Looking inside a circle: volunteer experiences of circles of support and accountability

Abstract

Circles of Support and Accountability (CoSA) is a promising reintegration framework whereby community volunteers support individuals convicted for sexual offenses to reintegrate safely into the community. CoSA has grown in popularity, yet little is known about volunteers’ experiences of CoSA, including how they cope with their core member being recalled or reoffending. The aim of the current study is to explore the experiences of CoSA volunteers in New Zealand. In-depth interviews were used to explore experiences. The findings relating to general experiences explored balancing risk management and social support, questioning the place of religion in CoSA, confronting stereotypes, and volunteering in a risk-averse society. The findings also revealed both the commitment to CoSA, and the stress and guilt felt by the volunteers following a recall or re-offense. Findings are discussed in terms of their implications for current and ongoing volunteers, as well as for community approaches to criminal justice.

In 1994, the pending release of an individual convicted for sexual offenses in Canada was met with protest, media involvement and community fear. A Mennonite pastor and several members of his congregation agreed to support the individual as he transitioned into the community; the group became what is now known as Circles of Support and Accountability (CoSA). Originating in Canada, CoSA promotes an inclusive approach to the release of individuals convicted for sex offenses so that a safe and successful reintegration may be possible and the individuals can lead a productive, responsible and accountable life (Wilson & Prinzo, 2002). A CoSA is a group of three to seven volunteers from the community that form a supportive group with the individual convicted for sexual offending, known as the core member. So far, evaluations of CoSA have shown a reduction in reoffending rates and positive cost versus benefit analyses (Bates, Williams, Wilson, & Wilson, 2014; Duwe, 2013; Höing, Bogaerts, & Vogelvang, 2013; Wilson, Cortoni, & McWhinnie, 2009; Wilson, Picheca, & Prinzo, 2005, Wilson, Picheca, & Prinzo, 2007b). There are now active CoSAs in the United Kingdom, the Netherlands, Australia, New Zealand and parts of the United States, and CoSA is gaining traction throughout Europe (Bates et al., 2014; Duwe, 2013; van Rensburg, 2012). Working with individuals convicted of sex offenses presents numerous challenges, not least working with one of the most stigmatized groups in society (Kernsmith, Craun, & Foster, 2009). Volunteers are crucial to the continuation of CoSA, yet despite CoSA’s steady expansion, few studies have investigated volunteers’ experiences of working with a population that many would rather never leave prison. The current study sought to explore volunteer experiences of CoSA, including how they deal with recalls or re-offenses.

Why is CoSA so important?

Social support has been identified as a key factor for reducing re-offense risk and promoting an offense-free life (Laws & Ward, 2011). Negative social influences are an established empirically supported risk factor for sexual recidivism (Mann, Hanson, & Thornton, 2010). In addition, there is empirical support for the presence of positive social influences protecting against sexual recidivism. In their retrospective evaluation of prison release plans, Willis and Grace (2008; Willis & Grace, 2009) found that non-recidivists were more likely to have had pro-social support from more than one independent group (e.g. friends, family, work colleagues) on their release from prison than matched groups of recidivists. The overall number of people in a support network did not predict non-recidivism. Whilst social support is important, it remains that individuals convicted for sexual offenses are likely to lack adequate pro-social support upon release; thus CoSA is a unique solution to the lack of pro-social support.

In CoSA, volunteers support the core member in a number ways as he/she transitions into the community; for example, social and practical support includes going out for coffee or a meal, showing the core member how to use public transport, open a bank account, or get a driver’s license. Support also includes emotional and spiritual aspects whereby the volunteers are available if the core member needs advice or someone to talk to. The volunteers also ensure that risk factors are appropriately managed and that the core member is accountable for their actions (Höing, Bogaerts, & Vogelvang, 2014; Wilson, McWhinnie, Picheca, Prinzo, & Cortoni, 2007). Accountability is defined as holding the individual responsible for their decisions, behaviors and the consequences of their actions on themselves and others. In CoSA, accountability is maintained through constructive feedback, asking questions and discussing the core member’s approach to different situations (Wilson & Prinzo, 2002). The support offered by a CoSA is invaluable during the transition from prison into the community; the effects of a lack of social support, relationship and life skill deficits, and the institutionalized result of long-term confinement are bridged through the social support and connection within a CoSA (Fox, 2013).

The benefits of CoSA for the core member and community safety are clear; however, the continuation of CoSA relies on volunteers. As CoSA gains international recognition and the number of CoSAs rise, it becomes increasingly important to give voice to the volunteers behind CoSA, and explore their experiences. Such an understanding will help inform volunteer training and support needs, and ultimately the continuation of CoSA around the globe.

Volunteering with individuals who have sexually offended

Research has shown that in general, volunteering offers benefits to both the volunteer and the community (e.g. Snyder, Omoto, & Dwyer, 2016; Thoits & Hewitt, 2001). Further, people who volunteer are generally happier and more satisfied with their lives than non-volunteers; these characteristics are both a cause and an outcome of volunteering (Thoits & Hewitt, 2001). It is undeniable that volunteering activities have significant benefits for volunteers; however, volunteering with individuals who have sexually offended is arguably unique. For example, the volunteer often has to deal with conflicting emotions, where on the one hand they are appalled by the offense, and on the other hand they are empathic for the individual as a person, and motivated to help prevent a re-offense.

Research examining the effects of volunteering with people who have sexually offended has shown both positive and negative effects (Höing et al., 2014). Positive effects include seeing tangible results for those they are volunteering with, enhancing their own skills, and making connections with other people, as well as the self-reward that comes from volunteering for a good cause. However, volunteering with individuals who have offended sexually is emotionally demanding, and the negative effects include burnout or burnout-related symptoms such as stress and fatigue. Further, the exposure to potentially traumatic material, such as offense details, can have a significant impact on the volunteers, as it does to therapists specializing in working with individuals who have sexually abused (see Moulden & Firestone, 2007). Negative effects might reflect vicarious traumatization (VT). VT is described as a change in the individual resulting from empathic engagement with another person’s traumatic experiences. In therapists working with people who have sexually abused, VT can be exhibited through intrusive thoughts, avoidance behaviors and hypervigilance (Moulden & Firestone, 2007). The primary characteristics of VT include the pervasive impact (affecting all aspects of the therapist’s life), the cumulative effect (each exposure to the trauma increases the impact of the response in the helper) and the potentially permanent effects (Moulden & Firestone, 2007). Given that therapists have had much more training and supervision to manage VT than volunteers have, such effects might be especially pronounced in volunteers. Positive and negative effects of working with people who have sexually abused can exist simultaneously and can be mediated by a number of factors, including personal characteristics of the volunteer, social and organizational support, and the complexity of the volunteer work (Höing et al., 2014).

