Since their emergence in Hamilton, Canada in 1994, following Charlie Taylor's release from prison, Circles of Support and Accountability (CoSA)1 have spread across the globe. CoSA have been implemented across Canada and in parts of the United States2 (Bohmert et al., 2018; Chouinard & Riddick, 2014; Duwe, 2013, Duwe, 2018; Elliott et al., 2013; Elliott & Zajac, 2015; Fox, 2013, Fox, 2014, Fox, 2015, Fox, 2016, Fox, 2017), the United Kingdom (Bates et al., 2007; Bates et al., 2012; McCartan, 2016; McCartan et al., 2014a, b, c; Nellis, 2009; Thomas et al., 2014; Thompson, 2015; Thompson et al., 2017; Winder et al., 2020), New Zealand3 (Lowe, 2017; Lowe et al., 2019; van Rensburg, 2012), parts of Europe (Höing, 2015; Höing, Bogaerts, et al., 2016, 2017; Höing et al., 2010, 2013, 2015; Höing, Petrina, et al., 2016; Höing, Vogelvang, et al., 2017; Petrina et al., 2015), and South Australia (Richards et al., 2020; Richards & McCartan, 2018). CoSA differ considerably by location; while the overview here is designed to offer a general introduction, specifics will inevitably vary.
CoSA have twin objectives: to reintegrate people with sexual offence convictions into the community; and to reduce the sexual victimisation. The aligned twin mantras of CoSA are: “no one is disposable” (or “no one is dispensable”) and “no more victims” (Hannem & Petrunik, 2007; Liautaud, 2010). CoSA also have twin methods for reaching these objectives: support and accountability. The collective mission of CoSA according to CoSA Canada is “to substantially reduce the risk of future sexual victimization of community members by assisting and supporting released individuals in their task of integrating with the community and leading responsible, productive, and accountable lives” (see http://cosacanada.com). In McAlinden's (2007, p. 168) terms, CoSA “have a dual responsibility to be caring for the offender often in the midst of wider public hostility, and at the same time to be concerned that community safety is not compromised”. (see also Heise et al., 1996).
How Do CoSA Work?
How, then, do CoSA work in practice? This is not an easy question to answer, especially as CoSA practice continues to evolve (see e.g., CoSA Canada, 2020). It is important to note here that CoSA are not a “program” as such, although CoSA advocates and scholars (myself included) often slip into using this shorthand. CoSA are not a “program” in the ordinary sense of the word – i.e. the weekly meetings of the core members and volunteers that form the basis of the approach are not extensively structured or formulaic. Rather, discussion commonly focuses on the core member's progress against his aims, struggles that he may be experiencing, and how these might be addressed. In other words, CoSA may vary considerably from one another, and there is no set content being delivered in the same way that there would be in, for example, a treatment program relating to sexual offending. Volunteers do receive guidance on tools they might use within Circles as well as from their training, codes of conduct and so on (e.g., https://www.edmontoncosa.org/volunteering-with-cosa/volunteer-code-of-conduct/). CoSA programs in some parts of the world also receive operational support from an umbrella agency. For example, an operational framework of good practice has provided guidance and support in Canada since 2017, and both Circles UK (www.circles-uk.org.uk) and CirclesEurope (https://www.circleseurope.eu/) provide support for good practice. For the purposes of this book, I use the term “CoSA program” to refer to the organisations that deliver multiple CoSA rather than to individual Circles (e.g., “the Fresno CoSA program”).
As CoSA have developed and matured around the world, they have come to rely on a variety of operating models, receiving funding from a variety of sources (including government and philanthropic) and having varying relationships with the respective criminal justice systems that form their local backdrop. While CoSA historically usually commenced once an offender has been released from prison, as was the case with Charlie Taylor and Wray Budreo, increasingly, staff meet with the offender prior to his release from prison (see e.g., Kitson-Boyce, 2018). This was the case in the flagship Minnesota CoSA (“MNCoSA”) model. The Safer Living Foundation in the United Kingdom has likewise recently adopted a model that commences in prison prior to an offender's release to the community (Kitson-Boyce, 2018). Despite such differences, CoSA programs operate on the same premise: that by providing core members with a Circle of community volunteers who provide practical support and accountability, offenders will be better equipped to lead law-abiding lives in the community. As Almond et al. (2015, p. 27) saw it:
The role of the CoSA is to develop interpersonal contact between the core member and the wider community in order to generate the kind of social capital that militates against future offending; core members openly discuss their behaviours and thoughts with the circle volunteers, and are answerable to them for any deviations from their own aspirations to live non-offending lives.
