Abstract
Youth justice in England and Wales has seen the increasing adoption of desistance thinking in recent years. There has been scarce academic debate of this development, despite the desistance evidence base focussing on adult pathways away from crime. This article examines the theorisation and application of desistance thinking with children, centring on the experiences and narratives of four ‘groups’ involved in the formal youth justice system in England and Wales, across two empirical studies. It challenges previous scholarship that denies the relevance of desistance theories to under-18s, arguing for progressive desistance practice that prioritises children’s healthy long-term development.
. . . we disavow the term and the concept of desistance when applied to children. One can only desist after having developed an adult identity (in this instance, as a so-called ‘offender’) and having evidenced a long-term trajectory of offending behaviour. Children are still growing, physically, mentally, emotionally for the first time so they cannot desist, they can only ‘become’ (Case and Haines, 2020: 12).
Introduction
Buried in the footnotes of a recent Youth Justice article was this thought-provoking statement on the applicability of desistance thinking to children, by Stephen Case and Kevin Haines. It is implicitly a critique of the increasing adoption of desistance thinking in youth justice policy in England and Wales over the past 7 years.
There has been scarce academic discussion of this move, despite the desistance knowledge base being based largely on research with adults. 1 Although there is a limited, albeit developing international knowledge base about the desistance pathways of children, these studies have typically focussed only on children understood as ‘serious’ or ‘persistent’ offenders (Halsey and Deegan, 2015; Johns et al., 2017; McMahon and Jump, 2018). Understandably perhaps, some scholars have questioned the theoretical and policy relevance of studying desistance among low-rate adolescent ‘offenders’, since such activity is normative (Laub and Sampson, 2001). However, with desistance thinking increasingly applied across the spectrum of youth justice sanctions in England and Wales, it has become necessary to question and understand the relevance of desistance thinking with a much broader group of children.
This article accordingly seeks to examine three interrelated questions: first, given that most desistance research is focussed on adults, what is the relevance of the concept to under-18s?; second, does desistance presuppose an offender identity?; and, third, if so, what does this mean for children?
In what follows, we begin with a brief overview of recent policy developments regarding the implementation of desistance thinking in youth justice in England and Wales. We then briefly discuss the criminological literature on maturation and age, as well as identity. Next, we outline the methodologies of the two studies on which our arguments are based. From here, we turn to the substantive findings of the article, which are split into four sections, each exploring the experiences of a different group of youth justice-involved children: the ‘conformists’, ‘uncertains’, ‘shifters’ and ‘survivors’. The article concludes that, contrary to Case and Haines’ ‘disavowal’ of the concept of desistance for children, desistance thinking has much to offer youth justice. However, our findings suggest that the application of desistance to children is context-dependent, meaning quite different things for different children. While our focus is England and Wales – in reflection of the location of the two research studies around which the article is organised, and the policy context with which we are most familiar – desistance terminology is employed in many other youth justice jurisdictions and thus the findings are likely to have international relevance (for a discussion of desistance and youth justice in other contexts, see, e.g. Bugnon, 2015; Droppelmann, 2017; Haigh, 2009; McAra and McVie, 2022; Mulvey and Schubert, 2016; Panuccio et al., 2012).
Desistance and Children – in Policy and Research
In policy
Over the past 7 years or so, desistance language and research has been manifest in youth justice policy in England and Wales. The Youth Justice Board’s (YJB) introduction of the revised youth justice assessment tool, AssetPlus, from 2015, was ostensibly intended to mark the departure from risk to desistance thinking. While its predecessor, Asset, was grounded in the risk paradigm, AssetPlus draws explicitly on desistance research by placing ‘more emphasis on strengths and on factors which support/hinder desistance from offending’ (Baker, 2014: 6). At the same time, it should be acknowledged that the new framework did not purport to eschew risk, which it was argued ‘should not be the only focus of work with young people but neither can it be completely ignored’ (Baker, 2014: 5). Moreover, reservations remain about the extent to which AssetPlus has led to practice that is informed by desistance thinking (see, e.g. Hampson, 2018). Nevertheless, the latest iteration of the national standards for youth justice further reflects this desistance orientation, asserting as a key aim that agencies ‘Build on children’s individual strengths and capabilities as a means of developing a pro-social identity for sustainable desistance from crime . . ..’ (Ministry of Justice and YJB, 2019: 6). Parallel developments are evident at His Majesty’s Inspectorate (HMI) of Probation (2016, 2021b), with desistance thinking threaded throughout its youth offending team (YOT) inspection framework, following a thematic inspection of desistance and children. However, some commentators have noted the continued dominance of risk-focussed thinking in policy (Day, 2023; Hampson, 2023).