Few studies have specifically examined volunteers’ experiences of working with people who have sexually abused. Haslewood-Pócsik, Smith, and Spencer (2008) interviewed eight English volunteers involved with IMPACT circles about their experiences – IMPACT circles are based on a similar model to CoSA but with a specific focus on employment. Positive effects reported by the participants included increased confidence, enjoying the connectedness to the other volunteers, professional experience and witnessing clients change. The main negative effect for the participants was the negative reactions they received from friends and family. However, other negative effects included having problems dealing with ‘difficult’ behavior, as well as mixed feelings toward the individual they volunteered with whereby they felt outraged and distressed about their offense, and also feeling sympathy for him/her as a human being (Haslewood-Pócsik et al., 2008).

Volunteering with CoSA

Research exploring the experiences of CoSA volunteers is in its infancy. Volunteering with CoSA involves working long term with the core member; functional and empathic relationships are formed, and sometimes strong friendships develop. However, unique to volunteering with CoSA is the fact that the core member has committed serious offenses that are discussed to varying degrees throughout the CoSA process. The volunteers are made aware of triggers and offending-related behaviors, so they are able to hold their core member accountable when required (e.g. Wilson, McWhinnie, et al., 2007). Thus, the volunteers have to deal with processing the information about the offense(s), and learn about triggers and offending patterns, while maintaining a supportive and accountable relationship with the core member (Höing et al., 2013). Further, volunteering with CoSA can be emotionally and practically demanding; for example, the core member may have complex needs, activities organized must adhere to parole conditions, and the volunteers may have to deal with family and friends who are not supportive of them volunteering with CoSA (McCartan, 2016; van Rensburg, 2012).

Höing, Bogaerts, and Vogelvang (2015) examined the outcome of volunteering with CoSA in terms of volunteer satisfaction, determination to continue and mental wellbeing. The study used a cross-sectional, quantitative design; 40 volunteers answered an online questionnaire. Results from the research showed high levels of volunteer satisfaction and mental wellbeing, and a strong determination to continue volunteering with CoSA. Further, the researchers found that co-worker support and connectedness was a strong predictor of a positive outcome for CoSA volunteers. Wilson, Picheca, and Prinzo (2007a) also used questionnaires to examine the experiences of 57 Canadian CoSA volunteers. The researchers reported that the volunteers felt enhanced feelings of connectedness and a sense of belonging, and had developed new friendships. However, as the volunteers continued with their CoSA, a decline in the volunteers’ expectations about making a difference in the life of the core member was reported (Wilson, Picheca, & Prinzo, 2007a). Further, research from 62 CoSA volunteers in the United Kingdom illustrated that while the volunteers were happy to work with their core member, they compartmentalized their work with CoSA as a protective mechanism against possible negative responses (McCartan, 2016). Using semi-structured interviews and online questionnaires, McCartan (2016) found that the volunteers were hesitant to introduce their core member to other people in their lives and were cautious when explaining what CoSA involved.

Snatersen (2011) conducted interviews about the impact of volunteering with CoSA with eight Dutch volunteers who had been involved with a circle for one year. Participants reported both positive and negative effects; positive effects included increased self-esteem and self-awareness as a result of being a volunteer with CoSA. However, negative effects included stress, ruminating about their core member’s risk in between circle meetings, feeling unsafe, an increased awareness of risky situations, and irritation about the social exclusion that their core member and other individuals convicted for sex offenses faced (Snatersen, 2011). The current study seeks to add to extant research, exploring volunteers’ experiences in New Zealand, including experiences of a core member recall or re-offense that has not been explicitly explored in previous research.

The New Zealand context

In New Zealand, CoSA volunteers are recruited from faith communities, Māori iwi (extended kinship group, tribe, nation, people, nationality, race – often refers to a larger group of people descended from a common ancestor and associated with distinct territory; Māori Dictionary, 2017), and academic staff and students. The criteria for volunteer selection include emotional and social maturity, having a non-judgmental attitude, problem-solving and conflict resolution skills, and compatibility between the volunteer and the core member (van Rensburg, 2012). The volunteer must have no previous serious convictions, especially for sexual offending, and they must have no unresolved issues regarding a sex offense committed against themselves or people close to them. The volunteers must maintain a balanced lifestyle in which they have interests other than CoSA. Financial resources and available transport are also necessary because the timeline for CoSA in New Zealand follows the Canadian model whereby volunteers commit for at least one year post release (van Rensburg, 2012).

The CoSA pilot project in New Zealand began in 2010 and focused on individuals who were serving an indeterminate sentence, or had complex needs and were on an Extended Supervision Order. In 2013, 10 circles had been established in which none of the core members had reoffended sexually, a couple had been recalled back to prison, and six core members had been living in the community for more than two years (van Rensburg, 2014). However, in 2014, a core member failed to return to prison from a temporary release and subsequently escaped the country. Since this incident, CoSA was subject to a review, and the funding for the volunteer organizations contracted to recruit volunteers was not renewed. In 2016, a new pilot program was established targeting men on determinate sentences (Pennington, 2016). Volunteers in the current study were engaged with CoSA prior to recent changes, providing unique insight into the volunteers’ experiences during challenging times such as the incident described.

Volunteering with CoSA can be a particularly demanding experience, both emotionally and practically (e.g. time and money). Further, whilst evaluations have shown a reduction in recidivism, it is a reality that a small number of core members reoffend or are recalled to prison. A close relationship is often formed as the volunteers work closely with the core member during their transition into the community, and often for a number of years after release. A recall or re-offense is a particularly difficult time for a circle and could place the volunteer under significant stress. As yet, there has been no research examining volunteer experiences of a re-offense or recall.

Research aims

The present research aims to examine the experiences of CoSA volunteers in New Zealand using in-depth, semi-structured interviews. The research aims to contribute to extant research by examining the experiences of CoSA volunteers in New Zealand where CoSA has received mixed government support. Further, the experiences of a core member reoffending or being recalled into prison are explored. A qualitative approach was adopted to explore the everyday and ongoing experiences of the volunteers in an in-depth manner. Given the exploratory nature of the research, no hypotheses were generated.

