‘A Poor Prospect Indeed’: The State’s Disavowal of Child Abuse Victims in Youth Custody, 1960–1990

Abstract

Child abuse in youth custody in England and Wales is receiving an unprecedented degree of official attention. Historic allegations of abuse by staff in custodial institutions which held children are now being heard by the courts and by the Independent Inquiry into Child Sexual Abuse (IICSA), and some criminal trials have resulted in convictions. A persistent question prompted by these investigations is that of why the victims of custodial child abuse were for so long denied recognition as such, or any form of redress. Drawing on original documentary research, this article aims to explain why and how state authorities in England and Wales failed to recognise the victimisation of children held in penal institutions between 1960 and 1990, and argues that this failure constitutes a disavowal of the state’s responsibility. We show that the victims of custodial child abuse were the victims of state crimes by omission, because the state failed to recognise or to uphold a duty of care. We argue further that this was possible because the occupational cultures and custodial practices of penal institutions failed to recognise the structural and agentic vulnerabilities of children. Adult staff were granted enormous discretionary power which entitled them to act (and to define their actions) without effective constraint. These findings, we suggest, have implications for how custodial institutions for children should think about the kinds of abuse which are manifest today.

1. Introduction: The Research Context

Since 2014, the Independent Inquiry into Child Sexual Abuse (IICSA) has been ‘consider[ing] the extent to which State and non-State institutions [in England and Wales] have failed in their duty of care to protect children from sexual abuse and exploitation’ [1]. Set up in the wake of widespread public and media coverage of historic child sexual abuse (CSA) allegations, one of its thirteen sub-investigations concerns children in custodial institutions. IICSA hearings on recent allegations were held in 2018; further ones relating to historic abuse may follow.

Many of the most notorious non-recent allegations relate to HM Detention Centre Medomsley. Described by prison inspectors in 1977 as a place which ‘has never hit the headlines’, where ‘nothing of any import ever occurs’ and which ‘is unlikely to cause any problems’ [2], Medomsley closed without fanfare in 1988. In the early 2000s, however, it did hit the headlines: two former officers, Neville Husband and Leslie Johnson, were convicted of sexually assaulting boys and young men detained there during the 1970s and 1980s. Substantial evidence then emerged that their abuses had been systematic and committed over the long term. Moreover, the Prison Service had missed opportunities to prevent them: in 1969, at HM Borstal Portland (where he then worked), Husband was caught at work in possession of child abuse images. He was charged by police with possessing obscene materials, but the charges were dropped, whereupon he transferred to Medomsley, grooming and abusing vulnerable boys with apparent impunity for well over a decade. It has also been alleged that some of Husband’s colleagues at Medomsley were aware of his activities but took no action [3,4,5,6,7,8,9].

Following widespread publicity and speculation, a police investigation was set up in 2013 to investigate further allegations relating to Medomsley. More than 1600 former detainees came forward to allege they had been the victims of physical and/or sexual abuse there, and in March 2019, while this paper was under review, five former officers were convicted of assaults, wounding and misconduct in public office during the 1970s and 1980s [10,11]. These were physical abuses; it is worth noting that other former officers were acquitted, and no allegations of sexual abuse in the new trials were proven. Even so, police elsewhere have investigated historic allegations relating to 25 more penal institutions which held children and young people [12]. New prosecutions may follow the 2019 Medomsley verdicts, but not in all cases: many allegations have proven impossible to link to identifiable perpetrators, or have been not been pursued because those accused have since died. The number and scale of allegations both suggest systematically abusive institutions rather than isolated instances of abuse. Out of this context comes the central question we address in this article: why were the victims and survivors of abuse in youth custody denied official recognition (or any form of redress) for so long?

Our analysis focuses on penal institutions which held children between 1960 and 1990. We select this period because it is now known that abuses took place in these institutions during this period without being officially recognised as such and also because this was before present-day safeguarding and child protection regulations came into existence. As such, the period belongs to a different era in official thinking: about children, about offenders, about abuse, and about custody. Our analysis offers insights into how and why custodial child abuse in this era was not recognised or responded to as such. We base it on documents examined during a historical research study [13] commissioned by HM Prison and Probation Service (HMPPS), which we describe further in Section 3.

1.1. Key Terms

As the meaning and use of the concepts ‘child’ and ‘abuse’ vary over time and by context, we begin by clarifying our use of these terms. We have adopted a definition ‘child abuse’ from the NSPCC: ‘any action by another person—adult or child—that causes significant harm to a child. It can be physical, sexual or emotional, but can just as often be about a lack of love, care and attention’ [14]. IICSA considers only sexual abuse, but our choice of this broader definition is deliberate: it allows us to bring into the analysis a wide range of staff conduct which could have been experienced as significantly harmful, but not all of which was sexual abuse. Such conduct might include peer abuse, or physical abuse by staff. Both are relevant to the phenomenon of sexual abuse because of their contribution to an overall culture in which care and welfare take a back seat to other considerations. IICSA itself sees the link between forms of abuse as relevant: it has concluded that a ‘culture of violence’ can ‘increase the risk of child sexual abuse’ [15] (p. 102).

