Researching inside police custody in four jurisdictions: ‘Getting in’,‘getting on’,‘getting your hands dirty’ and ‘getting through it’

Introduction

Police detention is an important entrance to the criminal process, where large numbers of suspects are held each year. ¹ Yet, it is relatively under-researched, especially in some countries (e.g. Ireland), albeit that the last decade has seen a revival of interest (e.g. Bevan, 2020; Dehaghani, 2019; Holloway et al., 2020; Skinns et al., 2020). Comparative research on police detention is limited to a few studies (see Blackstock et al., 2014; Skinns, 2019), and those writing about the principles and practices of research inside police detention are also scarce (e.g. Greene and Skinns, 2018; Phillips and Brown, 1997; Skinns et al., 2016), with none examining this research process comparatively. As a result, this article makes a unique contribution. I return to field notes written as part of a multi-method comparative police detention study conducted in 2007 and 2009 in England and Ireland and jurisdictions in Australia and the United States. ² With the benefit of hindsight, I look at what can be learned about the police detention research process. This research took place on the cusp of the proliferation of research with the police, that is, involving close collaborative relationships between the police and academics (Bradley and Nixon, 2009). By contrast this project was an example of research on the police, in which academics offer an external, sometimes critical, perspective and in which I was thus an ‘outside outsider’ (Brown, 1996), even an ‘outside outside outsider’ given my ‘foreign’ status in Australia, the United States and Ireland. By looking at due process rights in practice, and the dynamics of power, police culture and discretion in the context of neoliberal political economies, it was also more theoretical and did not have the impact focus of some of my subsequent research.

As argued in this article, this approach and the findings generated and presented by Skinns (2019) are still valid, and there is also much that can be learned about the police custody research process. Although research with the police has come to dominate, at least for the moment, I argue that it is only one part of what Bowling et al. (2019: 14) see as an increasingly ‘diverse’ and ‘plural’ police research agenda. In order to support these arguments, I offer a reflexive ‘fieldwork confessional’, in which I demystify the fieldwork process in police custody research (Van Maanen, 1988: 73), considering what can be learned from the challenges of ‘getting in’, ‘getting on’, ‘getting your hands dirty’ and ‘getting through it’. However, first, I review the literature on doing police research and, second, the politics of police research.

Doing police research

Police research presents a number of ever-changing challenges for researchers. The nature of the police role – which includes the use of coercion and controversial, if not illegal tactics – means that one of the peculiarities of police research is that it explores actions and information that research participants may prefer to hide (Reiner, 2000). This tendency towards being closed off is also rooted in aspects of police culture, such as suspiciousness (and also isolation and solidarity; Skolnick, 1966: 42–70), which may extend to researchers (Fassin, 2013: 18–19). At the same time, others have remarked on the surprising degree of openness that police participants display towards researchers (Reiner, 2000; Souhami, 2020), including in police custody (Phillips and Brown, 1997). However, this openness may be contingent, for example, on the actions of researchers including the frequency of their presence and their efforts to build trust (Fassin, 2013; Marks, 2004; Reiner, 2000; Souhami, 2020; Westmarland, 2016: xii), on their status and positionality (Phillips and Brown, 1997) and on the politics of police–academic collaboration.

In the past, police actions were also hard to uncover by researchers due to the ‘low visibility’ of everyday police work (Reiner, 2000: 219), especially in police custody, given its ‘back stage’ nature (Holdaway, 1980: 89). However, this has since changed. Police custody work has been rendered more visible to the public due to audio-visual enabled CCTV and growing numbers of criminal justice actors in police custody (Newburn and Hayman, 2002; Skinns, 2011: 82–84), and as part of ‘new visibility’ (Goldsmith, 2010), which may alter how the police respond to researchers in their midst. In short, the police are increasingly used to being watched, including in police custody, meaning that they may be more open to also being watched by a researcher and/or also more adept at hiding what they do not want to be seen.

