Access denied: Research on sex in prison and the subjugation of ‘deviant knowledge’

Abstract

Despite the fundamental necessity of gaining gatekeeper approval for prisons fieldwork, researchers rarely publicly acknowledge and analyse their failures to secure access. Drawing upon the Foucauldian-inspired literatures on the production and policing of new criminological knowledge, this article presents as a case study the Sex in Prison research project, instigated by the Howard League for Penal Reform, and for which permission to interview serving prisoners was refused. This denial of access, it is argued, resulted from a politically motivated attempt to prevent the acquisition of knowledge perceived as likely to be unfavourable to and critical of prison authorities, emanating from ‘deviant’ informants, using ‘deviant’ methods and obtained on behalf of ‘deviant’ penal reformers. The troubling implications for prisons researchers are explored. The article concludes by highlighting the potential of formerly imprisoned people, as participants in research, to circumvent the limits to knowledge imposed by prison authorities.

Keywords

Access ex-prisoners power/knowledge prisons research qualitative research

Introduction

Access to specified research sites and informants is an essential pre-requisite for the successful completion of much empirical social research. If access is unforthcoming, the research, at least as originally designed, may have to be abandoned: no access may mean the non-production of data and in turn, the non-generation of publishable findings. Yet, despite the centrality of access to the research endeavour, access remains relatively unproblematized in the criminological and socio-legal literature, beyond describing the increasingly bureaucratic procedures for applying for, and the various stratagems for securing, admittance (Mopas and Turnbull, 2011; Trulson et al., 2004). Such discussions nevertheless presume that the researcher was eventually permitted to ‘get in’ to her desired site and then ‘get on’ with her research.

Public admissions of failure to gain access are much rarer. While journal editors will not unreasonably decline to publish research about which there is genuinely nothing to report, the researcher herself may fear that any account which deviates from ‘the chronological lie’ (Cohen and Taylor, 1972: 32) and methodological misrepresentation of research conceived, nurtured and delivered with clinical deftness, undermines her reputation for professional competency. Scholarly attention needs to be expended, however, on the challenging issues and dilemmas that can variously arise during the research process of epistemology and methodology, politics and priorities, academic freedom, state transparency and public accountability (Mopas and Turnbull, 2011; Presdee and Walters, 1998).

The purpose of this article is to contribute to this discussion as it concerns (denial of) access to public sector prisons by independent researchers, defined here as researchers neither employed by, nor contracted on a time-limited project to, government agencies. While its focus is England and Wales, aspects of the analysis will resonate with the international research community.¹ Concretely, it offers as a case study the Sex in Prison primary research project, instigated by the charity, the Howard League for Penal Reform. Permission to interview serving prisoners was refused by the National Offender Management Service’s National Research Committee, allegedly following the intervention of its political overlord, the Secretary of State for Justice (Dunt, 2014a; Green, 2014). This denial of access, it is argued, resulted from a politically motivated attempt to prevent the acquisition of knowledge perceived as likely to be unwelcome to and critical of prison authorities, and obtained through qualitative research with prisoners, at the behest of campaigning penal reformers. In documenting these events, the article draws upon the intersecting literatures on the creation and legitimization of a societal ‘regime of truth’ (Foucault, 1980), the regulation of ‘deviant knowledge’ (Walters, 2003) and attendant marginalization or silencing of critical scholarship in the cause of institutional protectionism and organizational risk management (Hannah-Moffat, 2011; Watson, 2015), to show how particular epistemological and methodological paradigms come to be preferred and non-conforming research proposals, rejected.

The article first outlines the application process, the aims of the intended research and the efforts made to interview serving prisoners. The stated reasons for refusal are then critically reinterpreted, and their troubling implications for prisons researchers explored. Given how informants were subsequently recruited, the article also highlights, in conclusion, the potential of formerly imprisoned people, as contributors to research, to circumvent the limits to knowledge imposed by prison authorities.

Research Access to Prisons

By definition, secure custodial establishments are not freely accessible institutions to which unqualified rights to enter exist. For researchers, access to prisons and the people who live and work there is dependent upon the approval of ‘gatekeepers’: individuals with the authority to grant or withhold access. Historically, most researchers were government or prison employees. Independent researchers could petition for admission informally through personal contacts, sometimes aided – and this still applies – by their affiliation with influential academic colleagues (or, for research students, doctoral supervisors) (King and Liebling, 2008; Trulson et al., 2004). Following the significant expansion in scope and scale of prisons research internationally in recent decades, and corresponding increase in applications, however, two discernible trends have emerged: access to prisons in western jurisdictions has become progressively formalized and more tightly regulated (Mopas and Turnbull, 2011; Trulson et al., 2004); and penologists have experienced reduced support for their research, especially when involving ethnographic fieldwork and/or addressing sociologically informed issues (Jewkes and Wright, 2016; Wacquant, 2002). In combination, research access, while never particularly easy, has become considerably more difficult to secure.

In England and Wales, a standardized application form, comprising mandatory questions across multiple sections, was introduced in 2004. Concomitantly, and as part of the development of a strategic framework for psychological services in public sector prisons, regional and prison-based psychologists were assigned a central role in the scrutiny and approval of proposals (Crighton and Towl, 2008); the consequences of which for qualitative researchers this article will subsequently explore. The would-be researcher currently applies through the National Research Committee (NRC) of Her Majesty’s Prison and Probation Service (HMPPS), formerly known as the National Offender Management Service (NOMS).² In addition to the predictable questions about the research’s rationale, hypotheses or objectives, methodology, ethics, plans for dissemination and the researcher’s qualifications, the applicant must also explain how the research will ‘be of significant benefit to NOMS policy/business’ (Prison Service Instruction [PSI] 22/2014: 3.12)³ and specify the most relevant NOMS priority to which the research relates.