Method

The research context

Data collection for this study began mid-2014 when there were 55 active CoSA volunteers in New Zealand. The research presented in this article is part of a wider study exploring experiences and the motivation to volunteer with CoSA in New Zealand (Lowe, Willis & Gibson, 2017; Lowe & Willis, 2018).

Reflexivity

Reflexivity is an important part of qualitative research and helps establish rigor (Morrow, 2005). Reflexivity requires acknowledging one’s own perspectives on the subject matter, and reflecting on these throughout the analytic process. The primary author has a particular interest in restorative approaches to crime and justice, with CoSA being an example of such an approach. The primary author’s perspective of CoSA has been informed through involvement in CoSA information sessions, prison information sessions, and meetings with a number of the developers and key members of CoSA in New Zealand. In the present research, reflexivity with the data included inviting all participants to review and edit their interview transcripts to ensure the accuracy of transcriptions, and to give participants an opportunity to clarify or add to any of their responses. The primary researcher kept notes during all 18 interviews to aid self-reflection of the transcripts during analysis (Rubin & Rubin, 2011). Further, the primary author discussed developing themes throughout the analytic process with the co-author and used self-reflection to challenge analytic assumptions (Hollway & Jefferson, 2013).

Participants

Recruitment of participants occurred over two recruitment drives whereby the CoSA coordinator sent an email to all CoSA volunteers with information about the study. The information was sent out to 55 CoSA volunteers in the first email recruitment drive, and 49 in the second. Of the 20 volunteers that responded, 18 took part in an interview (14 men, 4 women). Most of the participants identified as New Zealand European, and two identified as New Zealand Māori. Participants ranged in age from 26 to 79 years, with an average age of 57 years. Most of the participants were religious, identifying themselves as Christian. Overall, participants were well educated (see Table 1 for all available demographic information).:

Table 1.

Volunteer demographic information.

Variable	n (N = 18)	%
Gender
Male	13	72.2
Female	5	27.8
Ethnicity
European	15	83.3
Māori	2	11.1
Other	1	5.6
Highest level of education
High school	4	22.2
Post-school certificate/diploma	4	22.2
Bachelor degree	3	16.7
Bachelor with Honors	2	11.1
Masters/Doctoral	5	27.8
Area of occupation
Retiree	7	38.9
Student	4	22.2
Education	3	16.7
Health	1	5.6
Tradesperson	1	5.6
Other professional	2	11.1
Religion
Christian	11	61.1
Spiritual	3	16.7
No religion	4	22.2

The length of time that the participants had been involved with CoSA ranged from less than 1 year to 7 years, and the circles were at varying stages of development (see Table 2). Most of the volunteers were involved with one circle; however, six of the participants were involved in two or more circles directly, or indirectly (assisting setting up new circles).

Table 2.

Details of the circles the participants were involved with.

Volunteer circle involvement	N	n	%
Stage of circle	25
Recently formed		5	20
Prerelease		6	24
Released within a year		4	16
Well-established		10	40
Length of involvement	18
< 1 year		2	11.1
1–2 years		6	33.3
2–5 years		7	38.9
>5 years		3	16.7
Recalls	18
Recall		4	22.2
No recall		14	77.8
Re-offenses	18
Re-offense		3	16.7
No re-offense		15	83.3

Procedure and research design

Data were collected as part of broader research project into the motivation to volunteer with CoSA as well as the volunteers’ experiences. Volunteers interested in participating in the current study were asked to make direct contact with the first author. In-person interviews were then scheduled and took place in private place (interview room at the University or the participant’s home). Interviews lasted between 40 minutes and 1 hour and 30 minutes. Recruitment and interviewing continued until the point of saturation (Mason, 200l. All interviews were recorded and transcribed verbatim with participants’ permission. Participants were given the opportunity to review and edit the transcription of their interview; 10 participants took up this opportunity.

Semi-structured interview

The interview schedule was informed by relevant literature as well as through consultation with the pioneers of CoSA in New Zealand to gain a deeper appreciation of the context. A semi-structured interview format was chosen to bring focus to the interviews whilst not compromising rapport and flow. Questions included, ‘How do you balance providing support to the core member and holding them accountable when they appear to be doing something inappropriate?’, ‘Did you set up boundaries between yourself and the core member? Were these boundaries ever an issue?’, and, when applicable, ‘How did the re-offense/recall of your core member affect you/your circle as a whole?’. At the end of the interview, participants were invited to raise anything that had not been discussed that they felt was relevant.

Data analysis

Following the method set out by Braun and Clarke (2006) the data were analyzed using thematic analysis. The first author transcribed, proof-read and reviewed all transcripts to become familiar with the data. The transcripts were coded using NVivo, a computerized system to organize data for qualitative research. Given that the data were part of larger study, initial coding involved extracting all data relevant to understanding volunteers’ experiences. Volunteer experiences were then coded separately according to two a priori categories: experiences relating to a recall and/or re-offense, and all other experiences. All codes relating to ‘volunteer experiences’ and ‘re-offense/recall experiences’ were extracted and analyzed for common themes. The themes were reviewed and discussed to ensure consistency, and then defined and named. To support the developing themes, quotes were extracted from the data throughout the review process. Analyzing the data was an iterative process of identifying themes and rereading transcripts to ensure clarity and consistency of themes or quotes identified. Discrepant findings or disconfirming evidence were discussed within the research team to avoid confirmatory bias and simplistic interpretations. Contact was maintained with the co-author and research team throughout the research to discuss themes and to facilitate a reflexive approach, minimize biases and ensure consistency of analysis.

Results

Findings are presented according to the a priori categories ‘volunteer experiences’ and ‘re-offense/recall experiences’. Main themes and subthemes identified are described in detail in the following sections.

Volunteer experiences

Most of the volunteers described their involvement with CoSA positively and were eager to share their experiences. Two main themes relating to volunteer experiences were identified: ‘life inside a circle’ and ‘life outside a circle’. Each theme incorporated a number of subthemes that are explored in depth.

Life inside a circle

You can’t have one without the other: risk management and social support

While volunteers were cognizant of the distinct concepts of support and accountability within CoSA, the majority of participants explained that in practice, support and accountability were not separate concepts. Rather, participants emphasized the notion that support and accountability can and should be practiced concurrently, as illustrated by the following quote: ‘Here’s the thing about providing support . . . and holding him accountable, they're not two different things’ (Participant 13).