Given the largely White, male, heterosexual and hierarchical nature of police organisations, a researcher’s gender, race and nationality are salient aspects of police researchers’ positionality. For example, female police researchers report being seen as unthreatening and harmless, thus enabling frankness in participants (Horn, 1997). ³ Indeed, some female police researchers take advantage of this. Diphoorn (2012: 207) notes that sometimes she was ‘purposely naïve and subordinate’ and other times ‘tougher’ and like ‘one of the guys’ or sometimes she was ‘flirty’ and other times ‘purposely . . . nurturing’ as a way of encouraging participants to share their emotional reactions to their work. However, this raises ethical challenges, and female researchers risk being seen as duplicitous and as putting the interests of the research over their obligations to be open and honest with participants, though of course this duplicity may be mutual (Punch, 1989). As for race, Milner (2007) cautions researchers to consider in what ways our racial and cultural backgrounds and those of research participants influence the research process, noting that this kind of reflexivity may bring to a researcher’s ‘consciousness explicit, hidden, or unexpected matters, which can have a bearing on an entire research study’ (p. 395). Certainly, engaging in such ‘racial and cultural introspection’ about my mixed-race but White assumed identity while writing this article led to some new epiphanies about my fieldwork (as discussed later; Milner, 2007: 397). Nationality is a further part of a researcher’s identity, which may be especially important in comparative research. Jardine (2019) notes that researching the police in one’s country of origin can influence researchers’ ability to question the taken-for-granted, while being ‘foreign’ and from a different country to police participants may enable researchers to ask naive questions.

Another way of thinking about positionality is in relation to the role of the police researcher in the field. Brown (1996) outlines four as follows:

As with other forms of positionality, these roles impact on all aspects of the research process. For example, ‘outside outsiders’ are most likely to experience the greatest barriers to both formal and informal access, though these are not insurmountable (Reiner, 2000). The roles played by police researchers can also create ethical dilemmas and conflicts of interests (Skinns et al., 2016).

One of the main anti-dotes to the challenges of positionality is reflexivity. This refers to ‘researchers recognizing . . . that their insight about social worlds and processes (as socially constructed, and mediated by tensions and intersections between agency and structure) also apply to themselves, the social worlds of the academy and to their own work’ (Armstrong et al., 2017: 12). In practice, this means researchers acknowledge their positionality and consider how they affect the research process (e.g. on participation, field relations, data, its interpretation and the knowledge that is produced) and how the research process in turn affects them. Such reflexivity is seen as particularly important, in light of the growth of research with the police (Henry, 2017; Lumsden, 2017), given the potential for such relations to erode the much strived for ‘interdependent independent’ relationships characteristic of strong collaborative relationships (Laycock, 2015; Rock, 1990: 39). Although it may be imperfect (e.g. undermining all claims to truth or encouraging navel gazing; Armstrong et al., 2017: 23), ultimately, reflexivity encourages ‘critical dialogue’ vital to democratic societies (Fassin, 2013: 29), which recognises that neither academics nor the police have a monopoly over the truth (Henry, 2017; McAra, 2017: 165; Wood and Williams, 2017). In this piece, I therefore offer an account of one version of this ‘truth’ about the police custody research process, though by locating this in relevant academic literature enables broader exploration of the meaning of this account, for example, for police–academic collaboration.

