Framing Perpetrators of Sexual Violence Who Participate in Circles of Support and Accountability (CoSA): An Analysis of Global Print Media Reporting

Introduction

Circles of Support and Accountability (CoSA) have become part of the criminal justice landscape across North America, the United Kingdom, Western Europe, and Australia (see Thompson et al., 2017 for an overview). CoSA are groups of trained community volunteers who support a person convicted of sexual offending (often against children) to reintegrate into the community after prison. CoSA have the twin objectives of reintegrating those convicted of sexual crimes (known in CoSA parlance as “core members”) into the community and enhancing the sexual safety of the community. CoSA operate on the premise that by providing core members with support from community volunteers, offenders will be equipped to lead law-abiding lives in the community (Almond et al., 2015). CoSA involve weekly meetings of the core member and his volunteers, supplemented by other interactions such as telephone calls, aimed at supporting the core member to reintegrate and reduce his risk of reoffending. Volunteers (with the backing of paid staff) provide both practical and emotional support to core members (Bohmert et al., 2018). They may, for example, help core members with obtaining identification documents, finding or settling into temporary accommodation, shopping for groceries, and so on (McCartan & Kemshall, 2020) as well as providing a “sounding board” for the core member (see Richards, 2022, for a recent overview). Volunteers also play a monitoring function, ensuring that the core member meets the requirements of his release from prison and is not engaging in any “pro-offending” behaviors (Quaker Peace and Social Witness, 2008).

Evaluations of CoSA in Canada (Wilson et al., 2007, 2009) and the United Kingdom (Bates et al., 2014) demonstrate significantly decreased rates of sexual and other reoffending among CoSA participants than among comparison groups. Most significantly, Duwe’s (2013, 2018) randomized controlled trial of CoSA in Minnesota, USA, found significantly less sexual recidivism among core members than a control group. (It should be noted that these studies used small samples of core members, and that they do not conclusively demonstrate that CoSA reduce sexual recidivism. As Clarke et al. (2017) argue, studies with larger samples and longer follow-up periods are required to more conclusively determine CoSA’s effectiveness (see also Kitson-Boyce, 2018; Richards, 2022; for recent overviews of the research evidence about CoSA)).

CoSA are radically different from other policies for dealing with those convicted of sexual offenses, which often involve long-term or indefinite detention, community notification and public offender registers, all of which have the primary effect of excluding and vilifying offenders (see e.g., Duwe, 2015). CoSA resist the dominant characterization of people convicted of child sexual abuse inherent in these approaches, operating according to a very different imagined offender—one who is able to desist from further offending and who ultimately can (and should) be accepted as part of the community. CoSA recognize the social context of offending by positing the notion of “community” as a powerful tool to reduce sexual offending.

Media coverage of child sexual abuse rarely focuses on prevention or solutions,- instead focusing on the salacious details of individual cases (Kitzinger & Skidmore, 1995; Mejia et al., 2012; Thakker & Durrant, 2006). It is therefore somewhat surprising that CoSA have received considerable media attention. Understanding how CoSA are portrayed in the media is important because the public obtains much of its information about crime, including child sexual abuse, from the media (Davies et al., 2017; Levenson et al., 2007; Malinen et al., 2014; Mejia et al., 2012). Public opinion can shape criminal justice policy in this area (Pickett et al., 2015; Schiavone et al., 2008). As Mejia et al. (2012) state, media coverage of child sexual abuse also plays an important educative function, “helping policymakers and the public understand what must be done to prevent future assaults” (p. 471).

While there has been no research to date about media representations of CoSA, there is an existing body of research on how child sexual abuse and those who perpetrate it are portrayed in the media. Sexual offending against children has been the focus of much media reporting (Davies et al., 2017; Galeste et al., 2012; Hove et al., 2013; Landor & Eisenchlas, 2012; Shelby & Hatch, 2014). While there has been only limited scholarly attention to media portrayals of child sexual abuse specifically (as distinct from sexual violence more broadly) (Mejia et al., 2012; Weatherred, 2015), there is consensus that such reporting is typically sensationalistic (Ducat et al., 2009; Galeste, 2010; Malinen et al., 2014), and focused on instances of child sexual abuse that are highly atypical (Ducat et al., 2009; Jenkins, 1998; Malinen et al., 2014; McCartan et al., 2015; Mejia et al., 2012; Thakker & Durrant, 2006; Weatherred, 2015), such as the abduction, sexual assault, and murder of children (Jewkes & Wykes, 2012; Landor & Eisenchlas, 2012).