The volunteers believed that outside the context of a CoSA and the supportive environment that a circle provides, encouraging the core member to be accountable may not be as successful. The volunteers believed that a sense of accountability evolved organically with the growing and strengthening relationship that being a part of a CoSA can foster. That is, accountability is a natural outcome of the supportive relationship, and, likewise, a supportive relationship comes from the accountability inherent to being a member of a CoSA. The volunteers described how once they had established a strong relationship, they felt that they had developed the credibility to comment on core members’ decision making and behavior:

The accountability comes down to establishing a strong, trusting relationship where we do get the information we need to be able to see where he’s at . . . it kind of ties hand in hand because the support gives us a platform to build that relationship, and the accountability comes with us staying vigilant and also checking. So actively checking if things are an issue and potentially challenging him on his behavior . . . and for the relationship to be strong enough so that it can survive that. (Participant 13)

The volunteers explained that they had to establish a method of maintaining contact without being intrusive, be able to question their core member without being accusatory, and allow them the freedom to develop a productive life. One participant clearly described the balance between accountability and suspicion, ‘It’s definitely about asking questions . . . it’s so interesting . . . how do I differentiate between asking a question to create a space of accountability versus asking a question as a really suspicious act that’s quite distrustful?’ (Participant 17). The participants described that accountability did not only come from the volunteers – the core members’ self-accountability was paramount to an effective CoSA. The participants believed that working collaboratively formed the foundation to an effective circle.

Developing relationships

All of the participants described how they believed that a major strength of CoSA was the focus on relationships. The participants described that the relationship between the volunteers and core member differed from that between a parolee and a parole officer because CoSA did not have the same institutional arrangement; that is, the dynamic within a circle allowed for a mutual relationship to develop that minimized the power difference between the core member and volunteers. Many of the volunteers became good friends with their core member and maintained contact after the required commitment time post release. Most of the participants described that their core member, and other individuals who had committed similar crimes, often had no family or friends to support them in the community. For many of the volunteers, the friendship that developed was a natural part of spending time with their core member, but also a fundamental process of CoSA. The following quote illustrates the importance of friendship in a CoSA and in the reintegration process: ‘He is a good friend and I see that as being really important . . . a really important part of his reintegration into society is to have people that he can trust and rely on and feel comfortable with’ (Participant 3).

As well as developing friendships with the core members, the volunteers often developed strong friendships with the other volunteers. Developing a strong relationship with the other volunteer members meant that if issues arose with their core member they were comfortable with talking to another member of their circle to offer or receive support. The friendships developed in a CoSA entailed a significant amount of communication and honesty, both between the volunteers themselves, and the core member with the volunteers. The volunteers all supported each other and communicated following difficult conversations with their core member to minimize the risk of manipulation and enhance their ability to hold their core member accountable. The participants described that their ability to communicate confidently with the other volunteers was affected by the strength of the relationships in the circle; some participants described their circle as being very close and cohesive, whilst others explained that their circle still needed some work to strengthen the relationships and ensure everyone was consistent: ‘It’s based on relationships; the whole thing is based on relationships and they're either good or bad and that's put pretty simply’ (Participant 5).

Different roles, same goals

Often the circles were structured whereby the circle had a leader, co-leader, and someone who kept minutes of the meetings; the volunteers described that they naturally fulfilled roles within the structure created. That is, some volunteers were happy to lead meetings, and others were happy to let others lead. The volunteers described differing approaches to providing support depending on available time, their relationship with the core member, and what they believed they brought to the circle. For example, some volunteers were more comfortable with providing practical support such as supplying clothes or appliances, whilst others were more comfortable with providing emotional support and friendship. The volunteers described that the diverse nature of a circle enabled the volunteers to balance what the other volunteers brought to the circle. The following quotation illustrates how this participant feels more comfortable offering emotional support than practical support:

A lot of it for me is hanging out and checking that he’s ok emotionally . . . if there is any practical bits and pieces I can do I'm not very good at ‘doing’ . . . more of the talking and just that kind of people stuff. (Participant 15)

Some volunteers developed a friendship with the core member, whilst others felt they did not need to be overly friendly with their core member. Indeed, some volunteers distanced themselves personally from core members, and treated their involvement with CoSA as a mentoring relationship. The difference in the approaches appeared to work well as the ‘mentor’ and the ‘friend’ could play different but supportive and complementary roles in the circle.

Most of the volunteers commented on the dynamics of the circle they were involved with. The participants described how different all the volunteers’ backgrounds were, and they believed that being involved with CoSA fostered a lot of diverse discussion. Most of the participants expressed their amazement that a group of people with different backgrounds were united by a single goal. The participants believed that their varied life experiences and different ways for dealing with issues created diverse and adaptable circles; the following quote illustrates the diverse nature of the circles:

We’ve all become friends, we’re a very diverse group, hugely diverse; I mean our interests are like chalk and cheese, our lives are like chalk and cheese. And so it must be incredibly interesting for him, and he’s very involved with all of us and I just think that’s wonderful. Yea, I think he’s lucky. (Participant 3)

Questioning the role of religion

Whilst the volunteers commented on the diversity of circles, a number of participants mentioned the over-representation of religious volunteers. Further, the participants described issues that had arisen between different members of circles with regards to attitudes toward core members who were gay. The majority of the volunteers felt that being a member of CoSA necessitated that the volunteers were open-minded and non-judgmental, and they had felt uncomfortable when they had been confronted by judgmental attitudes toward their core member because he was gay. Some volunteers shared that they had been asked to volunteer specifically with core members who were gay because of some of the issues with a few volunteers not wanting to volunteer with the gay community.

