This guest post comes from Ros Burnett of the Centre for Criminology at the University of Oxford. Anyone who knows desistance research will know and respect Ros’s work. She was one of the UK’s pioneers in the field, and has influenced and supported many other students, scholars and practitioners. In this post, she reflects back on her pioneering ‘Dynamics of Recidivism’ study.
Recently and not for the first time, I’ve been thinking about the word ‘dynamics’ in relation to ‘desistance’, what it really means and whether it is a concept which elucidates or obfuscates our understanding of pathways out of crime. Some twenty years ago I was the lead researcher on a study at Oxford, commissioned by the Home Office, called ‘The Dynamics of Recidivism’. That was its given title. It could just as easily have been called ‘The Dynamics of Desistance’ because it involved learning about the interaction of social and psychological factors which contributed to whether or not individuals reoffend in the period following their release from prison.
It relied mostly on their own accounts given in up to three long qualitative interviews, the first one being close to their release from prison when they were looking ‘over the wall’ towards their future. At that point, most of the cohort (130 male property offenders) wanted to ‘go straight’ but a lot expressed doubts about whether they could succeed. I managed to track and reinterview 109 of them, and about half of them again on a third occasion, during the following two years. On re-meeting they were impressed by the memories we shared of what they were experiencing last time, and touched by my recall of the details they had imparted in the previous interview. Suitably engaged, they became as keen as me to explore continuities and departures in the route their life was taking and to understand why things had worked out as they had and to explain their present directions, wishes and intentions.
Far bigger and more sophisticated studies along similar lines overtook the efforts of this small tracking study; but it did briefly make a splash – quite frighteningly at the time – when, the then Home Secretary, Michael Howard cited it to support his pronouncement that ‘prison works’. Yes, most of the cohort had told me, in each interview, that a determination to avoid further imprisonment was foremost in their reasons for desisting from offending, and a key disincentive when we conjectured about tempting opportunities to make money illegally. I and my colleagues were quick to shunt away the rational choice implications that were being drawn to support greater use of imprisonment. Indeed I felt some pressure to ‘bury bad news’ and the prospect of publishing the report was shelved.
A decade or so later, Shadd Maruna, Tom LeBel, Shawn Bushway and I analysed the subsequent reconviction and reimprisonment data for the cohort in relation to the original qualitative data. An alarming percentage had continued offending and served further prison sentences; but a few had ‘come through’ without further blemish on the official records. Whittling it all down (with clever statistical wizardry – thanks Tom and Shawn!) we proposed that, if anything, hope works. That was too nearly another slick answer, another slogan. Shadd and I batted around some alternative terminology. I think I suggested ‘self-efficacy’ or some inelegant labels like ‘optimistic expectation’, ‘confident intention’. Then Shadd introduced me to C.R. Snyder’s hope theory, and things clicked into place.
‘Prison works’. ‘Hope works’. Such slogans are memorable and capture the imagination, and wouldn’t it be nice if we could all agree we’ve hit on ‘the answer’ with something neat like that, instead of all these wishy-washy conclusions about interactive dimensions and trajectories to desistance. We should be careful though not to either accept or dismiss single factors on the grounds of their compatibility with our own ideology and value position.
When I commenced the 1992 study, it was not long after a previous career in the probation service, and talk of ‘dynamics’ sounded right to me because I knew a lot about the difficult, often prolonged, journey my ‘clients’ (as we used to call them) were making in ‘going straight’. I would have been less comfortable with a single theme study on the role of, for example, drugs, or employment, in crime and rehabilitation. I liked that it wasn’t advancing a hypothesised key factor but kept all options open, and that it effectively allowed that desistance is complicated.
At the beginning though I admit to being a bit bemused about what I was meant to be looking for: the concept of ‘dynamics’ is so rich in meaning. It is evocative of movement, change and forces. It has its origin in the Greek word dynamikos, meaning ‘powerful’, and in its singular, adjectival sense, ‘dynamic’ people and events gets things done. There can be a dynamic for something to happen or change: a motivator, a push, perhaps a sudden source of socio-psychological energy; or perhaps a more gradual intertwining of events, situations actions and reactions that leads to a particular outcome. Mostly though, I understood my task to be a search for that ‘dynamic interplay’ of all those factors that might influence a person to behave in a certain way and that might impact on the outcomes. With all those potential strands, it was going to be different for each person, for each further offence and each triumph of desistance, but there would be patterns to draw out and recurring factors that would stand out.
The dynamics of recidivism has a good academic ring about it, and ‘dynamic interplay’ connotes factor analysis and multi-dimensional scaling and regression, while the notion of dynamics embraces the complexity of causation, of relapse, of shifting change. The opposite of static and with its implications of movement and continuous activity or progress, it dissuades us from looking for finite positions and simple explanations.
On the other hand, it is the kind of jargon which becomes its own explanation and a stopgap for real understanding. It tells you everything but it tells you nothing; begs the question and defers the answer. You can accurately reflect that there are multiple interacting factors that influenced why this person stopped offending completely and that individual desisted for two weeks but then relapsed. But saying that there is a dynamic interplay is itself a kind of simplification; is reductionist. No-one can really argue with it but what does it tell us, and what do we do with it to improve policy, to change lives. It’s a long way from a pointed explanation that could become the centre of a new rehabilitation policy.
I still wonder about the individuals I got to know at that time and how long it might have taken them to become ‘secondary desisters’ or whether they went back round the relapse loop; and, for all my years of practical and then academic inquiries into rehabilitation and desistance I haven’t yet discovered an equation of ingredients that lead to one or other of those outcomes. I’m pretty sure though that there are some dynamic variables that are more powerful in the mix, and I’m left with a lingering unease that despite best intentions to report the data faithfully and with integrity, I didn’t manage to tell the general story as straightforwardly as I might have done, did not show the ways in which prison might indeed contribute to changes in behaviour and did not clearly enough explain in what senses and in what ways hope is a critical factor.