Consensus also characterizes the existing research on how people convicted of child sexual abuse are portrayed in the media. This scholarship demonstrates that people convicted of sexual offenses are characterized according to a range of recurring tropes. It has been consistently documented that perpetrators are depicted as compulsive, specialist offenders (Ducat et al., 2009) with a high likelihood of recidivism (Galeste, 2010; Galeste et al., 2012; Shelby & Hatch, 2014) and an inability to be rehabilitated (Galeste, 2010; Galeste et al., 2012; Shelby & Hatch, 2014). Thakker and Durrant’s (2006) analysis of newspaper coverage of sexual offending in New Zealand found that less than 3% of articles espoused the view that individuals convicted of sexual offenses can or should be rehabilitated, and Mejia et al.’s (2012) study of newspaper coverage of child sexual abuse in the USA found that only 1% of articles espoused this view.

Those who perpetrate sexual offenses against children are invariably positioned as “others” (Grealy, 2014; Greer & Jewkes, 2005; Jenkins, 1998; Jewkes & Wykes, 2012; Shelby & Hatch, 2014); that is, excluded from mainstream society due to being perceived as a threat to the social order (Barlow, 2015; Schuster et al., 2018). This othering takes the form of portraying those convicted of sexual offenses as animalistic or sub-human (Davies et al., 2017; Ducat et al., 2009; Jenkins, 1998; Kitzinger, 2004; Landor & Eisenchlas, 2012), evil, perverted, or sick (Davies et al., 2017; Ducat et al., 2009; Kitzinger, 2004; Thakker & Durrant, 2006). People convicted of sexual crimes are frequently pathologized (Mejia et al., 2012; Shelby & Hatch, 2014; cf Hove et al., 2013), psychologized (Grealy, 2014), and “homosexualized” by the media (Grealy, 2014; Kitzinger, 2004; Ostrowski, 2009). While a significant majority of child sexual abuse is committed by men against female children (Dowling et al., 2021), abuse against male children receives disproportionate media attention (DiBennardo, 2018).

Taken together, these tropes work to abjure the reality that sexual offending, like all other forms of criminality, is “the product of broader social conditions” (Mejia et al., 2012 p. 480). The literature demonstrates that the actions of people convicted of sexual offending are presented in media accounts as stemming from fixed, individual traits, without reference to the social context of offending (Grealy, 2014; Kitzinger & Skidmore, 1995; Mejia et al., 2012; Shelby & Hatch, 2014; Weatherred, 2015). Hove et al. (2013) argue that this is deeply problematic, because portraying sexual abuse in individualistic terms undermines the important role that the community can and should play in preventing it.

Much less research has examined media portrayals of policy or practice responses to (child) sexual abuse. The little existing research suggests that a common theme in this regard is the perceived failure of the criminal justice system to protect the community (Ducat et al., 2009). “Bureaucratic bungling,” the failure of experts and professionals (Kitzinger, 1999; Thakker & Durrant, 2006), and the need for a harsher criminal justice response to perpetrators (Ducat et al., 2009; Mejia et al., 2012; Thakker & Durrant, 2006) are also commonly highlighted.

Two key pieces of scholarship suggest some level of divergence from the overarching themes that characterize these interrelated bodies of literature. First, Hove et al. (2013) analyzed child abuse-related articles published in eight US-based newspapers between 2000 and 2008. They found that in contrast to existing research, societal causes and solutions were posited in the majority of articles on child sexual abuse (76% and 88% of articles respectively). Hove et al. (2013), p. 105) explain this surprising finding by arguing that individualistic framing has diminished over time, and that a lack of regard for social context no longer characterizes newspaper coverage of child sex offending.

Second, Cheit (2003) systematically analyzed newspaper coverage of individuals charged with sexual assault against a minor in Rhode Island between 1993 and 2001. While Cheit’s (2003) analysis to some extent affirms the findings of previous studies (e.g., that “stranger danger” is disproportionately highlighted), it also found that a majority of defendants were never mentioned in a newspaper article, and that nearly two-thirds of those who did appear in newspaper reports were mentioned only once in passing. Cheit (2003) found that the language utilized in newspaper articles was rarely salacious or sensationalistic, thus challenging popular and scholarly claims about “media hysteria” over child sexual abuse, which he likens to “an urban legend. . .told and accepted without appropriate skepticism” (p. 619).