The two people who dropped off the group was because they had some rather pointed views when it came to things like homosexuality and ah . . . I guess the nature of it which . . . definitely not a good fit for the core member, but also the rest of the group was kind of a bit . . . a little bit unsettled by some of the beliefs. (Participant 13)

The volunteers believed that having faith was a positive and often important part of their core members’ lives. However, some of the participants felt that religion created a barrier between some religious volunteers and core members who did not want to participate in a religion. The non-religious volunteers described feeling that there seemed to be a certain level of pressure to attend church events from a couple of the more religious volunteers. The following quotation illustrates the participant’s concern with religiously dominated circles and the possible pressure felt by the core members:

I don’t think anybody should be made to feel excluded if they're not a part of a religion, just like religious people shouldn’t be excluded from things . . . but I think you're always going to be prone to a type of pressure with that sort of thing, whether it is direct or not. I think maybe some of the guys inside would feel like maybe they need to associate with that religion in order to get the help . . . that would be just a gut feeling. (Participant 15)

Life outside a circle

Volunteering in a risk-averse society

The participants described that they felt that the broader correctional and societal perspective was to be overly cautious and restrictive in the approach towards reintegration. The participants felt that in general there was more concern with short-term community safety than with ensuring a safe re-entry into the community for long-term successful reintegration. The participants felt that the cautious approach could hinder the core member’s ability to become an integrated member of the community: ‘They are more concerned, I suppose, about community safety and they don’t want him in a situation where there may be an opportunity to reoffend’ (Participant 4). However, all participants were understanding of the wider socio-political context that criminal justice is practiced within and acknowledged that caution was unavoidable due to the population they were volunteering with. The participants acknowledged that community safety is extremely important and that being overly cautious was a way to mitigate possible reoffending: ‘I guess they have to sort of cover themselves because they have had a few people who have stuffed up on them, and they are left red faced and answerable to all sorts of people’ (Participant 4).

Justifying my involvement in CoSA to others

As a volunteer with CoSA it is inevitable that at some stage, family members, friends or members of the public will ask questions or want to know more about CoSA. The majority of volunteers were hesitant when it came to explaining what they did with regards to CoSA; only a few actually explained that CoSA was focused specifically on individuals convicted for sex offenses. Instead, they explained in general terms of volunteering in the prison, ‘they just see it as quote “prison ministry” really, the detail of how we go about it isn’t really widely known’ (Participant 5). Further, the participants were all acutely aware of how (un)receptive the public could be when it came to talking about individuals convicted for child sex offenses. Gauging how each person would react to information about CoSA defined exactly what they would say, with a number of volunteers never offering more information than the bare minimum. The following quotation illustrates how the participants assessed the possible reactions of people before talking about CoSA: ‘It’s so politically sensitive and it’s really a sensitive issue for a lot of people, so if they're not comfortable talking about or I don’t think they're going to be receptive, I generally don’t get into it’ (Participant 12).

Many volunteers found that when they did explain what CoSA involved, people were more understanding and wanted to know more. The volunteers were happy to talk about CoSA to people who wanted to know more; however, a number of participants shared that explaining what CoSA was often turned into reaffirming confidence in the criminal justice system. The volunteers explained that after briefly describing CoSA to a community member, they would often make an assumption that the core member would have few parole conditions because they were in a CoSA. The participants described that many people to whom they spoke felt that individuals who had sexually offended could not change, and if they were to be released, they needed very strict rules. Thus, the volunteers found themselves reassuring community members that their core member had been through treatment, was monitored by the Department of Corrections, and still had strict parole conditions they had to adhere to. This participant described how he would often end up reassuring people that CoSA did not mean absolute freedom for the core member:

It’s like they need to hear that he's got all these rules that he has to meet . . . and so they need to hear that he's still actually, like he might be outside but he still actually has got a lot of rules and regulations . . . so it’s kind of like you have to reassure people a whole lot about all that sort of stuff before you get to the bit you want to say which is, I actually think he is kind of worth it. (Participant 15)

Not all volunteers were hesitant about explaining what CoSA involved and gladly shared their involvement with CoSA, and specifically that they volunteered with people who have sexually offended. The following quote shows that other people’s judgements fueled how this participant shared their involvement with CoSA:

Where I think other people are just being really prejudiced or violent or something I’ll be heaps more confrontational and challenging and like really push them because I’m like ‘why do you get to think this and not have your ideas tested?’ (Participant 17)

The reactions from the public, once the volunteers had explained what was involved with CoSA, were mixed. Some participants received very negative responses, for example family members disowning them or neighbors avoiding them. Other participants received positive responses whereby some people to whom they had spoken wanted to become involved with CoSA. However, the majority of responses were relatively neutral, as illustrated by the following quotation: ‘I've explained it to them and they've said “look its great what you’re doing, I could never do it”’ (Participant 8). However, due to some negative responses, a couple of the volunteers had changed their approach and how they explained CoSA:

I’m a bit more cautious now about who I say it to, so originally I used to talk about it a lot, but now I’m a bit more cautious, I sort of suss [infer or intuitively figure out] the person out before I talk. (Participant 10)

CoSA is a big commitment and often takes up a lot of the volunteers’ time. The participants who did not share what was involved with CoSA with their family and friends described how they had to compartmentalize their life. That is, they kept CoSA separate from other aspects of their lives. The separation of a big part of their life led some of the volunteers to feel quite isolated about their involvement with CoSA, ‘It feels slightly odd . . . like I feel like I'm doing something really important but I don’t talk about it’ (Participant 15).

Confronting long-held stereotypes

For some of the volunteers, the initial thought of joining CoSA was quite confronting, and before they could commit to a circle they had to reflect on some long-held beliefs. The participants described that prior to meeting their core member and other people in prison, it was easy to keep this population as an abstract ‘other’, or dehumanized, with their only source of information being from the media. Thus, they described how their first meeting with their core member challenged what they initially believed about the population. The following participant described their first encounter with a man who had offended sexually:

I can recall the first time I went in and I met the first prisoner and I shook his hand and I thought ‘oh, I hope I don’t catch something, I better go and wash’ and that's the feeling that you have . . . but immediately after, or shortly after that though I met two other prisoners and told them who I was and what I was trying to do and their answer to me was ‘for god’s sake, get out there and tell them that we’re not all old men with long raincoats and pockets full of lollies’ [candy] and they're not. . . . I do not condone for one second anything that any of these guys have done, but there is a great variance. (Participant 8)

The volunteers shared that through CoSA they had learnt about the large intra-group variation of behavior patterns, risk level, offense profiles and motivations of persons who have sexually offended. For most of the participants, volunteering with CoSA led to a deeper understanding of the issues surrounding sexual offending and stereotypes, and opened their eyes to how the public views this population.

