to excessively long periods of time in one's cell, given that time 'locked up' is a basic feature of imprisonment, time spent in solitary reflection is not mentioned in relevant Scottish Prison Service documents as a mechanism through which change may be effected (Scottish Prison Service, 2011). James' account, then, did not depend on rehabilitation as provided by the prison regime.
A more striking illustration that transformation-through-imprisonment narratives did not depend on rehabilitative input was provided by Andy, the only respondent amongst those interviewed while on license to relate such a narrative. 1 His need to position the prison sentence as a catalyst for change meant that he drew on particular aspects of imprisonment to fit his individual situation and needs (or even vice versa). Andy spent a substantial part of his interview trying to explain how the discipline in prison and access to the gym had made him a better person. In essence, he felt the discipline had put him in his place, while the gym had given him much-needed self-confidence. Compare the following: I think it's discipline, I think that's what a lot of people's missing, I think that's what I was missing, eeehm, and the jail gi'd me that, because I'd nae choice but tae just dae what I was told. [. . .] My dad had left the hoose, I was like 15 and I just took control o' the hoose, like I mean, I was a big guy so I ran the show, I did whatever I wanted and when I got the jail it just really brought me doon a peg.
And the gym really did help, I mean I says tae (my social worker) and that as well when I first got oot 'I really feel the gym is a positive thing for guys, especially boys who feel the need tae carry a knife, because it can all tae be wi a lack o' self confidence'.
The contradiction in needing both to be put in one's place and greater self-confidence was grounded in the way Andy presented his past self in the interview: as 'a big guy who ran the show', but also as a 'wee boy' who needed a knife to 'even up the odds' against 'big guys out there'. These selves may be compatible, if one represents his public and outward self (the big and fearless guy, a force to be reckoned with); while the other is his internal experience of himself (a frightened wee boy who has to work hard to maintain his outward persona). It is striking that Andy found aspects of imprisonment to fit both of these selves and rehabilitate them, so that he was able to see himself as irrevocably changed by his time in prison.
Despite Andy, Colin and James attributing their transformation to specific (albeit different) aspects of their prison experience, their transformation story appears to be a product of their need to craft a story capable of explaining future desistance, rather than owing to intentional rehabilitative input by the prison. Of course, this might be different in other prisons. For example, Stevens (2012) has recently described how residents in Therapeutic Prisons in England and Wales, convincingly ascribed their transformation to intentional features of the regime, such as being given responsibility and the self-discovery facilitated by therapeutic meetings. However, the transformation stories presented here chime more with earlier findings that respondents perceived rehabilitation in the absence of meaningful intervention in order to give their sentence purpose (Ashkar and Kenny, 2008;Comfort, 2008). What the discussion above adds is that it is not only the wish for meaning that leads prisoners to imbue their imprisonment with rehabilitative power, but also the wish for a positive future; one that allows for a break with offending and further imprisonment.

Narrative possibilities and limitations
The discussion above is not meant to present a typology, or to suggest that the type of narrative the men told was determined by the number of their offences, by their experiences of imprisonment or by their resources on the outside. The narratives told were too complex to fit neatly into exclusive groups and not all the men took the expected view: Neil was neither a very persistent offender nor resigned to future imprisonment, and had precious few resources to draw upon on his release, but even so did not narrate a story of transformation. Field notes summarise his interview as, 'has had a long term sentence before, nothing about personal change, except for getting older and hoping that he realises it isn't worth it. Wants to avoid further sentences'. Equally, Gordon, who told a transformation story, had never been imprisoned before and might therefore equally have appealed to a 'non-starter' identity, especially as he maintained he was innocent of his index offence. This demonstrates that constructing any narrative remains a creative endeavour, far from wholly determined by circumstances. But narratives, while often associated in the literature with agency and creativity, are nonetheless shaped within the confines of our lives. As we use them to make sense of our circumstances, these circumstances inevitably have an impact on the narratives we tell.