The politics of police research: The rise of research with the police

Police research is shaped not just by personal identities, subjectivity or location but also by the politics and culture of academia (Henry, 2017; Souhami, 2020 171), including the growth of police–academic partnerships, particularly in the last 20 years in most parts of the Anglo sphere (Bacon et al., 2020; Canter, 2004; Engel and Henderson, 2014; Marks et al., 2010; Rojek et al., 2015). This is partly rooted in shifts in the academic climate, most notably the growth in the status of criminology and police studies since the 1960s (Rojek et al., 2015). Within academia, it is also rooted in the growing importance of university-community engagement, public criminology and evidence-based practice (Bartkowiak-Theron and Herrington, 2015; Loader and Sparks, 2010; Sherman, 1998), as well as in the need for social science research to make a demonstrable impact (Chubb and Watermeyer, 2016; Pearce and Evans, 2018). For the police, collaboration with academia has been stimulated by a lessening of the cultural divide between the police and researchers through growing police professionalisation (Canter, 2004; Greene and Skinns, 2018; Rojek et al., 2015). The growing complexity of the police role linked to the increasingly complex social conditions has provided further stimulus for police–academic partnerships, as have the demands for doing more with less, precipitated by austerity politics and cuts to police budgets (Innes, 2010). However, at the time of writing, fuelled by concerns about racialised police practices, questions have arisen in the media about the merits of police–academic collaboration (e.g. Jameela, 2021), suggesting that the current emphasis on research with the police may shift once again in the future.

For the time being, though, this growing emphasis on research with the police has implications for the conduct of police research and for what research becomes valued. The insider/outsider dichotomy becomes more porous, with ‘inside insider’ police officers, becoming more familiar with the language, tools and skills employed by researchers and ‘outside outsiders’, like academics, having more opportunities to work with the police, to familiarise themselves with police culture and to become ‘outside insiders’ (Davies, 2016). At the same time, this prioritising of research with the police also risks valuing crime control-oriented ‘what works’ quantitative research (Davies, 2016; Lumsden, 2017), over theoretical or qualitative research, or research that considers ‘what matters’, including ‘justice, equity, privacy, rights, use of force, accountability, governance, abuse of power, discrimination, values, etc.’ (Punch, 2015: 15). Others, however, see the current phase of police research as ‘diverse’ and ‘plural’ (Bowling et al., 2019: 14) and as therefore offering encouragement to police researchers to use a range of research methods and approaches, including research on and with the police.

The aim of this article is to explore four methodological challenges of conducting research inside police custody in cities in four jurisdictions: getting in, getting on, getting your hands dirty and getting through it. Although this research was an example of research on the police, in which I was an ‘outside outsider’, in the conclusion I reflect on the value of the research nonetheless as part of a rich history and a broad spectrum of research now and in the future, in spite of the current trend towards research with the police.

Research methods

This article draws on two studies, the broad aims of which were to examine police powers and citizens’ rights in practice, situating these understandings within the political economy of police work. The first was a mixed-methods study of two police custody areas in cities in the South of England, which explored the impact of inter-agency cooperation on suspect access to due process rights. Between February and July 2007, in each custody area, a week-long period of participant observation was followed by semi-structured interviews with a range of criminal justice practitioners and suspects and, then by the collection of quantitative data from police custody records. In the second study, data were collected in August, October and November 2009 in cities in Ireland, Australia and the United States. These countries were selected for inclusion in the research because of their shared common-law systems, and also because of their shared language and long-standing social, economic and political ties. This second project was more exploratory, charting similarities and differences between jurisdictions with regard to suspect due process rights and implications for police discretion, the law, police culture and state–citizen relations. In a month spent in a large city in each of the jurisdictions, ⁴ as much data were collected as possible through observing in police custody and by conducting semi-structured interviews with staff. In both studies, interviews were transcribed, while brief contemporaneous observation field notes were written up in full immediately after observing a shift. In total, the research involved 480 hours of observing in 17 custody blocks in five cities and in four jurisdictions and 71 semi-structured interviews mainly with police and other criminal justice actors, though also with detainees in the two English cities, all of which were thematically analysed for the purposes of the research. ⁵ In both studies, these data were accompanied by electronic field diaries.