This leaves the question of why we treat child abuse in custody as a ‘crime of the state’. This follows straightforwardly from the fact that compensation has been paid by the state: first to the original Medomsley victims [3]; and now, we understand, further payments are expected to follow the more recent trials [10,11]. Such payments acknowledge the state’s responsibility by omission for offences of commission by individuals who failed to fulfil a duty of care. Nevertheless, it is likely that these trials and compensation payments will not draw a line under the full range of custodial abuses perpetrated at the time, since many cases are unlikely ever to receive official recognition. The literature on the victimology of state crimes [16] has been particularly helpful in explaining this troubling possibility. We return to it throughout this paper, starting in Section 2.1.

We use the term ‘child’ to refer to anyone under the age of 18. This is the current statutory definition in UK law, and also the age threshold between the criminal and youth justice systems [17] (s. 142A). However, the terms ‘child’ and ‘children’ were not used in this way during the period we cover. Our findings refer to several different kinds of penal institutions, all of which held some under-18s, though many also held 18 to 21-year-olds, and a few also held over-21s. Greater precision in our analytical focus on under-18s is difficult, because contemporary usages were fluid and the administrative boundaries between adults and children flexible. We explore this point further in Section 2.2.

1.2. Article Outline

The discussion is structured as follows. In Section 2, we frame the analysis theoretically. Section 2.1 describes Christie’s concept of ‘the ideal victim’, Kauzlarich et al.’s proposals about the victimology of crimes of the state [18,19], and Cohen’s seminal work [20,21] on the politics of denial. All three were helpful in interpreting our data. We then present contextual materials drawn mainly from secondary literatures, describing first how the concept of ‘childhood’ and the regulation of sexual consent have changed during the period we cover (Section 2.2); then describing how cultural perceptions of ‘child abuse’ and risk to children have changed during the same period (Section 2.3); and then introducing shifts in institutional landscape during this period (Section 2.4). Together, these materials establish the broad cultural and institutional contexts in which custodial child abuse occurred and was understood. In Section 2, we occasionally use primary material from the archives, but the broad approach is contextual and shaped by secondary literature.

In Section 3, we give an account of the research process itself: how the work was conducted, the materials upon which it is based, and how we interpret them as evidence. In particular, we describe how our research methods evolved in response to the gaps and silences in the archive and reflect on the particular challenges of using official documents to examine child abuse, by definition a hidden and covert phenomenon.

Section 4 contains the bulk of the analysis of archival sources. We interpret these sources with reference to Christie’s and Kauzlarich et al.’s theorisations, identifying factors which combined to make abuse unlikely to be recognised as such. These were: the entitlement claimed by the state and its agents to carry out harmful practices (Section 4.1); the specific ways these were legitimised (and victims were blamed) (Section 4.2); and the ineffectual nature of the safeguards in place at the time, which failed to check these legitimising, blaming tendencies (Section 4.3). We show throughout why past safeguards against abuse failed in their purpose, showing in particular that it would have been extremely difficult for children in custody to seek redress.

In Section 5, we conclude by drawing these materials together and reflect on how further historical research might offer insight into the prevention of present-day custodial child abuse. We argue that the insights to be gained from archival research are worthwhile and difficult to obtain using other methods, and that the present moment is an excellent time to be gaining them.

2. Theoretical and Contextual Starting Points

2.1. Sociological Perspectives on Victimhood

Crime victims became increasingly prominent in professional, political and public discourse on crime and criminal justice from the mid- to late-20th century onwards [22,23,24]. However, the status of victimhood is not distributed equally to all who are victimised; it can be diluted or overwhelmed by other characteristics. The ‘ideal’ victim, according to Christie [19], is weak, and engaged in a respectable activity in a place where he or she cannot be blamed for being. Accordingly, the ‘ideal offender’ is ‘big and bad’ and unknown to the victim. A perceptual overlap between victims and offenders is problematic, tarnishing the ideal victim’s image: ‘victims that merge with offenders make for bad victims’ [19] (p. 25). The victim’s social standing is also relevant for the recognition of victimhood: victims must be powerful enough to make their case known and to successfully claim the status of an ideal victim. They must also not be opposed by counter-voices so strong that they cannot be heard [19] (p. 21). Recognition of victimhood therefore suggests a certain skill on behalf of the victim in public self-representation, or an alliance with others so skilled.

