Navigating moral dimensions and lateral power – The experiences of men with sexual convictions and histories of sexual abuse serving sentences in a therapeutic community

Abstract

This article examines the moral world of therapeutic penal communities through the lens of the experiences of men with sexual convictions and histories of sexual victimisation. Drawing on the concepts of lateral power and stigma, we discuss how Therapeutic Communities (TCs), which are intensely social communities, operate within three moral dimensions: the institutional set of therapeutic norms, the con code, and the broader sociocultural set of norms that intensely stigmatise sexual offending. Specifically, the focus of this article is how these norms shape the manner in which lateral power is used, navigated and resisted within TCs. We explore the manner in which TC processes are used as tools to enact lateral power over men with sexual offences, and how the three moral dimensions create a messy moral environment within the TC. Importantly, we explore how the residents navigate the morally complex dual identity of offender and victim. We also note that the moral messiness, and the conflicts between the moral dimensions, also raise challenges for how vertical institutional power is exercised. Finally, we point out that the neoliberal individualism inherent in today's sociocultural moral dimension further shapes how men with sexual offences experience the moral community of the TC.

Introduction

Therapeutic Communities (TCs) within penal settings usually comprise of small, regular psychotherapy groups of offenders (who are called residents), and a facilitator. TCs can either be mixed, holding people convicted of different types of offences, or those with only one type of conviction, for instance for sexual offences. As stated within the goals of TCs, residents are encouraged to discuss their offending histories and locate them within their life histories, within a positive social environment (Stevens, 2013). TCs are thus explicitly moral institutions. A sense of broader community – beyond the small therapy groups – is a key aspect of TCs (Akerman, 2010), and, thus, the therapeutic process is intended to enable residents to construct positive self-identities and experience a more positive life within the penal institution (Akerman, 2010; Ware et al., 2010). Critical to this article are two features of TCs (Akerman, 2019; Pearce and Haigh, 2017; Genders and Player, 1995). Firstly, TCs intend to foreground transparency and discuss both the offending itself and the context in which it took place in order to help residents understand why they offended, which may include histories of abuse. Secondly, the importance of community and positive interaction with other residents and with staff is emphasized – relationships are intended to be non-hierarchical and tolerant. We do not wish to imply that all TCs always operate in the way described above, but it is important to set out their stated goals before analysing how power dynamics in TCs operate in practice.

TCs thus have a set of institutional moral norms, which include an ethos of community, non-hierarchical relationships, etc. They intend to function in a way that helps to foster non-hierarchical interactions between residents but also staff and residents, and, to some degree, have been shown to succeed in these endeavours (Genders and Player, 1995). In doing so, TCs are intended to be a form of reinventive institution (Scott, 2010, 2011): an institution designed to empower and enable people to reinvent and reinterpret their identities. Within such institutions, ‘agency resides in choosing to adhere to the institutional regime rather than questioning its legitimacy’ (Scott, 2011: 4), but Waldram (2012) found that this is far from this binary: residents in the Canadian TC he studied actively negotiated and at times resisted the institutional norms.

The above discussion raises interesting questions about how lateral, informal power operates within these unusual penal institutions. Rhodes (2010: 460) has argued that TCs involve ‘the social world unfolding, messily, on the communities’. Social worlds are moral worlds (Waldram, 2012), and in analysing the nuances of power within TCs, we examine how the moral norms of a TC coexisted with the moral world of outside society and the moral con code of prisons; the ways in which these norms clashed in some situations, buttressed each other in others, and were resisted and manipulated. We see power as not a fixed thing possessed by some residents (e.g., those with violent offences) and not by others (e.g., those with sexual convictions), but something that is exercised, reacted to and negotiated within the TC context and shaped by these numerous moral dimensions.

Literature review

There is a rich and growing body of work on vertical or formal penal power. Penologists such as Foucault (1977) and Bentham (see Engelman, 2011) highlighted the panoptic nature of penal power and showed how it can work upon the psyche of the imprisoned person. Crewe (2011a, 2011b) argued that the nature of modern penal power is ‘tight’, with the vast array of prison rules, individualised conditions and bureaucracy creating a penal power that is psychologically entrapping rather than physically oppressive. On the other hand, Crewe and Ievins (2020) found that indifference and non-use of vertical penal power – described as looseness or ‘nonopticon’ – can also be experienced as painful when the person in prison feels they would benefit from the use of penal power.

There has been less research on lateral penal power, though there is a growing recognition that prisons are social environments with lateral power dynamics. Much of this work has focussed on how lateral power is used to resist formal power (e.g., Goffman, 1961; Clemmer, 1950). Recent research on the experiences of lateral penal power that operates within a prison for people with sexual convictions has highlighted the laterally tight nature of informal penal power. Ievins (2020) described how men with sexual convictions in a non-TC, single-offence prison engaged in lateral regulation of each other's behaviour, calling out and reporting on what they perceived to be inappropriate behaviour. This, she argues, created a sense that one was constantly watched by their peers in a ‘tight’, panoptic manner, driven by the salience of sex and morality to issues of sexual offending. Although Ievins focussed on regulation, she made it clear that she saw informal social relationships within prisons as rooted in power – the power exercised by people in prison upon their peers. Considering the dense nature of social relationships within TCs, we develop her work in this article by exploring in detail how lateral power operates within them and what shapes how it is utilised by residents.

