Questions, questions, questions: Desistance and probation

I was asked a week or two ago by the French Section of the International Prisons Observatory to provide written responses to a series of questions about desistance research and its implications for probation work (though this is relevant for prisons people too). Since they’ll be publishing this in French, they were happy for me to post the English versions on the blog. As with all (my) blog posts, these ‘answers’ come with a health warning; I’m expressing opinions here (albeit with an eye on the evidence)… Very happy to hear others’ views, including dissenting ones!   

According to desistance studies, which are the main factors that lead to stop criminal activity? 

Most reviews of the literature point to three main theoretical perspectives on desistance. The first draws on evidence about the relationships between crime and age. Noticing that crime is disproportionately a youthful activity – and that even persistent offenders seem to eventually ‘grow out’ of crime, these ‘ontogenic theories’ suggest that desistance can be explained in terms of age and the developing maturity that it usually brings.

The second perspective suggests that desistance can be best explained not by age and maturity per se, but rather by the changing social ties or social bonds that tend to come with adulthood. These ‘sociogenic’ perspectives point to evidence that desistance is correlated, for example, with securing meaningful employment, developing successful intimate relationships, investing in becoming a parent. People desist from crime because they acquire a stake in conformity.

The third perspective points not so much to the structural nature of these ‘turning points’ (linked to work or family life) but to the subjective dimensions of them. A new partner or a new job is more likely to provoke or support desistance if and only if the person values that new pro-social partner or that new job more than they value existing pro-criminal relationships or activities. These subjective dimensions take us into a consideration of how criminal identity can be cast off – and how new and more positive identities can become established. That process of ‘de-labelling’ – both by the person themselves and by those around them – seems to be especially important for people who have been involved in persistent offending, and whose criminalized identities are therefore more deeply entrenched.

Although they place the emphasis in different places, most desistance scholars now tend to agree that desistance can best be explained not from one of these three perspectives but by examining the interactions between these three sets of factors – age and maturity, social ties and identity transitions.   

Does that mean that supervision of offenders by probation officers has only a minor effect on criminal careers?

No, I don’t think so. It is true that one of the most important studies of probation and desistance – discussed in Steve Farrall’s (2002) book ‘Rethinking What Works – suggested that probation had little direct impact on desistance, and that the individual’s motivation to desist and his or her social context seemed to matter more. But even then, Steve argued that probation could have positive indirect effects, for example, by working to develop motivation and by addressing social problems. Also, since 2002, Steve and his colleagues have conducted several follow-up studies on the same probation cases, and these studies now suggest that probation did a better job in terms of ‘sowing the seeds’ for future change that he first thought. This finding echoes something I discovered in a small study of people who had been on probation in Scotland in the 1960s. Looking back from the vantage point of 40 years later, several of those I interviewed recognised that probation had a significant and positive impact on their lives – but not always immediately.  

How important is the quality of the relationship between offender and probation officer? Which supervision methods should be favoured? 

I think the quality of the relationship between the person under supervision and the supervisor is absolutely vital – my 1960s study showed that very clearly, and many larger and more robust studies have reached similar conclusions. Indeed, it is a common finding in studies across the human services (and in relation to psycho-social interventions of many different sorts) that the relationship is a critical factor in success. It seems obvious to me that people are more likely to be prepared to embark on a challenging change process when they feel cared about, valued and respected by those supporting that effort.   

More generally, there is now ample evidence that desistance is a relational process; it involves renegotiating identities and relationships. Though personal relationships are perhaps more fundamental to desistance in the longer term, supportive professional relationships have a key part to play – especially where the person has complex needs and faces many challenges.

I think it is probably impossible to answer the question about which methods should be favoured in a general or abstract way; it depends on the individual. One of the key implications of desistance research is that, since the process is affected by subjectivities and questions of identity, we need to have properly individualized approaches that tailor the support to fit the situation, the needs and strengths and the particular dynamics of each individual’s change process. 

It is possible to say that everyone will need motivation to change, the capacity (or skills) to begin to live differently, and the opportunities to do so – and that the supervision process therefore is likely to involve counseling, educative and advocacy-related aspects. But exactly what is required to support any individual depends on carefully exploring – in collaboration with the person him or herself – where they are on their desistance journey, what obstacles they face and what can help them to get moving or to keep moving forward.

How can probation officers work on questions related to the influence of couples in criminality, on the « social capital » of offenders, their lack of « positive » relations and networks ? Can the relatives of offenders be implied in the supervision and how (examples welcome if possible)?