Kauzlarich et al.’s theoretical propositions on the victimology of state crime [18] offer further helpful insights for our analysis. The victims of state crime are often victims of ‘the attempt to achieve organizational, bureaucratic, or institutional goals’; they are not simply victimised by ‘a few people engaging in immoral, unethical and/or illegal behaviour’ [18] (p. 188). Further, the state generally fails to understand or recognise the nature, extent and harmfulness of its policies, and if it does, this will usually be done in a way that asserts the state’s ‘entitlement’ to inflict harm, for example under the legitimising rubric of some higher moral aim. The phenomenon of how states deny and disavow the harms they perpetrate has also been typologised by Cohen [20,21], whose analysis draws our theoretical attention to the distribution of power, especially the aforementioned power of (self-)representation: as Christie suggests, the power to define what is ‘true’ is vitally important in claiming victimhood. Kauzlarich et al. argue that the victims of state crime are generally among the least powerful actors in society. Their powerlessness is particularly pronounced when it comes to using official processes (such as complaints procedures) to represent their experiences as harmful. As a consequence, they are more likely to be blamed for their suffering, because they will already have (or can easily be allocated) a master status (e.g., ‘offender’) which marginalises their claim to any rights they might possess. Thus for Kauzlarich et al., a particular challenge faced by victims of the state is that they generally have to rely on ‘the victimiser, an associated institution or civil social movements for redress’ [18] (p. 188).

This theoretical literature provided three focal points for our analysis: first, the extent to which children in prison were likely to see themselves (and/or to be seen by others) as victims of institutional abuse; second, the extent to which managers and staff of penal institutions recognised themselves (and/or were recognised by others) as potential offenders and victimisers; and third, the extent to which children who recognised their victimhood were able to convince others and obtain redress.

2.2. Changes in the Definition of ‘Children’ and in the Regulation of Their Sexual Activity

Since around 1800, the concept of childhood in England and Wales has commonly been associated with innocence, dependency, and vulnerability, and distinguished from the knowledge, agency, independence, and responsibility associated with adulthood. But the boundary between childhood and adulthood has also been vague, context-dependent, and mobile. It has been associated with different transitions: biological (e.g., puberty); legal (e.g., sexual consent); ritual and cultural (e.g., school graduation), social (e.g., increasing independence) and economic (e.g., entry to the workforce) [25,26,27,28]. Culturally, therefore, who is recognised as a child (and hence as innocent, dependent or vulnerable) is strongly context-dependent. During the period we cover (1960–c.1990), these broader shifts were institutionalised by various changes to the law. The age associated with adult roles, rights and responsibilities in England and Wales generally moved upwards [13] (pp. 6–8), converging upon 18 as the legal age of majority, a convergence formalised in law in 1989 [29]. Previously, however, there were two statutory categories of non-adults: a ‘child’ was someone aged under 14, and a ‘young person’ someone aged 14 to 17 [30]. Implicitly, this distinction also institutionalised an interim category of agency and independence, lying somewhere between childhood and adulthood.

It is clear from archival records that this ambiguity was deepened by the formal and informal classifications created by different branches of government. For example, in records produced by the Home Office Children’s Department, children are commonly described as such (although most such records cover the Approved School system, which was not within the scope of our research). But the Home Office Prison Department used its own formal classification: ‘juvenile’ and ‘young offenders’ for those aged 14–17 and 17–21, respectively. References to ‘children’ in its records are rare, but terms such as ‘inmates’, ‘trainees’, ‘offenders’ or ‘young prisoners’ are common, as are gendered colloquialisms such as ‘girls’ and ‘lads’. The inconsistency is striking partly because the same children often passed from one department’s jurisdiction into the other’s jurisdiction. It is clear that prisons did not stick rigidly to any particular system of classification, and that chronological age was secondary to other cultural categories under certain circumstances, for example if a child was considered ‘disruptive’ or ‘dangerous’. In short, whether a person under 18 in Prison Service custody was recognised as a ’child’ at all was at least partly contingent on administrative classification.

The statutory regulation of sexual consent also changed during this period, with implications for the official understanding of CSA in custody. From 1885 until 2001, the law defined adult sexual agency (and thus also sexual victimisation) principally in relation to gender. Age was a secondary variable affecting the degree of an offender’s culpability [31]. The (still-current) age of consent for females (16) was set in 1885, with severer penalties for an adult male convicted of sexual offences with a girl aged under 13 than one aged between 13 and 16 [31]. Gender was also influential through prevailing attitudes towards homosexuality. The law did not regulate sex between females at all until 2001, meaning that there was no lesbian age of consent until then [32]. But until 1967, all sexual acts between males were crimes. Afterwards, they were decriminalised with a different age of consent (21) than for heterosexual sex [32]. Only in 2001 was this differentiated regime harmonised at 16, with further amendment to 18 in 2003, for cases where the older party was in a ‘position of trust’ vis-à-vis the younger [33].

In summary, the legal regulation of sexual consent shifted from a preoccupation with specific acts (and with the gender and age of those party to them), towards a more explicit recognition that power imbalances are the central problem in sexual acts between adults and children: from 2001, the age of the parties concerned was the paramount consideration in determining victimhood. Before this, however, it is clear that even where sexual offences against children were successfully prosecuted, official recognition of victimhood was uneven and contingent on wider cultural definitions, whether the victim was male or female [34] (pp. 420, 423). In 1953, for example, a judge sentencing an adult man for buggery nevertheless named his child victims, remarking that by ‘soliciting’ his attention they had been ‘repulsive little pests’ [34] (p. 423). Such blaming also affected female victims, particularly those in adolescence, who could be ascribed a degree of agency and were thus blamed for their moral ‘waywardness’ [34,35,36,37,38,39,40,41,42,43].

