The pains of desistance

Desistance is generally presented in a positive light, with themes of ‘making good’ and generativity recurring in the literature. This article reports on two qualitative studies exploring the desistance journeys of two different groups of ex-offenders, drawing attention to the pains of this process. It examines the possible consequences of these ‘pains of desistance’ and how they are linked to three spheres of desistance: act-desistance; identity desistance; and relational desistance. The attempt to achieve act-desistance often led to the pain of isolation for our interviewees, while the clash between the need to achieve identity desistance and a lack of relational desistance (especially on the meso- and macro-levels) meant that they suffered the pain of goal failure. The pains of isolation and goal failure combined to lead to the further pain of hopelessness. Those interviewed were indeed ‘going straight’, but taking this path led many to a limited and often diminished life.


Introduction
The Liverpool Desistance Study (LDS) (Maruna, 2001), the touchstone for subsequent qualitative, phenomenological studies of desistance, consisted of narrative interviews with 20 persisters (those who continue to offend) and 30 desisters (those who had stopped), to examine how change was initiated, maintained and sustained over time. The findings highlighted the importance of 'making good', the process of the individual recovering their core good self and constructing a positive narrative about the future, and of generativity; the process of 'giving back' to society, and to future generations in particular. Other studies since this time reveal barriers to desistance, such as substance misuse (McSweeny, 2010), lack of housing (Edgar et al., 2012), difficulties in finding employment (Hlavka et al., 2015) and the individual's lack of motivation or their inability to achieve status or respect in a legitimate way (Farrall, 2002;Giordano et al., 2002;Healy, 2010Healy, , 2014. The Sheffield desistance study also exposes the adverse impact of financial hardship and the attraction of further offending for those trying to stop ). The desistance model arising from this research includes the meeting of obstacles, but holds that, in the absence of a relapse, the desister will find reinforcers and go on to establish a crime-free identity as a non-offender (Bottoms and Shapland, 2011: 70). This article on the other hand, demonstrates how some desisters refrain from offending for long periods but are not in a position to establish a new identity. The Sheffield study found a desistance strategy among its participants of avoiding situations in which offending might happen, which the authors call 'diachronic self-control' (Shapland and Bottoms, 2011: 274) and largely see as adaptive (but see Bottoms, 2013, on self-binding). The studies discussed here provide a crucial insight into how excessive diachronic control can lead to a restricted and impoverished existence.
Although as discussed the challenges to desistance are acknowledged in the literature, the maintenance of desistance is generally portrayed as a straightforwardly positive outcome, both for the desister and for the wider community (Barry, 2010;Jamieson et al., 1999;Laub and Sampson, 2003;Weaver and McNeill, 2007). In this article we argue that this is not the reality for all those who are, in fact, desisting from crime. Instead, the experience of 'going straight' can be painful and lead to limited lives. In order to explain why some people suffer the 'pains of desistance' identified, we introduce new subtleties of meaning in the desistance journey and the spheres in which desistance needs to be maintained and supported in order to achieve the positive destinations described in the literature. We draw on the findings of two qualitative studies with desisters in very different circumstances (young people with a short criminal career and men on licence after a long period of incarceration), which show that these two groups of participants suffered very similar pains of desistance.
It may be that research findings of desistance leading to redemption or 'moral heroism' (Maruna, 2001: 95) are largely a product of retrospective studies, where people who have long since desisted (and identify as such) are interviewed (Healy, 2010). For example, Aresti et al. (2010) concentrated their research on men who were either studying at university level or had secured a conventional career. This focus meant that the likelihood that such outcomes are (very) unusual after significant criminal involvement is not addressed. Similarly, Maruna explicitly acknowledges that his research excluded many ex-offenders, as he only recruited those who either identified themselves as 'successfully "going straight"' (2001: 48, emphasis added) or as 'actively persisting ' (2001: 48). As Maruna (2001: 44) notes, this means that those in between were not included in the research, despite the relative rarity of the desister and persister characteristics he describes.
