Abstract
It has long been recognised that changes in social networks (and the underpinning changes in personal and social identity) are strong predictors of both desistance from crime and recovery from substance use. Building on existing work attempting to measure and shift social networks and transitions to prosocial groups, the current study provides pilot data from prisoners and family members about a visualisation technique widely used in specialist addiction treatment (node-link mapping) to map opportunities for linkage to prosocial groups and networks. The data presented in the paper are from a small-scale feasibility pilot. This suggests both bonding and bridging capital in prisoner populations due for release and the diversity of community capital opportunities that exists in this population. The implications of this work are significant for substance users and offenders pending return to the community, and has implications around resettlement and reintegration support for probation staff in prisons and in the community. The paper emphasises the importance of mapping connectedness as a key component of planning for reintegration back into the community for those working with offenders who are aspiring to achieve desistance and recovery.
Introduction
Although definitions of recovery have been elusive and contested (e.g. Best, 2011), the Betty Ford Institute Consensus Panel defines recovery from substance dependence as a “voluntarily maintained lifestyle characterised by sobriety, personal health and citizenship” (2007: 222), while the UK Drug Policy Commission concludes that recovery should be characterised as “voluntarily sustained control over substance use which maximises health and wellbeing and participation in the rights, roles and responsibilities of society” (2008: 6). Writers such as Valentine (2011) have argued that ‘you are in recovery if you say you are’; however, there is a recognition that recovery requires activity and action, that it changes over time, and that it is socially mediated (Cano et al., 2017; Mericle, 2014; Laudet et al., 2002). In a review in the British Journal of Psychiatry of studies showing positive results from recovery interventions, a model was produced of ‘essential elements’ of recovery, summed in the acronym CHIME (Leamy et al., 2011). This stands for Connectedness; Hope; a positive sense of Identity; Meaning and Empowerment, and provides basic domains that recovery projects can be mapped against.
The impact of social factors is further emphasised by Longabaugh et al. (2010), in an analysis of alcohol outcome data, asserting that a strong predictor of recovery from alcoholism is shifting from networks supportive of drinking to networks supportive of recovery. Similarly, in the UK, Best et al. (2008) found that, while initial cessation of substance use was triggered by psychological change and trigger events, maintaining long-term recovery was more strongly predicted by transitions in the composition of peer groups from individuals who are using to being characterised by recovery.
Similar discussions and debates have taken place around desistance from offending. Desistance has been defined as a process that involves ‘the long term abstinence from criminal behaviour among those for whom offending had become a pattern of behaviour’ (McNeill et al., 2012: 3). Sampson and Laub (1992) used a developmental approach to demonstrate that desistance-oriented life transitions are dependent on wider social variables such as changes in social status and an expanding repertoire of life experiences. In a review of their life course model, Laub, Sampson and Sweeten (2011) assert that “we recognise that both the social environment and the individuals are influenced by the interaction of structures and choice. […] In other words, we are always embedded in social structures” (pp. 281–2), an issue that applies equally to desistance from offending and recovery from substance use.
In the recovery field, the synthesis of contributing factors has been defined as ‘recovery capital’ described by Granfield and Cloud (1999) as the sum of resources available to people to support their recovery pathways. A key aspect of recovery capital is social capital; social capital has been referred to as the value of aspects of social structure to social actors and how they can draw on different aspects of this as a type of resource (Coleman, 1988); the goodwill that is generated as a result of investment in social relations (Adler, 2002); and ‘investment in social relations with expected returns’ (Lin, 1999: 30). It is clear that social capital can be considered structurally, functionally, and in terms of the relational benefits it can manifest (Cattell, 2001). This has provided the foundation for examining key elements of recovery resources at the intra- and inter-personal levels as well as the community resources required (Best and Laudet, 2010), and has introduced the idea of ‘community capital’ to refer to those resources in the lived community that can support positive change, such as access to groups and organisations. The idea of recovery capital draws in part on the idea of social capital (Putnam, 2000), including the division between bonding capital (the strength of associations and relations between existing networks) and bridging capital (the capacity of the individual or group to link to new groups and networks). Putnam (2000) went on to suggest that it was second-generation connections (friends of friends) who could provide access to information about community resources that was critical in appraising the levels of community capital. Cloud and Granfield (2008) subsequently introduced the idea of ‘negative recovery capital’ to refer to things that could act as potential barriers to recovery – including stigma and exclusion both on an interpersonal and on a structural and societal basis.
