The Dynamics of Desistance

Charting Pathways Through Change

Deirdre Healy

• Publisher: Willan
• Year: 2017
• Print ISBN: 9781843927839
• eBook ISBN: 9781351544986
• Topic: Ciencias sociales
• Subtopic: Criminología

It is well-established that the majority of youth offenders cease to commit crime in early adulthood, but the mechanisms behind the shift from a criminal to a conventional lifestyle are not fully understood. The Dynamics of Desistance aims to contribute to this nascent area of inquiry by providing a phenomenological account of the psychosocial processes involved in desistance from crime.

Drawing on a variety of methods, including in-depth interviews with repeat offenders and their probation officers, police records and psychometric scores, this book charts the early stages of a journey taken by individuals who exist in the liminal space 'betwixt and between' crime and convention. A combination of quantitative and qualitative analysis is used to explore the shifts that occur in desisters' minds and lives as they make the often turbulent transition to a crime-free life, and the dynamic processes that occur at this psychosocial boundary are described.

The theoretical and practical implications of the findings in this book are explored in relation to key issues in desistance literature, and as such this book provides a key resource for academics and students working with the area of probation, as well as practitioners in involved in probation, social work and parole supervision.

Chapter 1

Desistance and reintegration

That the majority of offenders eventually terminate their criminal careers is a well-established finding in criminal justice research. For a long time, this phenomenon was dismissed as a function of maturation or ‘ageing out’ of crime (see Glueck and Glueck 1940; Gottfredson and Hirschi 1990) and little was known about the processes underlying it. As Maruna (1999: 1) concluded, ‘few phenomena in criminology are as widely acknowledged and as poorly understood as desistance from crime.’ Recently, this neglected and potentially valuable area of research has begun to receive increasing attention and there has been a renewed interest in studying the mechanisms, contexts and individual characteristics that are associated with reducing or stopping offending.

This book aims to contribute to this nascent area of enquiry by providing a phenomenological account of the psychosocial processes involved in desistance. It focuses on a number of key questions that have not yet been fully explored in desistance research. What prompts prolific offenders to periodically cease their criminal activity? Are there different factors involved in the onset and maintenance of desistance? What impact, if any, does desistance have on the minds and lives of ex-offenders? And finally, can probation supervision support individual efforts to change? Before addressing these questions, it is necessary to examine what existing theoretical perspectives and empirical findings reveal about desistance. The following review draws on a broad literature, including research into risk, recidivism and rehabilitation, in order to provide additional insights into the process of change.

Crime across the life course

Criminal career research has provided a wealth of information about pathways through crime. One of the best known and most methodologically stringent studies to explore the progression of criminal careers was initiated at Cambridge University during the 1960s (see Farrington et al. 2006). The Cambridge Study, as it became known, consists of a longitudinal survey of 411 eight-year-old boys from a working class area in London. The study, which is ongoing, uses a combination of self-report surveys, interviews and police records to measure offending and its correlates over time. Various techniques were used to minimise the loss of participants, resulting in an exceptionally good response rate, with over 90 per cent of the sample re-interviewed at age 48.

By that time, 41 per cent of the men had acquired at least one conviction and 93 per cent self-reported committing at least one offence during their lifetime. The study identified clear differences that reliably distinguished between individuals who offend and those who do not. Children who were most at risk of later offending were impulsive, lacked good parental supervision, experienced socio-economic deprivation, performed poorly at school, and had a family member who engaged in crime. Many of these differences began to emerge when the participants were children, and their influence remained evident well into adulthood. In particular, conduct problems in childhood predicted offending and antisocial behaviour in adulthood. As a result of these findings, Farrington (1997: 362) described involvement in crime as a feature of ‘a larger syndrome of antisocial behaviour’ that originates early in life and is evident over the life course. He observed, ‘typically, a career of childhood anti-social behaviour leads to a criminal career, which often coincides with a career of teenage anti-social behaviour and leads to a career of adult anti-social behaviour’ (ibid).