Individual Circles usually meet weekly in the first instance, although the frequency of meetings may decrease over time as a core member becomes reintegrated into the broader community. At least, this is usually the stated aim of CoSA that Circles last for 12–18 months, by which time the core member is sufficiently integrated into other social and community networks that a formal CoSA is no longer necessary or desired (Hanvey et al., 2011). In practice, however, some Circles, such as those run by the Minnesota Department of Corrections, last for a shorter time (Duwe, 2018). More significantly, many others last well beyond the envisaged 12–18 months period, placing significant resource burdens on CoSA programs (Richards, 2011). Some core members simply lack the necessary skills and attributes to form relationships and networks outside of their Circle, and often remain in a formal Circle, or at least in contact with a CoSA program, for a longer time. This is particularly so for core members who have cognitive impairments and/or mental illnesses and is no doubt exacerbated by the very rigid conditions under which some core members live. Such conditions may prohibit offenders from forming new relationships, such as romantic relationships, without disclosing their offending history. In other cases, Circles may end due to core members being recalled to prison for breaches of their statutory conditions or due to concerns raised about their behaviour by volunteers and/or staff4 (Lowe & Willis, 2019; Quaker Peace and Social Witness, 2005, 2008; discussed further below). Core members may also opt to stop participating in a Circle for a range of practical and personal reasons (Elliott et al., 2013; Quaker Peace and Social Witness, 2005, 2008; Richards, 2011). Importantly, CoSA were initially envisaged to last as long as necessary to mitigate the risk posed by a core member. As Heise et al. (1996, p. 16) stated:
[Ideally, a Circle] maintains the necessary supports for as long as necessary to reduce the risk effectively. Each Circle and plan has its own personality because it is designed around the unique profile of the Core Member. The life of a Circle extends as long as the risks to the community and the Core Member are above average. The intensity with which members of the Circle are intrusively involved in the life of the Core Member varies and will, hopefully, diminish.
In practice, CoSA last for varying lengths of time. Duwe (2013) initially reported that in Minnesota, CoSA generally lasted for at least one year, but later reported that most CoSA last between 6 and 12 months (Duwe, 2018).
Weekly meetings may be supplemented, especially in the early days of an offender's release from prison (Duwe, 2013), with regular contacts (e.g., 2–3 additional contacts per week) with individual Circle volunteers (Heise et al., 1996), at the discretion of a given CoSA program. This depends on the level of need, availability of other community resources, safety concerns identified by the site, and number and availability of volunteers (L. Joly, personal communication, 16 January 2021). Volunteers commonly help core members with the practical tasks associated with re-entering the community after prison, such as obtaining identification documents, finding or settling into temporary accommodation, shopping for necessary items such as clothing and groceries, and so on (McCartan & Kemshall, 2020). As many individuals convicted of sexual crimes have served very long prison terms, they often also require assistance with technological changes to which they have not been exposed in prison, such as electronic banking, smart telephones and electronic public transport passes. Volunteers also help smooth a core member's transition back to the community by simply offering support and encouragement, meeting for a coffee or meal and offering a listening ear. Historically, Circle volunteers may also have provided a liaison point between the core member and agencies such as police, as well as the media and concerned community groups (Heise et al., 1996); in Canada today, this role is undertaken by paid CoSA staff as a matter of good practice, or by others exceptionally with the agreement of staff (L. Joly, personal communication, 16 January 2021).
CoSA are underpinned by a written agreement between the core member and volunteers, sometimes called a “contract” or “covenant”. As Heise et al. (1996, pp. 16–17) explained:
The Core Member and other members of the Circle enter into a mutual agreement, or covenant. The covenant, signed by all Circle Members and the Core Member, establishes conditions of behaviour and associations for the ex-offender and clarifies the expectations to be met by the Circle. It establishes procedures for dealing with breaches of the agreement and failure to meet the conditions. Consequences may include notification of authorities or potential withdrawal of support. The covenant is reviewed at regular intervals and revised when necessary.
While the specific content of covenants (now called Circle agreements in Canada) will vary according to the individual Circle, these agreements typically cover confidentiality expectations, safety procedures and protocols for responding to “warning signs” in the core member's behaviour (Heise et al., 1996, p. 17). In Canada, a national Circle agreement template is now used as the basis of agreements, which each Circle is able to customise to respond to the specific circumstances and characteristics of the core member.
Who Volunteers in CoSA?
You may be wondering who on earth would volunteer their time to support a person convicted of sexual offences. You certainly wouldn’t be the first. Media coverage about CoSA consistently refers to CoSA volunteers in superlative terms, as “courageous, dedicated” (Laycock, 201...