Furthermore, one of the major barriers to recovery is the challenge of exiting from using groups (and identities) to recovery groups. This transition is outlined in the Social Identity Model of Recovery (SIMOR; Best et al., 2016) in which recovery is characterised as a change in attitudes and values associated with the transition from groups supportive of substance use to groups supportive of recovery. This group transition creates new social capital through engagement with groups who have access to more pro-social resources in the community and who are able to provide structures and support to the recovery pathway (Moos, 2007).
This notion of identity change and the role of social networks also plays out in the desistance literature. Farrall’s (2002) study of 199 probationers identified desistance as being closely related to the offender’s motivation to change and to the social and personal support networks that supported these changes. In Maruna’s (2001) Liverpool Desistance study, based on interviews with 50 former or current offenders, 30 of whom were classified as desisting and 20 as persisting offenders, Maruna argued that to desist from crime, ex-offenders needed to develop a coherent, pro-social identity. The underpinning rationale is that social network transition is associated with identity change and that this is linked to changes in access to resources.
In a more recent study, Cano and colleagues (2017) extended this finding in a study of recovery residences in Florida where there was an association between longer periods in residence and reduced barriers to recovery, and to improved recovery capital and wellbeing. However, this relationship was mediated by the extent to which residents engaged in meaningful activities. Thus, simply being a resident for extended periods of time is insufficient if the resident does not use that time to actively engage in their community and build their recovery capital in doing so. The Cano paper is part of a larger project (Best et al., 2016) that measures recovery capital (personal, social and community) and uses this to plan recovery and to link in to community resources, based on the acronym MPE (Measure, Plan and Engage).
It is this underlying rationale that provides the basis for the current work. The underpinning model is predicated on the assumption that people early in recovery typically lack substantial personal and social capital through the loss of prosocial friends and family and the loss of houses, jobs and others sources of self- and external esteem. In earlier work with the Salvation Army on the East Coast of Australia, Best et al. (2014) undertook community asset mapping (Kretzmann and McKnight, 1993) as a means of actively engaging the clients of therapeutic communities into local communities with a view to both building the recovery capital of the residents and also building effective bridges to the community for the Therapeutic Community (TC) more generally. This work was supplemented by further Australian work (Best et al., 2014) that involved using a visualisation technique to map social networks and social identity as a means of determining both the size of the social networks and the extent to which the members of each group were supportive of or hostile to recovery objectives (the method of social identity mapping is outlined below).
The aim of the current paper is to extend this social identity mapping process to incorporate a method for measuring access to desirable community resources – in other words, to assess the extent to which social networks are perceived to link to community capital. The project was a pilot study funded by Sheffield Hallam University as a feasibility study. The current findings are based on the second iteration of the pilot which was adapted from lessons learned from the first iteration. These aims are based on a small-scale pilot study undertaken to demonstrate the feasibility of the approach and the engagement and participation of all key stakeholder groups. Although the intervention was developed by researchers from Sheffield Hallam University, the delivery was jointly undertaken by university academics and probation staff within the prison, supported by peers from the first prisoner cohort to have taken part.
Method
Setting
HMP Kirkham is a large open prison in the North-West of England. It is for male prisoners who are predominantly coming to the end of long sentences and it is regarded as a bridge back to the community.