When desistance terminology was first introduced to youth justice in England and Wales, HMI Probation and the YJB notably articulated slightly different conceptions of desistance thinking. HMI Probation’s framework emphasised relationship-based and strengths-oriented approaches as key facets of desistance (HMI Probation, 2016, 2021a). In contrast, the YJB centred on the importance of self-identity change. For example, YJB resettlement guidance states that ‘the aim of resettlement should be to support a child to shift their identity from pro offending to pro social’ and encourages practitioners to consider the value of this approach with all youth justice-involved children (YJB, 2018: 3; authors’ emphasis).
In the interim period, there has been an evolution towards a more nuanced focus on ‘identity development’ rather than ‘identity shifts’ at the YJB, in line with the adoption of ‘Child First’ as a guiding principle for policy and practice in 2018. It should be recognised here that the recent YJB (2022b) case management guidance also gives significant attention to the professional relationship, with practitioners entreated to see it as ‘the basis of any work with children’. It is notable that the relationship is seen as playing a crucial role in identity development: ‘the care and concern you demonstrate will contribute to them feeling valued and developing a positive pro-social identity’ (YJB, 2022b; no page). This again reflects the Child First principle, which since it was first adopted has been elaborated to encompass four interrelated tenets: children’s best interests should be prioritised, practice should be focussed on nurturing children’s pro-social identity through supportive relationships, children’s active participation and collaboration in work should be encouraged, and all work should seek to minimise children’s contact with the system (YJB, 2022a).
While they do not explicitly refer to it, the four tenets are arguably consistent with at least some interpretations of desistance thinking, albeit that two of the architects of the Child First model appear, as noted earlier, to be sceptical about the applicability of desistance to children (Case and Haines, 2020). There is, in addition, a wider question as to whether the desistance turn is reflected in practice at all, with evidence that desistance thinking is often inadequately understood and implemented through an individualistic and risk-focussed lens (Day, 2022; Gray and Smith, 2019; Hampson, 2018; Johns et al., 2017). Nevertheless, desistance thinking now increasingly infuses the youth justice policy infrastructure. Yet reflexive consideration of the questions posed earlier – regarding the relevance of desistance to children–have been almost entirely absent from this shift. So, how relevant is the desistance model to children?
Desistance and Children – in the Research Literature
Case and Haines’ (2020) rationale for rejecting desistance with children is significantly rooted in the centrality to the concept of an ‘adult (offender) identity’ and ‘long-term trajectory of offending’. While their objection to the applicability of desistance to children on this basis is somewhat tautological, it raises an important issue about how desistance is and should be conceptualised for under-18s. We contend that desistance – particularly in relation to children–can be more broadly understood as describing the process of growing out of crime (albeit recognising that this phenomenon ‘indexes a range of variables’, Maruna, 1997: 2). Ultimately, the question of whether desistance thinking and identity change apply to children is an empirical one, and not a logical one as Case and Haines imply. And as is evident from the discussion below, the application of desistance to children does not presuppose any entrenched offending behaviour or identity.
Age and Desistance
That most children who offend eventually stop is well-established as one of the basic tenets of criminology. This, of course, refers to the evidence that offending typically begins during the teenage years, peaks between the late teens and early 20s, and then rapidly decreases; otherwise known as the ‘age-crime curve’ (Graham and Bowling, 1995; McAra and McVie, 2017). Life-course criminological research has shown that a significant proportion of the incline in the curve is accounted for by Terrie Moffitt’s (1993) famously termed ‘adolescence-limited offenders’. These comprise the majority of juvenile ‘offenders’, with offending beginning in adolescence and generally stopping by early adulthood, with the prevalence of such behaviour during this stage ‘so great that it is normative rather than abnormal’ (Moffitt, 1993: 692). In contrast, exists the so-called ‘life-course-persistent’ group, which is much smaller in number, and characterised by early onset and long-term offending (Moffitt, 1993).
While many subsequent studies have provided partial support for Moffitt’s thesis, most have identified a greater number of groups, 2 including a childhood-limited group (Odgers et al., 2007) and multiple, different adolescent-limited groups (McVie, 2004) (see also Ezell and Cohen, 2004). This fits with evidence that there are varied individual pathways within the general age-crime curve, such that, some children, with girls often a particular case in point, 3 will desist earlier than the usual down-slope in late adolescence/early adulthood (and others far later, for discussion, see Shapland and Bottoms, 2017). Thus, ‘desistance from delinquency can best be conceptualised as a process that takes places between childhood and adulthood, and that is not solely taking place during late adolescence/ early adulthood’ (Loeber et al., 2016: 89).
Early empirical research and theorising about desistance understood the phenomenon as an inherently natural process correlated with aging (Gottfredson and Hirschi, 1990) and the associated physical, physiological and psychological maturational processes (see, e.g. Glueck and Glueck, 1974). Later research sought to move beyond age and its correlates as the explanation for desistance, focussing instead on the relevance of social context to pathways away from offending. Nonetheless, what might be termed socio-structural maturation (Bateman, 2020) remains a thread in these explanations too, insofar as they focus on the relevance and meaning of conventional social bonds that are associated with the transition to and realisation of adulthood, such as movement away from ‘delinquent’ peers, family formation, and employment (Laub and Sampson, 2003).