I think having actually worked in the prison system made a big difference to me – made me realize that prisoners are people. I think until then I had had the attitude, which I think a lot of us have . . . ah you know, ‘he’s committed a crime, lock him up, throw away the key’. (Participant 3)

Involved in making a difference

The participants believed that the focus on relationships, support and future thinking positions CoSA as an inherently positive framework. As such, all the participants believed that CoSA was an effective framework for reintegration. The participants described their satisfaction with being able to make a difference to an individual core member, as well as contributing to creating a safer community. The volunteers believed that having a group that the core member could rely on could be the difference between a successful reintegration and an individual struggling to find their place in the community. The participants explained that having a group of supportive individuals for one core member showed the core member that the volunteers believed he/she could successfully desist from crime and reintegrate into the community, thus providing more incentive to live a productive life. One participant described how the community and future orientated nature of circles influenced their core member:

I think it was really great for him to see that people did care . . . and I think that’s been really effective in the way that he sees himself, like in the community I think he sees a way forward now, that he might not have before. And he’s got people he will be able to turn to . . . obviously that’s beneficial and I think he knows that too. It’s given him a real reason to be enthusiastic about things and really hopeful for the future. (Participant 12).

The participants described how they believed that the obvious stigma associated with a sexual offense meant that CoSA was essential to their core member’s reintegration, ‘society gives you that label, and the label is branded heaviest by those people who offend’ (Participant 3). The participants acknowledged how difficult gaining suitable employment or accommodation could be for someone with a sexual offense. Further, the participants believed that it was difficult for their core member to create an alternative identity when they were seen first and foremost as a ‘sex offender’. Many of the participants commented on the importance of living a meaningful and ‘good life’ believing the strengths-based support offered by the circle was fundamental to maintaining the identity shift away from crime. The participants believed that the ongoing support from a circle gave their core member a safe space to foster a productive and crime-free life. After having volunteered with CoSA, the participants believed that a framework such as CoSA would be effective for other populations of offenders: ‘I was looking at it and thinking wow this good, not just for sexual offenders but for other offenders um and I could see this happening I think . . . we’ve got to support it’ (Participant 9).

Uncertainty over CoSA’s future in New Zealand

As previously described, over the course of this research a significant event occurred (a core member fleeing the country while on temporary release), which ultimately led to halting of CoSA funding. Half the interviews took place after the event.

The participants shared that they were disheartened about what they saw happening to CoSA. Following the event, participants described how temporary releases and other reintegrative initiatives, such as work-to-release, were suspended. The participants explained that the biggest issue they had faced was the suspension of the reintegrative initiatives that had been previously available to the core members still in prison. The volunteers believed that the temporary releases were an important part of the reintegration process, and without completing the required number of temporary releases, the core members were precluded from release, ‘they would ask us if he had met the conditions for release but the conditions were temporary releases and stuff so . . . it was an impossible situation’ (Participant 12). Further, the participants who had core members still in prison described how insecure their core members were feeling now that their opportunity for release was increasingly difficult. The following quotation shows the disruption to the release plan for a core member still in prison, ‘When things were taken away – release to work and temporary releases – that was a huge knock for my guy, for the core member . . . so that was a bit of a couple of dark months for us as a group’ (Participant 12).

Some of the participants explained that it was difficult to support their core member when they did not know how long they would be in prison or how long CoSA would continue. Further, the participants also felt that there was a limit to the reassurance they could offer their core member, when they could not make concrete plans regarding temporary releases and their eventual release. The effect of not knowing what was going on with the inquiry process and the uncertainty about the release date for their core member culminated with some the participants feeling that they had lost some of the connection they had with their core member. However, the volunteers all wanted to continue with their core member and worked hard to offer them the support they needed whilst still in prison. The following quotation illustrates how one volunteer continued to offer support while their core member was uncertain about their future:

We talk about things with him and try to encourage him to . . . I don’t need to encourage him to stay positive because he hears that all the time and it’s like a default that he has to stay positive but . . . so yea I guess I’m just there to be understanding and listen to him and I try to plan the future with him but it’s kind of a bit more bleak now. (Participant 12)

Experiences of core member recalls or reoffending

Several participants shared that their core member had been recalled due to parole violations or displaying offending-related behavior, and a few shared that their core member had reoffended and was back in prison. CoSA is a small community in New Zealand so all the volunteers knew the people involved in the circles where the core member had been recalled or had reoffended. Three main themes relating to core member recall or re-offense experiences were identified, ‘the emotional toll of a recall or re-offense’, ‘commitment to support’, and, ‘a recall is not a sign of CoSA failing’.

The emotional toll of a recall or re-offense

The participants who experienced a recall or a re-offense committed by their core member described the experience as an extremely stressful time. The level of involvement in the recall differed whereby some of the volunteers actively worked with the probation officer up to the point of recall, while other volunteers felt completely surprised. Following the recall, the volunteers described feeling emotionally fatigued; this was in part caused by the large amount of guilt that they felt. The participants felt that they had failed their core member because he was back in prison; for example, some participants felt that their circle had not adequately held their core member accountable to prevent being recalled. Some of the participants shared that they felt distressed because they felt they should have done more to prevent their core member from being recalled: ‘I was really upset about it . . . I felt I had failed the core member . . . I just couldn’t get past that feeling that we had somehow failed him’ (Participant 11). To add to the stress of the recall, some of the participants described how their core member blamed their circle for their recall and return to prison: ‘he saw it as the circle’s fault that he was inside’ (Participant 11).

The stress and emotional fatigue were particularly acute for the participants whose core member had reoffended; the participants explained that one of the key principles of CoSA was ‘no more victims’, and because their circle had ended up with a new victim they felt like they had failed, ‘This is hard because you know . . . there’s someone been offended against and another guy in prison’ (Participant 12). The participants also feared that the re-offense would harm the trust and confidence that was being developed for CoSA. The volunteers stated many times that they believed CoSA was an effective framework for reintegration, thus when the core member reoffended the volunteers were devastated. The participants described different ways of coping, for example trying to find a specific reason for the re-offense. However, in doing so, some of the volunteers looked for someone to blame and turned on the other volunteers in the circle in an attempt to create accountability: ‘One of the circle members in that circle has . . . almost turned on the circle and said, “You know, we’re at fault, we should be in prison as well, not only him . . . because we didn’t do this or that or the other” . . . ’ (Participant 5).