It is clear, therefore, that gender norms obscured and reinforced the age-related power imbalances at play in CSA. It is also clear that victim-blaming discouraged disclosure. Disclosed abuse was often trivialised, explained away, or denied altogether [44], reinforcing power disparities and protecting abusers. Even so, it is unambiguously the case that throughout the period we cover, most under-21s engaged in non-consensual sexual relationships with adults could be recognised as the victims of offences defined as such by the criminal law, with the exception of all cases where both parties were female, in which no crime of any sort existed. The fact that prosecutions were rare [31] demonstrates simply that other cultural discourses were highly influential in defining sexual agency.

This material signals four reasons why custodial child abuse might have gone under-recognised. First, it is clear that cultural attitudes in wider society discouraged the disclosure of what is now recognised as CSA. Second, the morally and culturally ambiguous status of ‘young persons’ blurred the agency and independence of those in this category who had sexual relationships with adults. Third, such ambiguity meant that mainstream culture offered a repertoire of moral tropes (for example those of the ‘wayward’ or ‘promiscuous’ adolescent girl ‘seducing’ a helpless adult man) in which the denial of adult responsibility could be framed. Fourth, recognition that children who had had sex with adults were the victims of a crime was most likely where the victim was female and the perpetrator male, less likely where both parties were male, and impossible where both parties were female. (This last point partly explains the lack of historical CSA allegations relating to female institutions: since, until the late 1980s, most custodial staff in these institutions were themselves women and there was simply no legal category of a sexual offence between females, sexual victimisation was a legal impossibility.)

2.3. Child (Sexual) Abuse and the Cultural Construction of Risk

Public and policymakers’ perceptions about the principal sources of abuse risk to children shifted significantly during the period we cover [13] (pp. 9–11). During the 1960s and 1970s, sexual risk was perceived to relate mainly to girls, lying in public space and ‘stranger danger’, especially where the stranger was Black or Asian [38,45,46]. Even in these cases, responsibility was often deflected onto ‘precocious’ young women, who tended to be constructed as ‘risky’ rather than ‘at risk’ [43]. The label ‘child abuse’, first used in the 1960s, initially referred only to the physical abuse of ‘battered’ children [47,48]. The 1970s saw feminist campaigners turn the focus to intrafamilial sexual abuse. This was initially labelled ‘incest’ [49,50,51,52,53,54,55,56], but by the 1980s was increasingly framed using the distinct term ‘child sexual abuse’. By the end of that decade, this term dominated public discourse on child abuse generally [13] (p. 10), but there was also deep unease over how the state should respond to CSA: it still seemed clear that sexual abuse ‘occur[red] primarily within the family’ [57] (p. 20), but a series of scandals in which abuse had been misidentified or overstated (and children had been removed from their parents) suggested that social workers were misappropriating the state’s power to make overweening intrusions on family life, traducing parents’ rights in the process [58].

Also in the 1980s and 1990s, a series of scandals relating to children’s homes, boarding schools and other residential settings [59,60,61] shifted the focus onto abuse in institutions. One effect was to draw attention, for the first time, to the vulnerabilities of male children, especially teenage boys [13] (p. 10). The protection of children ‘living away from home’ [62] became an especially urgent policy concern, since they now appeared subject to distinctive kinds of risk. This had implications for youth custody, which became a major preoccupation among penal reformers during the 1990s [13,62,63,64,65,66]. All of these developments reflected a growing recognition, domestically and internationally, that children were a distinctive category of people possessing distinctive rights which required distinctive legal protections [29,67].

These shifts in how society understood children to be at risk suggest further reasons why child abuse in penal institutions may have gone under-recognised during the period we cover. First, the concepts of ‘child abuse’ and ‘child sexual abuse’ were not defined until the 1980s, and even then referred entirely to phenomena of the domestic sphere. Second, the risks of child abuse in institutions were not recognised at all before the late 1980s, and only began to be perceived in youth custody after this. Third, these two forms of ‘risk-blindness’ were especially likely in relation to boys (who then as now were an overwhelming majority of children in custody). Fourth, the statutory multi-agency child protection duties which now affect all state agencies were unknown before the mid-1980s, and only took their current form in 1989 [68], with some doubt about their applicability to youth prisons until 2002 [69].

2.4. Complexity and Instability in the Youth Custody Landscape

The policies directing life in youth custody changed considerably during the period we cover. We have described these changes (in both the penal and care systems) in greater detail elsewhere [13] (pp. 23–28), but we note here that these changes fit into a long-term pattern, evident since the late 19th century, of oscillation between the competing priorities of child welfare and punishment. In England and Wales, welfare was the primary motivation for establishing dedicated custodial institutions for children, but the punitive aims inherited from adult prisons were never completely abandoned [70]. Over the long term, punitive aims have generally proven ineffective, while rehabilitative aims have proven expensive and difficult to realise, so that custodial policy has cycled between retributivist disillusion and welfarist hope [22,71,72,73]. In particular, new institutional models have repeatedly been established, run into difficulties, and been reformed or abandoned [74].