In the USA, three longitudinal studies of delinquency involving over 4,000 participants largely confirmed the findings of the Cambridge Study. The studies identified a similar set of risk factors for crime, which included individual (such as intelligence, hyperactivity, gender), family (such as poor parental supervision), and structural (such as low socio-economic status, parental unemployment) factors (e.g. Browning and Loeber 1999). These factors had a cumulative impact on offending; that is, the higher the number of risk factors, the higher the probability that the individual would engage in crime. In England and Wales, a survey of almost 5,000 youths aged 12 to 30 identified drug use, problems in school and the absence of pro-social bonds as the most important risk factors for offending among males (Flood-Page et al. 2000). Again, these factors were found to have a cumulative effect on offending. Over half of males who had at least four risk factors were involved in persistent or serious crime at the time of the survey. Of course, this also means that just under half were not committing serious crime despite being at high risk, suggesting that the criminal pathway is not inevitable. Taken together, these studies suggest that certain demographic, behavioural, social and psychological factors increase the likelihood that someone will engage in crime but their effects are not inescapable. As with many social phenomena, pathways to crime are rarely clear-cut and adult criminality is not necessarily presaged by childhood experiences.

The idea that certain factors are strong predictors of reoffending suggests a somewhat deterministic view of criminality, where individuals possessing certain characteristics are almost fated to play out a particular destiny. In reality, there is always the potential for change as well as continuity in human behaviour. Longitudinal research on criminal careers has revealed that changes in offending trajectories are possible and frequently occur. In Rutter’s (1990) longitudinal study, children at risk often managed to circumvent troubled pathways if certain protective contingencies were present. These factors exerted a protective influence, shielding them from the damaging effects of early difficulties. For example, children who liked school tended to plan their futures better, and this was related to improved outcomes later in life, particularly in the areas of marriage and employment.

Furthermore, while a large proportion of adolescents engage in delinquent behaviour, few become involved in serious or sustained criminal activity. When Kyvsgaard (2003) examined the criminal careers of Danish offenders, she found that 6 in 10 desisted after their first conviction regardless of age, gender and employment status. In all, estimates suggest that fewer than 10 per cent of men become involved in serious or persistent offending although those that do are often highly prolific offenders (see Flood-Page et al. 2000). In the Cambridge Study, 7 per cent of the participants were responsible for more than half of all reported crime (Farrington et al. 2006). Nevertheless, even chronic or persistent offenders eventually move away from crime. In a longitudinal study of desistance among young offenders, Ezell and Cohen (2005) found that, despite evidence of continuity, there was variability in criminal activity among even the most persistent offenders. They described the juxtaposition of stability and change in the criminal career as the ‘paradox of persistence’ (Ezell and Cohen 2005: 6).

Longitudinal studies reveal that criminal activity tends to peak in adolescence and then tapers off in young adulthood. This phenomenon, known as ‘the age-crime curve’, shows that by the age of 28, most offenders have ceased to be involved in crime (see Farrington et al. 2006). In other words, desistance from crime is the norm rather than the exception. In fact, once they stop, many ex-offenders manage to create conventional lifestyles that are often indistinguishable from those of non-offenders. In the Cambridge Study, a ‘life success’ indicator was created to measure success in key life domains, such as accommodation, relationships, employment, alcohol and drug use, and criminality. The sample was divided into three groups: desisters (who had received no further convictions after the age of 21), persisters (who were convicted before and after the age of 21) and a group who were never convicted. Overall, although persisters had achieved the least life success, the majority of participants in all three groups were considered successful by the age of 48. There was very little difference between desisters and the never-convicted group, suggesting that desisters in particular had transcended the negative effects of criminality to lead successful conventional lives (but see MacDonald and Marsh (2005) for an alternative perspective on life success among ex-offenders).

Some dispute the notion that offenders can desist and have sought alternative ways to explain attrition in criminal careers. Adopting a developmental perspective, Moffitt (1993) attempted to account for observed heterogeneity in criminal careers by differentiating between two groups of offenders: ‘life-course persistent’ and ‘adolescent-limited’ offenders. She observed that ‘life-course persistent’ offenders were more likely to begin their criminal careers at an earlier age, engage in a variety of offences, and terminate their offending later than the ‘adolescence-limited’ group. ‘Adolescent-limiteds’ were mostly involved in delinquency during adolescence and some antisocial behaviours during early adulthood (for example, alcohol and drug use), whereas anti-social behaviour among the ‘life-course persistent’ group was fairly stable from childhood to adulthood and, for them, continuity was more likely than change. According to Moffitt, the ‘life-course persistent’ group suffered from ‘cumulative continuity’ where the psychological and social dysfunction they experienced in childhood was amplified over time. Unlike chronic offenders, ‘adolescent-limiteds’ had adequate social skills and resources. They committed crime during adolescence to obtain access to coveted adult behaviours and lifestyles. As they aged, they began to use their resources and skills to find conventional ways to obtain the things they had previously acquired through crime. They subsequently re-evaluated their view of crime and came to regard it as a threat to their legitimately acquired lifestyles and, as a result, stopped offending. Essentially, Moffitt proposed that offending patterns were stable for just a minority of offenders and that most of the people who became involved in crime engaged in relatively minor offending, which petered out as they reached adulthood. Her conclusions are not supported by longitudinal research, which suggests that most offenders have the capacity to change their criminal trajectories.

