Can Persistent Offenders Acquire Virtue?

Background to the Desistance Process

The Sheffield Desistance Study has both a quantitative and a qualitative dimension, and the interview schedules contained a mixture of ‘closed’ (tick-box) and more open questions. This mixed methodology has allowed us to include in the study a number of matters of relevance to those with an interest in ethics; so, for example, the abstract of one of our papers reports that the men in the sample had ‘surprisingly conformist values, for example with regard to future aspirations (employment, housing, etc.) and to the importance of staying within legal boundaries’, although a ‘dissonance between values and behaviour’ was frequently also observable. ⁶ That same paper includes a short section on ‘virtue ethics and the obstacle-strewn society’, ⁷ in which we identified virtue ethics as a valuable way of thinking about the processes of change that some of the men experienced, while also emphasising the many difficulties that they typically encountered when trying to change (for example, relating to employment, the pull of still-criminal friends, and the general difficulties of social reintegration for those with criminal records). The principal purpose of this article (to be addressed in its final sections) is to explain in greater depth why we have found a virtue-ethical approach to be a valuable way of conceptualising our data about processes of change in the men’s lives. Before we can address that issue, however, we must first (in the remainder of this section) say a little more about the significance of the fact that our sample consisted of young adults. Then, in the next two sections, we will offer three brief case histories that will hopefully bring to life some of the realities that we are trying to analyse; and we will also say more about the process of trying to change, which is the topic that above all else led us to an interest in virtue ethics.

In an initial, tentative, paper that we wrote about desistance (during the fieldwork stage of the Sheffield study), we suggested that that there might be merit in inverting the traditional theoretical focus of criminological studies, which has rested primarily upon ‘criminality’ and ‘becoming criminal’. Almost all children are socialised into pro-social norms, and even persistent offenders usually try to prevent their children from following in their footsteps. Therefore, people usually start from (broadly speaking) a position of conformity, and given the prevalence of desistance processes, this is where they return, usually after a criminal career lasting less than ten years. Accordingly, it might be more appropriate for criminologists ‘to start from the position whereby conformity is the anchor and the more important concept’. ⁸ That position was empirically strengthened when our later research results showed that the persistent offenders in our study had, even at the beginning of the research, predominantly conformist underlying values. ⁹

Looking at criminal careers in this way emphasises the importance of the period of adolescence and young adulthood, which is when, in most cases, criminality develops and then subsides. Persistent offending usually (although not invariably) begins in adolescence or before, ¹⁰ and those who have begun offending before puberty very often entrench their criminality in adolescence through the establishment of strong bonds with delinquent friends. ¹¹ For offenders as well as for almost everyone else, however, the subsequent period of young adulthood brings with it a process of reappraisal of one’s life. Questions are raised, such as: ‘What kind of a life do I want to lead as an adult? What kind of job will I get? Who will I share my life with? Shall I carry on with the kinds of activities I’ve been pursuing in the last couple of years?’ So it can be an unsettled time, but it is also one that often leads to significant changes in self-understanding.

A possible statistical pointer to the effects of this kind of generic life-appraisal can be gleaned from research into criminal careers by two Dutch authors. Their analysis confirmed the findings of many other researchers that specific changes in life circumstances (such as marriage or steady employment) did indeed have a crime-suppressive effect in the lives of persistent offenders. However, they also pointed out that these particular effects accounted for only a relatively small proportion of the overall reduction in offending in young adulthood. Accordingly, they concluded that ‘much of the effect age has on crime remains unexplained’ by existing research. ¹² It is a plausible hypothesis (although, of course, one that needs to be tested) that the generic self-appraisals often undertaken in young adulthood make an important contribution to the overall age effect.

When young adults who have become repeat offenders undertake self-appraisals of this kind, the gulf between their present situation and how they would like to live can be stark. Take, for example, the case of ‘John’, described in one of our previous papers. ¹³ When we first met him, at age 20, he was in a young offenders’ institution (YOI), having been convicted on nine separate occasions since a first caution at the age of 11. He had recently split up with a girlfriend; he had never had a job; his mother, whom he loved, had severed relations with him because of his offending and drug-taking (heroin and crack); and as a consequence, he had been homeless and sleeping rough, which he found both physically and emotionally difficult. Not surprisingly, when asked in the first interview to describe the good things in his life at that time, he answered ‘nowt’. However, to a further question asking him to say what kind of person he would like to be in a few years’ time, he answered very articulately: ‘Confident. Hardworking. Trustworthy. A good person to get in with’. In a later interview, he also revealed anxieties about a future that he feared, shuffling in and out of prison and having no friends. ¹⁴ From this kind of reappraisal, a determination to desist can emerge, and in John’s case it did emerge.

