Trapped in the abject: prison officers’ use of avoidance, compliance and retaliation in response to ambiguous humour

ABSTRACT The place of humour in organisational interactions has been the subject of long-standing interest. Studies have considered the positive role of humour in increasing social contact and promoting group cohesion, while warning it can be a means for expressing hostility and excluding group members. However, more ambiguous uses of humour remain underexplored and under-theorised. Using a single case study of employee experiences at ‘Hillside’, a high-security prison in the UK, we address this gap. Adopting Julia Kristeva’s ‘theory of the abject’, we conceptualise ‘abject humour’ as a disruptive activity, which is composite, shady and sinister. We show that, despite Hillside’s adoption of Challenge It, Change It as a UK-wide safeguarding policy, the liminal spaces abject humour opens and occupies, are difficult to regulate. Those spaces trap both perpetrators and targets, and necessitate the use of avoidance, compliance, and retaliation strategies by the latter, as ways of coping.


Introduction
From anthropology to sociology and philosophy, the functional modalities of humour in social interactions have been theorised across a wide spectrum of disciplines (Robert, Dunne, and Iun 2016;Critchley 2013). Workplace researchers have considered how uses of humour can create positive organisational environments (Lang and Lee 2010;Martin 2000), for instance, by providing a pressure valve through which tensions and conflict can be diffused (Kahn 1989;Keisalo 2018;Avolio, Howell, and Sosik 1999). Studies have also shown how humour can facilitate social contact and group cohesion through shared norms (Kahn 1989) and positive emotions (Köhler and Ruch 1996). At the opposite end of the spectrum, humour can be aggressive, subversive and schismatic (Westwood 2013;Warren and Fineman 2013). Humour can uphold established organisational structures (Westwood 2013) and amplify existing group dynamics, including those of sexist 'hegemonic masculinity' and heteronormativity (Plester 2016). At the same time, humour can disrupt and violate organisational order, turning established routines and expectations upside-down (the literal meaning of a 'catastrophe') (Höpfl 2013). Yet, what about those situations where the intent behind a joke or act is unclear? The capacity for humour to be ambiguous (cf. Billing 2005a) can disguise deviant and counterproductive work behaviour, for instance, by presenting ridicule, aggression and hostility (Caponecchia, Branch, and Murray 2020;O'Boyle, Forsyth, and O'Boyle 2011) as harmless and playful fun (Yang, Kitchen, and Bacouel-Jentjens 2017). key arguments and contributions to scholarly fields. We bring the article to a close by offering practical suggestions for researchers and organisations.

How humour works
The study of humour in organisations is 'serious business' and, as such, has attracted serious attention (Thomas and Al-Maskati 1997;Lang and Lee 2010). To date, studies have mostly focused on the role of 'positive humour' in organisational life, highlighting the benefits of having working environments where humour is encouraged, and managerial styles which incorporate humour (see for instance, Wijewardena, Samaratunge, and Härtel 2016;Fleming and Sturdy 2011;Kenny and Euchler 2012). Humour can cement collegiality (Holmes 2006), even by means of banter and verbal putdowns (Murphy 2017), while improving the overall happiness and wellbeing of employees (Robert and Wilbanks 2012). Humour can aid everyday interactions between employees (Korczynski 2011;Lang and Lee 2010), for instance, by challenging power imbalances and subverting dominant gender norms (Kenny and Euchler 2012). As an example, risqué banter may target ethnic or heteronormative differences, in order to signify the inclusivity and acceptance of the individuals in possession of those characteristics (Plester and Sayers 2007). Humour can offer a means of coping with tension and stress, for instance, by enabling parties to safely vent their reciprocal hostility (Plester 2009), rather than engage in open conflict.
Yet, uses of humour can also have the very opposite effect on employee experience. Rather than challenging established meanings (Plester 2009) and disrupting oppressive organisational structures and dynamics, humour can harass, 'marginalise' and 'supress' employees whose identities do not fit within the organisation's heteronormative environment (Butler, Hoedemaekers, and Russell 2015). In such instances humour, in the shape of jokes, mocking or teasing, can take on increasingly sinister overtones and even cross the line of what is deemed acceptable. In those cases, to paraphrase Barbara Plester, 'fun is no longer funny' (Plester 2009). It is to such sinister forms of humour we now turn.

When teasing hurts
A number of studies have recognised the dark side of humour and its more sinister uses, for instance, in the shape of counterproductive work behaviours, which can insult colleagues or sabotage the organisation's operation (O'Boyle, Forsyth, and O'Boyle 2011). William Kahn's (1989) seminal typology highlights this deployment of humour for the purposes of aggression and hostility, with jokes made in lieu of overt harassment, and in order to avoid incurring formal punitive measures. Humour can also become a device for social control when used, for instance, as an 'othering technique' (McCann, Plummer, and Minichiello 2010), or to discipline certain employees and groups (Godfrey 2016;Abedinifard 2016). In those instances, humour can constitute an inappropriate and counterproductive form of behaviour and even become bullying (O'Boyle, Forsyth, and O'Boyle 2011). Broadly defined, bullying occurs when there are uncivil, abusive and aggressive behaviours from one or more individuals, to other individual/s, usually in a context of power imbalance . Surprisingly, humour is under-conceptualised in the mainstream bullying literature (Matthiesen and Einarsen 2010;Zapf and Einarsen 2003;Zapf and Gross 2001) with only a few notable exceptions considering the experiences of employees within power structures of 'hegemonic masculinity' (Plester 2015, 541). Here, studies of lesbian, gay, bisexual and transgender employees have observed that darker and more liminal uses of humour, like mockery, can normalise masculine work culture (Godfrey 2016) and bolster heteronormativity in organisations (Abedinifard 2016). Specifically, 'organizationally-sanctioned hyper-masculine humour' (Plester 2015, 508) can have a marginalising impact on lesbian, gay and other minority identities (Plester 2015). When used as a marker of heterosexuality (Morgan and Davis-Delano 2016), and when seeking to achieve the 'cultural ideal' hegemonic masculinity (Plester 2015, 541), homophobic humour can also be used coverup (male) insecurity by camouflaging hostile comments as 'just a joke' (McCann, Plummer, and Minichiello 2010, 519). This is problematic for organisations, because such negative uses of humour are not always easily identifiable or sanctioned, even when bullying and harassment policies or codes of conduct are present. As already recognised, humour can be a liminal activity (Djurkovic et al. 2020;Keisalo 2018), which is hard to categorise as either positive or negative. Accordingly, current definitions conceptualise humour as a 'double-edged sword' (Billing 2005a;Billing 2005b), which is both 'collective and corrective' (Butler 2015, 5;Collinson 1988), 'transgressive and oppressive' (Pullen and Rhodes 2012), 'collective and divisive' (Rogerson-Revell 2007). Humour's ability to take on positive and negative forms, makes it possible for one form to be confused for the other. As a result, it not only transcends organisational boundaries (Kahn 1989) but also opens liminal spaces within them. Those spaces can escape managerial control (Plester 2015, 539), while allowing bullying and harassment to be practiced under the ambiguous guise of lewd jokes, teasing and innuendo (Adikaram and Liyanage 2021;Plester 2015;. It is to the conceptualisation of ambiguous humour and its liminal spaces we turn next.