Participants
The project was a part of a larger Kirkham Family Connectors (KFC) intervention designed to encourage family members to establish pathways to meaningful activities for prisoners in the period immediately prior to their release. Programme participants were selected by probation staff, who selected prisoners whose offences they believed would not negatively impact group participation and involvement with family members, such as domestic violence and abuse. The sample was not in any sense representative as it involved judgements by prison probation staff about who might be appropriate and then the willingness of those prisoners to participate. Following initial meetings with prisoners regarding their interest in the programme, participation was then promoted by probation staff and committed prisoners. Participants were a combination of offenders and family members. Of the 13 participating in the programme, 12 completed the evaluation forms – four prisoners and eight family members. None of the participants dropped out in the course of the project.
All four prisoners were male, as were two of the eight family members; the mean age of participants was 43.8 years (range = 20–66 years). Ten of the participants were married, one was widowed and one was single. Of the family members, one lived with a partner, two with children only and five with partners and children. The family members were a combination of wives/partners, parents and siblings and a mother-in-law.
Instrument
Best et al. (2014) outlines the technique of social identity mapping in which post-it notes are headed with the names of groups that individuals belong to. For each group, the participant then notes the first name of each group member they regularly engage with. Then each name is colour coded (using sticky dots) to indicate whether the person is ‘in recovery’, a non-user, an occasional user or a problem user of illicit drugs or alcohol. This then creates a visualisation of the recovery capital available to the individual through their social networks.
Wellbeing was assessed using two separate measures – the three wellbeing rulers from the Treatment Outcome Profile (Marsden et al., 2008) were used to assess physical and psychological health and quality of life (on a scale ranging from 0–20 with higher scores representing better wellbeing), and the Therapeutic Optimism Scale (Cardoso and Xavier, 2015) was adapted to measure extent of belief in the family’s capacity to initiate change. All of these measures were implemented and completed before the start of the session.
The current approach combines the social identity mapping with a visualisation approach that has been widely used in education and addiction treatment, called node-link mapping (Czuchry and Dansereau, 2003). A dedicated map was created that attempted to reconcile a visualisation map for social identity (Best et al., 2014; Haslam et al., 2017) with an asset mapping approach. This map provided participants with up to five groups to identify with and four types of asset in the community – sport and recreation; education, training and employment; mutual aid and community; and volunteering opportunities. For each group member identified within social groups, participants were asked to identify members who had access to community groups and resources that they could tap into. The aim of this mapping was to create a social and community asset map that participants would complete in terms of the groups they belonged to and how they linked to assets in the community that they wanted to access. To achieve this, participants were asked to complete up to five groups and list members that they had contact with and then code those members with coloured dots as follows: – red dots denoted that this individual was involved with drinking, drug use or crime – yellow dots denoted connections to sport and recreation groups – green dots denoted access to recovery or community groups – blue dots denoted access to employment, education or training resources
Ethics
The study received approval from both Sheffield Hallam University and from the National Offender Management Service. Informed consent was obtained from the prisoner and the family participants.
Results
The mean wellbeing scores were 14.0 for physical health (±4.3), 14.9 for psychological health (±3.7) and 13.6 for quality of life (±5.8). There were high rates of optimism reported – on a scale from 0–40, the mean score was 35.2 (±2.7).
On average, participants reported that they perceived themselves to be members of 4.1 groups (±1.4) and that they had a total of 10.2 contacts (±4.4). On average, per participant, there was a mean of 1.1 red dots; 3.1 yellow dots; 2.1 green dots; and 0.3 blue dots. In other words, the participants (both family members and prisoners) had a total of just over five connections to community resources. In terms of groups that they saw themselves to be part of, there was a mean of around one person linked to high-risk behaviours. Figure 1 provides an example of a completed visualisation map, on this occasion completed by a family member. What is illustrative about this map is the small social network that the family member has – with two groups (friends and family) and a total of seven people between the two. Although the yellow dots denote that the friends have access to community resources in the form of sport and recreation groups, the individuals themselves do not perceive themselves to have access to community capital.