The participants also shared that upon reflection, they felt that their circle had not connected well, had experienced conflict amongst themselves, and were not supporting each other to hold the core member accountable. At the time they felt that the circle was managing alright, but a clash of personalities amongst the volunteers meant they felt they did not offer adequate support or accountability. The participants explained that the dynamic between the volunteers in a circle can be positive or detrimental to the success of the circle. Most of the participants described that their circle worked as a cohesive unit both to support and hold their core member accountable. However, some of the participants explained that there were differences regarding what accountability meant; for example, some participants thought a few volunteers were being too intrusive, whilst others thought some volunteers were not utilizing the concept of accountability at all. One participant described some conflict in their circle.

So there was conflict within the circle and one of the . . . one of the rules we were given before is to not to let the core member dominate how things go but that was something our core member did very well. (Participant 1)

A couple of the participants still had lingering guilt that their core member had reoffended, but the participants shared that they had accepted that they were not at fault, and believed that their core member has to be responsible for their own behavior. The participants shared that whilst they would do everything possible, within reason, to support their core member not to reoffend, it was ultimately up to the core member to choose their behavior.

Commitment to support

Despite the stress, guilt and emotional fatigue felt by the volunteers, all the participants were committed to continuing their support of their core member. A couple of the participants described how initially they were going to walk away from CoSA; however, after some thought they decided they were firm in their commitment to both supporting their core member and supporting their fellow volunteers. The following quotation illustrates how one participant continued their support because they did not want to let down the other volunteers in the circle:

When he got recalled to prison it was really a question of ‘good luck, you’ve blown it’, sort of thing. But the rest of the group wanted to keep going . . . we’ve all got different skill sets so . . . pulling out would be letting down the rest of the circle. (Participant 2)

Whilst a couple of the volunteers were initially hesitant with continuing their support, the other volunteers’ support was unwavering. The participants said that they firmly believed in CoSA as a reintegrative framework, and when they had agreed to become a CoSA volunteer, they understood the risks involved. The participants said that even though they had not anticipated that their core member would ever be recalled or reoffend, they signed up to support their core member, and they were committed to continuing. The following quotation illustrates the unwavering support of some of the participants:

When he got recalled to the prison we were asked to disband . . . and I said you know, ‘If we’re going to support this guy, we’re going to support him through thick and thin. Sure, we’re disappointed, sure we feel we have failed somewhere along the line, but if he’s willing then we are . . . there for him’. (Participant 11)

A recall is not sign of CoSA failing

Despite the emotional toll of a recall and the feelings of guilt, the participants stated that a recall could be seen as a sign of failure. However, the participants all shared that they believed a recall was also a sign that CoSA was working. The participants explained that identifying triggers and behavior patterns, and through working with probation they were able to prevent potential reoffending, even if it meant their core member had to be recalled back into prison. The following quotation illustrates the idea that a recall is not a failure because they avoided a possible re-offense:

The circle had to have been effective because the aims of circles is for a safe community and no more victims. One, we had no more victims, and two, we recognized manipulative and controlling behavior before it could lead to grooming. (Participant 11)

The following quotation also illustrates the idea that a recall shows CoSA is working because not only did they avoid further victims, through the process of the recall they identified behaviors and attitudes that the core member still needed to work on:

The fact that he got recalled first time round . . . you could sort of say, some might say ‘well the circle failed didn’t it?’. Well I actually think that the circle worked because it brought out certain behaviors within him which hadn’t been sufficiently addressed. (Participant 2)

The participants whose core member reoffended felt that although they believed their circle had failed, CoSA as a program was not a failure. Further, they still believed that CoSA was a beneficial framework for reintegrating not only people who had sexually offended, but other offender populations. However, they also felt that their circle could be used as an educational tool for future circle members, and recruitment and training strategies. Further they believed it was essential that the circle was examined as a whole, as well as given a chance to reflect on what they did wrong and what they did right. Whilst they believed that the re-offense was not their fault, they also believed that as a circle they could have done some things differently and that recalls/re-offenses provided an opportunity for reflection: ‘It’s also meant that the circle has had to review what we did, what we didn’t do at the time, whether we should have been a bit stronger’ (Participant 5).

Discussion

The present research aimed to explore the experiences of CoSA volunteers in New Zealand. Using the a priori categories ‘general experiences’ and ‘experiences of recall or re-offense’, the transcripts were analyzed using thematic analysis. Two main themes were identified for general experiences: ‘life inside a circle’ and ‘life outside a circle’; four subthemes were identified within each overarching theme that spoke to the life-enhancing and sometimes challenging experiences that volunteering with CoSA entailed. Four themes relating to recall/re-offense experiences were identified that highlighted the seemingly contradictory experiences associated with re-offense and recalls: ‘the emotional toll of a recall or re-offense’, ‘commitment to support’ and ‘a recall is not a sign of failure’.

Life inside a circle

The ‘life inside a circle’ theme has subthemes that illustrate the complex and adaptable nature of a circle. The ‘you can’t have one without the other: risk management and social support’ subtheme highlights the ongoing negotiation required: negotiation amongst the volunteers, as well as with the core member. Further, consistent with research presented by Fox (2016), the findings of the present study explained that the concepts of accountability and support should be practiced concurrently. Further, the relationships developed through the CoSA process enable the volunteers to comment on the core member’s behavior whilst remaining supportive.

Whilst the volunteers commented on the diverse nature of the circles, the role of religion was questioned. Faith can play an important role in the reintegration process; previous research has noted that religion can provide a support network and access to pro-social peers, offer moral guidance and opportunity to seek forgiveness, and help bring meaning and psychological comfort (Kewley, Beech, & Harkins, 2015; Kewley, Larkin, Harkins, & Beech, 2016). However, findings from the present research also indicated that the pressure to endorse a particular faith may be counter-productive to the process of CoSA. Further, judgmental attitudes associated with some religious beliefs created uncomfortable situations for some volunteers and alienated core members. The volunteers in New Zealand are primarily recruited from faith communities (van Rensburg, 2012; Lowe et al., 2017) and a number of the volunteers were motivated to volunteer to offer a different perspective to religious volunteers (Lowe et al., 2017). Findings from the current study highlight that it is important to find a balance of perspectives as a reliance on religious volunteers could be detrimental to core members who may feel alienated by religious beliefs.

