Decision-Making and Offender Supervision: Our Belfast report

This post summarizes the progress made by the working group on decision making during the meeting we had in Belfast during the 23rd and 24th of October.

15 members of the decision making group made it to Belfast, and met in the Youth Action building, a very convenient and comfortable place for the COST meeting facilitated by our attentive host Nicola Carr. This meant that 12 jurisdictions were represented (Belgium, England, Germany, Greece, Ireland, Italy, Latvia, Malta, The Netherlands, Scotland Spain and Sweden). We have been working together for some time now and we are starting to know each others’ jurisdictions better. However, one always feels a sense of awe and excitement at being able to contrast the different legal, probation and parole cultures which make for rich and diverse European traditions.

In previous meetings we had decided to focus on decision making around breach because it is a situation which has not received much research attention and it is important in the three relevant decision making phases: pre-trial, sentencing and early release. Much work had been done in previous meetings discussing various available and workable methodologies in the different jurisdictions and finally deciding to explore the possibilities of vignette methodology for doing comparative research.

In order to do so, the group had previously created two common vignettes with John as a main character. We had one vignette with a 22 year-old unemployed John sentenced to community service and another with a slightly older John who had been released on parole. We had created various situations for John: John not attending work, John not attending an appointment with his supervisor, John getting a positive result in a drugs test, among others, and formulated various questions for the relevant decision makers about these.
So before we met in Belfast John had travelled to 12 jurisdictions and his situation had been presented by members of our group to judges, probation officers, immediate supervisors, parole board members and other parties involved in deciding his fate.

We had all done our homework before and we all brought to the meeting “process maps” with the relevant actors, norms, requirements and decisions, so we could all share an idea about how breach works in the different jurisdictions. And besides the maps, we had done pilots, most of us 3 or 4 interviews with the relevant decision makers using John’s community service and parole vignettes.

We used the first hours to present our findings and experiences using vignette methodology in the nine community service cases; after each two cases presented, we stopped for some discussion. Later on the same was done with the parole pilots in the five jurisdictions where it was tested.

Thanks to the great leading team made up of Miranda Boone and Niamh Maguire we managed to respect allotted time limits which allowed the group to hear presentations by all the members, clarify doubts and to engage in discussion.

As it was soon clear that we were going to work intensely, some members were kind enough to take advantage of one of the breaks and bring back wonderfully caloric pastries to keep us going during our passionate discussions.

The focus on previous meetings had been on methodology. In Belfast it was obvious that some more methodological discussion had to go on if we were to further our field work in the future, but results in the different jurisdictions were rich enough to start working on them.

Perhaps the main point around methodology was that vignettes had proved for most of us to be great conversation starters: they made decision makers talk and allowed researchers a good understanding of the decision making process, the roles of the actors involved and the criteria used, both at a formal and at an informal level. If we are to expand research beyond the pilots, however, it is clear that we shall have to revise the various situations we put John in, since not all made sense in all jurisdictions (eg. in Spain it is irrelevant if John arrives late to his work placement because he is taking his mother to hospital); we should add more information on context and we should revise the questions we posed, because some of them were too detailed and perhaps some of them were unnecessary.

Most of our discussion, however, went beyond methodology. We went through a social sciences equivalent to soul searching questioning, asking ourselves ‘what do we really want to know?’ and ‘what it is exactly we want to compare?’, ‘are our rich results comparable?’, ‘should we try to somehow quantify them to make them more so?’. It was clear that we had to make a decision on decision making, and this allowed us to clarify our research questions.

It is difficult to sum up all the points of discussion, but after some time it was clear that the results of the pilot were rich enough to show both expected variation and striking similarities amongst jurisdictions and amongst the two phases analyzed (sentencing: community service; and early release: parole). At the end of the last day, we managed to produce a list of the most relevant topics/research questions around which we could already compare our results. As well as a framework of analysis for comparison of our results we also agreed that we needed to develop a way to translate our qualitative data into quantitative data to allow for comparison across and within jurisdictions on a more abstract level. So for example, if we wanted to compare the level of discretion exercised at each stage of the decision-making process by different actors we could adopt a simple 3-point scale that might look like this: a lot, some, not at all. Applied across all the different measures we are interested in, this approach could help us to examine how breach works on a more abstract level and even allow us to develop models of the different types of breach systems in Europe. As Anders Persson and Alfredos Laurinavisius had already begun this work the group decided to delegate the job of further developing a way to quantify our data to them. Good luck Anders and Alfredos!

Lastly, I would like to highlight two very interesting themes that emerged in the pilots I conducted in Spain and that later emerged with even greater force in our discussions in Belfast. The first one is the observation that professionals at the various levels have developed the ability to circumvent sometimes rigid norms and often retain an important level of discretion. This level of discretion exists in spite of, or perhaps because of, the existence of protocols and guidelines. This raised an ethical question: what impact could this result have? If we say there is great room for discretion could this be used as an argument to justify the introduction of tougher legislation, aiming at restricting practitioners’ professional discretion?
Interestingly and in connection with the previous idea, the second theme was the existence of strong narratives of giving second chances, using Kristel Beyens’ expression, that emerged in many, if not most, of the jurisdictions. These narratives were based on the idea that if offenders had the right attitude, if they tried, they could fail, but their failures to comply could be interpreted as difficulties rather than as a cause for breach.

Many things were clear (again!) at the end of the meeting: that we were dealing with very exciting ideas we would like to keep working on, that it is possible to work hard and have fun at the same time and that ours is a very comfortable and fun group to work with, capable of orderly, and disorderly, but always fruitful and interesting discussion.

Ester Blay
Universitat Pompeu Fabra (with the careful editing of Miranda Boone and Niamh Maguire)