Table 1. Youth custodial models managed by the Prison Service, 1960–1990.

Name	Existed	Purpose	Successor
Borstals	1902–1983	Held ages 15–21; sentence of indeterminate ‘training’, regimes emphasised rehabilitative training over punitive confinement.	YCCs
Young Prisoner Centres (YPCs)	1954–1983	Held ages 15–21; separate residential units in adult prisons. Initially for ‘difficult’ inmates; sentences of ‘immediate imprisonment’ in YPCs later used by sentencers as deterrent alternative to discredited borstal training. Similar regimes to their ‘host’ prisons.	YCCs
Detention Centres (DCs)	1954–1988	‘Junior’ and ‘senior’ DCs held ages 14–17 and 17–21; punitive/deterrent sentences of ≤6 months the original aim, but significant local variation in regimes.	YOIs
Remand Centres	1961–1999	Dedicated local prisons holding adults and under-21s on remand, in or near large conurbations.	YOIs
Youth Custody Centres (YCCs)	1983–1988	Consolidation of borstal and YPC populations; regimes emphasised secure containment over rehabilitative training.	YOIs
Young Offender Institutions (YOIs)	1988–	Consolidation of YCC and DC populations and regimes; sometimes (as with YPCs) on same sites as adult prisons, but usually with differentiated regimes.	-

summarises how this long-term pattern played out between 1960 and c.1990. The general picture, in the penal system in general and youth custody in particular, is one of steadily declining faith in rehabilitative aims. This was partly under pressure of numbers: until the early 1980s, the number of children in custody steadily increased, while budgetary constraints and industrial strife (especially during the 1970s) limited attempts to respond to their needs [75,76]. In such circumstances, the welfare and experiences of individual children could easily go unnoticed, and across a range of different custodial models, rehabilitative aims were gradually abandoned in favour of punishment and secure containment. This was consistent with broader penal trends [13] (pp. 23–28).

In 1960, the Prison Service managed three institutional models (the borstals, the DCs and the YPCs) specifically intended for under-21s, and also held them in adult institutions according to need. By the late 1970s, Home Office files record that regimes (and populations) in all three had become so indistinguishable that a policy review was required to try and keep up with reality on the ground. Although the aims of borstal, DC and prison sentences remained different in law, the custodial experience in all was effectively the same [77,78]: all three institutions had ‘evolved into institutions for longer and shorter terms of something similar to [adult] imprisonment’ [79] (p. 249). This was reflected in reforms of the 1980s, when the three kinds of custody were amalgamated by stages into YCCs and then YOIs.

In this paper, we highlight Detention Centres (DCs) as a particularly important example of the muddled picture described above. Created in the 1950s as a deliberately and explicitly punitive form of short-term deterrent custody—a ‘short sharp shock’—for 14- to 21-year-olds, their original aims were diluted rapidly by attempts originating from within the Prison Department to introduce more rehabilitative regimes [80] (pp. 6–11). Such attempts introduced considerable confusion. The Warden of HMDC Whatton gave voice to some of this befuddlement in 1967, soon after taking command there: ‘As a newcomer to Detention Centres I soon encountered considerable confusion of thought on the administration of policy. “Discipline” is a word most used by those persons directly concerned with the sentencing of the inmate. And to some extent by the public who tend to substitute the word punishment for discipline. “Training” appears to receive little consideration—as though the two are diametrically opposed and not complementary to each other. The words of a Principal Officer at Whatton sum up the point I am trying to make. He said, “The Courts think one way when they sentence a lad to Detention Centre and the Prison Department another”. It is as though the word discipline has one interpretation—punishment.’

Throughout the 1970s, then, government policy was adjusted to keep up with the training-focused (and borstal-like) regimes prison managers had been trying to administer [81]. Yet DCs never fully realised these rehabilitative aspirations [82], and arguably never could. Archival reports from their Wardens (e.g., [83]) concur with internal and external policy reviews (e.g., [80,81,84]) in suggesting that it was difficult to achieve much with detainees whose needs were often complex and whose sentences lasted for an average of twelve weeks. The simpler rhetoric of a deterrent ‘short sharp shock’ thus retained a rhetorical appeal [76,82] long after managers had stopped trying to deliver it in practice. Indeed, it was regularly invoked throughout the 1960s and 1970s by the political opponents of more ambitious rehabilitative aims. These included politicians, staff unions and magistrates who sought a ‘tougher’ or more ‘military-style’ disciplinary regime. This return to the ‘short sharp shock’ became an explicit policy commitment in the Conservative manifesto for the 1979 general election. Yet research commissioned by the Home Office to evaluate the return to ‘tough’ DC regimes refuted the positive outcomes claimed for them [85], and the DC model was quietly abandoned thereafter.