Due to limited space, findings from these studies will not be considered in detail in the present publication and, in any case, they have been published elsewhere (see Skinns, 2011, 2019). Instead, here, I return to the relevant data from these studies relating to my experiences of collecting them. Given that more than 10 years have now passed, I inevitably reflect on them through the frame of what I know now. What I have done, what I have learned, what I have experienced personally and professionally in the 10 years have provided me with a new set of perspectives. These enable ‘insights which may have been unavailable at the time the research was conducted’ (Souhami, 2020: 5) and also facilitate the reinterpretation of prior lived experiences in a new context (Fleming, 2012). A crucial aspect of this new context is the growth of research not just on but with the police and my growing awareness of this, reflecting the prevailing dynamics of my ‘epistemic community’ of police scholars in the United Kingdom (Henry, 2017: 171). Notwithstanding the ‘curated’ nature of even private records like field notes (Souhami, 2020: 2), they provide an important resource for understanding the process by which one engages in research. They reveal something about my positionality at the time of the research and how this informed the research process. I am a mixed-race British Anglo-Indian woman who is routinely assumed to be White and at the time of the research I was in my early thirties and a Cambridge post-doctoral researcher engaging for the first time in a new area of study.

Getting in

After writing letters and filling in forms, formal access was granted to do the research in each of the cities by senior officers, though this did not guarantee access to the custody blocks when required. At three of the seven police stations I visited in the American city, no one at reception or among the supervisors on shift knew about the research or who I was. In two of these three police stations, the reception I received was nonetheless friendly and helpful. For example, in AMEPO7, while waiting to speak to a lieutenant in the operations room, staff offered me a slice of their pizza. In contrast, in AMEPO3, on arrival, I was told that ‘civilians were not allowed into the custody area’, and multiple photocopies were taken of the three forms of identification I had on me: my university card, driver’s licence and passport. In the Australian city, even though I had a police chaperone with me, I was similarly asked to verify my identity and I was also initially only allowed to observe in the first custody block for 4 hours while staff decided whether or not they were comfortable with my presence. After 4 hours, I left the police station and waited nearby until I received a phone call to say that I could return. Even my chaperone felt staff were being ‘overly cautious and a bit precious’. This highlights that even my chaperone – who was ostensibly an ‘insider’, as he was a serving police officer – was regarded as an outsider in the custody block.

Once inside the police station, the challenges of getting to where I needed to be did not end there. In the English cities, for example, while I was free to move around the main charge room, I was accompanied to the cells by staff, with whom I had to negotiate if I wished to do so and to access potential detainee participants. Owing to the busyness of staff, such requests were not always met, for understandable reasons and I depended largely on their good will to facilitate my requests. In the American city, if I returned to the custody block after having been elsewhere in the police station, when waiting outside to go in, I would be routinely asked who I was and what I was doing there, sometimes in suspicious, officious and aggressive ways:

After being upstairs in the detective division I went back down to the custody block . . . Waiting outside was an officer, who was extremely suspicious of me. He said to me ‘can I help you?’ I told him who I was . . . I also offered to show him my id, an offer which he did not take up . . . I then stepped towards the door again to see if I could go in. He said ‘Miss, you can’t go in there. Civilians are not allowed in there’. I then explained to him that ‘I’ve been in there from 6-10 pm and I’ve been to every other custody block in each of the divisional headquarters in this city’. He eventually relented and asked me a bit more about myself. The door opened and I walked in. (AMEPO7)

These examples illustrate the territoriality in and backstage nature of police custody (Holdaway, 1980; Skinns, 2011: 179–181), into which outsiders and even some insiders, like my police chaperone, were unwelcome. This combined with a sense of suspiciousness (Skolnick, 1966) and also a measure of security-consciousness. The difficulties of ‘getting in’ to the police station and also into the police custody block therefore confirmed my ‘outside outsider’ status.