That there is a role for agency in the crafting of transformation narratives is also demonstrated by the men who attempted to tell a transformation narrative, but struggled to maintain it. They, too, had few resources and desired to desist, but although they would start a story in which prison transformed them, they would come to contradict themselves in other parts of the interview. For example, different excerpts from Graham's interview contradict each other: Aye, I think I've changed, you know what I mean, I really do think I have changed. I think this sentence has opened my eyes up to a lot of things. Just, the way I look at life, stop being selfish, stop thinking about myself all the time. I've got a wee-I've got a family, I've got a wee girl, you know, I just want to make things different for her, I want to give her an upbringing, a better upbringing than I did, y'know? I hope to get my life sorted, but they can-they just keep on putting me in hos-it's not going to work. Something will happen, I'll do something again, y'know.
These quotes suggest that Graham would have liked to tell a story of transformation through imprisonment, but was not quite able to maintain its coherence. In this, he and others resembled Giordano et al.'s (2002) respondents who displayed an openness to change, but because of 'a hedge or a break in the storyline' (p. 1031), did not present a confident account of such change.
Leaving aside the question of agency, there are other possible explanations of the differences between the 'successful' transformation stories and those with internal contradictions. First of all, recalling our discussion of methods above, it is possible the difference is an artefact of the interaction. If different questions had been asked, James and Colin might also have partially discredited their own transformation narrative. Conversely, if certain questions had been omitted, inconsistent accounts of transformation might have remained consistent. However, close analysis of the interviews suggests that the men who told transformation stories were more invested in their story of change and worked harder to make it fit the facts. For example, when Gordon's language slipped at one point so that it allowed for a slight possibility of reoffending by using the word 'hopefully', he immediately corrected himself: 'I'm hopefully-I will manage outside, no, I'm definitely going to do it'.
By saying 'no, I'm definitely going to do it', he corrected himself and told himself to maintain his preferred narrative. He seemed to be training himself to think in the 'right' way -in the process giving credence to the theory that the narratives we tell ourselves shape our identities and our futures (e.g. Josselson, 1995;McAdams, 2008;Maruna, 2001;Pavlenko, 2002;Singer, 2004). The narrative Colin, James and Andy presented was about their transformation, the main theme of their story, in a way that Graham's and others' was not. The efforts they made to craft these stories in the face of unpromising circumstances (such as previous sentences not having led to desistance and limited effective rehabilitative input from the prison) shows the strength of their commitment to change and the reach of their 'agentic moves' (Giordano et al., 2002(Giordano et al., : 1026.

'Pretty scary': Prison-based narratives and post-release realities
Because the men at the end of their sentence were not re-interviewed after the end of their sentence, it is not possible to say how they fared upon release. However, it is perhaps telling that the men on license who were interviewed were generally less optimistic and hopeful than the men in prison who envisaged a future of desistance. Many isolated themselves in their own homes, in effect recreating some aspects of their imprisonment, because they found it difficult to cope with the unpredictability of social life outside (see also Calverley, 2013;Shapland and Bottoms, 2011) . They felt that limiting their social interactions was the only sure way to avoid trouble (and a return to prison). Their main problem was finding work. For some this was a more distant goal, but most wanted to achieve 'normal life' as quickly as possible, with a family, house and job. But finding employment with a criminal record in an economic downturn was very difficult. This left them without anything to do during the day, meaning that they spent even more time on their own: The last year I've been oot and I've maybe been sitting in the hoose and I'm pissed aff and I'm fed up wi everything, cannae get a job, you feel as if nothing's going right for ye. I just-I've had me sitting saying tae myself 'I feel like being back in the jail the now'. (Jack) The impact this lack of opportunities might have on the narratives of the men who saw themselves as transformed through imprisonment can be illustrated by Andy, the only interviewee on license who sustained such a narrative. He described himself as leaving prison with a desire to contribute to society: '[prison] turned me fae being just a wild, just an idiot really tae, tae comin' oot noo and try to change mysel', try tae dae a bit, try tae dae something for the good'. But despite his best efforts to achieve his ambitions -he went to college but left without qualifications and had applied to numerous jobs -he felt unable to break away from his previous life because, two-and-a-half years after his release, his attempts to establish a new identity had come to nothing. In this, he was not alone. As discussed, most of the men on license had not secured their desired future. However, Andy was one of the most despondent amongst this group, possibly in part because he had had such high hopes: So what's the point of trying to change if there's nothing you can dae tae move on, you know what I mean, so. I feel kinda bad telling you this, it's just a hopeless cause [slightly laughing] really for me I'm starting tae think, it's horrible. I came oot wi all these big ideas and I went tae college, 'I'm gonnae use my experiences, I'm gonnae try and change, I'm gonnae try an let people know what it could be like in my situation'. Pfff, it just didnae happen man. Like I was saying, the thing that's really frustrating me the noo is no being able tae get a job and it's so, so annoying, so frustrating honestly it's just [.] it feels like everything's just wasted, every bit of my sentence, every effort I've made after it just, pffff, been a waste o' time that's what it's starting tae feel like. Pretty scary.