Excepting access to a single establishment (for which the governor-in-charge, typically in consultation with her psychology manager, decides), the submitted application is assessed by the NRC with ‘input from business and analytical consultees … [and] regional lead psychologists(s)’ (PSI 22/2014: 2.2). Constructive changes to the project may then be recommended, and gratefully accepted by the researcher. Acceptance of, or at least acquiescence to, less welcome recalibrations or restrictions, however, may also be necessary, if the research is to proceed. Even when permission is granted by the NRC in principle, however, the researcher may find that unforeseen internal obstacles further complicate or frustrate the grant of access to a particular prison in practice. The governor-in-charge may eventually decide, for example, that the research cannot be conducted in her establishment at this time because of elevated levels of prison indiscipline, staff shortages or imminent changes in key personnel, existing ongoing research or more generic institutional ‘micropolitics’ and ‘local tensions’ (King and Liebling, 2008: 436). If the application is rejected, by either local or national decision makers, no avenue of appeal exists.⁴

The power of gatekeepers to permit, limit or deny access is thus monopolistic and absolute (Cohen and Taylor, 1977). Clearly, there are institutional costs to prisons research and the benefits, especially to front-line staff and prisoners, may appear at best nebulous and at worst negligible. Access accordingly appears to be contingent upon the proposed research’s perceived quality and utility, balanced against the actual or potential interruption to the prison’s routine; the additional imposition this places upon institutional resources, including the staff who must facilitate the research; and the duty of care every establishment has to defend its prisoners from unwarranted and unethical intrusions or investigatory techniques.

Yet, in deciding whether (and to what extent) to approve access, correctional gatekeepers, as civil servants, are necessarily cognizant of, bound to and protective of the interests and imperatives of their political paymasters. Political considerations accordingly mould the framing and scoping of criminological research agendas (Hannah-Moffat, 2011), so that ‘the questions asked and the answers reached within criminology have always been [and remain] subservient to, if not determined by, power’ (Hillyard et al., 2004: 383). Elucidating this dialectical relationship between what is permitted to be known and what is known was a unifying theme of the polymathic scholarship of Michel Foucault. His expression ‘power/knowledge’ (Foucault, 1977, 1980) denotes that power is realized by creating and regulating the production of knowledge and knowledge reinforces and develops power. This circularity arises because the exercise of power ‘produces reality; it produces domains of objects and rituals of truth’ (Foucault, 1977: 194) and this ‘reality’ enables the acceptance, then consolidation, of ‘the status of those who are charged with saying what counts as true’ (Foucault, 1980: 131). Accordingly, research which is labelled and legitimatized through dissemination as knowledge, and the ideologically framed discourses and practices of knowing they contain, becomes institutionalized and valorized as the appropriately ‘scientific’, ‘true’ way to understand and report the data or phenomena under study. Conversely, areas of enquiry, methods and findings which are not supported are ‘explicitly disqualified’ and disparaged as inferior, ‘subjugated knowledges’ (Foucault, 1980: 82). From this perspective, it can be anticipated that those who authorize, and those who seek to conduct, research may differ upon which topics merit investigation, the methodology to be employed and, ultimately, the determination of ‘what constitutes proper research’ (Cohen and Taylor, 1977: 77). This power/knowledge axis, and the sensitizing framework of critical analysis it supports, is now illustrated through a case study of access-denied prisons research.

The Sex in Prison Project

The biggest problem is that the people in NOMS who are saying sex in prison is not a problem, are not there, and no one asks the people who are there. (Lewis, SiP interviewee)

Through its advice and assistance to prisoners, the Howard League for Penal Reform became aware of the almost complete absence of research and policy on sex in British prisons. In June 2012, the League established an independent Commission on Sex in Prison, chaired by a retired governor-in-charge, with the aim of formulating evidence-based recommendations to improve prisoners’ sexual health, psychological well-being and safety. Simultaneously, NOMS was ‘sounded out’ about research access. Initial indications were encouraging: the Secretary of State for Justice, Kenneth Clarke, was supportive and the Chief Executive of NOMS detailed the Head of Women and Equalities to assist in securing approval. The League consequently recruited an academic consultant (the author) to conduct exploratory primary research, hereafter referred to as the Sex in Prison (SiP) project. Clarke was replaced as Justice Secretary by his Conservative colleague Christopher Grayling in September 2012, shortly before Commission members began to take evidence from statutory and voluntary stakeholders. The Head of Women and Equalities was also redeployed before substantive access negotiations began.

An international literature review on prison sex was conducted, and the merits of various methods employed by researchers noted (inter alia, Banbury, 2004; Gaes and Goldberg, 2004; Hensley, 2002; Richters et al., 2012). This review informed the proposal verbally presented to NOMS staff at a research meeting in February 2013. The SiP project sought to undertake short but intensive fieldwork in six prisons, to include the high security estate, a ‘closed’ facility for women, establishments for sexual offenders and ‘local’ prisons. By ‘being there’, however transiently, the researcher hoped to reassure potential participants about the project’s sincere, non-sensationalist purpose and the sensitive elicitation, secure retention and appropriate use of their confidential data. Prisoners would be asked about their sexual identities and personal or vicarious experiences and knowledge of consensual and non-consensual sexual activity, or non-activity, in prison, and could choose between a semi-structured interview or paper questionnaire with open-ended and non-categorical questions. While self-completion questionnaires afford more anonymity than face-to-face interviews, given the limited literacy skills of many prisoners, and the lack of privacy for cell-sharers in divulging personal information, it was anticipated that most would prefer an interview. Additionally, at each establishment visited, institutional data indicative of sexual activity⁵ would be collected, and interviews with prison managers with relevant responsibilities⁶ conducted.

This proposal was rejected. The NRC, the Commission was now advised, would need to be persuaded that the benefits of exploratory SiP research significantly outweighed ‘the high risks and scope for mischief’ and that the project would align with the Justice Secretary’s ‘priorities’ and have regard to his ‘sensitivities’. Specifically, the NRC would be concerned about interviews with an ‘unrepresentative’ sample of prisoners who ‘you won’t know are telling the truth’, including ‘manipulative’ sex offenders. Instead, it was suggested that the researcher visit four establishments to distribute a questionnaire, using closed-ended questions gathering (Likert scale) ordinal data, to a 10% ‘statistically representative’ random sample of prisoners, excluding sexual offenders. Although the questionnaire could conclude by inviting respondents to an interview, a chiefly quantitative research design was needed in order to compute ‘frequencies’ and provide ‘an estimate of prevalence’ and so place any findings from interviews ‘in context’. NOMS would support access to institutional data and interviews with managers ‘for balance’ (meeting notes, February 2013).

Further consultation by email followed, while Commissioners lobbied senior figures within NOMS and the Prisons Minister, who reported to Grayling, to seek their support. Expert advice was sought on probability sampling methods, which identified significant problems with the mooted design, and from Her Majesty’s Inspectorate of Prisons, whose experience of administering pre-inspection surveys highlighted the logistical difficulties for a lone researcher operating within a narrow timeframe. In combination, the Commission concluded that prisoners’ perspectives could best be elicited through interviewing those who self-identified as willing to discuss prison sex. The resulting findings could be satisfactorily placed ‘in context’ and supply the desired ‘estimate of prevalence’ by citing quantitative and mixed methods studies, with larger cohorts, which indicate that consensual and coerced sexual activity is uncommon in British prisons (Banbury, 2004; Edgar et al., 2003; Power et al., 1991; Strang et al., 1998); and through comparison with establishment data. One methodological conundrum remained: if prisoner participants could not be sought through the visible and facilitating presence of a fieldworker, how could they be reached?