This article focuses on how lateral power regulation is shaped by the moral nature of TCs (in its many complex dimensions). Waldram (2012), in his work on a Canadian therapeutic prison, highlighted the different moral dimensions at play in these complex, socially and morally messy institutions. He showed that there was a paradigmatic therapeutic model which structured and funnelled residents’ narratives about their offending behaviour and their treatment progress. These sets of paradigmatic norms consisted of the institutional expectations, the therapeutic goals, and so forth. These paradigmatic norms meant that, at times, residents had to speak the language of the treatment programme and structure their life narratives according to what was expected of them as opposed to their own subjective narrative as they understood it. These paradigmatic and narrative models, according to Waldram, influenced each other in a context rife with power relations. Importantly, the power relations within the TC (e.g., the power to challenge, or to silence peers) meant that residents dynamically negotiated power in an intensely moral community. Although Waldram was focussing on residents’ narratives rather than more broadly on social relationships and power dynamics, he usefully demonstrated that in a TC, power was not fixed, but rather flowed between people and situations, shaped by different dimensions of norms (institutional and con code, for example). He found that in order to survive, residents needed to develop a moral code relevant to their social world: the TC. As a result, the moral code of the TC (in terms of the stated institutional goals) could at times be at odds with the residents’ own moral codes because they needed to adapt to and negotiate a complexly morally loaded institution. At times, for example, the con code, which calls for secrecy around one's sexual offending and punishing of those with more disputable offenses, clashed with the transparency and dense sense of community of the TC. Here, we develop Waldram’s (2012) work to examine how these moral dimensions – institutional, con code and broader sociocultural norms that stigmatise sexual offending – coexisted in a tenuous balance and shaped power dynamics in a mixed TC for those with sexual convictions and histories of abuse.

It is crucial, at this point, to recognise the relationship between stigma and power. Stigmatisation is inherently morality-driven – to stigmatise someone is to indicate they are someone whose identity has been morally stained. However, stigmatisation does not happen in a vacuum; ‘it takes power to stigmatise’ (Link and Phelan, 2001: 375). In other words, ‘stigma functions as a form of power’ and operates within a specific setting of culture and power (Tyler and Slater, 2018). In the context of this paper, this culture includes the con code, which was discussed above, and broader sociocultural morality that deemes sexual offending to be especially heinous. It is important to unpack each of these in more detail.

People with sexual offences are often the most stigmatised people in prison, and the lowest within the offence hierarchy that is integral to the con code (Schwaebe, 2005; Crewe, 2009; McNaughton and Webster, 2018). They are also often stigmatised by staff in non-TC prisons (Ricciardelli and Moir, 2013). Lacombe (2008), for example, found that even within a Canadian penal programme for people convicted of sex offences, everything was ‘consumed with sex’: residents’ identities were defined entirely by the risk of reoffending. Other features of their identities became invisible – their offence became their ‘master status’ (Hughes, 1949). Genders and Player (1995), in their study of a mixed English TC, found that residents with sex offences were not completely accepted as full members of the mixed TC but did not delve into the detail of the social relationships and the lateral power dynamics at play. Waldram (2012) found that hierarchies of sexual offending were integral to the construction and sustenance of the moral community outside of the paradigmatic moral dimension, since a TC was a moral community well beyond the narrowly defined therapeutic sessions and their formal institutional norms. Here, we explore in greater detail how this played out within the therapeutic and everyday social interactions in an English TC.

Finally, there is the sociocultural dimension – within society too, sexual offending is deemed to be especially horrific, which likely feeds into the con code stigmatisation of sexual offending. Society, as well as the penal system, stigmatises those with sexual convictions as mythical monsters (Petrunik, 2002; Waldram, 2012) who will inevitably reoffend – even though this is not supported by empirical evidence (Hanson and Morton-Bourgon, 2004). Moreover, men who had been sexually victimised also experience stigma within society and are often shamed for not protecting themselves (Johnson et al., 2007; Hlavka, 2017). People who offend have long been recognised as being morally non-ideal victims – in the words of Christie (1986: 25): ‘victims that merge with offenders make for bad victims’. Moreover, there is a broader neoliberal sociocultural tendency to see crime as an individual ‘choice’ rather than as a social problem that occurs within a context of trauma, deprivation and other forms of inequality (Beckett and Herbert, 2010). Stigma, as Tyler and Slater argue (2018) operates as a political apparatus. Stigmatising people who offend serves to other and demonise them as especially morally tainted individuals. It is important for us that this broader sociocultural dimension is considered in how power and morality play out in a TC, since this neoliberal individualism might shape how residents with sex offences and histories of abuse negotiate and navigate this moral duality.

TCs are not insular utopias (Bennett, 2013) and operate within a society and a penal system where certain groups of people are disempowered and excluded. At the same time, those working within TCs explicitly attempt to challenge some of these hierarchical relationships (Bennett, 2013). Yet the research discussed above has shown that residents with sex offences, by virtue of having a highly stigmatised offence, often experience intense lateral scrutiny from their peers – a form of lateral power. In this article, we examine how these forms of lateral power operate, and how they at times manifest in a clashing or buttressing of institutional, con code and sociocultural moral dimensions.