The abject and the liminal
To fully understand ambiguous uses of humour (Thomas and Al-Maskati 1997) and the in-between, liminal, 'alternative' or 'anti-rite' spaces (Kenny and Euchler 2012, 307) they open, we adopt Julia Kristeva's (1982) 'theory of the abject'. Kristeva's contribution is part of the largely underutilised, psychoanalytic turn in the organisational studies literature (Rizq 2013;Fotaki et al. 2012), predominantly developed by gender and feminist scholars (Butler 2004;Fotaki 2011). In Kristeva (1982, 2), the abject is a boundary object, '(…) something rejected from which one does not part, from which one does not protect oneself as from an object.' In her example, the corpse is an abject par excellence, as it is neither the presence of life nor its complete absence (4). There are obvious parallels between Kristeva's conceptualisation and scholarship on the subjects of hegemonic masculinity (Connell 2020) and humour (Plester 2015). Through humour, hegemonic masculinity 'rejects both femininity and homosexuality', which qualities are 'expelled', in order for dominant heteronormative practices and identities to be enforced (Plester 2015, 541). This process is abjection. Abjection is the expelling of Self, or rather, those characteristics of Self deemed taboo, in order to re-establish the Self (Kristeva 1982, 3). We propose that, against the hegemonic masculinity culture in organisations, non-heteronormative traits are the expelled taboo, in order for masculinity ideals to be pursued.
We have already acknowledged the negative impact this process, which we will now refer to as abjection, can have on organisations and on employees who do not possess heteronormative traits. Like humour (Höpfl 2013;Kenny and Euchler 2012), abjection is a disturbance of organisational morality (cf. Kristeva 1982, 3), as codified in organisational policy. Unlike humour, which the previous section described as both-and (positive and negative), the abject is neither-nor (Kristeva 1982, 1). This makes abjection 'shady', 'sinister' and 'scheming' (Kristeva 1982, 4), because it neither accepts/complies with rules and norms, nor outright rejects them. Like mocking, jest and other 'weapons of the weak' (Scott 2008), the abject subverts organisational order but unlike humour, it does not openly resist it (Kristeva 1982, 4). Congruent with humour (Lang and Lee 2010), the abject occupies a liminal and boundary space, where ordinary rules and norms are obscured and where interactions and exchanges can be interpreted in a variety of ways (Keisalo 2018;Plester 2009). Contrasting with humour (Kahn 1989;Thomas and Al-Maskati 1997), there is no true separation between a speaker and a listener (cf. Kristeva 1982, 3). Yet, the lack of subject-object separation does not imply a meaningful bond, for instance, one leading to camaraderie or mutual respect. Instead, abjection traps the speaker and the listener in the liminal space it creates. Thus, even without constituting bullying and harassment, encounters with the abject can create feelings of unease for all, with more negative outcomes such as disgust and anxiety specifically experienced by the listener-target (cf. Rizq 2013).
In this way, having observed the emergence of abjection in contexts of hegemonic masculinity, we turn our attention to the experiences of employees at Hillside which, similar to the military and the police, was a site of men's work (Hsu 2005;Karp 2010).

Prison humour
The place of humour in prisons continues to attract ongoing interest. Several scholars have explored the use of humour in prisons including Crawley (2004a;2004b), Crawley and Crawley (2008), Nielsen (2011) and Liebling, Arnold, and Straub (2015). However, researchers have tended to follow the 'both-and' framing familiar from the wider humour literature. Thus, while studies have shown that humour enforces masculinity and heteronormativity among prisoners (Karp 2010;Hsu 2005); research on employee-prisoner interactions has mainly followed the much-used path of 'positive humour'. Humour has been shown to build rapport and respect between prisoners and prison staff (Liebling, Arnold, and Straub 2015), as well as foster conflict avoidance and smooth daily interactions, for instance, by navigating power differentials (Nielsen 2011). Prison humour plays a role in improving staff morale (Crawley and Crawley 2008), managing emotions in order to cope with serious incidents (Crawley 2004b) and creating a collective employee identity (Nielsen 2011). Two major UK publications on prison staff, The Prison Officer (Liebling 2011) and Understanding Prison Staff (Bennett, Crewe, and Wahidin 2008) touch on workplace uses of humour, noting prison officers' 'sense' of humour and how humour can 'substitute' formal orders (Liebling 2011, 55 and 139), but neither points are unpacked. Furthermore, the issue of regulating banter is only recognised in relation to prisoners (Bennett, Crewe, and Wahidin 2008;Liebling 2011), while the place of humour in prison officer interaction is underexplored.