The research application submitted in May 2013 proposed a startlingly straightforward solution: advertising for interviewees in the newspaper for British prisoners, Inside Time. Multiple copies of this not-for-profit publication are freely available in every prison and, it was hoped, readers of its website might alert imprisoned family or friends to the research. If necessary, for the manageability of the project and representativeness of prisoner interviewees, purposive sampling would be deployed, until data saturation was achieved.

Commissioners remained hopeful that access would be granted and, as the months passed, were sustained by intermittent reassurances that the proposal was ‘under consideration’ and ‘with Ministers’. Sixteen weeks after submitting the application, however, the research was vetoed. The NRC’s reasons⁷ were:

the potential benefits to NOMS … were limited … [and] did not sufficiently link to NOMS priorities … concerns about the robustness of the methodology, notably the self-selection of prisoners and the likelihood of a small, non-representative sample … [and NOMS] had recently approved the first stage of a research project looking at the response by prisons to adult male prisoners’ reports of sexual offences …

These criticisms will be dissected below but one notes here that the third reason relates to part-time PhD research. The student is analysing Incident Reporting System data on the nature of and institutional responses to reported prisoner-on-prisoner sexual victimization in men’s prisons. This research is thus distinguishable, in scope and methodology, from the SiP project, and never had the possibility of reporting its findings within the lifetime of the Commission. A predominantly quantitative study which was not scheduled to complete until 2019⁸ in part justified rejecting a predominantly qualitative study which had to be concluded by 2015.

The Commission decided not to re-apply but to advertise with criminal justice charities for interviewees among ex-prisoners, no longer under post-release probation supervision and hence for whom permission to interview was unnecessary. Subsequently, during the summer of 2014, the author interviewed 26 formerly imprisoned people, including 11 men convicted of sexual offences. Key findings from the Commission and primary research were widely reported by news media, and the latter, published in a peer reviewed journal (for which, please see Stevens, 2017).

The Subjugation of ‘Deviant Knowledge’

Yeah, I read they wouldn’t let you in. I wonder why! [tone of heavy sarcasm] They’ve got nothing to hide, have they? (Liam, SiP interviewee)

The decision to grant or withhold research access reflects and hence reveals the politically determined values and preoccupations, and the epistemological and methodological preferences, of correctional gatekeepers. For it is gatekeepers who, by reference to ‘NOMS business priorities’ (PSI 22/2014: 3.4) determine the scope of knowledge to be sought and through assessment of the application, the methods and sources afforded the authority and legitimacy to produce that knowledge. The logical corollary of establishing parameters of inclusion is the facilitation of exclusion. The creation of an approval process which, this article contends, systematically envisages and promotes specific forms of intellectual inquiry and knowledge construction simultaneously invites and enables the less positive, if not entirely negative, appraisal of other areas of, and avenues to, penological research.

In particular, research proposals which question, challenge or appear incongruent with correctional agendas and orthodoxies, and are deemed likely to generate findings or recommendations which are ideologically incompatible with, embarrassing to, unsupportive of, or in other ways politically inconvenient for, policymakers, politicians and prison services, risk designation not only as ‘irrelevant or useless’ but as ‘threatening’ or ‘risky’ (Hannah-Moffat, 2011: 446, 445). It is these proposals, with their potential to produce ‘deviant knowledge’ – defined by Walters (2003: 2) as ‘criminological knowledges that are unfavourable to, and/or critical of, agents of power’ – which conservative correctional bureaucracies are most concerned to ‘manage out’. Where this is not possible and the research proceeds, the resulting findings are liable to selective redefinition, interpretation and utilization; neutralization or, following Foucault (1980), subjugation; and in extremis, censorship and suppression (e.g. Cohen and Taylor, 1972;⁹ Hope and Walters, 2008; Presdee and Walters, 1998; Yeager, 2008).

The construction of ‘deviancy’ which facilitated the repudiation of the SiP project, and its wider relevance to prisons research, is now examined.

‘Deviant’ methods

The most overt difficulty encountered in achieving support for the research was its reliance upon ‘a small, non-representative sample’. Sample bias is a fundamental limitation of non-probability sampling. It is unproblematic, however, when the purpose is not to determine prevalence or incidence but to pursue participant-led exploratory research, most especially with ‘hidden’ and stigmatized populations whose members are not easily identifiable (Meyer and Wilson, 2009). The project needed to hear from people who had something instructive to say about sexual activity in prison, and thus if (purposive) sampling were to be undertaken at all, it needed to be from within this conceptually defined sampling frame and not from the total prison population.

At the research meeting, however, NOMS staff proposed randomly sampling 10% of prisoners, from four prisons, to include a London ‘local’, holding 1300 men. If replicated across three other similarly populous establishments, this would have yielded a sampling frame of 520 prisoners. Unsolicited, self-administered questionnaires on prison sex frequently produce very high refusal rates (Gaes and Goldberg, 2004), but if one had achieved a (typical) 25% response rate (Struckman-Johnson and Struckman-Johnson, 2000; cf. Barth, 2012), one might have garnered 130 respondents. Her Majesty’s Inspectorate of Prisons informed the Commission that from its confidential surveys of prisoners (completed by up to a third of the population, randomly selected, at every prison inspected), 1% of prisoners report sexual abuse from other prisoners or staff, rising to 2 to 3% for people who identify as disabled or Muslim (Howard League, 2012). Academic research further suggests 2 to 3% of male prisoners have consensual sex (Strang et al., 1998; Taylor et al., 2013). If these percentages had been reflected in the probable respondent sample, the researcher might therefore have heard from eight men with direct experience of consensual or coerced sex.

Such a ‘small’ (for quantitative research) sample size and response rate could result in selection bias and differential attrition, in which factors and characteristics which motivate participation or non-participation render the results unrepresentative, even though the sampling frame was originally randomly constructed (Bachman and Schutt, 2003). Following the research design advocated therefore would still not have produced the ‘statistically representative’ sample NOMS sought, from which generalizable conclusions about the extent of sexual activity in British prisons could be reached and qualitative findings reliably placed ‘in context’. To be able to provide ‘an estimate of prevalence’ would have required a sampling frame and research endeavour far beyond the Howard League’s financial, temporal and human resources and an exploratory study’s modest scope.