John was not alone. When respondents were asked the ‘desired self’ question in the first interview (‘what kind of person would you like to become?’), their answers overwhelmingly centred on normality, with the great majority of comments focusing on ‘going straight’, being a ‘family man’, ‘living a normal/regular life’ and being a ‘good person’. ¹⁵ Yet, for this sample, the road to ‘normality’ was never going to be simple, given their criminal histories, their lack of qualifications and work experience, and (in some cases) their histories of addiction. In the remainder of this article, we shall examine how they tackled this process of hoped-for change.

Case Histories and the Mechanisms of Change

In this section, we offer three varied case histories. Two of the men described claimed by the end of our research study to have desisted, while the third—despite some expressed good intentions—had relapsed into serious crime. After describing each case, we offer a brief interpretative comment. In the first two (desisting) cases, this comment focuses on the mechanisms that seem to have led to compliance in that particular case. ¹⁶ For this purpose, we will use the fourfold typology of mechanisms of compliance put forward a decade ago by one of us, ¹⁷ which can be summarised as follows:

‘Learning to Live a Non-Criminal Life’

Desistance research projects have been conducted using a variety of methodologies, and each approach has its strengths and weaknesses. In the Sheffield Desistance Study, we chose to adopt a prospective research design with a sample of (mostly) persistent young adult offenders, and this method has allowed us to focus in particular on the early stages of desistance. ²¹ From this perspective, it has become very apparent that, for those who have been deeply involved in crime, desistance is a process of learning to live a non-criminal life when one has been living a largely criminal life.

Attempts to desist usually start either with a realisation that one’s current situation is far from desirable (a ‘push’ factor, as in the cases of Len and John) or with a glimpse of a more satisfying life (a ‘pull’ factor, as in Kevin’s case). But this initial step is only the beginning of a process. The clear evidence of our study is that, if they are serious about changing, those who have been deeply involved in crime need to start making many adjustments to their patterns of daily living—for example, concerning where to live, whom to mix with, which places to visit or how to respond in particular situations of temptation. They will also often encounter obstacles, and sometimes (as in Len’s case) they will experience definite setbacks. But if desistance is successfully accomplished, gradually a different pattern of activities will take shape, which in time will become routine. Often, too—and of special interest for the purposes of this article—this will be accompanied by changes in the moral evaluation of past behaviour. An example of this occurred in John’s case. In the first interview, he said that ‘I don’t regret no crimes I’ve done …. not my victims or nothing’, but by the fourth interview his view was more nuanced. Looking back, he said that while he used to enjoy the offending lifestyle (‘money, having drugs and that’), he now felt remorse for the effects of his past actions: ‘burglaries and that—I feel sorry for the people I did it to, and people I’ve robbed’.

The important topic of ‘obstacles to desistance’ requires some elaboration. A key statistical finding of the Sheffield Study was that an index of ‘obstacles’, self-identified by the men themselves, was one of the most powerful predictors of the level of offending at the end of the study period (if more obstacles were self-identified, later offending was more likely). ²² Very interestingly, also, three of the four most frequently identified ‘obstacles to going straight’ were basically financial: (i) lack of money; (ii) ‘opportunities for easy money’; and (iii) lack of work. ²³ Employment is a critical issue; persistent offenders who are serious about desistance quite often experience a reduction in income, so if legitimate work cannot be found, the temptation to re-offend can be very high. But employment, of course, is not solely within the control of the would-be desister; hence, obstacles to desistance are at least in part a function of local economic and social conditions. This raises important ethical and political questions about the extent of society’s responsibility to assist would-be desisters to regain a role as full members of society. ²⁴ It also explains why, in the absence of social conditions facilitating employment, the availability of local ‘social capital’ (exemplified by Kevin obtaining a job with his step-father) can be so crucial to the life-chances of individuals wishing to desist. ²⁵

But obstacles can be surmounted, and what we see in the cases of both Ken and Len at the end of our study period is the apparent beginning of a habit of compliance which seems pleasing to the desisting subject. We will explore this more fully in the next two sections.

Why Virtue Ethics?