Although there is evidence for continuity in behaviour between childhood and adulthood, ‘the chain of adversity is far from inevitable’ (Rutter et al. 1998: 31). While early criminal career theorists emphasised childhood experiences as the primary determinants of adult criminal behaviour, it is now recognised that experiences throughout the life course exert an equally important influence on development. Santrock (1999: 9) characterised the life span as ‘lifelong, multi-dimensional, plastic, historically embedded, multi-disciplinary and contextual’. At birth, an individual is placed on a trajectory or path. This pathway may be interrupted by transitional experiences, positive or negative, and these can initiate turning points that alter the current life direction. While childhood risk factors show a strong and consistent relationship with crime, adult life events often exert a more important influence on behaviour (see Laub and Sampson 2003). While there is variability in the timing, nature and celerity of desistance, virtually all offenders cease to offend.

Different shades of grey

Pathways to desistance are rarely linear or straightforward but are better described as tumultuous, dynamic and uncertain. Desistance does not occur as an abrupt termination of offending but usually consists of a gradual reduction in the frequency, severity and versatility of offending (see Loeber and LeBlanc 1998). Even then, desistance is rarely absolute and many ex-offenders continue to engage in some form of antisocial activity. Nagin, Farrington and Moffitt (1995) found that some young men who had not been convicted since adolescence and were officially classified as having terminated their criminality, continued to participate in more socially acceptable forms of antisocial behaviour such as brawling, drinking and drug use. Nagin et al. (1995: 132), who endorsed the central tenets of social control theory, described this as ‘circumscribed deviance’. They suggested that these offenders avoided crimes with a high risk of conviction to escape the possibility of losing work and family ties. Interestingly, their employment prospects and spousal relationships were similar to those of the group that had never been convicted, suggesting that they had superseded the negative consequences of their earlier brushes with the law.

Leibrich (1993) interviewed 35 people classed as desisters (i.e. who had no official police contact recorded against them since completing a probation sentence 3 years earlier) and found that almost half were still involved in crime. Not all were engaged in the type of crime they had previously committed, and some continued to commit minor offences such as small-scale shoplifting. To further complicate matters, many of them regarded themselves as going straight and described themselves as desisting. Leibrich (1993: 1) concluded that the pathway away from crime is ‘curved’. She introduced the notion of improvement, where people were offending less frequently or less seriously. She noted the implications of her findings for policy, observing that ‘decisions will continue to be made on black and white figures, whereas it is the shades of grey that matter’ (Leibrich 1993: 42).

The distinction between persistence and desistance is frequently blurred and, for the majority of offenders, their identification with both categories is fluid. The offender ‘transiently exists in a limbo between convention and crime, responding in turn to the demands of each, flirting with one, now the other but postponing commitment’ (Matza 1964: 28). They belong to neither world and reside in both, depending on prevailing circumstances. It seems that criminal activity is intermittent rather than constant. Greenwood et al. (1978) discovered, contrary to their expectations, that persistent offenders actually committed relatively little crime. Overall, prisoners in their study reported committing violent offences at a rate of 4.6 per year and property crimes at a rate of 18 per year. Their analysis led Greenwood et al. (1978: 7) to conclude: ‘the picture that begins to emerge from these empirical data is not one of a large number of dedicated criminals consistently pursuing a pattern of serious crime. Rather, the majority of incarcerated offenders appear to commit serious crimes at relatively low rates.’

Periods of temporary desistance in criminal careers are relatively common. Despite their prevalence, the ‘zig zag’ path of criminal careers (Glaser 1964: 57) has been subject to little theoretical or empirical scrutiny to date (but see Piquero 2004). One theoretical perspective offers a potential way forward. Maruna and Farrall (2004) proposed a useful distinction between two stages in the process of reform: primary desistance, defined as any hiatus in criminal activity; and secondary desistance, a more long-term process, which results in the reframing of personal identity into a new conventional self (see also Maruna, Immarigeon and LeBel 2004). Secondary desistance is treated as a long-term process during which individuals must gradually develop a new sense of identity as a non-offender. The process is complete when this new, non-criminal identity is internalised. In the language of social learning theory, these concepts might correspond with the distinction between the acquisition of behavioural change and its maintenance (Bandura 1977). While secondary desistance has generated much empirical interest in recent years, little is known about the processes involved in primary desistance.