Nonetheless, NOMS’s preference for a randomly sampled survey of prisoners, using closed-ended questions, was unequivocally signalled at the research meeting. Current guidance (PSI 22/2014: 3.12) also states, ‘[r]esearch which focuses on outcomes through quantitative evidence of impacts is encouraged’. Pursuing a qualitative design for the SiP project, however methodologically appropriate for its research aims, therefore always risked being labelled as ‘illegitimate’ research which should be refused. Two, interlinking and mutually supportive, observations follow.

First, this partiality for quantitative methods reflects the disciplinary background of psychologists, whose endorsement of the scientific rigour and worth of proposals hastens access approval. Psychologists’ research training, anchored in positivist epistemology and objectivist ontology, is overwhelmingly focused on the functional achievement and application of quantitative and technical competence involving the experimental and hypothetico-deductive measurement and (re-)testing of ‘countable’ phenomena (Crighton and Towl, 2008). Rarely, however, do psychologists specialize in ethnographic research or other qualitative and person-centred approaches, which rely upon social interaction to pique interest and engender trust among prospective participants and employ inductive reasoning, to arrive at their ‘results’.

This learned ‘favourable bias’ (Martel, 2004: 162) towards quantitatively oriented epistemologies is evident in the application form.¹⁰ The ‘proposed methodology’ section obliges the researcher to select from a list of seven options, five of which involve ‘review’, ‘assessment’ or ‘evaluation’. The eighth, catch-all ‘other’, and corresponding free-text field, is the only opportunity by which to propose qualitative research such as interviews, focus groups or participant observation. Asking about ‘sampling and sample sizes’ anticipates quantification, while the requirement to provide precise dates for the data collection period assumes considerable certainty about the ease and speed with which the research can be conducted. Such confidence may well be possible for researchers working out of psychology departments and undertaking statistical analyses of prisoner databases to explore, for example, variables correlated with prison misconduct or changes in psychometric assessment scores pre- and post-treatment. For prison fieldworkers, however, seeking to secure the co-operation and participation in interviews of human beings, within an intrinsically low-trust and sometimes chaotic environment, such certainties are more elusive. Moreover, an honest, cautious answer may attract concern about the security risks that sustained interaction with prisoners is presumed to incur (Liebling, 2014).

The assessment of research proposals has therefore been devised so as to construct and replicate the taken-for-granted privileging of quantitative data collection and analysis, and equally, to render less visible and to problematize qualitative research and its accompanying interpretivisit epistemology and constructionist ontology. This helps to explain the previously noted decline in research internationally which requires approval for ethnographic, exploratory, sociological studies of individual prisons. It also illuminates why for claims to knowledge to be accepted as methodologically ‘robust’ and ‘sophisticated scientifically’ (Beyens et al., 2015: 73), the assessors may require the supplementation of qualitative research with the generation of numerical data (Martel, 2004). In Foucauldian terms, the discursive practices of the application form operate to expedite the approval of favoured, familiar research approaches and to inhibit, if not ultimately exclude, the ‘other’. Difficulties arise for qualitative researchers when their methods conflict with gatekeepers’ expectations and consequently, are subjugated: disqualified as inadequate to answering, by themselves, the research questions and to providing an evidence-base because ‘the techniques and procedures accorded value in the acquisition of truth’ (Foucault, 1980: 131) are quantitative.

Second, NOMS’s emphasis upon ‘outcomes’ and ‘impacts’ indicates governmental preference for research which is primarily administrative, technocratic and policy-related – not critical, sociological and theoretical (Walters, 2003). This orientation towards instrumentally applicable knowledge has become progressively more embedded in recent decades, due to the growth of in-house research and emergence of ‘market-led criminology’, whereby some academics tender for and accept government-funded research contracts. Either way, the research process is subject to close political scrutiny and overwhelmingly focuses upon assessing immediate problems and finding short-term solutions (Hood, 1987; Tarling, 2011; Walters, 2003). At this juncture, it is important to acknowledge that much in-house and commissioned research is of the highest methodological standard and has added immeasurably to a descriptive understanding of crime and governmental responses to crime. The point is, such research is descriptive.¹¹ The ‘customer’ confines the brief and research questions to atheoretical fact finding, information gathering and outcome-focused evaluations of performance and policy in which, as discussed, positivist, quantitative data, ‘measuring the easily measurable’ (Morgan and Hough, 2007: 54), are preferred; and for which ‘service providers’ are neither invited nor required to interrogate, let alone discredit, the underpinning political premises and ideological legitimacy of that policy (Hillyard et al., 2004; Presdee and Walters, 1998).

Circumscribed expectations of research, determined and demarcated by political imperatives, are to be expected of government and government-sponsored researchers. Of concern to other academic producers and consumers of knowledge, however, is that this cognitive schema additionally shapes assessors’ expectations of what credible research should look like and achieve. One sees this in the declaration that proposals ‘must be aimed at informing and improving policy formulation, analysis and delivery, clearly allied to strategic priorities’ (PSI 22/2014: 3.2). Note the word ‘must’. The apparent intention of this commandment¹² is to curb the criminological imagination of independent researchers by penning them into, and pinning them down to, the same evaluations of extant or proposed government policy and operational protocols and practice it requires of non-independent researchers.

Ironically, then, the Commission’s ambition was to enhance the evidential basis for policy and practice. To link the production of knowledge to increasing ‘the effectiveness of operational policy/delivery’ (PSI 22/2014: 3.1) therefore appeared to be not only achievable but entirely consistent with the project’s objectives and ‘NOMS priorities’.¹³ Indeed, the primary research discovered, among other things, that ‘the condom policy’, formulated in the 1990s to enable sexually active prisoners to obtain condoms, is inconsistently applied and interpreted across the prison estate, resulting in multiple inequalities and uncertainties of provision (Stevens, 2017). What is supposed to happen does not necessarily happen, and this prompted the Commission to make recommendations about what should happen, in order to restore the policy’s effectiveness. Here, then, was an example of ‘useful’ research.