On the face of it, one might predict media accounts of CoSA to be largely negative and sensationalistic, in line with how sexual offending is typically framed. An approach treating people convicted in relation to sexual violence as human beings in need of community support seems unlikely to garner sympathetic reporting. But is this indeed the case? The current research addresses this question by presenting the first analysis of media representations of CoSA. It also adds to the existing literature, outlined above, on how the media portrays people convicted of (child) sex offending more generally. Using framing theory, the study involved analyzing 167 articles about CoSA published in English-language newspapers around the globe. Overall, it identified that media accounts of CoSA depart in key ways from those identified in previous scholarship, suggesting that the nature of CoSA as an intervention may allow for more sympathetic and humanistic representation.

Conceptual Orientation

Media frames can be understood as the “conceptual tools which media and individuals rely on to convey, interpret and evaluate information” (Neuman et al., as cited in Semetko & Valkenburg, 2000, p. 94). According to pioneer Goffman (1974), frames help individuals to make sense of life experiences; they are “schemata of interpretation” used to “locate, perceive, identify, and label” (p. 21; see also Entman, 1993). As recent framing scholarship has been criticized as lacking a clear conceptualization of framing (Cacciatore et al., 2016), Goffman’s (1974) focus on existing schemata is important to highlight. Price et al. (as cited in Cacciatore et al., 2016) argue that to be a useful exercise, framing analyses must not focus solely on the information presented, but on how the “attributes of a message (its organization, selection of content, or thematic structure) render particular thoughts applicable” (p. 16). Price and Tewksbury (as cited in Cacciatore et al., 2016) call this the “knowledge activation potential of news story frames” (p. 16)—in other words, how messages resonate with existing cultural schema or “contextual frameworks of reference” (Lavie-Dinur et al., 2015, p. 327) to inform audience interpretation of the message.

Identifying and understanding frames is important since they are a “powerful mechanism that can help define and solve problems and shape public opinion” (An & Gower, 2009, p. 107; see also Semetko & Valkenburg, 2000). Frames help us understand how the public make sense of an issue or event (Khakimova Storie et al., 2014) and inform audiences’ perceptions of what should be done about a particular issue (Hove et al., 2013; Mastin et al., 2007; Mejia et al., 2012). Media frames are thus vital to understand in relation to CoSA, given both the strong relationship between media portrayals of those convicted of child sexual abuse and public opinion about them, and that CoSA rely on media as a key source of volunteer recruitment.

As set out in Semetko and Valkenburg (2000), there are five specific frames of analysis seen as consistently active across a wide range of news media. The conflict frame emphasizes conflicts among individuals, groups, and/or institutions to capture audience interest. The human interest frame seeks to personalize, dramatize and/or emotionalize a topic in order to capture audience interest (An & Gower, 2009); it emphasizes the “human face” of an issue. The economic consequences frame focuses on the economic consequences of an issue for individuals, groups, or countries as a whole (An & Gower, 2009; Semetko & Valkenburg, 2000). The morality frame contextualizes issues by reference to religious or moral tenets (Schulenberg & Chenier, 2014). This is often achieved indirectly (e.g., by including a quote from an expert) due to the news media needing to appear objective (An & Gower, 2009; Semetko & Valkenburg, 2000). Finally, the attribution of responsibility frame presents issues in a way that attributes responsibility to a particular person, group, and/or institution (An & Gower, 2009). For example, in news stories about poverty, individuals are often seen as the cause of their own financial distress (Semetko & Valkenburg, 2000).

Research Design

The Factiva, Westlaw, UKPressOnline, Proquest News Databases, and Australian and New Zealand Newstream databases were searched on 29 September 2021 for articles containing the term “Circles of Support and Accountability” that had been published prior to this date. In the case of Factiva, the search was narrowed to articles with “Circles of Support and Accountability” in the title or lead paragraph. For Proquest News Databases, the search was narrowed to the abstract, and for UKPressOnline, the search was narrowed to the body content, as this was the only available option. No narrowing function was available for Australian and New Zealand Newstream, and none was required for the Westlaw database. For all databases, the search encompassed all dates, authors, subjects and regions of the world. The following were excluded from the search: (1) Articles not in English; (2) Articles about CoSA not specifically relating to sexual offending; (3) Industry publications (e.g., Corrections Today); (4) Advertisements designed to recruit CoSA volunteers; (5) Announcements about government contracts such as Requests For Tender; (6) Letters to the Editor; and (7) Duplicates. (Articles that were largely similar but not identical were included in the study if they were published in different outlets, since this reflects the extent of newspaper coverage across different geographic regions).