Life outside a circle

The subthemes of ‘life outside a circle’ highlighted both the challenges and the rewards that the volunteers face as a result being involved with CoSA. The volunteers described having to navigate explaining to others what CoSA involves, volunteering with a stigmatized population and working in a risk-averse culture. Participants were met with the reality that many community members would rather persons convicted for sex crimes never leave prison, yet alone be supported (e.g. Craig, 2005; Kernsmith et al., 2009; Levenson, Brannon, Fortney, & Baker, 2007), which meant volunteers ended up reassuring others that being in a circle did not mean absolute freedom for the core member and he/she did still have restrictions. Whilst the volunteers were positive about their involvement with CoSA, many would gauge how (un)receptive others would be to the framework before sharing details about their involvement with CoSA. Consistent with previous research, the cautious approach to telling other people about CoSA suggests volunteers separate their work with CoSA and compartmentalize it as a protective mechanism (McCartan, 2016). Whilst compartmentalization may offer short-term benefits, as a long-term approach the implications for the volunteers may be negative. A significant amount of time is required of the volunteers, especially in the early stage of an offender’s release. Separating such a big part of their life has implications for when stressful events arise inside and outside of CoSA, and volunteers may find themselves lacking support themselves outside of CoSA.

Experiencing a recall or re-offense

The four themes identified relating to the experiences of a core member recall or re-offense emphasized that volunteering with CoSA can be very demanding. The themes highlight the different feelings of failure associated with a recall or reoffending; for example, the volunteers felt they had failed their core member following a recall because they felt personally responsible for keeping them out of prison. Further, the volunteers felt they had failed the community following a re-offense because one of the main tenets of CoSA is having no more victims. The volunteers had to deal with the complex situation with their core member, as well as their own emotions. The dichotomy of emotions described whereby they felt upset at their core member being recalled or appalled at the re-offense, but they were also committed to continuing with their support. The volunteers made it clear that they did not condone what their core member had done as a way to manage their feelings toward to recall/re-offense, whilst maintaining the view that their core member is still a human being who needs support to transition from an offending identity to a non-offending identity.

The current study is the first to explore volunteer experiences of recalls or re-offenses. Unlike the professional members of CoSA, or therapists working with individuals convicted with sexual offenses, the volunteers do not have strict professional boundaries or supervision requirements that must be maintained. The themes identified emphasize the need for appropriate supportive follow-up to situations where the core member is recalled or reoffends, as well as ongoing support during other challenging times. Findings highlighted the importance of preparing the volunteers for the possibility of a recall/re-offense (e.g. that it’s not their fault, they cannot control the core member, etc.). The demanding nature of being a CoSA volunteer necessitates adequate training and preparation for challenging situations, as well as ongoing and follow-up support.

Implications for community approaches to criminal justice

Whilst the dominant discourse in criminal justice is punitive, the results of this study illustrate that many people are in fact open to discussion regarding community engagement with criminal justice. The participants described some negative responses from the public, but the general response was relatively neutral; that is, many community members were encouraging of the volunteers and CoSA, but were not willing to volunteer themselves. Further, most people had never heard of CoSA, but thought it was beneficial that the core member had a dedicated group supporting them. The risk management paradigm dominates correctional practice, and public attitudes remain negative towards the reintegration of individuals convicted for sexual offenses, thus positioning CoSA as a risk management tool enables a greater level of acceptance with the public.

Silverman and Wilson (2002) suggest that a viable solution to community violence is found when the criminal justice system engages the community. CoSA engages the community to take an active role in managing risk while supporting individuals who have abused to become productive members of the community. The response from the public shows that perhaps not everyone is as unwilling to engage in discussion about community management of individuals convicted for sexual offenses being released from prison. Indeed, whilst community members may never fully accept this population, the volunteers’ experiences provide evidence contrary to the prevailing negative attitudes towards people who have sexually abused (e.g. Willis, Malinen, & Johnston, 2013). The volunteers’ experiences of explaining what CoSA involves demonstrates that community members can and are able to move constructively beyond the understandable negative emotions elicited by sexual offending.

Volunteers’ involvement in CoSA provides a platform for discussion about sexual violence prevention in the community: CoSA as a restorative justice approach is inconsistent with prevailing public discourse around sex offending – for example, the focus on labelling and stigmatizing language used for individuals who have sexually offended or the wide support for punitive policies (Harris & Socia, 2016). The experiences of the volunteers, as well as the success of CoSA, shown through the reduced risk of reoffending and promotion of desistance (e.g. Duwe, 2012; Wilson et al., 2009), illustrate the value of restorative justice initiatives and volunteer-driven frameworks. Further, the reduction in recidivism demonstrates that restorative justice can be used effectively in a risk society in a way that empowers communities to include ex-offenders and build social support networks rather than reject and isolate them.

Limitations

As is common to qualitative research in general, the analysis and interpretation of the transcripts were undoubtedly influenced by the researchers’ personal experience and perspectives. Further, people’s experiences are understood as the product of a broader context, and in an interview context, the experiences conveyed are the product of both the interviewer’s and participant’s experiences as well as their interactions during the interview (Stephens, 2011). A number of steps were taken to avoid biases and to ensure consistency; first, the first author conducted and transcribed all interviews. Second, the participants were given access to their transcripts to review and make any edits to ensure they reflected what the participants wanted to say. Finally, the analysis included immersion of the data by transcribing, proof-reading and rereading the transcripts, and further ongoing discussion with the co-author and reflection of codes and themes were used to challenge analytic assumptions and ensure consistency.

An additional limitation includes the context in which the study was conducted – but could also be seen as an advantage. It is possible that the core member absconding while on temporary release that occurred during the course of this research might have impacted what information or how much information the volunteers divulged. By contrast, the publicity surrounding the absconding and impact on CoSA could have motivated some volunteers to participate to have their voices heard. The latter explanation seems more likely given participants’ eagerness to talk and the richness of data obtained.

Conclusion

CoSA started as a grassroots initiative in Canada in 1994. It is unlikely that those initially involved would have anticipated the level of international traction CoSA has gained since then. An individual convicted for sexual offenses reintegrating into the community could be seen as a risky liability; however, community volunteers chose to take a chance on someone most people would rather never left prison. It is hoped that findings from this and related studies will support the continuation and expansion of CoSA internationally, through supporting the people that make CoSA possible.

Ethical standards

Declaration of conflicts of interest

Giulia Lowe has declared no conflicts of interest.

Gwenda Willis has declared no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee [University of Auckland Human Participants Ethics Committee] and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study