Yet, the relationship between research and its translation into policy and practice is rarely linear, in part because the sensible-sounding term ‘evidence-based policy’ (and its twin, ‘evidence-led’) conceals some problematic assumptions. Many criminologists welcome engagement with policy concerns and hope their research will, in conjunction with cognate scholarship, incrementally influence, even if only indirectly, the substance and direction of criminal justice policy by cumulatively building a body of knowledge and evidence (Ericson, 2003). In relation to penal policy-making, such evidence profits from empirical analyses of the prison experience, not least to establish any disconnect between official policy and operational delivery. The championing of evidence-based policy, however, presupposes that policymakers want evidence – all evidence, regardless of whether it supports current government policy or otherwise accords with their political and epistemological standpoint – and that they will subsequently act upon that evidence in ways which respect and reinforce, rather than ‘rupture’, this knowledge-policy ideal (Goldson, 2010). Equally fundamentally, it implies that knowledge consumers agree upon what constitutes ‘good’ evidence and will recognize such evidence as evidence when presented with it. The fate of some politically ‘unhelpful’ criminological research demonstrates the fallacy of such assumptions (Glaser, 1965; Hope and Walters, 2008).

That the SiP project was neither intended nor able to provide ‘quantitative evidence’ of ‘outcomes’ ensured its envisaged benefits were deemed ‘limited’ and contributions to operational policies, insufficient. Worse still, the research’s manifest potential to inflict substantial reputational damage (‘mischief’) and generate (‘sensitive’) political controversies about the need for strategic action, characterized it as liable to create ‘deviant knowledge’: unfavourable to and critical of NOMS and by extension, the Justice Secretary. These are not the research ‘impacts’ which correctional gatekeepers value.

‘Deviant’ informants, ‘deviant’ penal reformers

Semi-structured interviews with prisoners were unappealing to NOMS for reasons other than their inability to produce quantitative data. Becker’s (1967) lauded hierarchy of socially recognized credibility, and its implications for the differential ability to make claims to epistemological authority, help explain why.

People who live in prison know about life in prison in qualitatively distinctive ways to people who work or research in prisons, because ‘serving time’ is irreducibly different – more visceral, more extreme – to ‘spending time’ in prisons (Earle, 2014: 432). Consequently, prisoners acquire a specific form of penological expertise about the ‘doing’ and surviving, or not, of a term of imprisonment. This is not to assert, of course, that only prisoners are qualified to pronounce and arbitrate upon definitive penal ‘truths’: prisoners’ understandings are as partial and situated, temporally, spatially and culturally, as anyone else’s. Their first-hand testimonies of incarceration, however, need to be sought out, utilized and built upon if researchers are to challenge the ‘established status order’ and its ‘knowledge of truth’ (Becker, 1967: 242), and thereby co-construct a more comprehensive, more nuanced, ‘penology from below’ (Sim, 2008: 190; see also Brown, 2008).

Despite – or because of – prisoners’ superior experiential knowledge, their evidence remains routinely absent from public discourse about, and other forms of official narratives of, the prison (Liebling, 2014; Scraton et al., 1991). Exiled and incapacitated, prisoners’ legal rights to contact with the outside world are restricted, and their ability to be heard by penologists is reliant upon gatekeepers not only granting research access but also to approving forms of research by which prisoners’ perspectives can meaningfully be captured. Yet even when such permission is forthcoming, and as illustrated by the concerns raised by NOMS staff about the capacity of prisoners to manipulate and lie, contentious research findings based on interviews with (formerly) imprisoned offenders are sometimes disparaged on the grounds that they are inherently deceitful individuals and hence unreliable informants (Martel, 2004; Sim, 2008); notwithstanding that, if this were to be systematically true, much of the empirical edifice upon which penological knowledge rests would crumble. Nevertheless, by pathologizing prisoners in this way, their fitness to dispute superordinates’ definitions of ‘the way things really are’ (Becker, 1967: 241) is questioned; their accounts, if heard at all, more easily discredited; and in consequence, their claims to status as expert witnesses, undermined.

Testimony from any interviewee demands from the researcher some appraisal of plausibility and internal consistency. Most criminologists (e.g. Liebling, 2001) would concur, though, that there is no evidence that offenders, uniquely among any other category of informant, volunteer to be interviewed and self-report experiences with the intention of deliberately misleading the researcher, nor that they are any more likely to dissemble than, for example, politicians. Still, such hierarchical ordering was evident in the different ways in which NOMS and the Commission approached the reasoning for preferring interviews with, to surveys of, prisoners and to conducting staff interviews.

The Commission considered that, as a conduit to giving ‘voice’ to prisoners, semi-structured interviews would be more effective. These allow the interviewer to respond spontaneously and intuitively to answers needing further clarification with supplementary prompts, and afford interviewees similar latitude to answer as they wish, by elaborating in detail but also by speaking to or prioritizing other issues of tangential concern. Such flexibility and autonomy therefore offer a more empowering (and more enjoyable) research experience for prisoners than ticking boxes on a questionnaire, whose terminology, theoretical orientation and range of responses have been pre-established by the researcher and cannot yield to the redefinitions of the researched.

For gatekeepers, however, flexibility equates with unpredictability: one cannot anticipate what additional – and unapproved – questions may be asked and what confessional disclosures may emerge when interviewer and interviewee co-operate in a creative search for knowledge construction and meaning-making (Holstein and Gubrium, 2004). Such uncertainty invites the very potential for ‘risk’ which the defensive scrutiny of research proposals is intended to obviate. That NOMS would agree to interviews for prison managers is therefore explicable as its attempt to re-exert some control: governors might more reliably be entrusted to remain ‘on message’. Moreover, prison managers – unlike prisoners – were not required to complete a questionnaire as a mandatory precursor to interview. The potential effect of this distinction was to position staff, given the greater time and freedom of expression interviews confer, as the more authoritative source of knowledge about prison sex. For while it was important to the Commission to ascertain and appreciate managers’ differently situated understandings of the reasons for consensual and coercive sex and the concomitant responsibilities and ethical dilemmas this created for them, it did not think that doing so could provide ‘balance’: a word which implies that prisoners’ accounts require correction or at least, corroboration. Governors cannot comment upon sexual activities and cultures which are purposely hidden from the managerial gaze by both prisoners and officers (Stevens, 2017). Yet clearly prisoners, with their hierarchically inferior ‘naïve knowledges’ (Foucault, 1980: 82), were not to be automatically afforded the same opportunity or equal power as prison staff to make ‘true’ claims for knowledge, and findings arising from prisoners’ testimonies were to be treated more cautiously than those sourced from managers. Thus, they were to be intentionally marginalized and differentially divested of ‘the right to be heard’ (Becker, 1967: 241). In sum, prisoners’ spoiled identity spoils their credibility.