Some research studies have examined (in passing) what the experience of desistance is like for those who are less 'successful' than Maruna's and Aresti's interviewees, but who still seem to be managing to stay away from crime. In French research, at The University of Edinburgh on March 29, 2016 crj.sagepub.com Downloaded from most desisters were found to live a 'normal' life, concentrating on themselves and their families, rather than trying to contribute to society at large (Herzog-Evans, 2011). Healy (2010) also found little evidence of generativity in a sample of desisters. However, these desisters, while far from 'moral heroes' still appeared content with their lot. Other research participants' experiences were closer to the ones we report here -with real pain inherent in the maintenance of desistance. Some of Appleton's (2010: 117) lifers on licence had had to settle for 'a menial and lonely existence' in which there were 'many desperate and monotonous periods'. Calverley's (2013) black and dual heritage desisters also struggled, with peers still involved in criminal activity and families of origin and formation often dispersed, they tended to isolate themselves in order to maintain their desistance. While there is evidence, then, that the journey of desistance is not necessarily a pleasant one, little in-depth analysis has focused on the pains of desistance, their causes and likely outcomes. These are the issues this article aims to start to address.
In order to elucidate how different aspects of desistance cause different pains the concepts of 'primary', 'secondary' and 'tertiary' desistance are drawn upon. Maruna and Farrall (2004) developed the concepts of primary and secondary desistance to mirror those of primary and secondary deviance; primary desistance refers to a period of nonoffending, and secondary desistance to a change in self-identity where the person no longer thinks of themselves as an offender. The term tertiary desistance is now being used by McNeill (2016) to highlight another aspect necessary for long-term change: the recognition by others that one has changed and the development of a sense of belonging. If we accept that how we act depends on how we see ourselves, and that this depends, in turn, on how we see ourselves reflected in the eyes of others (Weaver, 2013), which in turn depends on how we act, then these terms delineate desistance in different spheres (the world outside, within ourselves and in relation to others) rather than at different times. This extends the analysis and understanding of deviation and desistance to be more than a linear process. For this reason, we propose using the terms 'act-desistance' for nonoffending, 'identity desistance' for the internalization of a non-offending identity and 'relational desistance' for recognition of change by others. We argue that this terminology describes and differentiates between the different aspects of desistance better than 'primary', 'secondary' and 'tertiary' desistance, as it does not suggest sequencing in time or importance. This article furthermore distinguishes between different levels of relational desistance, where the micro-level relates to the individual's immediate social setting, the meso-level to their wider community and the macro-level to society as a whole. For example, while employers are situated on the meso-level, the prevalent way in which criminal records are dealt with in recruitment decisions would be on the macro-level. Those we interviewed were sometimes met with a measure of relational desistance on the micro-level, but when relational desistance was not forthcoming at the meso-and/or macro-level this undermined their act-and identity desistance and thereby threatened progress. It is important to note here that act-and identity desistance are, while dependent on others, something the individual can achieve. Relational desistance, on the other hand, is within the control of others. For example, it is up to the mother whether she will accept her son has changed, as it is up to the employer and society at large. However, the greater the individual's social and bridging capital (Putnam, 2000), the greater the at The University of Edinburgh on March 29, 2016 crj.sagepub.com Downloaded from likelihood that act-and identity desistance will be recognized and supported, especially on the meso-level. Therefore, individuals can take steps towards relational desistance by extending and drawing on their formal and informal connections (Weaver, 2015), but this article argues that, given the pains of desistance, our interviewees struggled to do so.

Research and Methodology
This article is based on two qualitative studies. Discussed first are the results from Schinkel's narrative cross-sectional study of 27 long-term prisoners at different stages of their sentence, conducted in 2009-2010. The discussion here focuses on the interviews with nine men who were on licence after serving a long-term prison sentence. Many of these men had long histories of offending (for more information see Schinkel, 2014), with the youngest three in their late 20s, and the majority in their late 30s, 40s or 50s.
The second research project, carried out by Nugent in 2012-2014, consisted of interviews with young people who could be defined as being in the transition to adulthood. Discussed here are the findings from the interviews with five young people who had offended (four males and one female), aged between 16 and 21. All had received many years of support from a third sector organization (Includem) and at the first interview were about to end their contact with the service because they felt they were now in a stable position. All five were in the process of beginning act-desistance after a short career in offending. An attempt was made to interview each participant four times over a fieldwork period of 15 months, spaced at the two month, six months and 12 month follow-up stage. In total, 15 interviews took place with this group and four followed up in the final stage.