Confused and unachievable aims were therefore particularly characteristic of DCs. The combination of a readily-invoked punitive rhetoric, scant time for rehabilitative work, vulnerable populations, frequent overcrowding, and the imperative to run a busy and ‘brisk’ regime fostered cynicism and purposelessness: an atmosphere in which, according to one later analysis, ‘failure never matter[ed]’ [82]. Daily life was characterised by bullying, the illegitimate delegation of power by staff to children, and the routine use of violence by officers (and by inmates on their behalf) [86]. Moreover, short sentences limited the time available for solidarity (and therefore effective resistance) to develop among detainees. It is unlikely to be a coincidence that most historic abuse allegations relate to DCs.

The material in this section has a number of implications for the (in)visibility of abuse. First, staff conduct was more likely to be influenced by organisational culture and local contingency than by any official safeguarding policy. Second, to the extent that there was mission drift and resource pressure, institutions for children were likely to be affected by significant internal conflict, deflecting attention from the individual child and expressed through a demand for more punitive treatment. Third, it is especially likely that these dynamics would apply in declining institutional models where rehabilitative credibility was dissipating (i.e., borstals) or absent (i.e., DCs). In short, the recognition of abuse and victimhood was highly contingent upon local cultural factors, and not on a clearly-defined understanding of what kinds of treatment were acceptable.

3. Research Materials and Methods

Objectives for this project were set in the summer of 2017 and research was carried out between November 2017 and January 2018. Our report to HMPPS (the commissioners of the research) [13] was published in 2018, and this paper was written in early 2019. Our original aims were framed by nine orienting research questions, agreed between the research team and HMPPS; these are detailed in the appendices to our report [13].

We began by reviewing existing literatures on institutional child abuse, and academic histories of youth justice and youth custody. These informed our understanding of penal change during the period under consideration and framed our reading of materials we encountered subsequently. We also compiled a list of establishments of interest: those in which historic CSA allegations have since come to light. All held males. To avoid artificially narrowing our search based on what might have been contingent accidents of disclosure, we added the small number of institutions which held females to the list.

We then carried out systematic catalogue searches in two repositories: the Radzinowicz Library in Cambridge, and The National Archives (TNA) in London. The former holds one of the largest collections of criminological material in the world; the latter holds the archival records of central government, including the Home Office Prison Department. We searched catalogues for keywords relating to children, child protection, child safeguarding, complaints, disclosures and the different types of custody; we also used keywords such as ‘sexual misconduct’ and ‘indecency’, which were commonly used before the 1980s to refer to CSA [34,44]; and we searched for the names of establishments ‘of interest’, reviewing all records that appeared to overlap with periods covered by historic allegations.

These searches generated uneven results, largely because of significant (and unquantifiable) gaps in the archive. TNA retains only a sample of the records lodged with it by government departments. By default, these are closed for 30 years after the record’s creation, or 80–100 years if they contain information liable to harm or embarrass named individuals or their families [87]. These rules, and the fact that some of TNA’s holdings have been destroyed over time to create space, mean that (according to one recent estimate) some collections retain as few as 1% of the records originally deposited in them [88] (p. 22). Most of what we found related to the 1960s and 1970s; materials relating to the period between 1980 and 1987 (the latest year available at the time of our research) are sparse, for reasons that are unclear. Also unclear was why several files, including some relating to allegations of staff ‘indecency’, were listed in the TNA catalogue but turned out to be missing, presumed lost.

The matters targeted by our systematic keyword searches were therefore unevenly covered. Others could only be approached through files dealing with system-wide issues (i.e., reviews of policy affecting adult prisons and establishments for children); records on staff disciplinary procedures and prisoner complaints, for example, both fell into this category. Our systematic search strategy therefore evolved pragmatically: we browsed the TNA catalogue for records in the same series as those we identified through systematic searches; we followed up cross-references; and where necessary we reviewed records covering both adult and youth custody, assuming (unless there was reason to do otherwise) that Prison Department policies applied equally to both. Full details of the TNA records we inspected and those we could not are given in our report [13] (pp. 73–5, 83–92).

Alongside archival materials, we quote selectively from recent and contemporary newspaper reports, which despite their selective reporting and contemporary prejudices remain one of the most important non-official sources of evidence on historic child abuse [34,38]. Most recent citations come from coverage of the Medomsley allegations, which to our knowledge are the only historic ones to have been reported in depth, albeit decades after the fact. Where press coverage contains information unavailable elsewhere, we quote it to make more sense of what was available. Older newspaper citations in this paper mostly come from contemporary clippings which we found in archival files. Their inclusion in those files justifies their inclusion here: they were significant in the thinking of officials.