Single factor causal theories can be over-simplistic and, as such, dangerous or annoying, and they add to the hostilities between opposing positions on criminal justice policy. We should be wary of supporting pet theories with equivocal research findings. Equally though, we should not shy away from acknowledging the role of some factors because they offend our sensibilities. It’s complicated. By all means let’s not be simplistic, but let us continue the quest to discover which factors are more important than others and to understand the circumstances in which they come into play.
[References provided on request.]
Ros Burnett
Email: Ros.Burnett@crim.ox.ac.uk
Twitter: @rosburnett
Discovering Desistance 
Ros, thanks for this insightful and reflective post. I will confess to always having conceived of your earlier study as a study of persistence and desistance and I have consistently returned to this study which, despite its comparative vintage in the desistance literature timeline, drew critical insights that remain acutely pertinent to policy and practice today. That these findings have, to varying degrees, been echoed in subsequent studies would not only support their veracity but suggest that we have been slow to respond, to listen to these voices. Perhaps our own process of change is similarly just that and can also be characterised in terms of lapse and relapse – ambivalence and uncertainty – as you have previously said. Perhaps similarly our hopes as practitioners have been beset with and sidelined by insecurities and feelings of powerlessness which reduced our sense of agency – our ability to effect change – and which stood in the way and discouraged us from embracing the change that studies such as yours have implied.
Anyway I’m deviating but I would entirely agree that the idea that prison works is inherently over simplistic if not pathologically flawed although as you rightly point out – and as Sarah Armstrong and I discovered in our Users Views of Punishment study – it can have positive effect for some people some of the time. However how, when and why it happens remains as much an enigma as understanding why some people persist and others desist. For one person – prison might pose a deterrent effect after one sentence, for another after twenty which in policy terms is a costly and unpredictable gamble I would say. You don’t need me to persuade you of that but I think what I am trying to contribute in my usually long winded way is that policy is always constrained by a desire to lay down a formulaic approach – which is kind of understandable in some senses but which as you say – reduces everything to its most simplistic and therefore almost impotent level in any meaningful sense – and then reduces practice, at times, to a slavish and unreflexive adherence to delineated regulations and procedure and smothers the types of creative practice that elude measurement. Perhaps policy development should orient itself further to facilitating professional agency/autonomy and incorporate the voices of both practitioners, ex offenders and service users, whose voices have rarely entered – except indirectly – policy domains. Is it overly optimistic, hopeful or naive to venture that two decades on perhaps we may, belatedly, be learning to listen? I am still writing up my research but hope I can incorporate your insights and demonstrate your reflexive sensitivity and integrity in my own work.
LikeLike
Hi Beth, I’m very touched by your generous comments on my 1992 Dynamics study, and glad it has been of some use. It’s embarrassing that it was never properly published, except as a ‘desk top’ report by Oxford’s Centre for Criminology – and I still get people contacting me saying they want to buy “the book”.
You are spot on in your observation that the same process of change applies to us all. That’s perhaps one of those obvious truths that we specialists – researchers, practitioners, policy-makers – can too easily overlook when we are focused on answers to a specific problem, behaviour or life situation in which others and not ourselves are the subject. On that basis, it is obvious that we should not expect the explanations for desistance or recidivism to be any less complex and variable than the explanations for changes we make ourselves or try to make in our own lives.
Thanks too for understanding so well my argument about the dilemma of being either too meaninglessly general or too simplistically specific in our attempts to generalise about causes. In the 1992 study my own attempt to get round that dilemma was to use a typology of several ‘ideal types’, none of them the exact replication of real people, but each capturing similar patterns that applied to a section of the cohort. The real impact of imprisonment as a deterrent applied much more to one of those ideal type characterisations. Some of my colleagues did not want me to say that and, as it happened, they were right to be concerned about how such a ‘finding’ could lead to over-generalisation.
I will read up on your work with Sarah Armstrong on this, and I look forward to hearing more about your own research.
LikeLike
Beth, I’m very touched by your generous comments on my 1992 Dynamics study, and glad it has been of some use. It’s embarrassing that it was never properly published, except as a ‘desk top’ report by Oxford’s Centre for Criminology – and I still get people contacting me saying they want to buy “the book”.
You are spot on in your observation that the same process of change applies to us all. That’s perhaps one of those obvious truths that we specialists – researchers, practitioners, policy-makers – can too easily overlook when we are focused on answers to a specific problem, behaviour or life situation in which others and not ourselves are the subject. On that basis, it is obvious that we should not expect the explanations for desistance or recidivism to be any less complex and variable than the explanations for changes we make ourselves or try to make in our own lives.
Thanks too for understanding so well my argument about the dilemma of being either too meaninglessly general or too simplistically specific in our attempts to generalise about causes. In the 1992 study my own attempt to get round that dilemma was to use a typology of several ‘ideal types’, none of them the exact replication of real people, but each capturing similar patterns that applied to a section of the cohort. The real impact of imprisonment as a deterrent applied much more to one of those ideal type characterisations. Some of my colleagues did not want me to say that and as it happened they were right to be concerned about how such a ‘finding’ could lead to over-generalisation. I will read up on your work with Sarah Armstrong on this, and I look forward to hearing more about your own research.
LikeLike
‘We can merely plant a seed’ is something that, echoing what you write, Steve, that several French PO told us in the course of my little Desisting in France research. There definitely is more than a metaphor here.
LikeLike