A fuller analysis of these dynamics will help not only to further our understanding of penal power but also inform the socially dense and morally messy work of TCs. Although here we focus on lateral power, we show that the clashing of moral dimensions can also raise challenges for how vertical power is utilised by staff. In examining these relationships between morality, power and stigma, we not only develop existing work (e.g., Waldram, 2012) but also demonstrate the manner in which TC processes could be used and abused to exert power, and how this is a morally loaded practice. We also expand the analysis to show how the moral world of the TC has multiple dimensions: paradigmatic, con cod and broader, macro-level sociocultural.

Studying penal power dynamics through the experiences of men with sex offences and histories of abuse serving sentences in a mixed TC provides a unique opportunity to explore lateral power and the complex, morally loaded ways in which it operates within a TC. They are a niche but by no means insignificant group, as research has found that a significant minority of men with sexual offences also had histories of abuse (Jespersen et al., 2009; Seto and Lalumiere, 2010; Garbutt, 2018). It is worth noting that although this article focuses on a very specific group of residents, the themes discussed here provide an insight into TC's moral and power-laden world more broadly and can also be applied to other groups within penal settings, such as men convicted of killing children (see Jacobs and Shuker, 2019). Although the fieldwork was conducted in the UK, penal TCs of the type discussed here exist across the world, making this article of interest to academics and practitioners in other countries.

Methodology

This article draws on seven interviews with residents and seven interviews with staff living or working in a TC in England. This was a mixed TC, where men were not segregated from each other by offence type. Only one wing housed men with sexual offences exclusively, but they were still free to attend gym, education, work, social events and religious services, and otherwise interact with other residents. Other wings also held men with sexual offences, among those with other types of offences (e.g., violent offences, drug offences). Although this was a small sample, the data were rich and detailed, and the fact that staff (who work with large numbers of men with sexual offences and histories of abuse) and resident interviews yielded very similar results suggests a good degree of generalisability. This should not be taken to imply that further research into different groups of men and women in TCs is not needed. Exploring the overarching themes of power and morality within the TCs is likely to yield further interesting findings in these contexts.

The research aimed to explore the lived experiences of the residents, and the subjective views staff members had of the experiences of this group of residents. This required a qualitative approach, since it allows for in-depth exploration of individual subjective experiences. Staff interviews were included because not only do they provide additional perspectives on the social lives of these residents but could also be used to triangulate data gathered from the residents themselves. Both uniformed and non-uniformed staff, such as therapy staff, were interviewed. All the men were convicted of sexual offences, and all had self-disclosed previous experiences of sexual violence in the TC setting.

The first author conducted the data collection. The second author, who works in the TC as a non-uniformed member of staff, assisted on the project and was named on the participant information sheet as the project contact within the prison. Her involvement allows for intriguing ethical issues to be discussed. Firstly, in order to ensure that what participants said during the interview remained confidential, it was not appropriate for the second author to conduct the interviews. There was also the risk that her direct involvement in the research process would inhibit participant candidness and openness if they were to be critical of the TC. Moreover, we wanted to ensure that participants did not feel that taking part or not taking part would affect their living or employment conditions in prison in any way.

In order to address these concerns, the first author conducted and transcribed all the interviews. The second author did not have access to the raw data or any personal data such as signed consent forms. The information sheet named the second author as a contact for all participants but stressed the research was not conducted for or funded by the Prison Service. It also stressed the voluntary nature of the research and that there would be no consequences for anyone whether they decided to take part or not. The legal limits of confidentiality were set out – for example, it was made clear that the researcher would need to report any risk of harm to the participant or anyone else. The second author only had access to fully and thoroughly anonymised transcripts, stripped of direct identifiers but also indirect ones such as the wing they were on, the number of years they lived/worked in the TC, their exact roles on the wing, and so on. The transcripts were organised numerically, with the order scrambled so that the interviews were not labelled according to the order in which they were conducted.

The second author's help, in hindsight, was useful as it ensured that the call for research volunteers went out widely and because there was a nominated and trusted person within the prison who was very familiar with the project. It perhaps also helped that she was a non-uniformed staff member, but it is not possible to know whether her involvement specifically hindered anyone from taking part. We were reassured that two men who initially expressed interest decided to not take part – this was interpreted by us as them feeling no pressure to participate.

The first author attended wing meetings to introduce herself and explain the purpose of the research and answer any questions. Prison staff then liaised with her via the second author to inform her when participants offered to take part in the research and when they were available to be interviewed. Although this meant residents’ participation was known to some prison staff, staff involvement was unavoidable because of the impossibility of direct contact with the participants. There was little risk of ‘outing’ resident participants as men with sexual offences and histories of abuse as these facts would have already been known to the staff and communities, as the TC setting meant the men had already discussed their offending and context of offending within their groups. Staff interviewees were recruited via an email shared within the TC.

The interviews themselves were conducted on the participants’ wings (offices or meeting rooms), with only the first author and the participant in the room. They lasted between 30 and 60 minutes. All participants were asked to sign consent forms informing them of the purpose of the research and their right to withdraw from the research after the interview. Some interviews were audio-recorded with participants’ consent, and detailed handwritten notes were taken during others if the participant was uncomfortable with their voice being recorded. After the interview, the researcher checked up on the participants’ well-being, and the information sheet sign-posted them to support networks.