Andy saying 'I feel kinda bad telling you this' indicates that he feels this ending is spoiling his story, the one that should have progressed from him making the most of his prison sentence to him having the job and life he envisaged. Instead, he feels dragged back to his previous life and that 'everything's just wasted', even his transformative prison sentence. His final comment in this quote ('pretty scary') sums up well how he portrayed his life throughout the interview: as careering out of his control. Ironically, he would most likely have told a much more positive story if, like Colin, he had been interviewed before release. Sometimes imprisonment promises endings that the realities of freedom cannot deliver.

Discussion
This article has focused on some of the underlying reasons why some men serving long-term prison sentences come to construct a narrative of transformation through imprisonment. In doing so, they are able to have confidence in their ability to go straight and tell a progressive narrative in the absence of any other resources or promise of improvements in their lives. Recent literature has made much of the role of subjective changes, including confidence in one's ability to go straight, in the desistance process (Burnett and Maruna, 2004;Giordano et al., 2002;LeBel et al., 2008;Maruna, 2001;Presser, 2008). In his seminal work, Making Good, Maruna (2001) found that ex-offenders who had desisted from crime told a 'redemption script', which contained three elements: an optimism about one's own ability to overcome obstacles, a motivation to contribute to causes greater than oneself and a belief in one's own, essentially good, true self. In contrast, persisting offenders told a 'condemnation script' in which they portrayed themselves as powerless to overcome their problems and therefore 'doomed to deviance' (Maruna, 2001: 74). The book strongly suggests that developing a redemption script has a positive impact on the process of desistance. Maruna and co-authors asserted in a later article that 'these narrative patterns seem to distinguish successful from unsuccessful ex-prisoners, predicting successful reform after imprisonment' (Maruna et al., 2006: 181, emphasis added).
The results reported here suggest that the connection between optimistic narratives and desistance perhaps also runs the other way, with desistance strengthening such narratives. The transformation stories discussed here in some ways resembled the redemption scripts described by Maruna (2001): those telling them ascribed much power to themselves to overcome obstacles and presented themselves as essentially good in the present. However, it is possible that some of the men who told these stories will be unsuccessful in their move away from crime and will have to give up their transformation story once they have reoffended. Only those who do manage to sustain desistance will still tell a story of transformation years later, capable of being picked up by retrospective research. If the transformation story is one of the (few) ways of making desistance coherent open to prisoners like James, Colin and Andy (i.e. those who have few resources on the outside but are also not resigned to a life of imprisonment), then it might not be (only) that those who develop redemption scripts desist, but also that those who desist are the ones able to retain their redemption scripts.
It is possible that Colin and others who told a narrative of transformation through imprisonment will be more likely to desist than those with less coherent storylines, because their narratives suggest or signal their commitment to this goal. In her research, Burnett (2004) found that those imprisoned property offenders who were more confident about desistance when they came to the end of their sentence, were less likely to reoffend after release. However, those who were the most confident about their desistance also had the fewest problems facing them upon release. A later quantitative study using the same data combined with 10 year reconviction data, aimed to separate the effect of social problems from the effect of subjective factors (LeBel et al., 2008). It found that (the number of) social problems faced upon release had the greatest impact on whether desistance was achieved, but that (1) the perception that being an ex-offender would make desistance difficult, (2) identifying oneself as a 'family man' and (3) regretting past offending also had some impact. The level of confidence in one's ability to desist only impacted on eventual desistance through reducing the number of social problems upon release (although a more likely explanation, acknowledged but not tested by this study, is that expectations of social problems upon release made the offenders in this study less confident of their eventual desistance). When these findings are applied to those telling a transformation through imprisonment narrative, it would appear that they should have little advantage over others as they faced multiple problems upon release, including unemployment, being housed in a hostel or with parents and financial difficulties, issues their transformation story had no power to resolve.