I have already alluded a little to this above. The concept of ‘social capital’ refers to the networks of relationships and reciprocities that we all rely upon in our lives. Family and close friends are our usual source of ‘bonding social capital’. Workmates, classmates and other people with whom we have some interest or activity in common provide ‘bridging social capital’. Of course social capital can licit or illicit – it can function to support desistance or to support criminality.

Both bonding and bridging social capital – obviously of the licit or pro-social sort — seem to matter in desistance. Recent studies in Sheffield and Tubingen, for example, point to the importance for young men involved in persistent offending of repairing their relationships with their parents, so as to secure their support in moving away from crime and criminal networks. Similarly, I have also suggested the role that forming new families may play in enabling desistance. But bridging social capital is very important in terms of securing involvement in work and in succeeding in education – it is critical for the social mobility that desistance requires.       

I think that the main implication of this is that probation can’t work with people under supervision on their own.  Probation should work with and through families and peer support networks – at least where these are thought to be supportive of desistance – in order to support the change process. Practically, that means that probation staff need to get out of their offices – to explore and understand the social contexts in which people live, to build relationships with families, communities, employers and NGOs, so that change can be supported, obstacles removed and pathways opened up.

One interesting and very promising example of this sort of work is the development of ‘Circles of Support and Accountability’ in which people who have committed sexual offences and are considered to be at high risk of doing so again are placed within a ‘circle’ of trained volunteers who are supported by criminal justice professionals. The volunteers act a bit like a proxy family or friendship network for people who are otherwise highly isolated, but they also provide an accountability mechanism – monitoring as well as supporting the person. This is a more artificial mechanism for building social capital than might be required in most cases, but it shows what can be achieved by mobilizing non-professional actors in the process even of supporting and supervising the most difficult cases.  

Which are the recommendations that desistance researchers make as regards working on human capital, particularly cognitive dimensions, offender’s values and representations linked to criminal activity (examples welcome if possible)?

To succeed in a change process probably does require the development of ‘human capital’ or ‘skills’ – or at least the redirection of existing skills and strengths in a more positive way. There is evidence in the ‘what works?’ literature that many people involved in offending have learned attitudes and behaviors that need to be un-learned, and that CBT programmes can play a big part in supporting this process. ‘What works?’ tend to be focused on very specific ‘criminogenic needs’ or ‘dynamic risk factors’ – those attributes of the individual (or to a lesser extent of their situation) which are most strongly correlated with reconviction. Three of the main ‘targets for change’ that emerge from this kind of research are anti-social attitudes, anti-social associates and substance use problems. 

I think that where these problems exist, probation certainly should work to address them. However, I think it is a mistake to assume that ‘fixing’ these intra-individual problems will resolve offending. As I  said above, desistance is a relational process; no amount of ‘fixing’ the individual will repair the breaches in relationships (between the ‘offender’, the victim and the community) that crime creates. So probation work has to support the development of the human capital of the people involved, but it has to do more than that if it is to support desistance. 

Should individual interviews be favoured, or group activities, or both?

Again, there is no simple answer to this question. I said already that the process of supervision needs to be highly individualized, but that doesn’t mean that it can’t involve elements of group work on particular issues (like the skills development discussed above). If there are some common needs across the probation population, then it can make both financial practical sense to bring people together to work on those issues – and to support each other’s change process. ‘What works?’ group-based programmes have always used peer support and challenge in the process, but they are now beginning to learn more about this from other kinds of ‘mutual aid’ – like Alcoholics Anonymous. So, for example, in Scotland we now run ‘rolling (groupwork) programmes’ where the programme has modular elements and different people can be working on different issues and be at different stages in the process. In each session, one or two people become the focus of the group discussion, working on their issues and accounting for their progress with peer support. In a sense this approach allows group work to become more individualized.   

That said, in my view, group work always needs sit within the context of an individualized supervision process, and the whole process needs to be embedded in an understanding of desistance, if that is its objective.

Could you accurately describe the SPP programme, its content? Has it been assessed and what are the results?

The SSP (the Structured Supervision Programme) borrowed the structure, preparatory training and cognitive behavioural techniques and tools from group work programmes, as well as their procedures for managing interventions to ensure that they are delivered with integrity.  However, it also drew on desistance theory, attachment theory (which concerns people’s early life experiences and how these shape their capacities to relate to others) and the Good Lives Model of Offender Rehabilitation pioneered by Tony Ward.

After receiving training, probation officers delivered SSPs through a one-to-one supervisory relationship rather than in a group context. The SSP was initially developed for the Romanian Probation Service was later re-developed for London Probation; it included 12 one-to-one sessions which made up 5 modules dealing with motivation; problem solving and assertive communication; goal-setting and the cycle of change; perspective-taking; and relapse prevention.  