In both these research projects, narrative interviews were understood as being (more or less loosely) based on past happenings and pointing to future outcomes (Bell, 2002). Having an inherently temporal thread, narratives were treated as 'interpretive devices, through which people represent themselves, both to themselves and to others' (Lawler, 2002: 242). As a consequence, narratives are influenced by the encounter in which they are told, with the audience and its reaction to what is said especially important (Pasupathi and Rich, 2005). As both projects were carried out by female researchers who were ostensibly not involved in offending themselves, there may have been some pressure on the interviewees to distance themselves from their past behaviour (see also Presser, 2008).
As both studies are based on fairly small samples, no attempts are made to identify how common or typical these desistance narratives are among the larger populations from which our participants were drawn (young people with a short offending career and men on licence after long-term imprisonment). Instead, we focus on theoretical generalization (Lewis and Ritchie, 2003): using our findings to identify concepts in these narratives to allow us to make connections that may apply beyond the immediate research setting. The commonalities in the findings of the two different projects are at least suggestive that the pains of desistance we identify may be commonly experienced by different groups of people trying to desist: by young people with limited offending careers and by older men convicted of much more serious (and usually numerous) crimes.

Desistance after Long-Term Imprisonment
The men on licence after long-term imprisonment experienced three main (but interconnected) pains of desistance: isolation and loneliness; goal failure; and, increasingly, a lack of hope. As these men had been imprisoned long-term, it is difficult at times to distinguish between the pains of desistance and the effects of institutionalization or their life-long history of deprivation, which produced their own pains, as well as exacerbating the pains of desistance.

Pain of isolation
Isolation and loneliness were described by six out of the nine men. First of all, their attempt to avoid offending meant that they tended to isolate themselves. There were several reasons for this (more carefully elaborated in Schinkel, 2014). Needing to stay away from former friends (and enemies) who were still involved in offending, in order not to get pulled back into such behaviours, was one factor. Second, a history with individual police officers (who might want to 'get them one more time') coupled with being on licence (and thus subject to immediate recall to custody) meant that they felt much more vulnerable to criminal justice interference than others. The need to stay away from temptations, provocations and risk, combined with the effects of institutionalization, meant that they often just stayed in their accommodation, in effect reproducing some elements of their imprisonment: '[i]f I get a couple ah' drinks in me I go aff ma heid in two minutes. So (.) I find mysel' tae be confined to the hoose a lot this time' (Jack).
For some, this new-found introversion also meant that they did not recognize themselves, that they had lost their sense of self and personal identity. For almost all, there was a sense of displacement, a sense that they were living a life with which they were unfamiliar: I've tried tae just thingmy away fae all that. It isnae an easy thing tae dae because my life, I'd been that used tae and, I could probably live my life back then wi my eyes shut, cause I knew everything and everything was just second nature to me. But trying tae come away fae that […] (Lino) Before the attempt to desist most had had fuller lives -full of associates, substance abuse, offending and excitement; not necessarily full of positive things, but full nonetheless. While in prison, many had become used to their lives in there, now sometimes looked upon with nostalgia, so that institutionalization again interacted with the need to desist, to bring about a sense of displacement. Furthermore, as is evident from the above, the men did not feel they had much control over their new, relatively barren, lives. The findings here suggest that in cases where diachronic self-control is made necessary by external circumstance (see also below) and not combined with more active forms of control, such as changing one's behavioural patterns and resisting opportunities to offend, it is not experienced as agentic and can lead to isolation and hopelessness.