Maturation continues to be seen as a key facet of desistance – one of the three points of McNeill’s (2003: 151) ‘desistance triangle’, alongside social bonds and identity/agency. There has been renewed interest in the concept of maturation in recent years, in recognition of growing research evidence demonstrating its significance, particularly as regards the correlation between neurodevelopment and desistance (Mulvey and Schubert, 2016; see also Bottoms and Shapland, 2016; Rocque, 2015). That said, the adverse effects of de-industrialisation together with wider socio-structural inequalities, means that the opportunities to transition to adulthood (e.g. finding work) will be more limited for some children, particularly those from deprived and minority ethnic backgrounds (Nugent and McNeill, 2016). This highlights the importance of ensuring children’s entitlements 4 are met, such that, they have genuine equality of opportunity and outcome (Williams and Daniel, 2021).
Our argument here is that given that much of desistance thinking relates to processes of maturation, to deny its application to children altogether is illogical. Nevertheless, there remains an important question of whether and, if so, how identity change – a key tenet of contemporary desistance thinking with adults–is relevant to under-18s, especially given that most children’s offending ‘careers’ are limited to adolescence.
Identity, Agency and Desistance
Over the past 20 years or so, a growing body of scholarship has established that the development of a non-offending identity is central to adult desistance. Desistance theorists, nevertheless, have different views about the nature of identity change in the desistance process. Maruna’s (2001: 87) Liverpool Desistance Study with adult ex-prisoners emphasises ‘more self-reconstruction than amputation’ of one’s offending past: desisters exhibited ‘redemption scripts’, viewing themselves as essentially good persons, empowered to realise their true-self, usually ‘with the help of some outside force, someone who ‘believed in’’ them. Other theories suggest that individuals undergo a clear rupture from their previous ‘criminal identity’. Paternoster and Bushway (2009: 1106–1133) propose that desistance requires that one’s old identity is ‘cast[s] off’. By their account, desistance is motivated by ‘a crystallisation of discontent’ with offending, together with a desire to avoid a ‘feared self’, and an envisioned positive self (Paternoster and Bushway, 2009: 1133). Grounded in adult women’s narratives of desistance, Giordano et al.’s (2002) theory of cognitive transformation places greater emphasis on the interplay with the individual’s social context. In this view, desistance grows from a cognitive shift towards openness to change; exposure to ‘hooks for change’, which have subjective meaning to the individual; the envisioning of a conventional ‘replacement self’; and a changed (unfavourable) perspective of offending (Giordano et al., 2002: 1001).
Yet the above empirical research focusses on adults in and leaving custody, with extensive and often ‘serious’ recent criminal histories. Although studies with probationers have similarly attested to the role of self-identity change in desistance processes (King, 2013), with several exceptions (Farrall et al., 2014; Goodwin, 2022) few have explicitly considered this in relation to the diversity of adults within this ‘population’. Goodwin’s (2022) micro-longitudinal Sheffield-based study with 15 desisting adult women found that identity concerns were largely absent in their narratives, arguing that perhaps the centrality of identity change in desistance has been overstated by the focus on high-frequency adult male desisters. Farrall et al.’s (2014) longitudinal research with 199 probationers notes the significant minority of single-conviction ‘offenders’ who desisted ‘relatively quickly’, with little need for support, which they liken to the ‘drifters’ of Matza’s (1990 (1964)) work whose offending is transient and uncommitted. Also of note is Shapland and Bottoms’ (2011) mixed-methods study with 113 repeat offending young adult males in Sheffield; the majority did not perceive themselves as an offender, but importantly many acknowledged this was how they were regarded by others, for which they often felt shame and regret. The authors nonetheless observe that young men’s efforts to live in line with their identity and conformist values entailed agency/cognitive transformation, in line with Giordano et al.’s (2002) theory. This indicates that desistance does not presuppose an offender identity, but cognitive change – and being no longer regarded by others as an offender – remains of particular relevance.
However, this body of work is drawn from research with adults. This raises several questions about the relevance of such work to children. Do they have ‘offending identities’ and ‘feared selves’? Does the formational stage of identity during childhood mean that they have greater susceptibility to thinking of themselves as ‘offenders’? Or are such labels likely to be more transient during this life stage? What is the relationship between self-identity and agentic change for children who are by dint of their age often restricted in their capacity to exercise agency?