Ethical approval for the research was given by the first author's institutional ethics committee (University of Birmingham), and approval was also given by the TC's own research committee, which included academics, staff and residents. All wings where interviews took place also had to approve the research. There were thus numerous layers of scrutiny, and no specific concerns were raised. The wings were very much used in research processes as the TC habitually hosts numerous research projects at any one time.

Data were analysed thematically, using NViVo. Some themes, such as offence hierarchies, stigma and lateral power inequalities, were drawn from the existing literature. Other themes emerged organically from the interview data, such as that of TC processes being used to stigmatise residents and the negotiation and rebalancing of penal power. Thus, although some overarching themes were identified from the existing literature, the under-researched nature of this group – this is the first study focussing on men with sexual offences and histories of abuse in the TC – lent itself to a more organic and explorative analytical approach.

Discussion

Lateral regulation via TC processes

One of the most notable themes to emerge out of both staff and resident interviews was the manner in which power was laterally exercised over men with sexual offences by residents with other types of offences. The TC was described by all as a relatively safe place where residents with sexual convictions were not habitually physically assaulted. However, they were still low within the prison hierarchy, with interviewees reporting hearing derogatory terms such as ‘nonce’, ‘rapist’ and ‘child molester’. One staff interviewee recalled that a resident had ‘rapist’ scrawled on his cell door. This was a clear example of lateral power being weaponised to stigmatise; residents with other types of offences were higher up on the offence hierarchy and therefore used stigmatising language to belittle residents with sexual offences.

Ievins (2020) described vividly how the behaviours of men with sexual offences were tightly scrutinised by their peers. Violence was not condoned and if a resident was violent to a peer or staff member, they risked losing their place in the TC, and thus their actions were constrained by the penal power of the institution to expunge them. TCs are socially ‘dense’ and involve regular and structured social interactions such as small therapy groups, and wing meetings. As a result, lateral regulation might take very specific forms – those of the TC processes. As a result, institutional TC processes themselves were being hijacked to enact lateral regulation and exercise power over men with sexual convictions in a formal therapy setting. This issue was something Ievins touched upon, but since her study was not conducted in a TC, the lateral scrutiny was more informal in process (e.g., reporting inappropriate behaviour to an officer). In this study, it became clear that formal TC processes became tools for enacting power over men with sex offences – for example, when residents with sexual convictions were not given community backing for a job, when other residents targeted them with especially difficult and/or numerous questions during therapy, or when other residents did not engage with their concerns during a wing meeting. Resident 5 recalled challenging the presence of ‘hooch’ – illegal alcohol – on the wing and other residents turning on him, and he felt this was in part because of his offence. ‘I think what you’re in for has a lot to do with how you’re treated’, he concluded. These sorts of subtle, non-physically violent targeting practices were briefly mentioned by Jacobs and Shuker (2019) in their study on men convicted of killing children but were a key theme to come out of the interviews conducted as part of this research. In a mixed TC, therefore, lateral power was uniquely funnelled via TC processes, which is a key difference to lateral power described by Ievins (2020) in a single-offence penal setting.

Additionally, it is also important to understand how the TC morality, as opposed to purely the TC processes, can be exaggerated by the con code and used to exercise lateral power over residents in a particular manner. The TC placed an enormous emphasis on accountability and challenging, and residents could use these to target residents with sex offences. This could be done by asking difficult questions and subjecting residents with sexual offences to greater scrutiny over their actions, and/or during the whole-wing meetings, where decisions about residents’ jobs and other community involvement were taken. For instance, having a role of responsibility – which had to be approved by all the residents on the wing – was a key and expected aspect of TC norms, and as Staff 4 noted, this meant that ‘if you want to make someone feel like shit, you will not back them for a job’.

Sometimes, therefore, men with sexual offences felt they were not backed specifically because of their offence rather than because of a lack of suitability for a job; not backing someone for a job was thus a morally loaded way of enacting power over someone. This morally loaded manner in which men with sexual offences were targeted via questioning was echoed by Staff 3:

…bandwagoning, you’d see. Sex offenders’ minutes [discussions about their progress] would come up and lots of people on the wing asking questions, and someone else [with a non-sexual offence] – there's nothing. You ask the question, ‘what's going on?’. I think there's that… It [the offence hierarchy] does play out.

Ievins (2020) argued that morality was especially salient in non-TC prisons holding men of sex offences; this research shows that morality is also especially salient in TCs but existed across several dimensions. TCs are reinventive institutions where residents embark on moral trajectories of reinventing themselves (Scott, 2010) and are also moral in that they are imbued with goals such as acceptance, community, etc. However, residents also acted in accordance with the con code and sociocultural moralities, which stigmatise sexual offending especially harshly. All of this meant that small actions such as voting for someone to undertake a job or not were experienced as morally layered and laden with lateral power. In this way, the moral norms of the TC and the con code and outside morality (which deems sexual offending as being especially repugnant and deserving of punishment) simultaneously clashed and buttressed each other. Residents used the TC processes, thus adhering superficially to the TC ethos of challenging and community, but on a deeper level, those with sexual offending were at times subject to disproportionately intense lateral scrutiny and belittlement.