Getting on

This ‘outside outsider’ status hindered, but sometimes helped the research process. My gender, for example, affected how seriously I was taken and how some of the officers responded to me, though, as found by others, this was something I could use to my advantage (e.g. Diphoorn, 2012). In AMEPO2, I was mistaken by two police officers, one male and one female as being an undergraduate or master’s student. For example, one of them asked me ‘what are you majoring in at college?’ Another officer in AMEPO6 who drove me home after observing a shift was shocked when I told him that I had my doctorate. He falteringly asked me, ‘so, how . . .’, ‘how old am I?’ I asked, finishing his question for him. ‘I’m 32’. He said ‘I can’t believe you are older than me, I thought that you were about 27’. While this failure to recognise my credentials and my experience was frustrating, it did not seem to hinder the research as noted by others (e.g. Belur, 2014). To the contrary, in a male-dominated policing environment, my gender facilitated the ‘naive researcher’ role and elicited carefully explained information. Moreover, the flirtatious undertones to my conversation with one of the detectives in the American city – on his part, not mine – meant that I was able to talk to him at length about his work. As noted earlier, being a female researcher was advantageous because of stereotypes about women as unthreatening, harmless, trustworthy and understanding.

With hindsight, my White assumed ethnicity provided an additional layer of harmlessness and trustworthiness, when it came to matters of race. It explains, for example, the openness with which some of the White officers, particularly in the Australian jurisdiction, aired racially prejudiced views and used racially prejudiced language in front of me. For example, an officer familiar with the main ethnic groups of the United Kingdom described Lebanese Australians as equivalent to ‘Pakis’ (AUSPO3). These officers presumed that, ethnically speaking, I was one of them and therefore would not mind their racial prejudice, but of course I did, given that similar racial slurs have been a consistent feature of the lives of my maternal family in the United Kingdom.

My nationality had an inhibitory impact on field relations, however, particularly in Ireland. Being English made me doubly outsider, a kind of ‘Outside Outside Outsider’. Unlike Jardine (2019), I did not have the advantage of being an occupational insider, though a cultural outsider:

The fact you’re English probably doesn’t help . . . If you told the average garda that an English criminologist from Cambridge is going to come and talk to you about your work, oh no, that’s probably their worst nightmare, to be frank, so that is a set of barriers that you’ve got to try and work your way through. (IREI5)

In the American city, being doubly outsider was evident in the way participants responded to the way I spoke. They found my accent hard to understand and asked me whether I had a strong accent. I found this puzzling, as while I grew up in Yorkshire, any traces of the accent I had while living in my home town have been rubbed away by living in a number of cities in the United Kingdom – Edinburgh, Cambridge, London and Sheffield – and by being surrounded by friends and colleagues from around the world.

Sometimes my outsider status prompted officers to express concern about me obviously taking notes in custody for fear it would arouse suspicion about how those notes would be used and the potential for the disclosure of evidence, particularly in sensitive criminal cases (ENGFD). It also prompted officers to interrogate me about ‘whose side I was on?’ For example, detectives in AMEPO3, on hearing that my research intended to use comparative methods to identify improvements to police detention practices, immediately asked ‘who will the police custody process be improved for?’ I was clearly being asked about ‘who’s side was I on?’ ⁶ I said that improvements would affect both suspects and staff, pointing out that these matters were different sides of the same coin. For instance, reducing the time spent in custody was better for suspects as it met their need to be released as quickly as possible, while also easing the burden on custody staff. My approach in this instance differed from other researchers in similar situations who sought to side with the police, at least superficially (e.g. Souhami, 2020; Westmarland, 2016). By saying that I appreciated both perspectives, and was thus on both sides, I expressed ‘critical empathy’, in which researchers ‘understand and explain the perceptions and practices of others while simultaneously maintaining a measure of distance that allows – indeed, demands – questioning and critiquing of these same perceptions and practices’ (Jauregui, 2017: 63). It is hard to say how not siding with the detectives affected the research. Perhaps, they appreciated my honesty or at least saw my response as acceptable. Either way, the detectives continued to talk to me at length in a frank way. Being an outsider, seeing oneself as doing research on the police and being honest about this may not always be damaging in the ways that are presumed.