A small number of studies have considered children’s desistance, although few report on the role of identity change and rarely in any detail. Research findings with children whose offending has been officially categorised as ‘prolific’ and/or ‘serious’ commonly align with the ‘adult’ literature perspective regarding the centrality of identity change to desistance (Halsey and Deegan, 2015; Johns et al., 2017; McMahon and Jump, 2018; Nugent and Barnes, 2013). For example, the six desisters in McMahon and Jump’s (2018: 9) English study of 21 ‘persistent and serious’ offending YOT-involved boys aged 13–17 ‘explicitly discussed shedding “offending identities” and developing new, pro-social non-criminal identities’. However, other studies with arguably more diverse samples suggest a broader spectrum of experience as regards to the role and nature of identity shifts in children’s desistance (Beyond Youth Custody (BYC), 2017; Haigh, 2009; Rajah et al., 2014). An English interview-based study with 100 young custody-leavers aged 14–25 found that a minority described a ‘very clear’ shift to a ‘new sense of self’, yet many were ‘resistant’ to the idea of identity change. However, participants often articulated the importance of ‘a crystallisation of discontent’ and ‘imagined future selves’ (Paternoster and Bushway, 2009) in the desistance process, indicating that cognitive change played some part (BYC, 2017: 18–23; see also Haigh, 2009). Rajah et al.’s (2014) US-based short longitudinal study with adolescent custody-leavers notably found that their narratives shifted in response to life circumstances, describing a new ‘positive’ sense of self and outlook on release, which had developed into ‘subversive’ accounts 6 months later, in response to the stigma and challenges faced. It is notable that most of the above studies have been with children leaving custody, and thus the desistance experiences of the wider youth-justice-involved population remain little understood.
Methodology
The data presented in this article are drawn from two studies based in England and Wales in which the authors were, respectively, involved. Neither of the studies was designed to explore the relevance of desistance to children, but they provide useful data for informing thinking about this question. The first project was a 17-month ethnographic study undertaken in two English YOTs, between December 2014 and April 2016 by the first author (Wigzell, 2020). This sought to examine the meaning and helpfulness of youth justice supervision. While the study involved over 700-hours observation of everyday YOT practice, and interviews with 46 YOT workers, this article draws primarily on the data gathered through semi-structured interviews with 25 children, analysis of the children’s case files and the ‘following’ 5 of 11 of the children during the supervisory journeys. For this study, the first author employed a purposive sampling strategy that aimed to capture the diversity of children involved with the YOT on community sentences, in terms of background, characteristics, as well as sentence type and length. The children ranged in age from 13 to 18 years; most were male (24 of 25), the majority were White (16 of 25), with just over a third from minoritised ethnic communities (nine of 25) and three fifths were undertaking community orders of 12 months or more (15 of 25). Data were analysed using an adaptive theory approach (Layder, 1998), involving the use of orienting concepts from existing theory and literature, as well as inductively searching for new meaning within the data.
The second project was a mixed-methods study, in which the second author was involved, which explored the relationship between care and youth imprisonment, by comparing the experiences of a sample of looked-after and non-looked-after children who had been incarcerated (Day, Bateman and Pitts, 2020). For the purposes of this article, we will be drawing on only the qualitative data gathered in this study. This comprised semi-structured interviews with 48 children in or who had recently been released from custody; analysis of 45 of the children’s case files; and semi-structured interviews with 19 professionals. All the children interviewed were male, the majority were aged 17 years or above (34 of 48), nearly half were looked after at the point of sentence (22 of 48), 6 one third were from minoritised communities (16 of 48) and most (39 of 48) were imprisoned in a young offender institution. It was evident from interview data that most children had been involved in offences of violence, robbery and domestic burglary. The sample was thus broadly representative of the youth justice custodial population.
Findings
In what follows, we examine the experiences and narratives of four ‘groups’ of children involved in the formal youth justice system across the two studies: the ‘conformists’; ‘uncertains’; ‘shifters’ and ‘survivors’. The first three groups are based only on data from study 1, while the last group is principally based on data from study 2. We have adopted this categorisation for two related reasons: first, to demonstrate the range of starting points evident among children and, accordingly, the different types of desistance approaches that are likely to be helpful in youth justice; second, to highlight the widely differing circumstances of children subject to youth justice intervention, which serves to challenge the bifurcation between ‘persisters’ or ‘desisters’ that is often evident in the literature. However, although this means of grouping children is a useful heuristic device, it carries a number of limitations too. Several children did not fit comfortably into any group, raising the possibility that a larger sample would have resulted in the identification of additional categories. Children may shift between groups over time; indeed at least two of the children who were ultimately categorised as ‘shifters’ would have initially been identified with the ‘survivor’ group. This highlights the benefits of even ‘short longitudinal’ approaches that are better able to capture such developments than cross-sectional studies. Perhaps, most importantly, we are not suggesting that there are distinct groupings of children to be found in youth justice according to which children can be classified, or advocating for prescribed desistance responses to their needs; our aim is to highlight the importance of a contextual, bespoke approach to each child.