In recognition of such issues, and in response to calls to study diverse uses of humour in a variety of organisational contexts (Thomas and Al-Maskati 1997;Lang and Lee 2010), we approached Hillside prison. The study was part of a wider research project investigating the prevalence of discrimination, bullying and harassment of lesbian, gay and bisexual (LGB) employees in a wide range of UK workplaces, as well as ways in which those negative behaviours are practiced. Hillside, a male highsecurity prison, employing over 1,000 members of staff, the vast majority of whom (approx. 70%) were male, was selected for the current case study on account of being an institution, which had implemented the Challenge It, Change It programme. This, we hoped, would enable us to study not only the bullying and harassment experiences of employees in sites of hegemonic masculinity, but the effectiveness of organisational initiatives aimed at preventing and managing them. Lastly, we sought to compare the experience of lesbian and gay employees with that of heterosexual employees, who interacted in the heteronormative working environment of a prison.

Designing the hillside case study
The case study design we adopted for our Hillside study is a method enabling researchers to investigate a desired phenomenon (e.g. bullying and harassment experiences of workers) in a 'real-life' context (e.g. the heteronormative environment of a prison) (Yin 2003). As case study designs seek to achieve 'holistic' (Eisenhardt 1989) understanding of complex environments, we felt it would be particularly suitable for the organisational context of a prison, where UK legislation on bullying and harassment, and in situ interventions of the Challenge It, Change It programme merged to jointly shape employee experience.
Our engagement with Hillside aimed at understanding the everyday working experiences and challenges of prison officers. Chronologically, we began by evaluating wider UK attitudes towards the Challenge It, Change It programme through online searches of an anonymous prison service blog (prisonofficer.org.uk). We also constructed three scenarios on lesbian and gay employee experiences, generated from previous interviews and based on themes published elsewhere (Hoel, Lewis, and Einarsdóttir 2014), as a tried and tested method to aid conversation about potentially sensitive topics (Ward and Winstanley 2005). None of the scenarios pertained to dynamics in Hillside and contained a degree of intentional ambiguity to stimulate conversation. Our research 'on the ground' involved eight semi-structured interviews with lesbian and gay prison officers, and four focus groups with a total of 20 heterosexual members of staffboth prison officers and governance (managers) (see Table 1 for participant profile).
The focus groups were managed by two researchers. Each group was presented with the three scenarios, allowing approximately 20 min of discussions for each. We opened the discussion by asking participants to state their initial reactions to Amir's story (outlined in the 'Producing Liminal Space' section), followed by a series of more targeted questions. The focus groups were organised on the basis of roles and institutional hierarchy, as indicated in Table 1.
In turn, our semi-structured interviews (Denzin and Norman 2005;Fontana and Frey 2000;Mason 2002) with lesbian and gay employees sought to understand their workplace experiences, the broader organisational culture and environment, as well as relationship dynamics with colleagues. Interviews ranged from 60-90 min and we recruited participants firstly, via general calls to employees and poster displays, then, via a Hillside-based lesbian, gay bisexual and transgender (LGBT) employee network distribution list and snowballing. As abject humour is a 'neither-nor' act (cf Kristeva 1982, 1) which 'traps' the subject-as-speaker and the object-as-target in the liminal space of abjection, we felt it was important to capture the experiences of abjection not only of those LG employee we anticipated to be targeted by it, based on our wider study but also of the perpetrators of abject humour.
Our original intention was to hold focus groups with heterosexual prison employees, whereas the interviews were primarily focused on lesbian, gay and bisexual employees. Yet, perhaps due to issues with heteronormativity and homophobia observed in prison culture research (Sit and Ricciardelli 2013), as well as the 'private' nature of homosexuality in prisons (Javaid 2018), we struggled to recruit gay men and failed to recruit any bisexual employees. We further failed to recruit anyone from the black, Asian and minority ethnic communities. Problems like these are well documented amongst 'hard to reach' groups (Browne 2005;Einarsdóttir, Hoel, and Lewis 2015;Einarsdóttir, Hoel, and Lewis 2016;Heaphy, Weeks, and Donovan 1998;Hephy, Smart, and Einarsdóttir 2013) such as this, particularly in masculine, male-dominated work environments (Rumens and Broomfield 2012). In the end, we were only able to interview two gay men (Andrew 1 and Ulrick) who both held healthcare roles within the prison, but only one of them was open about his sexuality at work. Recruiting lesbians proved easier. In total, we interviewed four: three prison officers (Bethany, Ingrid and Tracey) and one healthcare worker (Morgan). All of the women were open about their sexuality at work, but in two cases, the women had been outed (their sexuality being disclosed) by colleagues.