Pain of goal failure
This lack of control was also a feature of another pain of desistance: the pain of goal failure. For most of the men, their identity was in a liminal state -they were no longer prisoners, but had not yet fully served their sentence, or achieved a new identity. Some were able to draw upon identities that had preceded or followed their imprisonment, such as the identity of father or partner, but most (also) wanted to achieve other goals, usually employment, which would allow them to be the men they wanted to become (see also Schinkel, 2014). Here, their wish to achieve identity desistance conflicted with their criminal record and their previous experiences of deprivation (which had left most with no qualifications or work experience) to produce the pain of goal failure: since leaving prison (which for many was years in the past) most had not been able to take any meaningful steps towards their desired identity: I was just away for another job interview there, two weeks ago, eehm, wi my wee brother, he got the job, I got knocked back because of my conviction. That's what I found for the whole two years, two and a half years, I don't really know if I'm ever going to get the chance to move on. (Andy, emphasis added) Importantly, their criminal record (and for some, being known in their neighbourhood) impeded their progress through the effect it had on relational desistance. Many of these men had achieved a measure of success on the micro-level of this sphere of desistance; they had family members or a partner who believed in them and who provided much needed support: quite extended time of the licence to provide enduring links to the meso-and macrolevel, especially in relation to employment: I realize like [name of social worker]'s good. But he's no really/ he's no really got the powers there tae change it or anything for me, tae help me really. He's helped me wi' small things like when I first came out […] so I could get money and things like that. But now when it's getting tae the nitty gritty, my licence is going to end, I've got no job, it's looking like I'm no going to be able to get a job, even though I've got quite a bit of a qualifications now, that's nae use tae me as well, so I'm kinda worried aboot it. (Andy) Without any brokerage or leverage from others, when the men applied for jobs they were confronted with a lack of relational desistance on the macro-level -being forced to disclose their records meant that they were unsuccessful in reaching their goals of gainful employment, because society at large saw them as (potential) offenders rather than desisters, and rejected them as such: '[y]ou've got the disclosure, it's/ as soon as people know that you've been in so long and they don't want tae gi you a chance, so that's the hardest bit, for me anyway, I feel, getting employment' (Tim). This, in turn, could have ramifications for their micro-level relationships, as some of their partners found it difficult to accept the lack of progress.

Pain of hopelessness
The final pain of desistance was a product of the first two pains: not feeling able to fully take part in life or to move towards their goals meant that many of the men were slowly giving up hope for anything other than a life of merely existing. This hopelessness in turn undermined any zeal or motivation they might have had to achieve their original goals, leading to helplessness and apathy for many. They were losing sight of the happy endings they had wished for themselves: I would love to have kids, I would love to be married, have a full time job, settle doon, you know just have money, to have an easy life, a good life, you know […] But to be honest I don't think I'm gonnae get any legitimate work, not for a while anyway […] Eeehm, just with my criminal record, there's that many people looking for work just now that haven't got a criminal record. (Mohammed) It is of interest that the three men on licence who did not describe isolating themselves, and seemed to suffer less from the pains of desistance had other projects to occupy themselves. Mark and Smitty were still learning to live without alcohol and drugs, while Stephen was caring for his ill wife: To come outside and actually test myself in my own wee flat, get up in the morning and actually face a day withoot a drink, you know what I mean, myself it's great, and naebody else tae depend on, […] I'm quite happy. I've no got much, I've no got much at all but I'm quite happy. (Mark) Mark, Smitty and Stephen had something that provided their lives with a sense of progress, and were not, at the moment, aspiring to goals that proved impossible for them to at The University of Edinburgh on March 29, 2016 crj.sagepub.com Downloaded from achieve. They did not face the frustration experienced by the other men, because they were happy to (perhaps temporarily) confine their lives to the micro-level. The others wanted to link into the meso-and macro-levels, but because of their lack of goal achievement on these levels, were starting to withdraw more and more: I wasted all that time in prison and when you're in prison you want to get out so badly, but then when you're out you think 'what's the point?'. That's what it was like and then I thought 'do you know something, there's that much I'm no doing, I mean I could be going paintballing, swimming, fishing, I could be going abroad, I could be doing this when I'm/ I'm not, I'm just doing nothing basically, I'm just no living my life'. (Mohammed) It is difficult to assess what the long-term outcomes for these men will be. Mark, Smitty and Stephen might eventually want to broaden their horizons, and either achieve their goals or become as frustrated and hopeless as the others. Equally, they might just find peace within the micro-level world they live in. The others, confronted regularly with the difficulty (or impossibility?) of living a fuller life, might well keep withdrawing to lessen the frustration, and eventually accept their lessened and limited lives. Or, if they are lucky, their circumstances might change, they might achieve some goals, and start to progress towards true relational and identity desistance. This is unlikely to happen for many though, without significant changes in policy, the way in which offences are disclosed and perceptions of ex-prisoners in society. In the face of the strain created by their inability to achieve their aspirations, they might also return to 'innovation' (Merton, 1968) and crime.