These, then, were the stated reasons for refusal. There existed, however, an additional potent factor which although delicately alluded to in the research meeting, was never, and could never be, fully articulated by NOMS: the Justice Secretary’s interest in ensuring that the primary research did not proceed. The SiP research threatened to produce and disseminate ‘deviant knowledge’. The instigator of this unbidden and unwelcome knowledge was the UK’s oldest penal reform charity, the Howard League. While watchdogs are supposed to bark, it is arguable that from Grayling’s perspective, the League barked too loudly, too often: from a once comfortable relationship, in gentler penological times,¹⁴ it had become a vociferous and trenchant critic of the penal system and those responsible for it. From the League’s perspective, there was much to criticize, vociferously and trenchantly. Still, Grayling’s ‘sensitivities’ extended beyond penal reformers. The prospect that the project might uncover evidence of prisoners enjoying consensual sexual activity, for example, neither cohered with his concern that prisons should not be perceived as ‘comfortable’ environments nor his desire to be identified as a ‘harsh disciplinarian’ dispensing ‘tough justice’, ideologically opposed to the reformist agenda of his immediate predecessor. In this he succeeded: Grayling soon acquired a reputation similar to that of the famously hardline Conservative Minister from the 1990s, Michael Howard, who, it is instructive to recall, was always more concerned with mediated public opinion of the ‘toughness’ of his penal policy than in effecting positive change through criminological research (Burnett and Maruna, 2004; Lewis, 1997). Grayling’s reported response to the Commission’s work was accordingly to demand a ‘crackdown’ on ‘cosy, domestic relationships’ between cell-sharing prisoners because ‘prisoners aren’t going to have sex on my watch’ (cited in Dunt, 2014a; Mason and Winnett, 2012); this, rather than learning about prisoners’ sexual health needs, was his ‘priority’. After access was denied, it was further alleged that Grayling had directly intervened to prevent the SiP research in part because of his ‘personal animosity towards the pressure group’ (Dunt, 2014b; Green, 2014). If influential connections can help researchers to secure access, then logically the reverse applies: with whom one is affiliated can make access more difficult. Or as one Guardian¹⁵ columnist (Fogg, 2014) colourfully summarized:

The explanations offered for the obstruction range from the asinine to the bizarre. Grayling, we are told, is in something of a huff with ‘leftwing pressure groups’ such as the Howard League … Blocking this important research may be an act of petty retaliation.

If Grayling thought that denying access would extinguish the SiP project, he underestimated the Commission and in particular, the indomitable resolve of the League’s Chief Executive, Frances Crook. He also failed to anticipate the exasperation of formerly imprisoned people, many of whom, in volunteering for interview, categorically denounced the refusal to permit research and cited their ‘frustration’ and ‘irritation’ as motivating factors in their participation. Clearly, (ex-)prisoners understand experientially better than most the symbiotic nature of the power/knowledge relationship. To deny research access to prisoners may represent an attempt to foreclose investigation into and debate about normally invisible aspects of prison life, and thereby to silence and subjugate offenders’ carceral realities and perpetuate the structural exclusion and disenfranchisement of the imprisoned. But just as power/knowledge intertwines, so power and resistance co-exist: ‘there are no relations of power without resistances’ (Foucault, 1980: 142). Every power relationship is dynamic, with possibilities for active dissent and struggle, not merely passive acceptance. The SiP research afforded participants the opportunity to question the accuracy of official, sanitized discourses about sexual activity behind bars. Liberated, physically and psychologically, from the prison, interviewees were empowered to voice oppositional knowledges and counter narratives which, in some important respects, subverted the regime of truth structuring acceptable understandings about prison sex, and exposed troubling gaps between the rhetoric of policy and the realities of operational practice. Testimonies ‘from below’ are thus essential, not least for the alternative perspectives they can cast on the view ‘from above’.

Concluding Remarks

How dare NOMS say no! Okay, so it’s not the nicest of subjects, and I’m sure they’d rather not know about it, but that doesn’t give them the right to try and stop it. And they haven’t, have they, in the end? Here we are [having an interview]. (Miles, SiP interviewee)

Politics represented a dangerous riptide flowing through, and against, the Sex in Prison project from the outset. It seemed navigable while the (relatively) liberal-minded Clarke was in office but with Grayling’s arrival, its gravitational pull quickened and ultimately submerged the intended research. This case study serves to document one episode indicative of Grayling’s inglorious tenure, but more substantively, by deconstructing assessors’ methodological preferences, epistemological and ontological bases, and politically dictated priorities, this article has shown how ‘risky’ research proposals can be effectively neutralized; their ultimate rejection, facilitated; and the production of ‘unwanted’ knowledge, intentionally obstructed. The provision of ‘guidance’ sounds supportive but closer examination reveals that these non-negotiable instructions stake out the boundaries within which the prisons researcher is permitted to frame her theoretical and empirical enquiries. Since access to prison(er)s is conditional upon approval of her research design, institutional gatekeepers can therefore directly determine, and effectively curtail, the possible forms, domains and trajectories of knowledge which are created (Hannah-Moffat, 2011; Martel, 2004). This power to define ‘what talk is allowed in prisons’ (Cohen and Taylor, 1977: 76), when, where, how, with whom, and in what circumstances, threatens to shape and sustain an insular, short-termist, and theoretically impoverished penological research agenda which fails to produce knowledge about which it may be politically expedient, or simply more comfortable, not to know.

Following their difficulties in disseminating unpalatable contract research findings, Presdee and Walters (1998) concluded that there are only two fail-safe strategies: to do criminological research in which either state agencies are not interested, or alternatively, they regard as valuable or harmless. Prisoners’ experiences of sex were not considered interesting or valuable to Secretary of State Grayling, but neither was the quest to research them regarded as harmless. The initial enthusiasm for the SiP research, under Clarke, did not evaporate, and access approval was not withheld, solely because of the project’s requirement for a qualitative research design with an in-built response bias towards interviewing people with actual or vicarious knowledge of prison sex, rather than a random and quantitatively inconsequential survey. The denial of access resulted from a politically motivated attempt to prevent the acquisition of ‘deviant knowledge’, perceived to emanate from ‘deviant’ informants, using ‘deviant’ methods, and obtained on behalf of ‘deviant’ penal reformers. Researching sex in prison, however, should have been too important for politics. The possible health consequences of unprotected sex for prisoners and their sexual partners in prison and after release are too serious, and the certain trauma of sexual assault, too appalling, to be occluded by and submerged within political point-scoring.