One of the conditions for being granted access to Hillside employees was to cause minimal disruption to the day-to-day operations of the prison. This made it necessary to initially conduct all interviews and focus groups on location, which may have impacted employee trust (if researchers were seen as 'siding' with Hillside) and caused an unwillingness to discuss sensitive issues. There were also a range of practical limitations we had to navigate, including available room and security measures. As a result, we opted for an interview room in one of the training units on site, which allowed the use of audio recorders. However, we encountered issues during our research. As an example, the Hillside human relations (HR) team who organised the interviews and focus groups showed limited understanding of the sensitivities involved in researching sexualities, including the need for confidentiality and anonymity. For instance, the interview room was booked in the name of the LGBT employee network, as opposed to the university as requested, and the interviewer was forced to disclose the identity of one of the interviewees when the receptionist refused to let the person into the building. Considering these obstacles, it is perhaps not surprising that, in the course of the study and as we became known to employees, half of our lesbian and gay interviewees opted for being interviewed off siteoutside of working hours and away from the all-consuming prison environment where participants could draw too much attention to themselves. No requests to be withdrawn from the study following interviews were made. We adopted a thematic approach to our data analysis (Boyatzis 1998;Braun and Clarke 2006;Guest, MacQueen, and Namey 2012). In order to code the scripts, we looked for descriptions of unpleasant and uncomfortable episodes, experienced as a result of being targeted by ambiguous humour. Morgan's comments about being 'pissed off' with having her sex life openly discussed but having to hide it, is an example of this. In addition, we looked for instances of employee's coping strategies when exposed to abject humour and the availability of 'escape routes' out of the abject. Bethany's approach of seeking to comply and adopt a 'sick sense' of humour to better fit in the (heteronormative) working environment at Hillside, illustrates this. As we anticipated abjection to open a liminal space, difficult to manage and regulate, we wished to also consider the attitudes towards abject humour of those in power. In Hillside this included both hierarchical (managers), as well as heteronormative power, which lead us to extend our study to heterosexual employees.
As case study designs require the interpretation of closely overlapping phenomena, in our first reading of the data (interview and focus group transcripts), we sought to understand not only the experiences of employees, but their perceptions of the Challenge It, Change It initiative, as well. Here, we were particularly interested in representations of Hillside as a 'special case' workplace of such extreme conditions (note for example, Bethany's comment about seeing 'people's insides hanging out where they've [been] cut' in the Results section), that it did not fit into everyday norms and behavioural rules. Thus, at the outset of interviewing and following our initial visits, we started to wonder whether Hillside itself was a site of liminality. A place of total institutionalisation and hegemony, the prison was part-fascinating, part-disturbingan accurate signifier of Kristeva's (1982) abjection. What is more, the prison environment had already started to influence and manipulate our meaning-making (Westwood 2013) so we, as observers, had no better way of understanding the experiences of the abject trap, than to enter it. This allowed us to become gradually desensitised to the stark conditions which entomb the casual visitor at Hillside, triggering selfpreservation concerns through an overwhelming sense of unease and danger. We were certainly aware that prisons were a harsh environment (van Ginneken et al. 2019), not least because of the extremely dangerous individuals within them, but immersion in Hillside allowed us to accept that this environment was also a workplace, and a site for employee interactions.

Results
In line with interpretive case study research (Diaz-Adrade 2009), in this section, we construct our impressions of abject humour, as they emerged during our interviews and safety-induction at Hillside. The process of conceptualising the narratives we heard was gradual and, since the abject is not an abstraction but a phenomenon occurring in a specific context, our experiences were not unlike that of a journey. Accordingly, this is the format we chose for the section and we now proceed to interpret the experience of abjection in three stages: Producing Liminal Space, Witnessing Abjection and Trapped in the Abject. As researchers who were temporary guests at Hillside, this 'entrapment' was not permanent and we were able to leave at the end of the study. Our impressions, however, reflect the experiences of those employees who remained in it.

Producing liminal space
Following Kristeva (1982, 4), we conceptualised the abject humour as a 'neither-nor' phenomenon, which opens a difficult to regulate, liminal, 'composite' and 'in-between' space. To understand this liminal context, it was important to first study the wider organisational context within which it would entrenched. At Hillside, this included the provisions of the Challenge It, Change It programme, which was launched locally at the same time of its national roll-out in 2010. The programme covers the full range of negative acts, from outright bullying and harassment to unwanted and unwelcome behaviours, and seeks to provide feedback mechanisms for those staff targeted by them. However, despite intending to push negative behaviours outside the boundaries of acceptability, the programme opened an ambiguous, liminal space.
As a starting point, the programme's messaging failed to make clear what behaviour exactly could be challenged, by who, and what the outcomes of such a challenge might be. This placed the onus of interpreting witnessed behaviour on the employees, whoas we show belowhad a number of reservations regarding taking action against negative acts they experienced and witnessed. Programme posters (see Figure 1) offered an example of this ambiguous messaging.
As can be seen in Figure 1, these posters showed photographs of uniformed and non-uniformed prison staff, conveying personal experiences involving a single or multiple perpetrators of negative acts, and in one case, the witnessing of such acts. Messages such as: 'Should I have to choose between my religion and my work? What do I do?' and 'When I announced my Civil Partnership, people stopped being civil. What do I do?' were, we felt, overtly ambiguous, implying that acts of sexual harassment, discrimination or prejudice were, in some sense, a personal issue, despite the very public humiliation of the experience. In addition, those messages appeared to place the responsibility on the individual target and not the manager or the organisation. Back to the posters, the response to the central question 'what do I do?' seemed downplayed, as it was presented in small font at the bottom of the poster, stating 'if you are affected tell your manager, a colleague or contact Employee Support'. Furthermore, the Challenge It, Change It message, rubber-stamped in red, seemed to target witnesses alone, given the 'if you see or hear it-challenge it. You can change it' catchphrase. There was also a lack of clarity between the provisions of the central programme which, according to the Ministry of Justice required those prison employees who experience a negative act to 'challenge the perpetrator and ask them to stop' (National Offender Management Service 2011, 3), and the poster campaign at Hillside, which placed the onus on the witnesses, expecting them to contact HR or alert a manager. Such ambiguity made it difficult for employees to know not only when to invoke the programme and who to inform when they felt concerned, but also what behaviours were covered by it. The perceived use of Challenge It, Change It as at best, a surveillance tool and, at worst, a command-andcontrol device by management, contributed to employees' general 'distaste' towards the programme, as well as a perceived gap between intended and actual outcomes. Our further exploration of comments on the prisonofficer.org.uk blog offered further insight into such a (liminal) 'say-do' gap. As an example, it was possible for senior members of staff to directly retaliate against an employee who had challenged them through the Challenge It, Change It. Blog user 'DoctorPsycho' (a self-assigned profile name), elaborated on what happens when Challenge It, Change It was invoked against a superior officer: Challenge a Senior Manager these days it would seem, would change only one thing and that would be your job security. (…) Let's not forget, for every protocol and procedure, instruction or directive [such as Challenge It, Change It], there is a direct access route to the code of discipline.