Desistance after a Relatively Short 'Criminal Career'
At the time of the first interview the ages of the desisting young people ranged from 16 to 21. They had started offending around 13, with their offences increasing in seriousness, until between 15 and 20 years old. In the initial interviews they revealed conventional hopes of wanting to get a job, have their own place, meeting someone and 'settling down' (see also Barry, 2006Barry, , 2013Jamieson et al., 1999;Shapland and Bottoms, 2011). Embarking on act-desistance was seen as the first step towards having a better future.

Pain of isolation
They had all been in gangs and spoke about this as though they had been part of a family, not just in terms of loyalty (Warr, 2002), but also in terms of the emotional connections established, opportunity to share concerns, get advice, be vulnerable and achieve a sense of belonging. This emphasis on friendship is similar to past research with female gang members (Batchelor, 2009), pointing towards a more nuanced understanding of gang activity for both genders: 'We could be really close and if I had any problems I would say it to my pals […] it is not as if they are bad boys, because they aren't' (John).
In the first interviews all participants had recently 'knifed off' (Maruna and Roy, 2007) from 'territorial gangs' (Pickering et al., 2012). They found this painful but necessary in order to stop offending (Farrall, 2014). They had made a conscious decision to do so (Barry, 2006), and were proud of the strength they had shown in doing this. However, moving away from these peer groups in an attempt to achieve act-desistance meant that they experienced the pains of isolation. With few new relationships taking the place of the ones they were forced to end, like the men on licence the young people described spending their time by staying mainly indoors. They did not experience the same levels of surveillance from the state as the men on licence (or face the same consequences for any incident), but suffered similar loneliness. Tony admitted that he was scared of being 'out', saying that 'everyone where I stay is either involved in the gang or the gang goes after them'. While not unjustified in this view, it was apparent that as he became more secluded his fears became more entrenched, reaffirming the link between isolation and fear (Gadd and Jefferson, 2007). By the final interview, he seemed to be wary of everyone: I want to have a life that I don't have to watch all the time over my shoulder. I want to be able to walk about freely. Basically, be able to walk to the shop without looking over my back and all that stuff.
The other young people also admitted to being fearful and 'looking over their shoulders'; their locality had been transformed from a place of sociability and fun, mixed in with fear (McVie, 2010), to a place only of fear of the potential risk of reprisal. Desistance research highlights the importance of spatial dynamics (Farrall, 2014) -these young people viewed the spaces in which they lived as increasingly threatening, which they 'coped' with through self-isolation.
The transition to adulthood is recognized as a time when social networks usually expand (Cotterell, 2007). For this group though, their networks continually shrank, exacerbating the pain of isolation. Family connections were tenuous and particularly rare for the young men, as their family had given up on them when they had been offending. Notably, three of them made no attempts to re-establish those relationships: I used to have contact with my dad, but now I don't bother with that, he keeps to himself and I do the same, I don't try to put myself out there, I would be embarrassed if he didn't want to know. (Mark) In contrast, the young woman, Amy, worked hard to repair connections with her family, who acknowledged that she had turned her life around. Therefore her act-desistance had been accompanied by relational desistance at the micro-level. Her family's changed view of her had a significant impact in helping her to see herself as someone who could have a better life, thereby supporting her identity desistance: 'I feel like I have been more wanted since I did a big massive change. I think that I needed to change to prove to myself as well as everyone else' (Amy). This supports the finding in previous research that women may be more 'relational' in their attempts to move towards desistance (Barry, 2006(Barry, , 2010. However, crucially the young men interviewed did not avoid their families because they did not want to reconnect with them, but because they were fearful of rejection. By the final interviews, Kevin described having no support at all in life and three others were entirely reliant on their girlfriend for company and to act as a motivator to at The University of Edinburgh on March 29, 2016 crj.sagepub.com Downloaded from maintain their desistance. Their access to relational desistance (even on the micro-level) therefore was extremely limited.