We are now in a position to consider whether, and if so how, the resources offered by the broad field of ‘virtue ethics’ might be of assistance in the social-scientific quest for a deeper understanding of desistance processes. For those committed to a strong version of the fact-value distinction, this might seem to be an unlikely possibility, given that social science is in the business of explanation (based on facts), while ethics necessarily involves value-based discourse. However, while one must certainly always pay close attention to the type of claims being made, we would argue that there can be a helpful mutual interaction between these two fields of study. Thus, for example, the ethical theorist Valerie Tiberius (to whose work we shall return) has used some findings from psychological research in her quest to develop a more ‘empirically informed [normative] ethics’: ²⁶ and, given that human beings are sometimes motivated by ethical concerns, it seems clear that work in the field of normative ethics might in principle offer some helpful insights that are relevant to the development of social scientific explanations. ²⁷

As is well known, in the first half of the twentieth century normative ethics was dominated by the divergent traditions of consequentialism (especially utilitarianism) and Kantian ethics (deontology). A series of contributions by various moral philosophers (not all of whom shared the same ultimate commitments) ²⁸ then led to a remarkable renaissance of interest in classical moral philosophy and its virtue-centred approach, in particular Aristotle’s Nicomachean Ethics. Thus, by the end of the century a major primer on normative ethics had already placed virtue ethics alongside consequentialism and Kantian ethics as one of Three Methods of Ethics. ²⁹

The Cambridge philosopher Raymond Geuss has suggested that:

one of the great merits of Bernard Williams’s work ³⁰ is to have pointed out that ‘ethics’ has traditionally meant two distinct things: (a) the attempt to say something about what the ‘good life’ for me would be, and (b) the attempt to say something general about how people should regulate their behaviour toward one another … Western philosophy begins [in classical Greece] with the attempt to show the close connections between plausible answers to the two questions: the only way for me to live a truly happy life (it is claimed) is to do so in the context of acting towards others in morally well regulated ways. ³¹

What the late twentieth century debates showed clearly was that the dominant moral philosophy of the preceding period had largely lost sight of the first of these questions (the individual’s quest for virtue), and that much would be gained by reconsidering this tradition. Exactly how one might best return to a virtue-centred approach was and is, however, philosophically a very complex issue—and one that is beyond both the scope of this article and the competence of its authors. Instead, we shall consider here three aspects of a virtue-centred approach that, we believe, are particularly relevant to the study of desistance, namely: (i) its primary focus on dispositions rather than acts; (ii) its relationship to the question of how people should act towards others (as highlighted in the quotation from Raymond Geuss, above); and (iii) its alignment of reason and desire.

Consequentialism, and some versions of Kantianism, are primarily focused on specific situations, and on the appropriate moral action within that situation (perhaps expressed as a moral obligation). For people like those in our research sample, that kind of act-specific focus is indeed sometimes highly relevant: in Zed’s case, for example, things might have turned out differently if he had declined his fellow hostel residents’ request for heroin. But over and above such specific situations, the men in the Sheffield sample were, during the time that we were studying them, repeatedly asking themselves the virtue ethicist’s more general question: ‘how should I live my life?’. ³² The study of the early stages of desistance, it seems clear, will make little progress unless social scientists recognise the importance of this question for many offenders, as a prequel to attempted change.

But secondly, we must note again Geuss’s summary of the classical position: ‘the only way for me to live a truly happy life … is to do so in the context of acting towards others in morally well regulated ways’. In the Sheffield study, the importance of this comment was repeatedly exemplified. Desistance certainly begins with the question ‘what kind of life should I lead?’, but it is not long before this evolves into a better understanding of how the subject’s life interconnects with those of others. Accordingly, we found that desisting respondents would often speak about becoming more mature, feeling a stronger sense of responsibility for others (such as partners or children), or feeling remorse about the harm they had done to past victims.

Thirdly, there is the important question of the relationship between reason and desire, which in turn raises the issue of what we mean by the term ‘virtue’. In his admirable textbook on ethics, Jonathan Jacobs has defined a fully ‘virtue-centered theory’ as one which ‘maintains that moral value is grounded in certain excellent states of character (the virtues) and that actions have moral worth through being exercises of those states of character’. ³³ He further explains that in Aristotle’s view:

Neither reason alone nor sensibility alone is adequate to explain morally relevant human action. The virtuous person enjoys acting well, finding it naturally pleasing. Given the agent’s desires, his conceptions of worth, and the way that he appreciates situations, he is concerned to do what virtue requires because that is what informs his conception of the good … He does not have to resolve conflicts between morality and desire, or morality and feeling. ³⁴

This sounds admirable, but in its pure form it also seems rather remote from the lives of most of our desisting research subjects, who did not aspire to ‘excellent states of character’, and who—despite acquiring greater ‘maturity’, ‘responsibility’ and the like (see above)—had mostly not reached the stage where they could honestly say that there were, for them, no conflicts between morality and desire. Do we therefore have to conclude that ‘virtue’ is out of their reach, and hence that a virtue ethical approach is irrelevant to the explanation of desistance?