If Challenge It, Change It could fail to both protect the targets of negative behaviours and prevent subsequent re-targeting, how successful was it against the much harder to detect abject humour? Not very, it seemed. In the course of our study, we became increasingly aware not only of the topics narrated by employees, but also those, which were not. As an example, we found that asking the perpetrator of distasteful joking to stop, was not among the coping strategies discussed by our participants (further outlined in Trapped in the Abject). This seemed the case for our lesbian and gay (LG) interviewees in particular, and since those who witnessed the joke being made never intervened in their defence, the LG targets of abject humour chose to hide their own vulnerability by laughing along. Was this a case of LG employees going through the process of 'enduring' (Kristeva 1982, 2) abjection? Were targets of abject humour protecting themselves by laughing, and thus 'expelling' (Kristeva 1982, 3) the very personal characteristics (e.g. gendered identity) others were shaming? We felt that these explanations fit with the understanding of abjection we proposed earlier and it is to the experience of abject humour we turn to in Stage 2 of our narrative.

Witnessing abjection
Referenced to Challenge It, Change It as a tool, used by management to police behaviour was a common thread in focus groups discussion with prison employees. However, we did not feel that employee surveillance was problematic in itself, especially if it illuminated the dark, liminal areas where abject humour could be practiced with impunity and, as a result, prevented negative acts. Thus, to understand whether the programme achieved the expectations set by its name, we presented the three scenarios discussed in the preceding section to focus group members, turning them into third-party witnesses and share their reactions to Amir's story in this section. Our intentions in doing so were to observe whether instances of heteronormative thinking still occurred at Hillside (rather than having been Changed), and if they didwhether employees felt the availability of the programme helps in addressing (Challenge) them. Amir's story is as follows: Amir, a gay man in his thirties is 'out' to everyone at work. His colleagues generally describe him as 'loud', mostly because Amir is talkative and he does not hold back when he describes last weekend's adventures. When asked about his experiences at work, Amir says that he gets on with most people, but some of his colleagues do ask pushy questions about his personal life. Amir finds this both intrusive and upsetting. These same colleagues also make derogatory remarks about gay men and tell the odd joke about them. Amir admits that he does not challenge this and most of the time he joins in the laughter. A few years ago, Amir made a couple of serious workrelated errors, which were both confronted and dealt with at the time. Since then he has received positive appraisal, but is often reminded about his errors. This troubles Amir and he cannot help comparing his own professional trajectory to many of his colleagues who have recently been promoted. He has come to the conclusion that he is being discriminated against.
Based on responses from focus group participants, it seemed that Challenge It, Change It was not able to prevent abject humour from targeting sensitive topics in a risqué manner. Almost without fail, Amir's 'weekend adventures' were assumed to be of sexual nature and a few participants made the point of Amir deserving this treatment for failing to 'know his audience' and not being careful in 'choosing who to speak to'. Furthermore, it seemed that Amir's laughter was misunderstood. In the above example, Amir's laugh was a defence mechanism and a way of self-protection through which, to paraphrase Kristeva, he became one of the group by being prepared to destroy (through ridicule) his identity (Kristeva 1982, 3). However, for Hillside employees, his laughter opened a space of ambiguity, which connected the person being laughing-at and those laughing, in an uncomfortable experience for all. Participant views did not appear to fully reflect organisational policy, nor were they fully outside it. As a result, abject humour obscured the normative compass for appropriate behaviour and created a liminal space, the dynamics of which participants had to interpret for themselves. Consequently, focus group participants, as hypothetical 'witnesses' of abject humour, failed to invoke Challenge It, Change It requirements in a consistent, or unified way.
Those aware of the policy, applied it in a way which suggested Amir should have self-regulated. He was sharing too much of his 'life outside' the prison and thus failed to comply with the accepted code of conduct. In exposing colleagues to his comments, he was 'asking for it'.
It [sharing of personal stories] shouldn't happen in the workplace, and therefore management should enforce that. So therefore no one should be telling them sort of jokes, whether it's acceptable or not, and therefore they're just inappropriate whether anyone agrees with it or not. (Focus group 1, M1) In a different focus group, a Hillside manager detailed how he would prevent Amir from being the target of abject humour. Ironically, this included disciplining Amir himself. I would be having a quiet chat in the office and explaining in the workplace, those sorts of conversations aren't appropriate, you know, 'you've gone beyond the boundary as far as I'm concerned, that's it. Do your job', I'm not for', I'm not against whether it's a bit of banter, a bit of, yeah, I've had a good weekend, I've done this, that and the other, but there's a level as to where you take that, although people might not say they don't find it appropriate, there's people there that are quite quietly being offended. So, when you get to a certain point, let's draw a line. (Focus group 2, M3) Interestingly, Challenge It, Change It did not seem to unilaterally empower employees, as participants felt it was down to management (rather than witnesses) to determine if there was 'foul play', and take the appropriate actions required by the situation. When participants did adopt a thirdperson perspective, as a witness would, they did not feel there was anything needing reporting: [Amir] doesn't challenge it but laughs along with it, and I think that gives people then the wrong perception. If you're unhappy with something then … it's a natural feeling to show that you're upset. But when you start laughing along with it then it would be a natural thing for people to continue. (Focus group 1, F2) Some inverted the sentiments in DrPsycho's blog (see Stage 1 section above). Rather than viewing Challenge It, Change It as an inherently useful provision but subject to management retaliation, those participants (like M3) complained that the role of managers was unnecessarily 'burdened' by policies, which required them to consider 'harmless' everyday banter as potential bullying and harassment. Note the gender differences in the below excerpt, where the female participant (F1) expresses a legitimate concern, which is interrupted and invalidated in a display of hegemonic masculinity from M3: F1: [Amir] will have to go to management or whoever's in charge and say, 'I feel as though I'm being bullied by them [because] unless you say', if you've got a grievance with somebody, unless you sort of take them to one side and say, 'look, I'm not happy with this', it's never going to stop, is it? M3: I know that but F1: It's like bullying anywhere; it is a form of bullying. If it goes too far and there's three or four and one person, whatever you are, (gay, straight, male, black) M3: (No, no, but it's used far) too loosely for me, bullying though. You can just walk past someone and not say morning and you're [accused of bullying]. (Focus group 3, F1 and M3) Lastly, a number of participants sought to neither interpret, nor challenge abject humour as it occurred. For them, Hillside itself was a 'unique' place, which required a 'unique' sense of humour amongst prison staff. Furthermore, both humour and the ability to 'take a joke' were largely framed as a critical component of the job and, overall, a desired personality trait to fit in with the prison culture. Since abjection is separation from a threat (e.g. unwanted, lewd and crude humour), without the ability to remove oneself from it (e.g. either by reporting or exiting) (cf Kristeva 1982, 9), we considered those participants to be the ones 'trapped in the abject'. It is to them we turn next.