While there are limitations to focusing solely on newspapers and excluding other forms of media such as television and social media, newspapers are important to analyze as a first step since they often set the agenda for other media (Davies et al., 2017; Mejia et al., 2012). Print media are also the most accessible form of media and enable researchers to examine coverage over longer periods (Mejia et al., 2012).

All articles were imported into qualitative data analysis software program NVIVO for coding. Following Khakimova Storie et al. (2014) and Ozascilar and Ziyalar (2015), a qualitative content analysis was used to explore the content of these frames. Both headlines and article text were coded deductively. While many scholars have sought to identify “issue-specific frames” (e.g., Ostrowski, 2009), key framing theorist Iyengar has argued that framing analyses should be at least partly deductive to “explore frames that previous research has shown to resonate well with culturally shared schemas among audiences” (Cacciatore et al., 2016, p. 14; see also Semetko & Valkenburg, 2000). As such, a process of deductive coding was undertaken to determine the presence of the five common media frames outlined above. In tandem with this, and in accordance with good practice in qualitative research (Elliott, 2018), inductive coding was also used to identify more finely-grained ways in which offenders were represented within and across these five frames.

Framing CoSA in the print media

A total of 167 print media articles from national, regional and local newspapers—both broadsheet and tabloid—were located as a result of this search strategy. While it is undoubtedly the case that not every print media article on CoSA was captured, the articles analyzed nonetheless could be considered a representative spread. Slightly more than half the articles (n = 86; 51%) were published in Canadian newspapers, followed by the United Kingdom (n = 61; 37%). The remainder were published in Australian (n = 8; 5%), Irish (n = 6; 4%), American (n = 5; 3%), and Latvian (n = 1; <1%) newspapers. The articles ranged in length from 61 words to 5,394 words with a mean of 777 words. The remainder of this section sets out the ways in which print media representations of CoSA reflect the five (sometimes intertwined) media frames identified by framing scholars.

Discussion

Media portrayals of CoSA defy overarching characterization, ranging from salacious and pronounced resistance to evangelical support for CoSA (although in the main, inflammatory media representations of CoSA core members were found to be infrequent and in general to have appeared early in print media coverage of CoSA). This finding provides support to Cheit’s (2003) claim that scholarly claims about “media hysteria” about child sexual abuse and those who perpetrate it ought to be subject to a higher degree of skepticism.

Nonetheless, newspaper accounts of CoSA mark a change in the portrayal of those who perpetrate sexual abuse in the media. While the five generic media frames are used frequently throughout the international print media on CoSA, supporting framing theorists’ contention that most content can be categorized as belonging to one of these (Cacciatore et al., 2016; Schulenberg & Chenier, 2014; Semetko & Valkenburg, 2000), our analysis reveals features of CoSA print media coverage that challenge existing cultural schema. Taken as a whole, the framing of CoSA in newspaper articles signals a departure from three key cultural frameworks, explained in turn below.

First, while the schema of the untreatable offender is not entirely absent from newspaper coverage of CoSA, it is evoked surprisingly infrequently. Our analysis also documents an important counterpoint to this schema that emerges in CoSA print media. An examination of the framing of responsibility for child sexual abuse revealed that in CoSA newspaper articles, perpetrators are commonly represented as “pedophiles”—that is, as having an immutable sexual interest in children. Importantly, however, these men are depicted as actively managing their sexual selves with the support of their CoSA. Core members are thus positioned as active participants in a policy of rehabilitation—rather than passive and likely resistant recipients of control measures such as sex offender registers. This framing both resists and provides an alternative to the schema of offender untreatability, positioning offenders as actively seeking desistant lives. This is in stark contrast to the usual construction of the offender as a habitual abuser, helpless in the face of his innate illicit desires.