Social identity map and access to community capital from Family member #1. * = links to sports and recreation groups.
The same lack of connections to community resources applies in Figure 2 as well, with the family member having a total of eight connections – 6 family members and two friends. Only one of the friends is perceived to have a connection to a community group, in this case around recovery. What this indicates is that family members may see themselves as lacking sufficient connections to community resources and groups.
Example of a completed map by family member #2. * = Recovery group.
While the numbers are small for inferential analysis, there were no significant differences in wellbeing between prisoners (n = 4) and family members (n = 8) across a range of measures, as shown in Table 1. Across the entire group there were, however, a number of significant correlations in relation to social networks and connections: – Greater number of links to drugs, alcohol and crime (red dots) was associated with markedly lower quality of life (r = -0.74, p < 0.05) – Greater number of green dots (links to community and recovery groups) was associated with better reported physical health (r = 0.78, p < 0.01)
Wellbeing differences between prisoners and family members.
Discussion
There was very mixed access to community resources in both the prisoner and the family member groups, with most of the connections to sport and recreation resources, and very few connections to individuals who could support attempts at accessing employment, training and education. In essence, what the innovative model outlined in this pilot has done is to measure access to community resources that are critical to understanding their capacity to mobilise these resources as a protective effect against substance use and offending on release from prison. What is more, the use of visualisation methods (Czuchry and Dansereau, 2003) has meant that the approach and the process are accessible and meaningful to all participants and provide them with a strong visual representation of their work.
In spite of the small sample size, there were significant associations in the expected direction between the types of community connection and wellbeing. While the general principle of the ‘social cure’ (Jetten et al., 2012) is that more group involvement is associated with better outcomes and health, Jetten et al. (2013) have argued, based on a study of a homeless population, that all groups are not equal and that excluded and stigmatised groups can reduce wellbeing. The current findings fit with this model in identifying a negative association between connections to alcohol, drugs and crime and quality of life, on the one hand, and a positive association between connections to recovery and community resources and physical wellbeing on the other. This is also consistent with Cloud and Granfield’s (2009) notion of ‘negative recovery capital’ where bonds to excluded groups may prevent goals based on resettlement and reintegration.
What the current findings also show clearly is that, for this population, connections to groups cannot be assumed to equate to bridging social capital to community resources. Thus, of the 41 groups and the 102 members identified by participants in the study, only around half (n = 56) had any access to any of the forms of community resource (sport and recreation, recovery and community groups, and education, training and employment) with only one person having a link to the employment, training and education category.
The model that has underpinned this programme of research is that effective community linkage is contingent on first identifying resources in the community (the territory of Asset Based Community Development; Kretzmann and McKnight, 1993), linking these to what the community needs are and the accessibility of identified resources, and then creating the mechanisms for connecting individuals into those relevant groups (McKnight and Block, 2010). This provides a model for building social or recovery capital (in a separate piece – Hall et al., 2018 – we are considering the term ‘resettlement capital’ for those exiting prison) in which assets in the community and mechanisms of bridging to them constitute resources that build capital. What the mapping method outlined above attempts to do is to link the bonding and bridging capacity of existing social networks to a wider process of creating opportunities for assertive linkage to community groups (e.g. Manning et al., 2012). There is clear evidence that assertive linkage into mutual aid groups assists with initiating and maintaining this kind of contact (although there is less clear evidence in other areas such as sport or education), but we do not have good evidence around how blanket processes of assertive linkage can or should be reconciled with individual level social capital and resources. There is considerable potential for practitioners, who have little time or resource to act themselves as community organisers, co-opting family members and friends to take on this role where governance issues can be addressed satisfactorily.