Trapped in the abject
At Hillside liminality was a multifaceted phenomenon and seemed to occur at a number of levels.
Hillside itself existed as a liminal site, neither fully outside of society, nor fully integrated. Abject humour interactions also opened internal liminal spaces, with their own dynamics which were difficult to interpret, navigate and regulate. This explained the difference in the attitudes of employees put in the position of 'witnesses' of abject humour, as responses to Amir's story showed. Was there a difference in attitudes towards abjection when the participant him/herself was the target? Most interviewees described Hillside as a 'testosterone fuelled' and 'macho' environment, an altogether 'different kind of place'. For some, this meant the acceptance of abject humour as part of the job. Bethany seemed to suggest that, since Hillside constituted a traumatic working environment (cf. Maschi, Viola, and Koskinen 2015), common everyday norms and codes of conduct did not apply. People, including herself, 'took the piss' [mocked one another]. That was 'just the way it [was]' and no further explanation seemed necessary: You sort of get like, not a sick sense of humour cause that sounds bad, but cause you're working with many different types of people and you like, you see people like fighting, you [see] people cutting up, you see like people's insides hanging out where they've [been] cut, and you've had to take people to hospital, and you see people like swinging, hanging themselves, you just see things which a normal day to day job you wouldn't see … (…) you take the piss out of one another to be honest and that's just the way it is, it's weird. Therefore, Bethany seemed trapped in the abject through her practice of retaliation. Further in the interview, she related a complex incident with her manager, Jane, which was managed through the neither quite friendly, nor fully hostile uses of abject humour of her own: This is how bitchy it can be right, I'm not arsed, I let it all go over my head me, but she [Jane] was working on my wing for a little bit and we were talking about going out to … a lesbian gay bar and she was saying that how she feels old and that in there, cause she is, she's about thirty odd, forty odd or something and she was saying that she goes in there with her partner but she looks at all the baby dykes or something she said and we were in an office full of people and she's like, 'oh I bet you're a baby dyke aren't you, with like your knickers above your arse type thing, and your vest top on and your short hair and all that lot', so I was laughing saying, 'like whatever', I said, I said 'does that', no that's it she said 'oh I used to be a baby dyke', so I laughed and said, 'oh as long as I don't turn out ought like you when I'm older'.
Experiences of abject humour oscillated between teasing and bullying, e.g. from poking people in the stomach and telling them they are fat, to circulating photographs of male genitalia amongst colleagues, from colleagues being told they smelled disgusting to jokes aimed at large-breasted colleagues. As a result, another way through which staff responded to abject humour, was through practicing compliance and 'going along' with what was being said. Morgan, a nurse, explained: They [management] have this Challenge it Change it, right, [laughs] which, on paper, is a phenomenal idea. You've got the opportunity to speak your mind, challenge it, and then if it doesn't work, you've got a set procedure that you'd go through to change it. Great on paper. What do you think, realistically, would happen if there was a group of officers and they kept making comments about wanting to have sex with my girlfriend or myself or wanting to know about my sex life or how it is to be a lesbian or whatever and I was getting offended by it? If I went up to them and said, 'You're really pissing me off now,' it wouldn't have the same effect as it does on paper. It wouldn't … you just wouldn't challenge it because, for one, they'd be like, 'Oh, that new nurse, she's a right arse. She can't even take a joke.' That's what it would be.
In actual fact, Morgan was repeatedly asked about her sexuality and sex life and her reaction was as she described above. The fear of being trivialised or labelled as a killjoy stood in the way of her willingness to confront and oppose any inappropriate remarks. Instead, when encountering abject humour targeting her directly, she acknowledged the joke, but 'laughed it off'. Morgan felt that challenging any of the received comments would also put her 'on her back foot' and distance herself from her peer group. With the escape route neither desirable, nor possible, Morgan was also trapped in the abject.
When practicing compliance was unpalatable to the target, a final, avoidance strategy was used, instead. Following regular offensive remarks by one of his colleagues, Andrew tried to protect himself in this way, by keeping interaction with her to a minimum. He explains: I'll say 'hello'; I'll say 'morning'. If she asks me a question, if she comes into my office, I'll chat with her, but I'll try and keep it short to get her out. Reporting her, speaking about it, I don't think would make any difference whatsoever other than she would then get in trouble and then there is the whole thing about if I … It's like I said before. If I complained about her, she would then tell everybody that I complained about her and everyone would just think I'm too sensitive.