Though I’m not aware of any available evidence about the effectiveness of the SSP in reducing reoffending, it was very well received by probation staff in both Romania and London, and by probationers themselves. SSP seemed to allow for a focus on the individual’s concerns and hopes, without losing the rigour and structure of group-work programmes. More recently, the SSP informed the development of the SEED (Skills for Effective Engagement Development) programme which is currently being piloted in England and Wales. Like SSP, SEED focuses on relationship building, pro-social modelling, motivational interviewing, and on applying the risk-need-responsivity model and cognitive-behavioural techniques to a more structured form of one-to-one supervision.

I am part of a team, led by Prof. Joanna Shapland at the University of Sheffield, which is evaluating SEED. While it is too soon to report any results, there is no doubt that probation staff have responded very positively to SEED; they seem to like working with it very much.

The key practical point to take from this is that the emerging evidence, coupled with a concern to make the routine, high-volume business of one-to-one supervision more effective in fiscally difficult times, is leading probation services in some jurisdictions to invest more heavily in staff skills and one-to-one approaches, rather than relying on complex and expensive group work programmes to deliver reductions in reoffending. Such programmes are not being abandoned, but their number looks likely to become more restricted to those deemed necessary for particular groups of higher risk offenders.


6 thoughts on “Questions, questions, questions: Desistance and probation”

  1. The points about groupwork changing in their use is echoed in my probation trust. However, there is also a tension between this aspiration due to continuing centrally imposed targets around programme completions.

    In addition, SEED and similar training such as Thresholds and the Reflective Supervision Model (RSM) are exciting and as you mention greatly welcomed by Probation Officer (PO) practitioners. However, there is also a local drive towards non-accredited programmes in my trust based around locally produced workbooks and delivered by Probation Service Officers (PSOs), a significantly large staff group who have not had the benefit of the PO training.

    If these groups were limited to general offending, victim work and alcohol as examples, I would be more hopeful. However, there are also moves to deliver similar groups around domestic abuse and anger management. I am particularly concerned about the domestic abuse groups as these are being delivered based on a workbook designed for individual intervention. I am not critical of the idea in principle but would be happier if the PSOs received the training sooner rather than later.


    1. Yes, I’ve been hearing from various people emerging concerns about ‘de-professionalising’ some aspects of probation work. It’s a tricky issue. I think, from a desistance point of view, there is a lot of evidence that aspects of the required support can be more informal — for example, that the importance of the support of families and of mentors (including ex-offenders) needs to be recognised and given more weight. Probation needs to work with and through these kinds of desistance supporters (where they exist). At the same time, people with more complec problems may often need more complex and professionally skilled types of support. One of my worries here is that managers and policymakers assume that ‘lower risk’ (of harm) people don’t need highly skilled support, and can be supported to desist just by ‘sign-posting’ and onward referral. I think that’s a misintepretation of both ‘desistance’ and ‘what works’. The ‘risk principle’ in RNR is about risk of reoffending not risk of harm; its about probability not seriousness. Assuming that skilled, professional support can be abandoned in those situations is a bit foolhardy, I think, and a false economy.


      1. Fergus, thanks for giving us ‘la primeur’ of your OIP interview.
        You write in one of your reponses: “One of my worries here is that managers and policymakers assume that ‘lower risk’ (of harm) people don’t need highly skilled support, and can be supported to desist just by ‘sign-posting’ and onward referral. I think that’s a misintepretation of both ‘desistance’ and ‘what works’”

        This rings a very loud bell from my side of the English Channel: based on a rather basic risk assessment tool (if you can call it that…) French probation services are now dividing offenders in 3 categories, called ‘segments’. Segment 1 is supposed to apply to low risk offenders. The idea is that prison guards will be in charge of supervision of segment 1 and will just control whether obligations are observed, and tick boxes accordingly. As you know it is not as if probation services were really supporting desistance or doing much social work as it was. The idea of reformers is that they will have more time to do other things with segments 2 and 3. Not sure what since they’ve also been assigned lots of taks that in my opinion are still getting massively in the way of their core missions.
        We talk a lot about professional skills here, and the worry is deprofessionalisation… what about prison guards’ skills? What about probation limited to obligations checking – even with segments 1?
        As usual it will be very difficult to assess what is actually happens and what the outcome is, given the state of research here.


  2. “Looking back from the vantage point of 40 years later”

    I’m not sure the PBR pilots will go back that far…

    seriously though, so much of this stuff is so difficult to measure why do we even bother? why can’t our masters accept that not everything is quantifiable and calc-u-able? the desistance research is in some ways so stunningly simple: people are all different and will desist for many different reasons, therefore a simplistic measurable scale of desistance is surely impossible?

    Still that won’t stop them trying.


Comments are closed.