The ‘conformists’
For just under a third of the sample (seven of 25), interviews and observation indicated that they had consistently demonstrated conformist self-narratives. Examples of such narratives included maintaining the importance of keeping to the law and expressing conventional future aspirations (such as a job, home and partner) (Shapland and Bottoms, 2011). These children were usually first-time entrants
7
or had received one prior out-of-court disposal, typically regarding their offences as mistakes in a largely positive record. Chris’ (aged 17) account – who had been convicted (for the first time) for criminal damage to an unoccupied house and two counts of assaulting a police on his way home from a party – is illustrative: I didn’t think I needed any support with anything because I knew what I did was wrong, and knew where I had gone wrong, but I also could understand why other people might think that I needed some, so I thought I’d just go along with it.
The offending of these children was reminiscent of Matza’s (1990 (1964)) conceptualisation of ‘juvenile delinquents’ as ‘drifters’ whose offending is almost accidental. As such, their ‘deviant’ behaviour was of an experimental or situational nature that did not appear to form part of their identity (Lemert, 1951).
However, Chris notably reflected that immaturity was a factor in his behaviour: . . . it was all fine until I was about 13, then I did misbehave a bit, I didn’t really get on with my parents, but I think it was just being a normal teenager.
In this respect, around a third of the children (nine out of 25) partly accounted for their offending as a symptom of immaturity, explaining that they had since ‘grown up’ and now recognised ‘it was just stupid’. This accords with previous research findings that individuals often describe their shift towards desistance in terms of maturation (BYC, 2017; Farrall et al., 2014; Graham and Bowling, 1995; McIvor et al., 2004; McMahon and Jump, 2018). It seems that for some adolescents, like Chris, the perceived importance of maturation in their cessation or reduction of offending is indicative of a subtle, yet important, change in outlook, which might be understood as a cognitive shift.
The uncertains
Similar to the ‘conformists’, the five children identified with this category had minimal offending histories, although all of them had received one – two out-of-court disposals prior to their current (first) court-ordered sentence. However, unlike the ‘conformists’ who consistently demonstrated conventional values and aspirations (and confidence in their ability to adhere to or realise them), the children in this grouping expressed significant uncertainty about their values, who they were and who they could be. Two of the children (Ryan and Dylan), notably, expressed visceral shame for their behaviour and subsequent criminal justice involvement (evident in interviews and case files). Ryan, aged 16, who was completing a 12—month referral order for a public order offence (threatening behaviour), described feeling particularly distressed by his initial charge with attempted robbery and the protracted nature of court proceedings, which ‘made me start smoking weed again’. He wanted help with ‘managing my anger’, but regarded the YOT’s interventions as ineffectual: ‘. . . like doing child work . . . just wasting time, it’s not going to help . . .’. The perceived illegitimacy of the YOT work and connected sense of hopelessness contributed to his depression: I don’t mean it in a strange way, but I kind of feel like shooting myself when I come to sessions [at the YOT], like once a week it makes me want to [pause] yeah . . . this makes me feel depressed.
In common with a number of the children, Ryan was accordingly unsure of what his future would hold, seeing several possible paths: I could be in college on an apprenticeship or I could be smoking weed and doing drugs, that’s the other one . . . I definitely need to improve my anger because I won’t be able to get a job if I don’t sort it out because I hit things too much.
This is redolent of Paternoster and Bushway (2009: 1115) theory that individuals who offend imagine ‘possible future selves’, which comprise ‘a blueprint for self-change to lead one toward a positive possible self and away from a feared self’. But, as importantly, Ryan’s reflections also point to the uncertainty – the ‘hopefulness and hopelessness’ (Fitzpatrick et al., 2015: 178) – with which some children view their agentic control over their lives – and the potential role of YOT supervision in reinforcing this.
Dylan had been convicted for malicious wounding and various counts of assault, for punching and seriously biting another child as part of a preceding verbal and physical altercation. His case file stated that he had been upset by the offence and subsequently questioned his values, suggesting that it was affecting his self-identity. He reoffended shortly before the end of his order and was remanded to adult prison (as he had turned 18 by this point). While some previous studies have found that shame is an important motivation for desistance (King, 2013; Leibrich, 1994), these two cases resonate with Maruna’s (2001) contention that self-blame may have adverse consequences if internalised. Although Maruna (2001) sees this as a particular risk for individuals for whom offending is a lifestyle, the fact that the boys had no prior convictions may suggest that children, given the formational stage of their identities, may be particularly prone to such internalisation. (It should be noted that fluidity of children’s identity may also mean that positive shifts can occur just as rapidly, as we later discuss.) Yet, it is perhaps no coincidence that both boys’ apparent questioning of their self-identity in light of their offences, was set against the background of the deaths of their grandfathers. As Gadd (2006) notes, loss often brings into focus one’s projections onto others, identification and questions about the self.