Although a range of strategies had emerged as a result of being trapped in the abject, avoidance, compliance and retaliation responses failed to provide a means of 'escape' from the abject, because none of the above-stated incidents were reported. Consequently, on the whole, managers and HR were not aware of any 'immediate problems' at Hillside. In turn, since avoidance, compliance and retaliation responses were largely outside the remit of the Challenge It, Change It programme, we questioned to what degree the latter was successful in achieving either of the objectives in its title. In the next section, we discuss the implications for employees and the wider organisational studies literature.

Discussion
Although sometimes associated with levity and deemed unworthy of serious attention (Lang and Lee 2010) humour is a complex phenomenon, with multiple functions in organisations (Lang and Lee 2010;McGhee 1997). As a result, it has been subject to comprehensive and multi-disciplinary theoretical conceptualisation yet, with a few notable exceptions (e.g. Kahn 1989), there is a general absence of studies exploring darker uses of humour in organisational settings (Thomas and Al-Maskati 1997). This gap has, in part, been addressed by lesbian, gay and trans-sexual scholars, who have described how hegemonic masculinity (Plester 2015) can be practiced through jokes, teasing and mocking, in order to marginalise and 'other' non-heteronormative groups (Plester 2015). Less well-researched, however, are its ambiguous uses where, for instance, humour ceases to be a both-and activity with clearly identifiable dimensions but is instead, an experience of abjection (Kristeva 1982).
Addressing this gap and answering calls to study humour in a variety of organisational studies (Thomas and Al-Maskati 1997;Lang and Lee 2010) we extend Julia Kristeva's 'theory of the abject' into a conceptualisation of abject humour. In line with this we consider abject humour to be ambiguity (Kristeva 1982, 9), a dissociation from certain (taboo) personal characteristics as a way of self-preservation and protection (cf Kristeva 1982, 3). Abject humour opens liminal and in-between (Kristeva 1982, 4;Rizq 2013) spaces. As those are neither fully subject to organisational norms, nor in open rejection of them (Rizq 2013), we anticipated that abject humour would be difficult for its targets to detect, and organisations to regulate. Importantly, since abjection drew both the 'speaker' and 'target' together in a shared, uncomfortable experience, we proposed that abject humour would be difficult to escape.
Hillside offered fertile ground for our observations on several counts. As a high-security male prison, it was itself a site of liminality, neither fully outside society and social norms, nor fully participating in either. Furthermore, Hillside had adopted the UK-wide, Challenge It, Change It programme which was aimed at safeguarding employees from a wide range of negative behaviours in a wide range of contexts. It did not seem to have achieved its purpose. Although aimed at preventing negative behaviours from occurring and, at least notionally, attempting to offer guidance to staff experiencing them, the Challenge It, Change It programme appeared to be propagating ambiguity and entrenching liminality. One of the reasons behind this were contradictory messages regarding who was responsible for identifying inappropriate behaviour when it occurred, and what steps could be taken to stop it. Specifically, it was unclear if prison staff were expected to challenge the perpetrator directly, or 'tell [their] manager, a colleague or contact Employee Support' (both of which were stated on the information posters).
The Challenge It, Change It programme, therefore, had inadvertently produced a liminal space within Hillside, which made it difficult for those witnessing abject humour to act. To explore this further, we used three pre-constructed scenarios based on themes of our ongoing research and published elsewhere (Author, XXX). In those cases, we found that participants were more likely to adopt a heteronormative stance, which failed to notice a problem with the type of abject humour on display or empathise with its target (Amir). However, if noticed, the target of abject humour seemed likely to experience double victimisation -first thorough being targeted by abject humour and then, through being disciplined.
Whilst witnesses found it difficult to find an appropriate (or even sympathetic) response to abject humour, its targets and perpetrators appeared unable to escape it. The Challenge It, Change It programme was too ambiguous to provide a meaningful feedback channel which, even when used, left employees open to manager retaliation (as DrPsycho observed). Only one option remainedoperate within the liminal space, which abjection opened but stay trapped in it. In line with this, we identified the use of avoidance, compliance and retaliation strategies as means of navigating the dynamics of abject humour. Andrew avoided confrontation by keeping interactions with the person targeting him to a minimum. Bethany retaliated against her managers' crude and personal comments, while Morgan went along (complied) with abject humour, viewing it as part of the job. Our findings extend current understandings not only of humour in organisations, but also of the coping mechanisms which lesbian and gay employees deploy when they find themselves the target of jokes, teasing and mocking, at times, by members of their own groups.
Extant studies have identified uses of humour as a means of (heteronormative) control (Abedinifard 2016), especially in environments of hegemonic masculinity, such as prisons (Michalski 2015;Hsu 2005). In our study, uses of abject humour had several parallels with existing understandings. At Hillside, abject humour was made possible by policy ambiguity, which opened an equally ambiguous, liminal and in-between space. Abject humour was used as a way of interaction, yet offering neither the positive bond of camaraderie nor carrying the overt bullying overtones of exclusion. It was a way of caricaturing non-heteronormative traits (such as homosexuality), in a way which was neither in line with equality and diversity principles nor (always) an instance of overt harassment. At the same time, this was also a way for the owners of non-heteronormative characteristics to reject this caricaturing and safeguard their own identity, as well as position within the employee group. Interestingly, although those using and experiencing abject humour had found a way to navigate the liminal space they shared with each other through self-preserving avoidance, compliance, and retaliation strategies, neither were able to escape it.