The archival record on child abuse, like that of the popular press, is generally subject to significant ‘gaps, silences and distortions’ [34] (p. 421). If abuses went unnoticed or uninvestigated at the time, then there are, by definition, no documents about them. And if allegations were investigated, surviving files will usually name individuals, and hence will be closed for 80 or 100 (rather than 30) years. Identifying the archival traces of possible abuse therefore often entails reading documents imaginatively, in combination, and against the grain [34]. We did so by asking what the records we reviewed revealed about institutional culture, and what they revealed about the operation of past safeguards. This helped us build an impression of the institutions they described and the people who worked in, lived in, and monitored them. We considered questions such as: what were their cultural mindsets? How did they think about phenomena which are now understood to reveal vulnerability, such as self-harm and suicide? What did they see as the aims and purposes of custody, and did they think these were being realised? Our methods result in a synthesis of different sources. With reference to victimological theory, we present here an interpretive account of past official responses to child abuse in youth custody.

Thus far, we have shown that archival research is by its nature pragmatic (because it is shaped by what evidence is available). One of our co-authors on the original research report (a professional historian) has written with colleagues about the distinctive role that historians can play in addressing the specific issue of historic child abuse [34]. She has also argued elsewhere that historical methods are intrinsically ‘dialogic’ [89] (p. 6). We conclude this section by developing this idea and suggesting that there are at least two different kinds of ‘dialogue’ taking place when the penal past is analysed through the archive. First, the analyst is implicated in direct dialogue with the sources, with their broader understanding of period, place, and of contemporary versions of the phenomenon under investigation affecting how they read the documentary record. What we found significant and interesting in the archive and the library was undoubtedly shaped by our previous professional and research experience, in both adult and youth prisons. This experience influenced how we read and understood the sources, and will be evident in what we say below about the inherent contradictions of penal policy and the complex dynamics of institutional power. Second, we also suggest that the analyst is implicated in another kind of dialogue: they are cast in the role of an interpreter, who initiates and facilitates dialogue between different documents, whether these are classified as primary/archival/‘data’ or secondary/library/‘literature’. We assumed this interpreter’s role whenever we returned to a source that initially seemed insignificant, but which solicited our attention afresh after we read some other source which helped to make sense of it. Here, again, our backgrounds and our experience are significant: there was no necessary link between these documents save for the ones we perceived, although our conclusion relies on the implicit idea that people with comparable backgrounds and experience might perceive many of the same, and many similar, links.

Analytical iterations were therefore vital to our approach: we read archival and library materials successively, before returning to them selectively and reading them alongside one another. This method is reflected by the references for this paper, which are not neatly divided so that secondary ‘literature’ is confined to an introductory review, and primary ‘data’ appears in our subsequent ‘findings’. Rather, both ‘types’ of evidence are intermingled and juxtaposed throughout. In the developing field (e.g., [90,91,92]) of historical criminology, we therefore submit that a rigid methodological distinction between the categories of ‘literature’ and ‘data’ is false, at least if the ‘data’ come from the archive. We will return briefly at the end of this paper to what this means for how current practitioners might make use of archival materials (and we intend to return to this issue in greater depth in the future).

However, we lastly emphasise that value judgments and ethical choices are necessary when the topic is historic child abuse, especially if it concerns offences committed within living memory. The requisite methods are necessarily retrospective, and are therefore marked by the ‘unnaturalness’ of historical thinking [93], in which the analyst attempts to ‘transcend the finitude of individual existence by thinking in and across time’ [94] (p. 29). In doing so, we unavoidably use concepts and understandings that belong to the present. They shape the questions we ask, the kinds of answers that satisfy us, and the documentary records that interest us. We go further, however, in explicitly using these present-day understandings to make evaluative (and anachronistic) judgments about the past. This opens us to the charge of a particularly smug kind of wisdom after the fact. But: (a) most historic custodial abuses could be defined as crimes in the laws of the time; (b) various other factors meant they were ignored or defined away at the time, even by their victims, and (c) many of the latter are still alive and have received no redress. Considering the force of these points, we concur with others (e.g., [34,95,96]) that to suppress our moral emotions and refrain entirely from judgment would be unethical, colluding in the deflection of responsibility from those to whom it properly belongs.

4. Findings: Why Did Penal Establishments Fail to Recognise Child Abuse?

4.1. ‘Entitled’ Institutional Practices and Moral Hegemony

We now offer a more detailed exposition of the institutional practices and cultures we found in the archive. We begin with Kauzlarich et al.’s theoretical proposition that ‘victimizers generally fail to recognize and understand the nature, extent, and harmfulness of institutional policies’ and that ‘[i]f suffering and harm are acknowledged, [they are] often neutralised within the context of a sense of “entitlement” [18] (p. 185). After expanding on the concept of ‘entitlement’ and its theoretical implications using related literatures, we discuss the examples of routine strip searches in youth custody and staff disciplinary procedures, arguing that both institutionalised different kinds of staff entitlement.