Relatedly, the residents and staff interviewed as part of this study highlighted the pain of the ‘nonopticon’ described by Crewe and Ievins (2020) – which was entwined with the moral norms of the institution. Where TC processes were abused in order to subtly exercise power over residents with sex convictions, the residents concerned wanted staff to step in and challenge this behaviour because they felt that is what institutional norms demanded. They were acutely aware that the TC was intended to be a place of acceptance and non-hierarchical relationships, as described by Resident 7, who said he had expected the place to be ‘this perfect paradise, hippy commune. Loving, and everyone was gonna be helping each other’. Although formal penal power in the TC was much ‘looser’ than mainstream prisons, since it allowed residents comparatively more freedom to engage in activities such as jobs, organising social events as well as looking after fish and plants, in this instance the use of formal penal power was desired. When staff did not intervene during instances of belittling or targeting, some residents felt let down, as evidenced in the disappointed tone of the quote of Resident 7.

The discussion above highlights how lateral power dynamics within this TC did not simply involve residents internalising penal rules and operating as agents of the penal state to regulate the behaviours of other residents and/or being incentivised by their own morality to collectively regulate other residents’ behaviours (see Ievins, 2020). This is not to deny these dynamics did play out; some residents certainly discussed self-regulating their own and others’ behaviours, including calling out inappropriate behaviours. There was, however, an interesting separate but connected power dynamic; hijacking formal therapeutic processes and/or ethos to engage in behaviours that were contrary to institutional goals but very much in line with the con code. This was illustrated in the example of ‘bandwagoning’. Asking difficult questions was a process that all residents were encouraged to engage in, but the TC ethos did not intend it to be weaponised against certain residents on moral grounds, such as those with sexual offences. Hence, in such situations the institutional moral code and the con code were at odds.

Notable here is that although Scott (2011: 50) sees people within reinventive institutions such as TCs internalising the values and reinforcing them via ‘mutual surveillance in an inmate culture’, this happened imperfectly within the TC. Residents with non-sexual offences had to be seen as adhering to the TC values of non-violence and the importance of communication. Just as in Waldram’s (2012) study, they had to survive within the institutional moral code. However, the con code also manifested when they targeted residents with sexual offences. The con code and the institutional morality coexisted in this messy, to an extent contradictory manner.

The effect of this lateral regulation on residents with sexual convictions was that some reported stepping back, emotionally, from the therapeutic process. Resident 7, for example, said that while he ‘did not mind’ speaking of his sexual offence, he did not speak about the ‘detail’ of his offending within the group, out of feelings of shame and embarrassment. This highlighted how having a sexual conviction placed residents in a precarious state within the TC; a state that was further highlighted when they were belittled within the TC. Other participants also reported feeling a sense of distance or otherness from residents with non-sexual offences, a suspicion that was also deeply rooted in the stigma they experienced in mainstream prisons and within society. One discussed at length how people would show him kindness, like ask him how he is doing after a difficult therapy session, but still expressed suspicion about whether they meant it or not. To him, this was ‘the done thing’ in the TC, and it is difficult to know whether he was interpreting them through the lens of shame and stigma, but also the con code, which emphasized not trusting fellow prisoners. It may well be that his peers truly cared for his well-being. What is important is that the stigma and the con code limited the extent to which residents with sexual offences could meaningfully engage in the therapeutic process. They were part of the institutional community but did not truly feel part of the social community of the TC in a full moral sense. This was echoed by staff, who noted that residents with sex offences tended to exist on the periphery of the broader TC community.

As a result, although TCs aim to be reinventive institutions (Scott, 2010), disempowerment meant that men with sexual convictions found it more difficult to reinvent themselves. Scott (2010) described reinventive institutions as greedy places that allow residents to discard their previous identity; yet residents with sex offence histories found it difficult to discard the ‘sex offender’ label due to its intense ‘stickiness’. Scott (2010) also argued that even though reinventive institutions have a great degree of voluntarism, residents’ ability to reinvent themselves is constrained by lateral surveillance – a subtler form of control. This was highlighted for residents with sexual offences, since the comparatively low place assigned to them by the con code and the sociocultural stigma associated with the offending meant they were subject to more lateral surveillance than other residents.

Negotiating power and rebalancing power dynamics within TC processes

Although residents with sexual offences were subject to intense and often painful lateral scrutiny, it became evident that the TC environment, to some degree, relied on these morally loaded power dynamics manifesting in order for therapeutic work to be done. Hijacking of therapeutic processes allowed for the con code and social stigmas to be brought to the surface, and therefore to be challenged. Thus, the TC did offer an opportunity to dismantle some of the stigma and unpack the power dynamics at play in a given situation (as argued by Bennett, 2013, in the context of racialised power in TCs). Therapy staff, especially, discussed using such situations as a chance to address and challenge the con code and the offence hierarchies integral to it, and thus to reinforce the need to treat and respect everyone equally. Contrasting this with mainstream prisons, where there were few formal opportunities to do so, one staff member said – ‘at least it [the hierarchy] was being spoken about’. Both staff and resident participants noted that this challenging did not happen consistently within the TC and required staff experience and training, but as one staff member lamented, ‘we try, we try, we try’. This highlights the need for staff working within mixed TCs to be carefully trained in identifying and challenging instances where the therapeutic process is being misused in the manner discussed above, and to ensure that any such misuse can be redirected towards challenging such stigma.