Interviewer: And what are the things that help you to stay away from offending? Tony: Just Amanda that is all. 1 Although there is increasing evidence of the positive influence a girlfriend can have on helping the desistance process (Bottoms and Shapland, 2014;Laub and Sampson, 2003), there is less research on the effect that this may have on the person in question (Halsey and Deegan, 2015). Only one girlfriend agreed to be interviewed, which revealed the pressure she felt from being the single stable focal point to promote change, raising questions about sustainability of such support. This also means that the micro-level relational desistance relied on by the men on licence who only had their partner to support them might be less reliable than it appeared in the interviews.

Pain of goal failure
Like the men on licence, the young men struggled to attain a new identity and therefore suffered the pain of goal failure. The transition to adulthood is defined as a liminal stage in life where identity is still being shaped and formed (Barry, 2006;Turner, 1964). This group moved from being clearly defined as 'offenders' to being unsure of who they wanted to or could be. Both Tony and Kevin highlighted the detrimental impact of having a lack of access to 'acting out' an alternative identity. Kevin missed the status that crime had given him and had been unable to recreate a pro-social identity of any similar standing, a barrier to desistance reflected in other research (Healy, 2010). Tony had not re-offended and thus had been successful in act-desistance, but without a job had not achieved identity desistance and was just living day to day, merely existing. In past research with women offenders, the role of 'mother' has been identified as a crucial script to retain or build upon aiding desistance (Rumgay, 2004). This has been confirmed by more recent research by Sharpe (2015), but this also found that the stigma of past offending undermined identity desistance and those interviewed reported feeling judged as maternally deficient, even after they had not offended in a long time. As pointed out, being a mother ought not to be the only 'alternative' available for women (Weaver and McNeill, 2007). In this research, David and John were becoming fathers, a role that they felt defined them positively and was a major 'turning point' in how they regarded themselves and how they were regarded by others (Giordano et al., 2002). However, this new identity did not suffice, as both also wanted to get a job. Employment has been shown to be a means of doing masculinity and of achieving a sense of pride and fulfilment post desistance (Uggen, 2000;Weaver, 2015). Like the women in Sharpe's study, these men had precarious identities and their identity desistance remained fragile.
Achieving the goal of employment was as difficult for the young men as for the men on licence, with traditional paths to adulthood, such as from school to work which were readily available in the 1950s and 1960s in these areas no longer accessible in the same way (Coles, 1997). Like the men on licence, the young people found themselves at the very bottom of the employment ladder; on the outskirts of the 'the precariat': those who face uncertainty, instability and lack occupational identity (Standing, 2011: 8). The situation they faced as a consequence exacerbated the pain of goal failure. They felt that the courses they were forced to take as a condition of receiving Job Seeker's Allowance were meaningless, realizing that these did not lead to employment. Given that some had already completed the same courses previously, the requirement to attend them again made them feel like their time was unimportant. The significant challenges of surviving on benefits similarly contributed to the pain of goal failure: I just want to get a job and get a life. I don't want to live off the brew and get stoned and getting in trouble with the police and end up in jail, I am trying to dodge all that. (Tony) With fatherhood the only alternative identity to which the young men in this research had access, it is clear that both men and women are held back by limited opportunities and a lack of social and bridging capital to establish pro-social identities. Unlike past research with young people in the transition to adulthood who were optimistic about the future (Arnett, 2006), these young people felt 'stuck' (Pickering et al., 2012).

Pain of hopelessness
Like the men on licence, the young men were increasingly losing hope due to their loneliness and goal failure as shown by the following interview excerpt: Interviewer: Last time I spoke to you, you said that getting a job was a certainty. Do you still feel that way? David: I don't know, I just have to take every day as it comes and see how it goes.