Implications and limitations
Having outlined and discussed our findings, we are now able to comment on the effectiveness of employee safeguarding initiatives such as the Challenge It, Change It programme. Accordingly, we observed a range of difficulties in the way of regulating liminal spaces, opened by abject humour. Characterised by ambiguity, the hidden and more sinister undertones abject acts (from jokes to physical behaviours such as belly-poking) were only fully known to the perpetrator and target and were likely to escape the attention of witnesses. Thus, relying on the latter to 'challenge' inappropriate behaviour, in order to achieve a 'change' in the long-term seemed unlikely to produce the desired results.
This did not mean that inappropriate behaviours should be allowed to continue unchallenged, nor that Hillside employees should not be made aware of the damaging impact of uncomfortable humour. It did however mean that Hillside had overestimated the ability of witnesses to act as change agents when it came to abject humour. What of the targets themselves? Were they sufficiently empowered by the programme and able to challenge abject jokes? This requires a more complex answer. Using the strategies outlined above did enable employees to manage (but not fully avoid) exposure to abject humour and navigate the dynamics between themselves and the perpetrator, at times using abject humour of their own. Yet, Challenge It, Change It was ineffective in putting an end to the shady, sinister (cf Kristeva 1982, 4) and generally negative experiences, associated with abject humour. This, it seems, required changes and actions than went above and beyond reliance on employees' ability for self-regulation, or expectations that witnesses engaged in organisational citizenship. To improve the impact of the programme, we propose the need to add the currently missing dimension of compassionate leadership (West and Chowla 2017). Increasingly gaining traction in the UK's National Health Service, compassionate leadership is a framework promoting manager awareness of individual personal circumstances (Manolchev, Lewis, and Pascucci 2021). It seeks to overcome workplace cultures of victimisation and blame, through crossdepartmental collaboration towards worker support (de Zulueta 2016). It also places the responsibility for appropriate behaviours on both manager and employee groups, rather than reducing the practice of equality and diversity to a top-down process imposed on subordinates and policed by managers. Whilst this would not guarantee the success of an intervention programme, we feel that the introduction of compassionate leadership will enable managers to lead by example, rolemodelling behaviours which employees could use to better orient themselves in the course of their everyday encounters with unwanted acts and behaviours, such as abject humour.
It is important to note some limitations of our Hillside case study. Its interpretive design (Diaz-Adrade 2009) was adopted to allow us, as researchers, to co-construct the experience of abject humour, through observations and engagement with participant groups. Regrettably, the latter did not feature any senior management representatives. This angle would require exploration in future research. We also did not explore any supplementary training that may have been used to compliment the campaign, which opens further avenues for understanding the full impact of the Challenge It, Change It programme. We believe it would be beneficial for both components to be added to the design of future research into initiatives aimed at safeguarding LGBT + groups from the barbs of abject humour and associated negative acts.

Conclusion
There is little doubt that Hillside offered an unusual and 'different' working environment. Our interviewees shared their experiences of a hyper-masculine, aggressive culture, one very much aligned with descriptions of hegemonic masculinity observed in the literature (Plester 2015;Michalski 2017). The darker side of humour has indeed been linked to masculinity in previous studies (Godfrey 2016), for instance, as an attempt to reinforce heteronormativity (Abedinifard 2016). Our study takes this literature further. We use Kristeva's (1982) 'theory of the abject' to introduce the concept of abject humour as an unpleasant and uncomfortable act, which opens ambiguous and difficult to regulate, 'in-between' spaces. Although difficult to regulate by the organisation, those liminal, in-between spaces were not impossible to navigate.
As a result, we observed the use of avoidance, compliance and retaliation strategies by employees who were trapped in the abject by either perpetrating or being targeted by abject humour. This seemed to have been made necessary by the ineffectiveness of Challenge It, Change It both in terms of reporting provisions (who to contact and when), scope (failing to clarify the role of managers in the process) and perhaps, its fundamental principles (placing overt responsibility on front-line staff). As a result, and despite the emergence of bottom-up coping strategies, the programme did not seem to alleviate the mistreatment of lesbian and gay prison staff, nor did it prevent the teasing which targeted them. Instead, lesbian and gay employees were left, trapped in the abject with their abusers, often bereft of collegial empathy and managerial compassion.
So, how can we move the discussion forward? It seems that the standardised guidelines of programmes such as Challenge It, Change It, fail to address behaviours at the boundaries of im/propriety (Djurkovic et al. 2020). Furthermore, their inherent assumption that all staff are equally resourced to challenge inappropriate behaviour seems unfounded. Lesbian and gay prison officers in our study were particularly likely to experience daily teasing, unwanted jokes and ambiguous humour but were unable to prevent it. Ironically, Hillside recognised that humour can be problematic but was not able to regulate the liminal spaces where abject humour was encountered. This is concerning, given the tendency for bullying and harassment to escalate and offer progressively more traumatic experiences over time . Left unsanctioned, it is possible that abject humour, directed at sexual minority employees and already difficult to spot and sanction, can take on just such darker undertones.
These considerations suggest that abject humour will continue to pose a challenge to human resource (HR) professionals who would need to work closely with managers to not only design policy interventions, but role-model behaviours. Organisational policies can set expectations of conduct and regulate counterproductive work behaviour, yet we recommend that they need to be delivered with compassion, which avoids shifting the burden of responsibility solely onto employees. Expectations that prison officers would deliver on three frontscorrectly identify negative acts when witnessing them, feel empowered to challenge those acts even when being targeted themselves and maintain working relationships with colleagues, were not met at Hillside. Challenge It, Change It had seemingly forced negative acts out of bounds, yet left the door open for abject humour, which affected sexual minority groups disproportionately. Both targets and perpetrators of humour were trapped in the abject and forced to navigate the liminal space they co-inhabited. Our study identified some of the coping mechanisms used by employees at Hillside, but further research is required to illuminate the full range of strategies, both within heteronormative workplaces and beyond them. This is an urgent task for researchers because joking may be funny, but humour can hurt. Note 1. All names have been replaced with pseudonyms.