We believe that this conclusion can be avoided. We should note in this connection that the term ‘virtue’ in the current discussion has a technical connotation, which mainly serves to distinguish it from other approaches in normative ethics. Accordingly, we should not pay too much attention to the implication in everyday speech that to be ‘virtuous’ is to be close to an ethical ideal. We can also be greatly encouraged by Christine Swanton’s definition of a virtue as: ‘a good quality of character, more specifically a disposition, to respond to, or acknowledge, items within its field in an excellent or good enough way’. ³⁵ Two features of this definition seem to be of special importance for the study of desistance: first, that a response within a given field that is ‘good enough’ may still count as virtuous; and secondly, that we do not have to look for a fully formed ‘character’ to detect virtue—rather, a ‘disposition’ will suffice. On that basis, steps towards virtue were certainly being taken by men in the Sheffield sample; and we can also note that both Kevin and Len were beginning to ‘enjoy acting well’—that is, aligning reason and desire—and to ‘[find] it naturally pleasing’, as Jacobs (see above) suggests that virtuous subjects will.

We may develop these points a little further with the help of Valerie Tiberius. Similarly to Swanton, she suggests that virtues might be described as ‘more like habits and problem-solving strategies than like the robust character traits’ of Aristotelian ethics; and, that being the case, we are not precluded from ‘attending to the role of situational factors’, which may in time lead to character development. ³⁶ In our view, Tiberius is right—even straightforward situational compliance (such as diachronic self-control, as exercised by Len) can develop over time into a disposition in Swanton’s sense, as also, of course, can the kind of normative attraction that Kevin experienced in the presence of his girlfriend. Indeed, we noted when discussing these two cases that in each of them one could discern several mechanisms promoting compliance; but we can now see a little more clearly that these mechanisms were beginning to work together to generate a new mechanism, that of habit—which we have encountered before in a sociological context, but which in more ethical language might also be described as a disposition, or virtue.

Charting a Path Towards Virtue

The title of this article asks: ‘Can persistent offenders acquire virtue?’. The simplistic answer is ‘yes’, if only because most persistent offenders do actually desist and so come to lead a more virtuous life. Given, however, that at the beginning of the Sheffield Study most men held largely mainstream societal values while still offending, a more nuanced version of the question might be, ‘How do persistent offenders in early adulthood chart a path towards a more virtuous life, in which actions reflect values?’ Phrasing the matter in this way links the question to the criminological literature on the role of ‘agency’ in desistance; but it also evokes again the work of Tiberius, who, unusually among moral philosophers, aims to provide a ‘first-personal, process-based account of how to live’. ³⁷

Tiberius distances herself to a degree from the dominant Aristotelianism of virtue ethics, claiming instead, ‘in the tradition of Hume … to look to our experience as the only source of answers to our normative questions’. ³⁸ This approach will, naturally, draw criticism from philosophers committed to different foundations for ethics; but from the point of view of illuminating a social scientific quest, the stance is potentially helpful, precisely because it ‘addresses people on their own terms’, ³⁹ does not presume at what point the person begins the process, and is not teleological. Nevertheless, because individuals’ experience is treated as the sole criterion for normative judgements, the approach seems vulnerable to the criticism that it is subjective, and therefore relativistic. To combat this criticism, Tiberius argues that, in thinking about the question ‘what is a good life for me?’, we cannot adopt an ‘anything goes’ stance, because ‘a good life from your point of view is a life you can affirm upon reflection … [and] on the basis of the standards you take to be important’ (described as reflective values). ⁴⁰ Thus, for Tiberius there is a crucial distinction between ‘reflective values’ and other values; only the former are based on our all-things-considered judgements of what is normatively important to us. On this basis, and given the research results previously summarised, ⁴¹ ‘desistance’ and ‘normality’ can reasonably be regarded as the central reflective values of our sample. ⁴²

Tiberius argues that reflective values play two key roles in a person’s moral life: they ‘serve as action-guiding goals’, and they also function as ‘standards of evaluation … on how our lives are going’. ⁴³ The second of these functions is well illustrated in Len’s case; his fresh conviction caused him to realise that his action in taking the car was not in accordance with his reflective values, so he immediately took steps to ensure that there was no repeat. But of course, if a person’s reflective values are not strongly held (as, clearly, Zed’s professed wish to want to stop was not) and are in conflict with other (non-reflective) values (e.g. for excitement or a free-spending lifestyle), then the reflective values will not succeed either in guiding action or in evaluating progress.