In spite of my best efforts to cultivate and maintain relationships with participants and to overcome the effects of my outsider status, some were reluctant and reticent to participate, and perhaps even antagonistic towards me and the data I were attempting to collect. Some barriers could be broken down, for example, through laughing and joking (Marks, 2004; Souhami, 2020) or through sheer persistence and ‘just being there’ (Westmarland, 2016). However, on occasions, barriers to the research were insurmountable. For example, in IREPO6, nearly all staff on shift made it clear to me that they did not want to be observed or informally interviewed. They gave one-word answers to my questions, switched off CCTV monitors so I could not see what was happening in the custody block or would not acknowledge my presence. There were also instances of cautiousness among Garda participants, which were not encountered in the other cities, including sticking to the official line of what they thought they ought to be saying (see also Belur, 2014; Jardine, 2019; Rowe, 2007), more careful management of what the police wished for me to see through the presence of a chaperone, giving brief responses to interview questions and the denial of access to reports, at least initially. I therefore perceived there to be a greater sense of guardedness when conducting research on the police in Ireland relative to the other jurisdictions in the research, which made the research there more challenging. Yet, without such empirical research, gardaí will continue to be unaccustomed to researchers asking questions about what they do. Although research of this kind is growing (e.g. Charman and Corcoran, 2015; Ilan, 2018; Manning, 2012; Marsh, 2019; O’Brien-Olinger, 2016), the Garda Síochána were under-researched relative to police organisations in the other jurisdictions of interest at the time of the research.

Getting your hands dirty

The issue of participation largely came up in relation to when/whether to intervene or ‘whistle-blow’ after observing the mistreatment of suspects. Such incidents can be upsetting not just because of the distress caused to suspects but also because one can feel complicit, morally compromised and lacking in integrity if one does not act (Diphoorn, 2012; Hornberger, 2017; Souhami, 2020; Westmarland, 2001). However, whistle-blowing may ultimately compromise both formal and informal access to research participants. Resolving these ethical dilemmas in situ are a personal matter, in which ethical guidelines, rules and regulations may only have limited relevance. In the case of Westmarland (2001), she saw herself as more of an observer of police violence and as someone documenting events, rather than as a referee waiting to blow the whistle.

I adopted a similar stance in my research. One case particularly brought these issues to the surface. In the Australian city in the research (AUSPO2), a female suspect was left topless in a Perspex cubicle in full view of an all-male custody team for at least 20 minutes, where she fluctuated between being angry and distressed. A male officer described her as ‘just another topless Sheila’. This incident – along with others where officers aired racially prejudiced views (as described above) – left me uncertain of my role as a researcher, in particular, how far such forms of participation should be taken. For example, should I have discussed the incident with managers not just my chaperone? This question seemed particularly pertinent, as I felt that if staff behaved in these ways when researchers were openly observing them, how might they behave when I was not there? In the end, in part to avoid a sense of complicity with the police, I settled for making a detailed note of the incident in my field notes. I took some comfort in the fact that I would be able to give a voice to a female suspect, albeit in academic publications, whose distress would have gone otherwise unnoticed, while also maintaining the anonymity of police participants. This also meant that the individuals and the organisation would not be brought into disrepute, thereby also keeping access open for myself and for other researchers in the future. However, the difficulty of this decision cannot be under-estimated, given the fine line that has to be trodden between whistle-blowing, with the possibility of serious negative consequences for the police and for the researcher, and tacitly condoning, if not colluding with the police (see also Skinns et al., 2016, for further discussion of the ethics of this). Moreover, as Hornberger (2017) notes, no matter how much one converts this sense of complicity with the police into critical writing, it can be hard to erase. However, this is all part of the process by which researchers negotiate and balance their own interests and research agendas with those of research participants, whose views and attitudes may be inconsistent with those of the researcher (Milner, 2007).