The shifters
The narratives of a small number of the children (four of 25) – supported by the first author’s ‘following’ of their supervisory journeys – were strongly suggestive of deliberate development of a ‘pro-social’ identity (insofar as they regarded themselves as working on being a ‘good’ person, with a positive future ahead’). The children in this ‘group’ had at least one prior court conviction, including for robbery, burglary and assault, for which they had received a Youth Rehabilitation Order (YRO). All narrated a marked positive change in outlook and behaviour following a period of precarious self-identity, which was also observed by YOT workers and the first author. The account of 17—year-old Ernest is a case in point. When we met, he was on a 12—month YRO for possession with intent to supply class A drugs, which he had reportedly been coerced into holding. Yet, this seemed to have had a profound effect on his sense of self, despite the apparent absence of primary deviance here: . . . I thought it was going to be loooong, like ‘I’m doing this for something I didn’t do, it’s going to go on my record’, so at the time . . . I just thought to give up and go out there and just become a criminal, like, forget YOT, forget the court, forget my family, forget everything . . . And then I was like, ‘naah’, there is a better side to life in [town] than this, all these things that are happening for a reason and it’s taught me a lesson, so, just try it out, be good. Then, came to YOT, saw that [worker], like, was kind of like, a happening person.
Ernest’s account, importantly, paints a picture of someone who is becoming who they are as part of the maturation process, rather than making a signature departure from a ‘pro-criminal’ identity:
So how do you think you’ve changed, say, over the past year since you’ve been involved with the YOT?
I’ve gotten more mature . . . I [have] pulled my confidence back up, moved by bad friends away, what else, got myself together, and then I’m trying to get into college and I’m trying to do good things for myself and I’m thinking positive and being optimistic, so yeah.
He goes on to express especial optimism and aspiration for the path ahead: I want a meaningful life. . .I don’t want to settle, I wanna exceed. I wanna travel the world and see more things and know more people, help people out, and do better things, and do something that will actually contribute to your life and other people’s lives.
Ernest’s words strongly resembles the redemptive style scripts of Maruna’s (2001) work, whereby ‘making good involves taking control over one’s life and using that life to contribute, accomplish something, and leave a positive legacy’ (p. 150).
As is evident, Ernest identifies a confluence of factors in his assumption of a positive identity, including his self-agency. Yet, a consistent thread is the importance of the relationship with his YOT worker, who helps him to feel ‘believed in’ (Maruna, 2001: 87): I don’t know how [YOT worker] helps me, but he just helps me. Like every time I leave [YOT worker] I just feel different, I get out of here [YOT office] and look around [town] and I feel different, all positive.
The survivors
This ‘group’ comprised children whose backgrounds were, such that, they were fairly deeply embedded in (and potentially committed to) offending as a form of survival. All of the 48 children interviewed for the second study described what could be understood as ‘strategies for survival’ in their narratives of life before, during, and following their release from custody. However, children in care were more likely to incorporate this into their identity – a ‘survivor mentality’ – with adverse consequences for desistance from crime (Day, Bateman and Pitts, 2020: 63). Prior to custody, most of the children spent significant periods of time on the street with their peers, in reflection of difficulties at home and school, often involving exclusion. Yet, for children in care, ‘tight’ (Crewe, 2011) forms of surveillance and restrictions in ‘looked-after’ placements and feeling uncared for in such environments, combined to devalue their identity and undermine a sense of control over their lives. Often these factors propelled children in care onto the streets, an environment that simultaneously provided a sense of belonging and safety that was unavailable in care settings. Such experiences often served to exacerbate their offending due to its role in maintaining status and enabling survival (Day, Bateman and Pitts, 2020: 37). The words of one child in care illustrate this well (Day, Bateman and Pitts, 2020: 42): I guess I was looking for a sense of belonging really. I didn’t really feel like I belonged anywhere . . . so I was looking for a sense of belonging, and then I got a name for myself, and I just found I had to live up to it. Because people didn’t accept me before, but because I’ve got this name now, then I had to stick with it and people accepted me by my name, not by who I was (Looked-after child, 13).
There was evidence too that children’s emphasis on self-reliance was perpetuated by their deliberate emotional detachment from staff and foster carers to defend against disappointment and hurt (‘I’d prefer people not to care because I prefer to look after myself . . . Because then you can’t get hurt can you really’ (looked-after child, 8) (Day, Bateman and Pitts, 2020: 43).
On release, children in care were more likely than those not-looked-after to be faced with a ‘window of threat’, often related to the uncertainty and unsuitability of their accommodation in lieu of family involvement, in which, once again, their primary concern was survival and the need for self-reliance became further entrenched (Day, Bateman and Pitts, 2020: 56). The authors argue that these factors made it less likely that the children in care would desist from offending, at least in the immediate term: Looked-after children were not, therefore, lacking in resilience; indeed identifying oneself as a survivor requires considerable strength and confidence in one’s own abilities. At the same time, the particular form that this resilience took required a focus on the here and now rather than facilitating a future orientation; it meant that that children were more likely to be resistant to potential offers of external support because of a fear that they would expose themselves to being let down; and it encouraged a return, on release from custody, to an environment where the children felt comfortable and a lifestyle with which they were familiar. A survivor identity thus made desistance from offending less likely in the short- term because of the perceived risks of failure associated with the changes, and potential reliance on others, that an alternative future required (Day, Bateman and Pitts, 2020: 63).