Although Tiberius’ processual view of values is helpful, her analysis is perhaps weaker when it comes to how reflective values can change or be strengthened. Here, we may draw upon sociological analyses of the interplay of ‘agency’ and ‘structure’ during processes of change. As we summarised this (following Mouzelis) in a previous paper, ⁴⁴ at Time 1 an individual (D) will be living in a particular society with distinctive social structures; in that context, D will carry a set of dispositions, and perceptions of the social world (Bourdieu’s habitus), ⁴⁵ some of which will have been internalised from the social structures around him/her. So what happens later on, at what we might call ‘Time 2’? Tiberius’ account suggests that changes in reflective values occur through the increased salience of alternative values as external elements impinge on one’s life. ⁴⁶ Though certainly not denying the importance of external events and pressures, we would rather emphasise the continual interplay between the ‘action-guiding’ reflective values of individuals, their non-reflective values, and their evaluation of what they have recently been doing, as they encounter and perceive new surroundings (in Bourdieu’s terms, a fresh ‘field’). Those new surroundings can include ‘institutional structures’ (such as employment opportunities, or the lack of them) and ‘figurational structures’ (such as relationships with romantic partners, family members and friends). Immersion in this new field, and individuals’ reflection upon their experiences in it, will very likely slightly change their reflective values; and those refined values will then feed into further action. If that action is successful in terms of progress towards the reflective values (and hence towards a more virtuous life), then such actions may be repeated, eventually creating new habitual paths (a new habitus), and, hopefully (see previous discussion) a sense of pleasure as these new paths are taken.

An important empirical example of these processes from our study concerned friendships with peers. At the beginning of the research, most of our respondents said that they deeply trusted their closest ‘mates’ (male friends). Not surprisingly, however, the great majority of those close friends themselves had criminal records, so as desistance progressed, it became important for the individual—given the overriding reflective value of becoming a non-offender—to distance himself from them (as is well illustrated in Len’s case). Given also the importance of romantic relationships in many instances of desistance (illustrated in Kevin’s case), there was often a clash of values between a partner and old friends. Partners generally held pro-social reflective values and led pro-social lives; as a desisting offender spent more time with a partner, in order to be with her and to do what she wanted, so that offender was encouraged to internalise more of her values. Thus, pro-social activity was reinforced, and delinquent friends were gradually dropped. It was still an agentic choice, and at least initially it might be weak, or even on occasion lose out to the competing claims of peers, but for would-be desisters, the wish to lead a virtuous life often (some might think remarkably often) won through. Hence, as Tiberius put it, a person seeking true ‘reflective wisdom … learns from experience … and responds appropriately’. ⁴⁷ Early adulthood is a life stage at which such refining and hooking into the best forms of action to accomplish one’s virtuous life is perhaps most active and reflective. At the beginning of the study, aged 19−22, most of the men in the Sheffield Study had a general commitment to change, but they also had a very limited awareness as to how this change could be accomplished. But gradually, and often very hesitantly and with setbacks, many of them did acquire more virtuous dispositions.

In this article, we have not discussed specifically Christian ethics, mainly because ours is an empirical study, and none of the men in our sample referred explicitly to any influence of Christianity in their lives. ⁴⁸ But it might be apposite, in closing, to point to the analysis of the Christian ethicist Richard Burridge, who argues that in studying historically the ethics of Jesus, one must consider not only his ‘rigorous’ ethical teaching on ‘the major human moral experiences’, but also his practical example, which included an ‘open pastoral acceptance of sinners, with whom he spent his life’. ⁴⁹ On that basis, the study of offenders (‘sinners’) attempting to desist from crime (that is, to move towards a more ethically appropriate lifestyle) ought to be of interest to students of Christian ethics, especially as the initial steps along that path are usually (if often falteringly) made by the offenders themselves.