Getting through it

The incident described above, along with others, such as the first time I was aware of a (female) detainee deliberately headbutting the wall owing to her distress at being detained overnight until court the next morning (ENG2PO5) and which still evokes a vivid memory, highlight the emotive nature of doing police research, and the visceral reactions it creates, including anger, frustration, distress, apprehension, fear, loneliness, awkwardness, shame and embarrassment. Ten years on, I have begun to appreciate that, rather than only being something to be endured or ‘got through’, or something which is peripheral to the research process, these emotions are integral to it. This mirrors the current trend in social research to see emotions as a mode of participation and inquiry, which can be harnessed to, but also influential on all aspects of the research process (Diphoorn, 2012; Souhami, 2020).

In hindsight, this can be seen in different ways in the present research. During participant observation I was keenly aware of feeling awkward and as if I did not belong in custody, in light of my ‘outside outsider’ status. It was therefore difficult to know what to do in the field in order to look ‘natural, comfortable, engaged and welcoming, while not appearing bored, threatening or judgmental’ (Coffey, 1994: 73). This sense of awkwardness was one reason I felt routinely apprehensive before commencing a shift of observing in custody, especially in the Irish city, where I encountered more wariness, as noted above, and in the American city, where I visited the largest number of police stations, each time dealing with a new set of participants, a different layout and concerns about access:

I routinely feel apprehensive about doing this kind of observational research. This is partly because of the demands on me to be permanently interested and immune to social cues that people do not want me there. As fieldwork approaches, I am filled with a sense of apprehension and foreboding. What will it be like? Where will I spend my time? Will the detectives mind me going upstairs and speaking to them? Will I get a chance to speak to suspects? Will people be on their best behaviour? Will I be able to break-down some barriers? Will I be safe? Will I see things that upset me or depress me or make me want to cry? These are the many questions that I ask of myself as the day approaches for the start of the field-work. (AMEFD)

In most cases while in the field, I contained these feelings, focusing instead on the goals of the project and the demands of data collection. I aimed to ‘get through it’ by collecting as much data as I could, no matter how limited. In only one instance (IREPO6, described above) did I consider leaving the custody block, without completing the data collection.

However, in private, through conversations with family, friends and colleagues, at home and at the institutions I was affiliated during fieldwork, or with the chaperones in the research, I explored the meaning of these feelings. I found these conversations an invaluable way of reflecting on and therefore consolidating some of what I had seen. I often kept my notepad by my side to jot down ideas as often the conversations brought back memories of events or interactions that I had failed to record at the time. Sometimes these conversations were heart-felt. The distress and anger I felt about the treatment of the woman in the ‘topless Sheila’ incident and the private conversations I had about her were critical to how I understood, interpreted and wrote about this incident. Her powerlessness in this situation and, to an extent, mine too, drove my sense of obligation to give her a voice, while also illustrating the all-encompassing nature of police power in police custody and its gendered effects (Skinns, 2019: 174–176). In essence, this shows that police researchers’ emotional reactions to fieldwork help sharpen the senses, consolidate and fine-tune the data collected, and finesse analysis and interpretation.

Conclusion

In sum, even with a police chaperone present, police custody researchers face challenges in getting into police custody facilities, which are regarded as places that ‘civilians’ and indeed other kinds of ‘outsiders’ should not enter. They also experience challenges in getting on, again connected to their ‘outside outsider’ status, in my case as a female, English, Cambridge-educated academic. This status could sometimes help (e.g. being seen as a harmless woman enabled informal access). However, it could also be a hindrance, prompting questions, for example, about whose side I was on, and encouraging guardedness and a reluctance to participate in the research. As for getting your hands dirty, dilemmas about participation related largely to ‘whistle-blowing’ which had to be resolved in situ on a personal basis. Finally, police custody research must sometimes be ‘got through’ due to its visceral nature. Yet, these emotions can also be harnessed, shaping and informing data collection and interpretation.