Other research further illuminates the precariousness of the narrative of transformation through imprisonment. Giordano et al.'s (2002) found that, given that structural variables that have been credited with leading to desistance, such as employment and high quality marriages (Sampson and Laub, 2003), are now hard to come by, offenders need a real openness to change before any change can happen. Because structural changes do not fall in their lap; offenders have to work hard to leave their old lives behind. Giordano et al. (2002Giordano et al. ( : 1000 write, 'At the point of change, this new lifestyle will necessarily be "at a distance" or a "faint" possibility. Therefore, the individual's subjective stance is especially important during the early stages of the change process'. However, they also note that some of their respondents lived in such deprived circumstances that even the most strong-willed and motivated might not manage to desist. They argue that, while openness to change is necessary, it is not sufficient. It needs to be accompanied by 'hooks for change', which will allow the person to turn their wish into a reality. While for some of their respondents imprisonment provided such a hook, Giordano et al. (2002) note that the prison experience might not be as good a 'hook for change' as others, such as religion and parenthood, because it neither provides a blueprint for life after imprisonment nor a new identity to replace the old one of offender.
On the one hand, those who tell a transformation story may be more likely to desist than those who are less committed to a changed life, but their use of prison as a peg on which to hang change might not be as secure as other possible hooks they might have used to construct their narratives, had they been available. For example, if these men had found religion in prison, this may have been a more powerful predictor of desistance. As it is, they still have to leave prison and find a new way of living on the outside, rather than stepping into a ready-made identity (and, in the case of religion, a ready-made community). Giordano et al. (2002) also emphasise that, as those with the most resources are unlikely to need much motivation to escape their pasts and those with the fewest are unlikely to escape it at all, 'on a continuum of advantage and disadvantage, the real play of agency is in the middle' (Giordano et al., 2002(Giordano et al., : 1026. The question is whether Colin and others like him will have sufficient resources and encounter enough opportunities to fit into this 'middle' and to make their desistance a reality. Andy was starting to despair of his chances to develop activities and roles to take the place of offending. He was desisting, but his desistance consisted merely of the absence of offending, rather than the new, fuller, life he desired.
Some pasts lend themselves better to certain kinds of stories and, conversely, different types of stories might be associated with different futures. In the research discussed in this article, it was those with few resources and a significant criminal history who were most likely to position their prison sentence as transformative just before their release. The evidence suggests, however, that if these men's belief in their own transformation through imprisonment has any power to help them desist, it is modest. It does not provide any blueprint or identity for life after imprisonment and other identities, such as those provided by meaningful work, are very difficult to move into for those with a criminal record.
To return to the question of imprisonment's utility in supporting desistance, the apparently paradoxical finding that imprisonment can be transformative may turn out to be partly true dependent on the prison experience, partly a post hoc rationalisation, and partly a type of selection effect. For those whose transformation does survive release, transcending imprisonment may remain an important part of a redemption narrative of heroic struggle, but one in which other social resources probably played an important part. By contrast, those whose prison transformation narratives fail to survive the barren landscape of the 'second sentence' are, of course, destined to appear as 'persisters' and not as 'desisters' in desistance studies. The relationship between imprisonment and desistance seems to depend, therefore, on the nature of imprisonment itself as a lived experience; the personal and social resources the prisoner brings to imprisonment; the opportunities and resources s/he accesses during imprisonment; the narrative s/he constructs in and through imprisonment; and, perhaps most importantly, the personal and social resources with which and to which s/he returns post-release. As an institution and as a sanction, imprisonment therefore is at best a highly unreliable incubator for change -perhaps not so much a 'hook for change' as a 'shaky peg' -and for that reason a very risky investment for those charged with reducing the economic costs of reoffending.