Despite the offence hierarchies not being completely flattened within the TC, attempts to belittle residents with sex offences allowed for therapeutic discussion and moral reflection to take place (including explicit discussions of the conflicts between the con code and the TC norms). This resulted in moral conversations to take place in a manner not often discussed by Waldram (2012). He focussed primarily on the detail of cognitive behavioral therapy processes and narratives in an individual sense, but within the TC studied here, staff often raised broader questions about sociocultural and con code hierarchies and stigmas. A therapy staff member discussed challenging stigmatising terms like ‘nonce’. She recalled asking the person using such terms – ‘well, what does this mean?’, thereby using the incident as a gateway into reflecting on labels and hierarchies prevalent within mainstream prisons and society. To give due credit to the staff and the residents, at times the process worked well, as discussed by Staff 2 –

Because they’re straight out of mainstream, they come in with these anti-sex offender, hierarchy views, they’re dropped at the door and they’ve got to get on with people. And to be fair to them, it's amazing, how well they do work together. In the therapy spaces. Because they do. They speak about how difficult it is [to share a space with someone who committed a sexual offence], but they try and look beyond it.

Interestingly, the rebalancing work described above also created power-related challenges for staff. As representatives of the penal institution, staff members were tasked with ensuring that the TC processes ran smoothly. However, the hijacking of these processes in order to exert power over men with sexual offences meant that staff who facilitated therapy groups were forced to tackle some difficult questions about how and when to utilise their own (formal, institutional) power. They were not simply agents of the penal state unquestioningly exercising vertical power in order to enforce institutional norms – this was not the ethos of the TC. The staff-resident hierarchy was thus, to some degree, flattened, albeit imperfectly and not consistently. One therapy staff member lamented that challenging belittling behaviours could pitch the other residents against the person with the sex offence and damage the dynamic of the group. This helps further understand the moral nature of the TC small group community, one that goes beyond the narrow TC norms. The men needed to coexist outside of the session, and what staff did or said in the session could affect the community as a whole. Staff could be seen as ‘siding with’ and ‘protecting’ these residents unfairly, which reflects the offence hierarchies inherent in the con code (Schwaebe, 2005; Crewe, 2009). Some resident interviewees corroborated her discussion in their interviews, noting that from their points of view, staff members did not consistently challenge bullying behaviours in groups. This also highlights how the non-use of vertical penal power can be experienced as painful (Crewe and Ievins, 2020). However, it also shows that staff in TCs are moral agents wary of putting particular residents at risk of retaliation or further exclusion.

It is also important to note that none of the residents spoke about staff engaging in disempowering behaviour. None of the staff members interviewed expressed overt disgust or hated, unlike in Spencer and Ricciardelli’s (2016) study, in which some staff called men with sex offences ‘sick’, for example. This was likely because to express such strong feelings would be very much contrary to the TC ethos. Some staff, however, did note it was especially difficult – emotionally – to work with residents with sexual convictions. One staff member noted that ‘most people never have sex with a child’ (Staff 5). A sex offence against a child was, to him, something so alien and difficult to understand that he said he struggled with those who committed such offences, which reflects the prevalent social attitudes to sexual offending. He then noted that, in his mind, it was not out of the realm of possibility to hit a woman if one was very angry – this, according to him, would be bad, but not impossible to comprehend. Another one discussed struggling emotionally to facilitate therapy groups with men convicted of child sex offences because he himself had young children. Although the TC ethos of inclusion and acceptance appeared to moderate the stronger feelings of hatred and revulsion, TC staff also existed within the moral dimension of outside society – the two dimensions thus clashed. The TC ethos of acceptance, however, appeared to moderate the more intense staff reactions to men with particular sexual offences. Since sexual offending carries strong stigma within both mainstream and prison societies, it would have been unrealistic to expect staff (and residents) to shed such views completely, which echoes Bennett’s (2013) work on racial power inequalities within TCs.

The above discussion shows that within TCs, power can be fluid and negotiated. It was not simply a case of residents with non-sexual offending disempowering residents with sex offences. Rather, the power dynamics were negotiated using the ethos and the processes of the TC, with facilitators challenging belittling behaviours, and, when the process worked well, some offender hierarchies indeed being flattened within some groups (as discussed above). This was especially the case within the small therapy groups, where, both according to the residents and the staff, there was a greater degree of acceptance, perhaps because they provided more opportunities to do the difficult work of rebalancing power inequalities compared to the whole-wing meetings. This was made clear by Staff 2, quoted on page 10, and by the residents who said that their small groups were supportive to the degree that they would sometimes ‘cry together’ when the resident was talking about his sexual abuse histories.