Second, CoSA print media coverage begins to challenge the schema of individual responsibility for child sexual abuse, and to frame abuse in social context to a greater extent than might be expected given previous research on media portrayals of this issue (see especially Mejia et al., 2012; Shelby & Hatch, 2014). In contrast to most previous research, we found a mix of explanations, both individual and social/environmental, in newspaper coverage about CoSA. While individualized explanations were apparent at times, on the whole the articles examined present a more contextualized portrayal than previously suggested, often resisting more directly individualizing discourses. For example, offending was at least sometimes characterized as stemming from adverse childhood experiences such as prior victimization or exposure to dysfunctional family environments. Perhaps more significantly, even where individual-level causes are posited, community-level solutions are espoused in the CoSA print media. One subtle example of this is the construction in newspaper accounts of those responsible for child sex offending as modern-day lepers. This moral framing is noteworthy because instead of a focus on individual offenders as devils, it is the social response to them—treating them as lepers—that becomes a motif in the print media. Here, it is the community’s response to abusers, rather than the abuser himself, that is framed as problematic. This departs from typical media portrayals of those convicted of child sexual abuse as pathological or evil, and instead emphasizes the social context of child sexual abuse, thus representing a significant departure from media content about child sexual abuse. On the surface this finding may appear to support Hove et al.’s (2013) position that societal rather than individual explanations of child sexual abuse have been become increasingly commonplace over time. However, as newspaper accounts of CoSA adopting a social view of causality were published from as early as 2003, we posit instead that the nature of CoSA as a practice engenders this type of narrative. This is significant as it demonstrates that other narratives about abusers are possible in media portrayals, and that the media can produce alternatives to dominant narratives. It suggests that as a progressive practice, CoSA can give rise to new ways of portraying child sexual abusers that are more likely to foster support for effective preventative policies and practices. Thus it appears that while the image of the child sex offender as monster (McCartan, 2004) is not entirely absent from CoSA reporting, it is not monolithic or inevitable, again lending support to Cheit’s (2003) analysis. There is evidence that more complex public narratives—potentially leading to more balanced public opinion—are more commonplace than might be imagined. This follows on from studies that show that media audiences are both more informed and more actively engaged in creating messaging around sexual offending than has previously been documented (Richards & McCartan, 2018).

Third, CoSA are framed in such a way in the global print media as to challenge the cultural schema of the failed criminal justice system. As noted earlier, previous scholarship has identified that the notion of a failed system is a key motif in media coverage of policy and practice responses to child sex offending (Ducat et al., 2009; Kitzinger, 2004; Thakker & Durrant, 2006). This narrative is likewise often invoked in CoSA newspaper articles; importantly, however, in some instances it provides a platform for the consideration of progressive alternatives, including CoSA. This is noteworthy as it demonstrates that the conflict frame can be co-opted in support of progressive policy and practice responses.

None of these departures from existing cultural frameworks in the print media is pronounced or absolute. Nonetheless, given the dominance of particular ways of framing sexual offending in media accounts previously, these nascent challenges are notable. There are two potential explanations for them. First, it is possible that newspaper coverage is simply becoming better informed about people who perpetrate sexual violence. It is noteworthy for example that while previous scholarship has identified the “homosexualization” of offenders (Grealy, 2014; Kitzinger, 2004; Ostrowski, 2009), this is virtually absent from newspaper accounts of CoSA. However, as noted above, even very early news coverage of CoSA challenged dominant cultural schema. Second, the nature of CoSA as a practice may have enabled these challenges to dominant cultural frameworks. CoSA provide a community response to the problem of child sexual abuse and are premised on the assumption that those with sexual offense convictions can be supported to create offense-free lives in the community. These features of CoSA may give rise to media coverage that emphasizes offenders’ capacity for change and the role of social determinants of and responses to sexual offending. As such, media about CoSA provide an opportunity not only to inform the public about CoSA but about child sexual abuse—its causes and potential responses to it—more broadly.

Limitations and Future Directions

The current study examined the use of media frames in English-language newspaper articles since the inception of CoSA. While this is an important first step toward understanding media portrayals of CoSA, other approaches could be undertaken in future. For example, differences across newspaper type (broadsheet or tabloid) have not been considered in detail here but may shed light on important aspects of media coverage of CoSA.

Conclusion

This article presents the findings of the first study into how CoSA have been represented in the media. Newspaper portrayals have been mixed, ranging from salacious and caustic resistance to evangelical support. Understanding the ways in which CoSA have been framed in the media is critical for the international CoSA community, as the media provide perhaps the main avenue for informing the public about CoSA. Strategic approaches to interacting with the media can be developed on the basis of the findings presented above. The media provide a unique opportunity for the CoSA community to inform the public about this practice, and about child sexual abuse more broadly. The research presented here provides a platform for the development of a strategic agenda in this regard, as well as for future research about the portrayal of CoSA, and other interventions for those who perpetrate sexual violence, in the media.