To the wider difficulties discussed here, there are no easy solutions. Gatekeepers cannot be compelled to grant research access or to appreciate the merits of extended, exploratory and person-centred fieldwork. Prisons researchers, however, are neither entirely without influence nor without alternative options. In encouraging the Prison Service to be(come) ‘literate and enlightened users of, commissioners of, and partners in meaningful research’ (Liebling et al., 2017: 1007), senior scholars in particular may yet be able to persuade gatekeepers that there is no justification for the automatic preferment of quantitative methods or assumption that qualitative approaches must always be yoked, however awkwardly, to quantitative data collection and analysis in order to be perceived as rigorous and credible. They should, in short, contest what Martel (2004) terms the tendencies towards paradigmatic ‘bipolarization’ and ‘banalization’, in which the deification of quantitative methods seemingly demands the denigration of qualitative approaches. Less ambitiously, the combination of tenacity, patience and resourcefulness can sometimes eventually enable the successful completion of research by, for example, relocating the project to a more amenable jurisdiction or correctional service provider (King and Liebling, 2008); utilizing freedom of information protocols (Yeager, 2006); or analysing publicly available policy or institutional documents (Hannah-Moffat, 2011) and prisoners’ autobiographies (Morgan, 1999) and letters (Vannier, 2020). And while the involvement of formerly imprisoned people has frequently enriched studies of release and resettlement, re-offending and desistance, and parole and community supervision (notably, Appleton, 2010; Maruna, 2001; Petersilia, 2003), this article has drawn attention to the potential for formerly imprisoned people, as compelling contributors to research about (rather than conducted in) prisons, to resist and transgress the limits to knowledge imposed by prison authorities. For as the SiP project exemplifies, by interviewing ex-prisoners, suppressed voices can be heard, the lived realities of a hidden world exposed and the ‘insurrection’ of their ‘subjugated knowledges’ (Foucault, 1980: 81) enabled, when research access to prisons is denied.

Footnotes

Acknowledgements

I gratefully acknowledge the constructive and supportive comments of Neil Chakraborti, Yvonne Jewkes, Nuala McGrath and two anonymous reviewers on earlier drafts of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

ORCID iD

Alisa Stevens

Author biography

Alisa Stevens is a senior lecturer in Criminology at Cardiff University. A penologist, she is particularly interested in the experience of imprisonment for, and the potential for rehabilitation and desistance among, people serving long sentences for violent or sexually violent offences.

References

Appleton

(2010) Life after Life Imprisonment. Oxford: Oxford University Press.

Bachman

Schutt

(2003) The Practice of Research in Criminology and Criminal Justice. Los Angeles, CA: SAGE.

Banbury

(2004) Coercive sexual behaviour in British prisons as reported by adult ex-prisoners. The Howard Journal of Criminal Justice 43(2): 113–130.

Barth

(2012) Relationships and sexuality of imprisoned men in the German penal system: A survey of inmates in a Berlin prison. International Journal of Law and Psychiatry 35: 153–158.

Becker

(1967) Whose side are we on? Social Problems 14(3): 239–247.

Beyens

Kennes

Snacken

, et al. (2015) The craft of doing qualitative research in prisons. International Journal for Crime, Justice and Social Democracy 4(1): 66–78.

Blom-Cooper

(1976) The role of voluntary organisations as pressure groups in penal reform. In: Walker

(ed.) Penal Policy-Making in England: Papers Presented to the Cropwood Round-Table Conference, December 1976. Cambridge: University of Cambridge, 1–10.

Brown

(2008) Giving voice: The prisoner and discursive citizenship. In: Anthony

Cunneen

(eds) The Critical Criminology Companion. Sydney: Federation Press, 228–239.

Burnett

Maruna

(2004) So ‘prison works’, does it? The criminal careers of 130 men released from prison under Home Secretary, Michael Howard. The Howard Journal of Criminal Justice 43(4): 390–404.

10.

Cohen

Taylor

(1972) Psychological Survival: The Experience of Long-Term Imprisonment. Harmondsworth: Penguin.

11.

Cohen

Taylor

(1977) Talking about prison blues. In: Bell

Newby

(eds) Doing Sociological Research. London: Allen and Unwin, 67–86.

12.

Crighton

Towl

(2008) Psychology in Prisons. Oxford: BPS Blackwell.

13.

Dunt

(2014a) Grayling accused of obstructing prison sex inquiry. Politics.co.uk. Available at: http://www.politics.co.uk/news/2014/04/23/grayling-accused-of-obstructing-prison-sex-inquiry.

14.

Dunt

(2014b) Khan: I’ll be open to prison sex inquiry if we win election. Politics.co.uk. Available at: http://www.politics.co.uk/news/2014/05/13/khan-i-ll-be-open-to-prison-sex-inquiry-if-we-win-election.

15.

Earle

(2014) Insider and out: Making sense of a prison experience and a research experience. Qualitative Inquiry 20(4): 429–438.

16.

Edgar

O’Donnell

Martin

(2003) Prison Violence: The Dynamics of Conflict, Fear and Power. Cullompton: Willan.

17.

Ericson

(2003) The culture and power of criminological research. In: Zedner

Ashworth

(eds) The Criminological Foundations of Penal Policy: Essays in Honour of Roger Hood. Oxford: Oxford University Press, 31–78.

18.

Fogg

(2014) Why is Chris Grayling blocking research into rape in prison? Guardian, 6 May. Available at: https://www.theguardian.com/commentisfree/2014/may/06/chris-grayling-blocking-research-rape-prison-sexual-assault-jail.

19.

Foucault

(1977) Discipline and Punish: The Birth of the Prison. London: Allen Lane.

20.

Foucault

(1980) Power/Knowledge: Selected Interviews and Other Writings 1972–1977. Harlow: Harvester Press.

21.

Gaes

Goldberg

(2004) Prison rape: A critical review of the literature. Washington, DC: National Institute of Justice.

22.

Gentleman

(2015) UK prisons won’t let journalists in – what do they have to hide? Guardian, 10 February. Available at: https://www.theguardian.com/commentisfree/2015/feb/10/uk-prisons-journalists-what-hide.

23.

Glaser

(1965) Correctional research: An elusive paradise. Journal of Research in Crime and Delinquency 2(1): 1–11.

24.

Goldson

(2010) The sleep of (criminological) reason: Knowledge–policy rupture and New Labour’s youth justice legacy. Criminology and Criminal Justice 10(1): 155–178.