The fact that pathways to desistance are not linear or straightforward creates difficulties for defining and measuring of desistance. There is currently no agreed approach. Desistance is variously regarded as an event (e.g. Weitekamp and Kerner 1994) or a process (e.g. Maruna 2001). Laub and Sampson (2001: 11) tried to resolve the dilemma by differentiating between the event of termination of a criminal career, that is, ‘the time at which criminal activity stops’, and the process of desistance, that is, ‘the causal process that supports the termination of offending’, claiming that both need to be studied. There is also disagreement among researchers about how long an offender must be crime-free before being considered a ‘desister’. As already noted, many offenders experience lulls in their criminal activity, and it is important to be able to distinguish between ‘true’ and temporary desistance. Some researchers have argued that a year is sufficient to determine whether desistance has occurred (e.g. Graham and Bowling 1995), while others have observed that ‘the true age of desistance can be determined with certainty only after offenders die’ (Farrington 1997: 373). In the evaluation literature, it is generally accepted that a follow-up period of 2 years captures the majority of reconvictions, except in the case of sex offenders, who tend to take longer to reoffend (e.g. Cann, Falshaw and Friendship 2004).

The absence of a standardised definition of desistance means that research may generate inconsistent findings and make it difficult to draw accurate conclusions (see Kazemian 2007). Dichotomous measures may obscure the important changes in offending behaviour that occur as offenders progress towards reform. A purely quantitative approach may miss some of the important subjective changes that accompany behavioural change. A limited follow-up period may not capture all episodes of reoffending. These issues have elicited increasingly sophisticated methodological responses, and researchers now often use advanced statistical techniques to capture nuanced changes in offending behaviour. In their longitudinal analysis of criminal career trajectories, Bushway et al. (2001: 500) defined desistance as ‘the process of reduction in the rate of offending from a non-zero level to a stable rate empirically indistinguishable from zero’. This definition was based on findings from Laub, Nagin and Sampson’s (1998) longitudinal study, which described two groups as desisters because their rate of offending was so low as to be indistinguishable from that of non-offenders.

Bottoms et al. (2004: 383) hypothesised, ‘rather than a steady progression, we suspect that, whilst moving generally in a conformist direction, people oscillate on what we might visualise as a dimension, or continuum, between criminality and conformity. On such a continuum, complete criminality and complete conformity are for the vast majority, points never likely to be reached.’ It is important to be cognisant of these issues when attempting to research desistance.

Understanding desistance

Desistance research is fairly new and, although several models have been advanced to account for the termination of criminal activity, few have been subject to stringent empirical testing and several key questions remain unanswered. There is, for example, disagreement about the roles played by the various factors in the desistance process. Early explanatory models fell into three broad categories: those that attributed desistance to maturation, those that prioritised informal social control, and those that stressed the importance of personal agency (see Maruna 1999). Most researchers now endorse an interactive perspective, where maturation, agency and structural factors are seen to play a role.

Adult transitions, self-control and maturation

One of the best known and most stringently tested models of desistance was proposed by Laub and Sampson (2003). They reanalysed longitudinal data on 500 delinquent boys who had been incarcerated in a reformatory school during the 1940s. The authors subsequently contacted and reinterviewed 52 of the original participants, who were by then aged 70. Over their lifetimes, the men’s criminal careers were characterised by both stability and change. While childhood events and individual characteristics encouraged stability in offending behaviour, adult life events were capable of altering their criminal trajectories in either a positive or negative direction. Two factors consistently emerged as having strong correlations with desistance – employment and marriage. However, it was not employment or marriage alone that facilitated desistance, but it was their strength and quality that was important. Only jobs and relationships that the men had invested in emotionally were capable of prompting desistance. The authors proposed an age-graded theory of informal social control to account for these findings. In their view, desistance...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Table of Contents
• List of tables and figures
• General Editor’s Introduction
• Preface
• 1 Desistance and reintegration
• 2 Issues and challenges
• 3 Person and place
• 4 Thinking, attitudes and social circumstances
• 5 Multiple roads to desistance
• 6 Into the crucible
• 7 A catalyst for change?
• 8 Looking forward
• 9 Betwixt and between
• Appendix
• References
• Index