Finally, some residents with sex offences themselves used the TC processes to reclaim power and reinvent themselves, though this appeared to be difficult and happened infrequently. TC ethos puts much emphasis on resident involvement and taking on roles of responsibility. This allowed some residents with the sex of offences to break through the strong master status of ‘sex offender’. Violence within the TC was unaccepted, so one way of exerting power was verbal belittling, as discussed earlier. Some residents with sexual offences used this ‘gift of the gab’ to break through the con code and place themselves at the centre of the moral TC community. As one resident discussed in his interview, if someone was exceptionally eloquent and confident, the TC allowed them to gain more power within the community, more so than physical power or violence could. In fact, one resident with a sex offence had a role of significant responsibility on his wing and was very much liked and respected by the other residents, according to staff. This again highlights how lateral power within TCs is indeed complex and fluid, and subject to some negotiation – even if that negotiation might be especially difficult for residents with sex offences.

The disempowering effect of a dual identity

Despite sexual victimisation, especially male sexual victimisation, being stigmatised within society, none of the residents discussed being overtly stigmatised or shamed (by peers or staff) for being victims of sexual abuse. However, many still reported feeling ashamed of and uncomfortable about being victims of sexual abuse, which did reflect the social taboos surrounding male sexual victimisation (Hlavka, 2017). As one resident noted, when explaining his reluctance to discuss his history of abuse in small group settings, ‘no-one wants to hear about kids being abused’. There was a degree of self-stigmatisation here, imported from a society that shames men who have been sexually abused.

The dual identity further highlighted how the institutional moral norms clashed with the con code and sociocultural codes that view sexual offending as especially heinous. This dual identity resulted in the offenders having to navigate a nuanced, and often narrow moral ‘tightrope’, in the words of one resident. They were faced with the need to engage with the context of their offending, as per the moral norms of the TC, which included their histories of abuse, and which were difficult and painful to talk about. At the same time, all the residents interviewed expressed an intense desire to not be seen as minimising their offending behaviour and were concerned about what other residents would think of them if they were seen to be minimising. Some reported feeling caught in-between, in a precarious, liminal moral space where they were both offender and victim, but not quite knowing how to negotiate this identity. This metaphor of walking a tightrope between offender and victim, used by one resident, highlights the complex ways in which the moral dimensions of TCs intersect and clash.

The TC ethos, thus, pulled these residents in two directions. On the one hand, they were expected to discuss their offending openly and take responsibility for it. On the other hand, they were expected to also engage with the context of their offending behaviour, which would make them vulnerable to the accusation that they were ‘making excuses’ for their offending. All of the residents and staff interviewed pointed out that most of the men in this group tried to frame their past abuse as providing the context, but not an excuse or cause, for their offending behaviours. For example, Resident 2 was explicit that his history of abuse did not make him commit his offence:

For me, it's not a case of using it as an excuse. For me, it's about learning how it shaped my mind to… How I then distorted what was going on for me when I was younger. To put me in the state where I was, when I offended. […] I understand that there are people who can be quite sceptical towards that.

His conclusion that people ‘can be quite sceptical’ about him not using his history of abuse as an excuse is significant. It indicates that the TC norms – of transparency, engaging with histories of offending – clashed with the con code that stigmatised sexual offending and emphasized hiding one's history of sexual offending. To some extent, it could be that this was also exacerbated by the stickiness of the sex offender label. Residents with sex offences (who were used to being stigmatised within mainstream prisons) had perhaps also internalised the ‘trust no one’ part of the con code and were thus sceptical of others’ motives. This was a problem for the interviewees when it came to reinventing their identities and re/interpreting their life histories – prevalent social stigmas and the con code constrained their engagement in the TC. Resident 6 also discussed the lateral power other men exercised to challenge him even when he, apparently, simply mentioned being a victim of a sexual offence.

I don't want to just paint myself as victim, because I know I’ve made victims. So, there's always this strange tightrope between understanding being a victim, but understanding the pain we’ve caused as well. […] Some of the lads here who haven't had that experience [of abuse] do sometimes challenge us. ‘Oooh, you’re always playing the victim’. Now, I am quite happy… Well, not happy, but prepared to talk about my offending as well.

The dual identity appeared to give residents with non-sexual offences further ammunition with which to exercise power over the residents this study focussed on. One question to ask is whether the residents with sexual convictions had internalised the stigma of sexual offending and were interpreting all social interactions through the lens of their internalised ‘master status’, and whether their struggle to navigate the dual identity was unique to them. It is well known that many people in prison are drawn from underprivileged/challenging backgrounds, characterised by histories of poor mental health, histories of physical and sexual abuse, drug and alcohol addiction, exclusion from education and homelessness (Social Exclusion Unit, 2002; Ministry of Justice, 2012). The participants all recognised that all residents in the TC often shared such histories during therapy, so it was by no means unusual for a TC resident to explore how adverse histories shaped their later behaviours. The interviews conducted for this research suggested that there was something unique about having a sexual conviction; it was so stigmatised that mentioning histories of abuse was especially sensitive and more likely to be interpreted as ‘making excuses’, thus activating lateral power driven by this stigma. It is therefore significant that it seemed especially difficult for men with sexual convictions to discuss their histories of abuse, as highlighted by Resident 5:

…they’d [other residents] say, ‘yeah but you’d gone through all this abuse, and you grow up and you’re committing the same sort of abuse on other people. How could you do that? You know what it's done to you, so you know what it's done to other people?’ […] I’m sat there and I’m thinking – ‘you talk about the violence that has been committed against you, and you grow up and you’ve done the same thing to other people. But, because it's not a sexual offence, it doesn't matter, it doesn't count.’