25.

Green

(2014) Chris Grayling blocks inquiry into sexual assaults inside jails. Independent on Sunday, 4 May. Available at: http://www.independent.co.uk/news/uk/crime/chris-grayling-blocks-inquiry-into-sexual-assaults-inside-jails-9321406.html.

26.

Hannah-Moffat

(2011) Criminological cliques: Narrowing dialogues, institutional protectionism, and the next generation. In: Bosworth

Hoyle

(eds) What Is Criminology? Oxford: Oxford University Press, 440–455.

27.

Hensley

(ed.) (2002) Prison Sex: Practice and Policy. Boulder, CO: Lynne Rienner.

28.

Hillyard

Sim

Tombs

, et al. (2004) Leaving a ‘stain upon the silence’: Contemporary criminology and the politics of dissent. British Journal of Criminology 44(3): 369–390.

29.

Holstein

Gubrium

(2004) The active interview. In: Silverman

(ed.) Qualitative Research: Theory, Method and Practice. London: SAGE, 140–161.

30.

Hood

(1987) Some reflections on the role of criminology in public policy. Criminal Law Review 527–538.

31.

Hope

Walters

(2008) Critical thinking about the uses of research. London: Centre for Crime and Justice Studies.

32.

Howard League (2012) Seminar with key statutory stakeholders, 20 November 2012. London: Howard League.

33.

Jewkes

Wright

(2016) Researching the prison. In: Jewkes

Bennett

Crewe

(eds) Handbook on Prisons. Abingdon: Routledge, 659–676.

34.

King

Liebling

(2008) Doing research in prisons. In: King

Wincup

(eds) Doing Research on Crime and Justice. Oxford: Oxford University Press, 431–459.

35.

Lewis

(1997) Hidden Agendas: Politics, Law and Disorder. London: Hamish Hamilton.

36.

Liebling

(2001) Whose side are we on? Theory, practice and allegiances in prisons research. British Journal of Criminology 41(3): 472–484.

37.

Liebling

(2014) Postscript: Integrity and emotion in prisons research. Qualitative Inquiry 20(4): 481–486.

38.

Liebling

McNeill

Schmidt

(2017) Criminological engagements. In: Liebling

Maruna

McAra

(eds) The Oxford Handbook of Criminology. Oxford: University of Oxford, 990–1009.

39.

Loader

(2006) Fall of the ‘Platonic Guardians’: Liberalism, criminology and political responses to crime in England and Wales. British Journal of Criminology 46(4): 561–586.

40.

Martel

(2004) Policing criminological knowledge: The hazards of qualitative research on women in prison. Theoretical Criminology 8(2): 157–189.

41.

Maruna

(2001) Making Good: How Ex-Convicts Reform and Rebuild their Lives. Washington, DC: American Psychological Association.

42.

Mason

Winnett

(2012) Crackdown on sex in prisons to stop ‘cosy’ relationships between cell mates. Telegraph, 23 November. Available at: http://www.telegraph.co.uk/news/politics/9699532/Crackdown-on-sex-in-prisons-to-stop-cosy-relationships-between-cell-mates.html.

43.

Meyer

Wilson

(2009) Sampling lesbian, gay, and bisexual populations. Journal of Counselling Psychology 56(1): 23–31.

44.

Mopas

Turnbull

(2011) Negotiating a way in: A special collection of essays on accessing information and socio-legal research. Canadian Journal of Law and Society 26(3): 585–590.

45.

Morgan

Hough

(2007) The politics of criminological research. In: King

Wincup

(eds) Doing Research on Crime and Justice. Oxford: Oxford University Press, 45–72.

46.

Morgan

(1999) Prison lives: Critical issues in reading prisoner autobiography. The Howard Journal of Criminal Justice 38(3): 328–340.

47.

Petersilia

(2003) When Prisoners Come Home: Parole and Prisoner Reentry. Oxford: Oxford University Press.

48.

Power

Markova

Rowlands

, et al. (1991) Sexual behaviour in Scottish prisons. British Medical Journal 33(6791): 1507–1508.

49.

Presdee

Walters

(1998) The perils and politics of criminological research and the threat to academic freedom. Current Issues in Criminal Justice 10(2): 156–167.

50.

Richters

Butler

Schneider

, et al. (2012) Consensual sex between men and sexual violence in Australian prisons. Archives of Sexual Behavior 41: 517–524.

51.

Scraton

Sim

Skidmore

(1991) Prisons under Protest. Milton Keynes: Open University Press.

52.

Sim

(2008) An inconvenient criminological truth: Pain, punishment and prison officers. In: Bennett

Crewe

Wahidin

(eds) Understanding Prison Staff. Cullompton: Willan, 187–209.

53.

Stevens

(2017) Sexual activity in British men’s prisons: A culture of denial. British Journal of Criminology 57(6): 1379–1397.

54.

Strang

Heuston

Gossop

, et al. (1998) HIV/AIDS risk behaviour among adult male prisoners. London: Home Office Research Findings No. 82.

55.

Struckman-Johnson

(2000) Sexual coercion rates in seven Midwestern prison facilities for men. The Prison Journal 80(4): 379–390.

56.

Tarling

(2011) Relations between government researchers and academics. The Howard Journal of Criminal Justice 50(3): 307–313.

57.

Taylor

Munro

Allen

, et al. (2013) Low incidence of Hepatitis C virus among prisoners in Scotland. Addiction 108: 1296–1304.

58.

Trulson

Marquart

Mullings

(2004) Breaking in: Gaining entry to prisons and other hard-to-access criminal justice organizations. Journal of Criminal Justice Education 15(2): 451–478.

59.

Vannier

(2020) The power of the pen: Prisoners’ letters to explore extreme imprisonment. Criminology and Criminal Justice 20(3): 249–267.

60.

Wacquant

(2002) The curious eclipse of prison ethnography in the age of mass incarceration. Ethnography 3(4): 371–397.

61.

Walters

(2003) Deviant Knowledge: Criminology, Politics and Policy. Cullompton: Willan.

62.

Watson

(2015) Research access barriers as reputational risk management: A case study of censorship in corrections. Canadian Journal of Criminology and Criminal Justice 57(3): 330–362.

63.

Yeager

(2006) The Freedom of Information Act as a methodological tool: Suing the government for data. Canadian Journal of Criminology and Criminal Justice 48(4): 499–521.

64.

Yeager

(2008) Getting the usual treatment: Research censorship and the dangerous offender. Contemporary Justice Review 11(4): 413–425.