These findings have implications for how we understand the situating of researchers in the field and the factors influencing researcher reflexivity. These include not just the influence of a researcher’s social origins but also the institutional dynamics of the police and the cultural norms of the academy and the politics of police research (Armstrong et al., 2017; Lumsden, 2017; Souhami, 2020). In terms of social origins, in the case of my research, my gender, race, but also my nationality (both my Englishness and my accent) all influenced the research. The latter may be particularly important in comparative research settings. As Jardine (2019) notes, a researcher’s nationality may have ‘baggage’, which can hamper access. For example, had she been American not Australian, she wonders whether this would have hindered access for her research in Vietnam. Given the relatively recent colonial history of Ireland and ‘The Troubles’ that followed, my Englishness may have amounted to a similar kind of baggage, as alluded to by one of the interviewees.

In terms of the institutional dynamics of the police and of police custody, these are beyond the control of researchers but still have a bearing on the research process. For example, the police may hide the nature of their work. Indeed, this may be particularly so given the territorial nature of police custody. In some senses, like other areas of police work, police custody is more visible as a result of the introduction of CCTV (Newburn and Hayman, 2002: 153; Skinns, 2011: 82–84), but access to police custody or the footage of it is still restricted and thus limited. Combined with my positionality, as an ‘outside outsider’, particularly in the cities in Ireland, the United States and Australia, getting in and on were therefore an inevitable challenge.

In contemporary times, reflexivity in police research must also take into account the culture of academia and politics of knowledge production. At the time of the research, there was less of a focus on research with the police. However, since then, there have been significant shifts towards police–academic collaboration, combining academics’ desire for impact and police organisations’ drive for ‘what works’ evidence, in which certain quantitative methods are valued (Davies, 2016; Lumsden, 2017). By contrast, the present studies were exploratory and qualitative but also theoretical in nature. They were also predominantly research studies on not with the police, and I had made no explicit plans for these projects to impact on stakeholders, such as in my research grant applications, this being an era before ‘Pathways to Impact’ statements. ⁷ Had I attempted to conduct these studies, even 5 years later – as I found with my subsequent police custody study, which received Economic and Social Research Council-funding in 2012 – there would have been a greater expectation from participants and also funders that the research would have been conducted in a more collaborative fashion with the police in order to deliver meaningful changes in the real world. This suggests that the culture of academia and the politics of knowledge production are highly relevant to what research is done and acknowledging it should be a part of police researchers’ reflexivity.

Given these shifting politics of academia and knowledge production, retrospectively, how might the value of the research be assessed? Like many of the ethnographic studies of the 1960s and 1970s which were examples of research on not with the police, the present study can be seen as paving the way for further research which may fit better with the current climate of academia and police studies. However, this treats academic knowledge as something instrumental and as something from which there must be some kind of measurable gain whether for the police or the public. Particularly in nascent fields like comparative police custody research, it is enough to understand the nature and effect of police custody or to observe, ⁸ document and explain the lived experiences of those who work in or are detained therein.

Although conducting research on the police may go against the grain of the current received wisdom in police studies, and may also result in a study that is hard to conduct, given the likely challenges of access, it is a perspective that should continue to be valued. Indeed, in the current climate of concern about racialised police practices, research on the police may gain greater currency again. Regardless, research on the police should be seen as part of a ‘diverse’ and ‘plural’ police research agenda, which Bowling et al. (2019) encourage (p. 14). This enables police scholars to continue to grapple with critical issues which are foundational of the field and which are harder to examine in research focused to a greater extent on impact and on collaborative relationships with the police. It enables researchers to continue to examine the police role in reproducing social order, structures of power, inequalities and injustice, and to grapple not with ‘what works’, rather with ‘what matters’ (Punch, 2015: 15). In some cases, including this one, this is more than enough.