While Standing (2011) emphasizes the potential danger in the anger generated by the disappointment or 'strain' faced by the 'precariat', these young people had simply a loss of hope. By the final interviews three of the four men had generally given up on getting a job. The new identity offered to John and David by fatherhood provided some hope, but it did not fulfil their own goals, notably employment. Whereas John dealt with this by accepting his situation and thus limiting his ambitions, David continued to be frustrated. His hope had become a source of pain. The possible and dire consequences of growing hopelessness as a result of isolation and goal failure were illustrated clearly by Kevin. He struggled to name anyone he could rely on after he had left Includem. He tried desperately to make something of himself and to prove to his family that he was worthy of a second chance. However, unable to get a job and being repeatedly confronted with his lack of education and prospects, his attempts at act-desistance in a relational vacuum, eventually became too much for too little, and he cracked. By the final interview he had re-offended and had committed his most serious offence to date. His exasperation with his mere existence was exposed as he said the night of the offence he had had enough and went 'looking for a fight' and took his chance when he got it.
Interviewer: Is there anything you fear losing if you went to prison? Kevin: Nothing. Interviewer: Nothing? Kevin: Nothing.
Conversely, despite the posited limited availability of scripts other than motherhood to desisting women, Amy was forging ahead in discovering a new identity (Maruna, 2001), illustrating how the achievement of early goals and recognition by others can sustain desistance. By the final interview, because of the self-belief built as the result of reconnecting with her family and significant support from the Includem worker, she was seriously considering going to college. Through help with access to funding and making the option accessible, her worker acted as what Nugent (under review) defines as a 'meso-broker', supporting Amy to reach beyond the micro-level to begin to create a realistic blueprint on which a new 'possible self' could be operationalized (Paternoster and Bushway, 2009). Through being accepted as a student on the meso-level of relational desistance, it seemed likely that Amy would be able to further consolidate her identity desistance.
The above discussion shows that hope is fragile and requires markers, events and other people for it to be built, maintained and sustained. In the face of a lack of relational desistance and limited opportunities to achieve identity desistance, it is not surprising that hope is rapidly lost.

Discussion
Despite the two groups of interviewees discussed in this article being very different in terms of their 'criminal career' and stage of life, the similarities in their attempts to desist and the pains endured are striking. These studies show that maintaining actdesistance can lead to the pain of isolation, while the need to achieve identity desistance in the face of a lack of relational desistance or social capital leads to the pain of goal failure. Although similar in terms of the pains they suffered, the two groups of interviewees faced different barriers contributing to these pains. For example, while the young people spoke in-depth about the way in which spatial dynamics contributed to their isolation, the self-isolation of the men on licence was a result of institutionalization and a sense of surveillance as well as the desire to achieve act-desistance. Underpinning these pains of desistance and largely left unacknowledged in this article, is the context of inequality, structural limitations and poverty in which the participants in both groups had always found themselves (Corr, 2014;Farrall, 2014;Healy, 2010Healy, , 2014MacDonald et al., 2011).
The research offers a valuable insight into the interdependence of act-desistance, identity desistance and relational desistance; without each of these in place 'success' will remain limited or fragile. The interviewees' main action in establishing act-desistance was to 'knife off' relationships with their peers, offending opportunities and their anti-social past (Maruna and Roy, 2007;Weaver and McNeill, 2014). The inability to connect to prosocial networks or new activities meant that most experienced the pain of isolation; in this way 'knifing off' contained an element of self-harm for these people, and one that may have serious repercussions. Loneliness can lead to a downward spiral of decreasing selfesteem and self-confidence, and has been causally linked to serious mental health problems and an increased risk of suicide (Burns, 2012;Cotterell, 2007;Griffin, 2010).
While some of the men on licence were met with relational desistance on the micro-level, thus having some support for their identity of non-offender, for others this level of relational desistance was absent or fragile. With the exception of one young woman, no one had encountered relational desistance on the meso-or macrolevel, which was experienced as painful by those who needed such connections to achieve their goals, notably employment. In the face of this inability and that of being recognized as non-offenders by the world at large, most participants were stuck in a liminal position: neither offenders nor those who have 'made good'. Achieving an alternative pro-social identity was difficult, with fatherhood, caring for a loved one or being in recovery from drug-or alcohol addiction the few accessible options. Most were focused on gaining employment and failure to achieve this meant that identity desistance remained out of reach.