Even though the TC was intended to be a reinventive institution, the extent to which the residents interviewed for this study felt they were able to reinvent themselves fully was constrained by this conflict between the TC norms and the con-code and sociocultural stigma surrounding their type of offending. This, once again, shows how the paradigmatic moral dimension clashed with the moral community of the TC and broader sociocultural views of sex offending. As Bennett (2013) argued, TCs operate within a social context characterised by power inequalities and therefore cannot collapse them entirely. Residents with sex offence histories found it difficult to shed the identity of ‘sex offender’, which in turn made their status as a victim more tenuous and made it difficult to talk about their victimisation in therapy. This was even the case within a setting explicitly designed to help convicted men to address the context of their offending. Their already tenuous (due to their later offending) victim status was apparently further damaged by the nature of their offence, which, as discussed earlier, is highly stigmatising.

On the other hand, there was once again the opportunity for the TC to rebalance the flow of power in the context of the dual identity. As discussed earlier, both sexual offending and sexual victimisation of men are highly stigmatised – and the residents this article focuses on were especially exposed to lateral scrutiny in numerous ways within the prison society. The TC, however, offered the first opportunity for the residents to fully engage with the context of their offending, and to actually discuss how their abuse affected them and their later behaviours. Although talking about abuse was complex, as evidenced above, they still had a formal forum within which to tackle this within a discussion facilitated by trained therapy staff and although imperfectly, inappropriate uses of TC processes were at times challenged. Most of the interviewees acknowledged that until the residents entered the TC process, very few had engaged with this aspect of their life experiences.

There is also the broader overarching social moral dimension to be considered. The modern neoliberal criminal justice system stigmatises people who offend as individuals who made the (moral) choice to commit that offence (Beckett and Herbert, 2010) and sees offenders as imperfect victims (Christie, 1986). In this way, the moral world of the TC (and specifically its value of engagement with the full context of offending) clashed with society's view of individual morality in the context of offending. This tension between individualistic and contextual understanding of offending and individual work and group/community therapy work is recognised in the context of TCs by Pearce and Haigh (2017). In this study, residents were at pains to describe their offending as a ‘choice’ and were worried about being seen, by other residents and perhaps by the researcher, as ‘playing the victim’ if they were to speak about their histories of abuse in what might be interpreted by others as too much detail. This was likely especially difficult because of the highly sticky nature of the sex offender label; other residents seemed to be able to discuss their adverse experiences without the same degree of concern or challenging. Some of this was to do with the stigmatising reactions of other residents, described above. However, it was also evident that residents had internalised the notion that they were entirely responsible for their offending and that it would be especially morally wrong to discuss their status as victims beyond some ill-defined level of acceptability. The highly stigmatised nature of their offence made it especially difficult to locate their offending within their broader life histories in ways that seemed not to be the case for residents with other types of offences. It is suggested that further research on TCs and how the neoliberal individualistic perceptions of offending conflict with the TC norms is needed.

Conclusion

TCs operate within several moral planes. Firstly, there is the TC moral dimension that emphasises acceptance, community and transparency. Secondly, there is the ‘con code’ – within this dimension, sexual offending is deemed to be especially horrific and those with sexual offending are seen to be as more morally tainted than men with other types of convictions. Moreover, a key feature of the con code is low levels of trust when it comes to other men in prisons. This con code is, in turn, to a degree buttressed by social morality which also sees people who commit sexual offences as unambiguously morally repugnant and all offenders as individually and thus morally responsible for offending regardless of the context within which this offending takes place (including histories of abuse). These moralities often play out in TC communities in ways that result in residents with sexual offences being subjected to intense lateral scrutiny.

These findings indicate that TCs ought to ensure staff are adequately trained to navigate these moral dimensions, which play out not only during therapeutic sessions but during everyday interactions within the TC. They also suggest that TC work might be naturally limited by socio-cultural stigmas and, perhaps, ought to recognise and debate these constraints both at a policy and at a practitioner level more explicitly. More awareness of these complexities could ensure that TCs are able to tackle some of the limitations discussed in this research, especially in relation to disempowering behaviours targeting men with sexual offences. Moreover, this research indicates we need to have a broader discussion within the criminal justice system, one about individualism and the role adverse histories play in people's offending. If TCs are to work well, we need to recognise and be frank about the fact that most people in these TCs have complex and adverse backgrounds – a culture of neoliberal individualism appears to inhibit the degree to which TCs are able to help residents engage with their offending histories fully.

Although earlier research (Crewe, 2011a, 2011b; Crewe and Ievins, 2020) has indicated that penal power is in itself complex, this research shows that TC processes and ethos add an additional layer of complexity to both lateral and vertical penal powers. Power can be funnelled via these processes to target certain residents (in this study – men with sexual offences, though this might also be the case for men who killed children and other groups of people in prison). It is important to continue exploring these unusual penal institutions, since they offer a useful lens for understanding penal power, the manner in which it flows within TCs, and the ways in which it can be used, negotiated and reclaimed.