For this ‘group’ then, offending and self-reliance is a means of a means of sustaining status, a sense of safety and survival. Increased independence and self-determination is not necessarily associated with desistance. And a shift in identity is a complex and risky or, at least, a riskier undertaking that continuing as they are.
Discussion
Returning to our starting point, what can we make of Case and Haines’ (2020) disavowal of the concept of desistance to children? The fact that adult discussions of desistance often tend to assume that there is an offending identity that needs to be superseded does not imply that there is no equivalent or similar process for children. Our findings point to the differing relevance of identity shifts to children, suggesting that there are in fact different trajectories for children with different starting points, and different support required. In other words, desistance is context dependent and will mean different things for different children. In the paragraphs that follow, we consider each grouping in turn, considering the potential implications for desistance thinking and practice. Readers familiar with youth justice in England and Wales will likely identify some overlaps with Child First tenets, discussed earlier. While we acknowledge these commonalities, our deliberate focus here is on developing desistance thinking with children. This focal point reflects that desistance theories and their influence on youth justice policy in England and Wales predate the implementation of Child First (with the formal adoption of desistance with AssetPlus in 2014 (YJB, 2014 (2012))); that Child First and desistance have slightly different remits (the former focusses on children’s universal positive outcomes whereas desistance thinking centres on the reduction or cessation of offending); and hopefully allows identification of implications for desistance of children that does not rely on the particular policy context of a single jurisdiction.
At one end of the continuum, there is evidence of a group of children – the ‘conformists’ – typically first or second time entrants to the system, for whom the applicability of identity shift is tenuous. To the extent, there is a shift in identity – it is about the adoption of an identity of independence – so that, there is an interaction between identity and social status, and potentially cognitive maturation. In line with Matza’s (1990 (1964)) work, these children drift out of offending as quickly as they drifted in.
This is particularly relevant given evidence that an increasing proportion of children subject to supervision are first time entrants, more than quadrupling from 9 to 42 per cent between 2008 and 2020 (YJB/MoJ, 2020, 2021, table 10.9; Bateman and Wigzell, 2019). Added to this, there is good evidence that well over 90 per cent of children and young people self-report offending by age 24 (McAra, 2018: 5). The vast majority of those will do so fairly infrequently and make the transition to a (relatively) law-abiding adult lifestyle without difficulty. Perhaps, precisely because lawbreaking among children is much more prevalent than among adults – then desistance is pretty straightforward for the large majority. The more so, to the extent, that there is adequate mainstream support for children within the particular society (Bateman, 2020). Here, the idea to ‘leave the kids alone’ (Schur, 1973: 55) can arguably be equated with the promotion of desistance since system interaction will interfere with the natural maturation process. This is ‘not a clarion call for radical non-intervention’ but rather an invitation to consider the potential for further diversion away from the formal youth justice system, so that, as far as possible, children are offered supportive services that encourage positive behaviour outside of its stigmatising and labelling sphere (Haines and Case, 2015: 49).
Moving along the continuum, the tentative evidence for another ‘group’ of children – the uncertains – whose self-narratives were suggestive of fragile and faltering identities, despite minimal recorded offending, indicates that the relevance of identity shifts cannot be reduced to entrenched offending or the cumulative effect of criminal justice system contact. In this regard, it may be that criminal justice intervention is felt more sharply if such intervention comes on top of prior exclusionary and stigmatising experiences (e.g. social disadvantage, exclusion from school) and against a backdrop of deeper emotional turmoil (e.g. bereavement, family break-up), which can prompt questions about one’s self-identity and self-worth. The risk is that supervision (and wider criminal justice contact) may reinforce doubts about self-worth, especially if experienced as illegitimate, deficit-focussed or stigmatising. Thus for these children, perhaps the YOT’s principal task is actively to recognise and develop their incipient pro-social self-identity, rather than seeking to change it. In this context, progressive desistance practice does not insist on identity shift (unless it is in the child’s best long-term interests), but is equivalent to nurturing healthy child development, moving beyond the negative aspiration of reducing offending.
Next come the ‘shifters’, as represented by Ernest’s experiences. The narratives of this group more closely represent the ‘rhetoric of redemption’ (Maruna, 2001: 85) observed in the adult desistance literature. In contrast to the almost passive desistance journeys of the ‘conformists’, the children in this group tell stories of an active and conscious decision to embark on a conventional path and identity. Almost always, the children described this in terms of a ‘looking-glass recovery’, whereby their YOT worker ‘believed in’ them and helped them to realise their own self-worth and envisage a different life (Maruna, 2001: 96). It is through such relationships of recognition (i.e. where one feels understood and appreciated by a significant other) that one’s self-identity is formed: ‘I am not what I think I am and I am not what you think I am; I am what I think that you think I am’ (Cooley, 1902, cited in the work of Korgen and White, 2007: 58). This aligns with the evidence for the crucial importance of caring relationships in fostering hope, self-agency and providing vital socio-structural support to aid desistance (Bateman, Melrose and Brodie et al., 2013; BYC, 2017; Haigh, 2009), which may be particularly important at this life stage (Panuccio et al., 2012).