Both studies highlight the importance of hope, reaffirming the conclusions of past research (Burnett, 2000;Farrall and Calverley, 2006;McNeill, 2012). Based on the findings presented here it is clear that enduring failure and a lack of recognition from others that one has changed undermine hope. Without hope, not only does sustained desistance become less probable, but life becomes less fulfilling (Burns, 2012).
Unfortunately, the studies discussed did not include long-term follow-up interviews, but Kevin's return to prison shows that act-desistance sometimes does not survive a lack of identity and relational desistance. While the men on licence were sustaining act-desistance, the fact that many of them had been released for years and not managed to achieve lives that they felt were meaningful shows that long periods of lives may be blighted by the pains of desistance. Overall, this article shows that desistance for some is not just a 'process' but rather more like an endurance test with little to no reward for their efforts. As a society we should question whether mere act-desistance is really sufficient for our purposes (as it seems to be) or whether we want more than empty, limited lives for people who are making the considerable effort to desist from crime. If so, we need to do more to those who are trying to help themselves, to not just 'go straight' but to 'go somewhere new'.

Ways Forward?
This article suggests that one of the ways in which identity desistance should be supported is through connections beyond the micro-level. Halsey and Deegan (2015) similarly found that any move from a spoiled to a legitimate identity ideally should reach beyond the family sphere. McNeill (2006) has also argued that the odds of sustained desistance improve when ex-offenders develop social links with people in different social hierarchies because it enables them to access wider social resources. Getting jobs, taking up new hobbies and being exposed to new experiences assist desisters in 'moving on' and building a new life, rather than merely existing (Healy, 2010;Laub and Sampson, 2003;Weaver, 2013). Such experiences also mean that the 'new' identity is acted out and affirmed by a wider audience and identity desistance consolidated at a deeper level. To achieve this, desisters need access to social and bridging capital (Putnam, 2000), to move beyond the micro-level which can be provided by meso-brokers helping individuals to find in-routes and pathways to break through to the meso-level.
However, the suggestions above place all the responsibility for breaking through to the meso-level with the ex-offender and those close to them. In order to make access to the meso-level and beyond available to all desisters (e.g. access to employment and active citizenship), a cognitive transformation about 'ex-offenders' is required within society. Reformation of the Rehabilitation of the Offenders Act 1974 would be one step forward, so that where appropriate, those who have offended and fulfilled their sentence can actually move on with their lives and are no longer second class citizens in the job market. 2 Beyond this, there is strong evidence from countries such as Finland that a change of mindset in society led by those in power is possible and can have lasting results (Lappi-Seppala, 2006).
Employers also have a role to play and arguably some have already been exercising their own form of leadership and autonomy on the meso-level. The company Timpson actively recruits from prisons so that those coming out have support, and a job and direct access to the meso-level. Furthermore, organizations like Recruit with Conviction are encouraging more employers to do the same (as well as trying to create a societal change on the macrolevel in how ex-offenders are perceived and treated). This is an example of those from above reaching down to those below to give them a step up, thereby acting as meso-brokers.
To tackle the pain of isolation and the lack of micro-level relational desistance in the current context of social inequality and injustice, sharing the difficult journey of desistance might be another way forward. If desisters connect with others who have the same hopes and ambitions, and who have to overcome the same obstacles, the pain of isolation will be lessened and identity desistance may be better sustained. The idea of 'mutual aid' or co-desistance is fairly new, and undoubtedly will have opponents who deem it a risk. However, it can and has been successful (Nugent, 2015;Weaver, 2013Weaver, , 2015Weaver and McCulloch, 2012). By believing in and supporting each other, desisters might play a valuable role in helping each other towards a new identity. This article has brought together two studies of desistance: one with young people who had a relatively short career in offending and one with older men who were on licence after long-term imprisonment. The narrative methods adopted allowed for a valuable in-depth examination of the challenges and pains of desistance endured, with most participants becoming increasingly isolated, lonely, lacking in hope and losing sight of a happy ending. Indeed, it is almost difficult to rationalize participants' continued attempts at 'making good' in the absence of any rewards. We hope that this article will open up dialogue about the destinations we want for those who are travelling the road of desistance, and how these can best be supported.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: Briege Nugent'