There is an additional issue around the prison system that further emphasises the need for family involvement in this process. The Farmer Report (Ministry of Justice, 2017), which has reviewed the role of families in supporting the rehabilitation of prisoners, argues that not only does enhanced contact with families reduce reoffending rates, but increased family contact may help to break inter-generational transmission of offending and imprisonment. Visits from family or friends have been shown to provide the opportunity to establish and enhance social support networks and can assist the formation of a pro-social identity (Duwe and Clark, 2012). The Farmer Report also suggests that maintaining family contacts can have a positive effect on preventing inter-generational transmission of offending and substance use and can also provide a continuity of contact and support that is the enactment of social capital. What the current project attempts to do is to build social capital through establishing and maintaining relationships through shared goals and tasks that may also provide social support and community engagement for the family members themselves.
Families of those incarcerated are known to experience emotional distress as a result of the physical separation from a loved one, and have also reported feelings of guilt by association (Codd, 2008); the importance of increasing the amount of quality time prisoners and their family member could spend together, and also supporting family members’ attempts to contact groups in the community, is of paramount importance to the programme rationale. The programme therefore included time allocated to a visit between prisoners and their family member outside of the workshop setting, and supported relationship building between the group as a whole. As family members may suffer from secondary stigma and social isolation, the intention was that the programme may also benefit them as a result of this process of assertive community engagement and from being a part of the community connectors group. This is, however, not without risks, and it is important to acknowledge that for some families, the programme may add to the burdens they face and may place burdens on some families that they do not have the resources to cope with. This may be particularly difficult for partners who are attempting to look after children and pay the bills without the support of their incarcerated partners.
Further, as Best and Savic (2015) have previously argued, one of the aims of identifying and mobilising community assets is not only to create pathways but to strengthen the total stock of capital that exists in local communities through coalescing a group of community connectors and providing a sense of shared purpose between professionals, family members and prisoners. The explicit purpose of the community connectors initiative is not only to strengthen the bridging social capital available to prisoners (Almedon, 2005) but also to strengthen the social capital in the community and to build a bridge between the two. The long-term aspiration of this project is to create pathways between prison and community that builds the resettlement capital in each and that creates a ‘therapeutic landscape’ (Wilton and DeVerteuil, 2006) that prisoners returning to the community can actively engage with to support their effective resettlement and reintegration.
There are marked limitations to this paper that need to be acknowledged. The data presented here are of a very small pilot project and we have run only two cohorts of the community connectors project to date, and the data presented here involve only four prisoners (of the six participating in the event) and eight family members (of nine). Additionally, we have no evidence about the impact of the project or even the extent to which prisoners will manage to engage with the community resources that are identified in the project. Category D (open) prisons are not representative of the prison estate and there was an extensive screening process undertaken within HMP Kirkham so we cannot assume that, even within the prison, the sample is in any way representative. Finally, we have no data beyond self-report to test the accuracy or validity of the self-reports provided in the maps included here, and the data for evaluation were collected by the same team who were responsible for delivering the intervention.
Nonetheless, the work presented here represents an important and innovative contribution to the development of a rehabilitative culture in prison and to addressing some of the concerns identified in the recent Farmer Report. Crucially, it also provides a sense of hope and purpose to prisoners (see also Hall et al., in preparation) and to family members about what may be possible for them to achieve after release. This hope can be harnessed by professionals (including but not restricted to probation officers) who may themselves benefit from the process of engaging in this kind of strengths-based social capital building process. There are significant limitations – including the limited access to community capital in the family group – that will require further exploration, but there is also the clear hope that family members can also be supported to build bridging capital and to create pathways to meaningful activities and the linked positive social networks to support rehabilitation and reintegration of prisoners returning to the community. This potential should alert probation and prison staff to the possibilities of co-produced approaches to effective reintegration and resettlement. While significant further work is required, the paper has shown an innovative methodology for mapping access to community resources and an approach to building bridges to community resources to meet the personalised needs of individual prisoners.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received financial support from the university.