At the far end of the spectrum, are the survivors for whom identity is woven into the fabric of their continued offending. However, for these children, continued offending is a means of asserting agentic control over their lives. As Halsey et al. (2017), insightfully observe, for some, offending represents a rejection of the ‘mainstream moral order’, in favour of an alternative, more authentic ‘morality’, in which belonging, value, and safety can be more readily achieved. It is perhaps for these children in particular for whom the pains of desistance (Nugent and Schinkel, 2016), identified in the adult literature are likely to be most pressing, including the loneliness of breaking away from peers and the safety of a survivor identity. (The friendship circles of other children may present no obstacle to desistance.) Thus, in stark contrast to the first group, children whose offending is entrenched may need support to be able to mature in the same way. This will be of a practical nature, ensuring children’s ‘entitlements’ to services and rights, and responsibilising adults in this respect (Haines and Case, 2015). Extending the latter notion, as McNeill and Graham (2020: 13–14) assert, there exists a societal responsibility that involves both ‘informal social recognition’ of rehabilitation and ‘acceptance of the returning citizen’, raising critical questions about the labour market exclusion and stigma that criminalisation erects.
It is worthwhile reflecting too that both the ‘conformists’ and ‘shifters’ invoked their immaturity as a key explanation for prior offending, almost acting as a protection against a deviant identity (see also Coyle, 2019). In this regard, the ethos of responsibilisation observed in youth justice (Haines and Case, 2015; Phoenix and Kelly, 2013) potentially risks undermining desistance by understanding such neutralisation as evidence of behaviour minimisation, to be challenged, rectified and responsibilised. Alongside, it is important to recognise and nurture children’s sense of agency to shape their destinies, perhaps particularly for those approaching adulthood. As Coyle’s (2019: 1194) research with young adults highlighted, their subjective notions of developing maturity played a ‘critical, active element’ in desistance. This underlines the delicate balance to be struck in youth justice policy and practice in simultaneously fostering children’s agency and recognising its environmental constraints, through responsibilising adults (Haines and Case, 2015). As noted above, the classificatory model we have deployed in the article is not intended to suggest that children neatly fall into discrete groupings with corresponding implications for desistance; rather it is designed to illuminate the diversity of children’s circumstances, pathways and needs. Instead, we propose that the notion of progressive desistance practice, in which the emphasis is on fostering the child’s long-term healthy development, offers a constructive way forward for applying desistance to youth justice, and arguably complements the Child First philosophy.
Concluding Comments
These research findings suggest the critical importance of nuanced understanding and application of desistance thinking with children in the youth justice realm. We do not pretend to have arrived at definitive conclusions, rather our principle aim has been to open up the space for debate, so that, we may progress understanding.
At a policy and practice level, the implications of the adult-focus of the desistance evidence base for children must be more explicitly acknowledged and considered. There is an opportunity here for the YJB and HMI Probation to develop a shared understanding of desistance and its distinct and context-sensitive application to children, which informs YJB case management guidance and inspection standards. Our findings might suggest that a principle focus for youth justice practice should be caring professional relationships that provide personal and structural support, rather than on direct attempts to change and develop children’s identities. While this might, on the surface, seem to echo the YJB’s ‘Child First’ tenet of ‘building children’s pro-social identity’ through supportive relationships, we are advocating something slightly different here. The YJB’s promotion of ‘pro-social identity’, while an improvement on the original focus on the importance of identity shift, still appears premised on the idea that there is a fault in the child’s identity that needs addressing, which as our findings show, often may not be the case. Thus, our argument is that desistance practice with children should be more broadly focussed on long-term healthy development – of which identity development is just one potential part that may or may not be relevant – and this should be facilitated through caring professional relationships.
There is a strong case for further research regarding desistance and children. The focus should be on advancing knowledge about how children move away from offending; the relevance of identity shifts to children and how, if at all, this varies by biographical experience (including history of offending and stage of system involvement), as well as deepening understanding of how desistance thinking is understood and being implemented in youth justice practice.
Finally, we propose the development of a child-focussed theorisation of desistance centred on progressive desistance practice. A disavowal of desistance with children risks overlooking the clear connections and fruitfulness of the desistance knowledge base for our understanding of children’s pathways away from offending. Yet, an uncritical acceptance of the lessons from adult desistance research to youth justice risks the adultification of children. What is needed, is a conceptual framework that draws on the vast body of knowledge about adult pathways away from offending, but explicitly distinguishes the distinct developmental and psychological features, as well as the age-based and socio-structural limitations associated with western childhood.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors are grateful to the The Dawes Trust and the Nuffield Foundation who respectively funded the first and second research projects on which this article draws.