The principle that a criminal conviction can deprive someone of their rights (particularly their liberty and security of person) is enshrined in UK domestic law. This principle is so central to criminal punishment in liberal democracies that some version of it also features in major human rights covenants (e.g., [97] (Art. 5)). Prisoners are defined, then, partly by their loss of certain rights, a consequence of the moral censure communicated by their punishment [98]; and custody itself is defined as an environment in which the applicability of broader standards (e.g., norms against the use of coercive force or against non-consensual nudity) is at best ambiguous. For our purposes, it is helpful to think about the entitlement the state thereby claims as taking two forms. First, the state claims an entitlement to act or to carry out certain custodial practices (such as the use of coercive force, or of strip-searching) which violate near-universal norms governing normal interpersonal behaviour. Second, it claims an entitlement to define in which moral criticism of these practices is deflected by reference to some other higher purpose. Thus a strip search can in principle be necessitated by reference to the need for security, or the use of force by reference to the need for an ordered and safe environment.

Sociologists of imprisonment (e.g., [99,100,101,102,103]) have long argued that closed institutions develop moral cultures and forms of social order which can become quite detached (but which are never wholly isolated) from those of the outside world. Of course, prisoners can and do resist penal power on moral grounds, despite the costs of doing so. As a result, apparently monolithic regimes of penal power are in practice riven by weakness, compromise, accommodation, and the need for constant (re)legitimation. The aims and priorities of official policy are often interpreted by staff to suit their own interests, especially when they perceive prisoner resistance as a hazard to their safety. A full discussion of this literature cannot be provided here, but it is worth noting two particular factors strongly associated with effective prisoner resistance: (a) prisoner solidarity and (b) skill in the use of bureaucratic systems which regulate staff power. To the extent that solidarity is difficult to achieve, or safeguards and checks do not exist or are difficult to access, what we call the prison’s moral hegemony (i.e., the efficacy of its entitlement to define) will be pronounced: staff will be able, effectively, to determine what is right and wrong, and to suppress or defeat counter-claims.

This is especially relevant to a discussion of child abuse, because custodial practices that transgress wider moral standards may in themselves be experienced by their targets as a transgression or assault and thus as a form of victimisation. Why does such victimisation go unrecognised? Many inquiries into institutional child abuse in different settings suggest that power dynamics and the entitlement to define are critically important (e.g., [15,60,61,104,105,106,107,108,109]). Once wider moral constraints (for example about the acceptability of coercive force) have been overcome once, a precedent is set, and can quickly become institutionalised, with two effects. For staff, it can make licit or illicit violence seem an effective means by which to secure legitimate ends such as orderliness, intensifying their sense of an entitlement to act beyond the limits originally intended for it. For abused children, experiencing harmful or abusive treatment which staff/the institution defines as acceptable can entrench negative attitudes towards staff, safeguards, and perhaps power in general, on the basis that they are uncaring and not to be trusted. Abuse inquiries have suggested over and over again that punitivity plays a pronounced part in this process, insulating power holders from moral criticism by framing hard treatment as the offender’s just deserts. The special relevance punitivity in penal institutions hardly needs stating.

Even so, archival evidence of ‘entitlement’ is generally implicit. Tracing it involves comparing internal and external perspectives on the same phenomenon, which can be difficult because the majority of documents are produced by officials working within the institution’s own moral universe. Matters which form part of their daily experience might appear unusual or problematic to outsiders, but only when the archive records outside perspectives does the contrast between internal and external moral standards become clear. Because it privileges the former and erases the latter, the archive facilitates few such comparisons. One example, however, can be found in a newspaper clipping on HMDC Aldington, a Junior DC holding 14- to 17-year-old boys, which is appended to a Home Office file. Among other things, the article describes the strip-search of a new inmate, which appears to have been conducted in front of the reporter: “Now your clothes, lad” [says the officer]. The last traces of civilian identity are peeled off. “Legs apart lad to be searched.” The last traces of civilian dignity go too and the pain far deeper than physical hurt begins. The boy stares ahead dumbly, trying perhaps not to cry. “We have to search there, lad, it’s prison regulations. Last week a lad had some pound notes on a string. It could have been drugs.” [84] (emphasis added). Whoever appended this cutting to the file added a handwritten note: ‘[i]f anyone asks why we want to get rid of junior detention centres, they have only to read this’. Two things are clear: first, that a reporter visiting Aldington sensed that his readers might find the search remarkable enough to describe it for them in some detail; and second, that a civil servant who added the report to the file was aware that this account could undermine support for the detention centre concept.

Routine custodial practices such as the strip-searching of children appeared self-evidently to insiders to fall within the entitlement to act. Yet equally self-evidently, they appeared to outsiders as problematic, perhaps especially because Aldington’s residents were young. Yet what is striking about archival sources in general, whether written by prison staff or by ‘independent’ external monitors and inspectors, is how seldom they mention such matters. Strip-searches of children were neither monitored in official statistics (e.g., [110,111,112]); nor mentioned in contemporary inspection reports (e.g., [2,113]); nor noted in the annual reports of Boards of Visitors (e.g., [114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132]) appointed to monitor places of custody; nor discussed by the reports of contemporary non-governmental organisations with an interest in penal reform (e.g., [80]). It is therefore very difficult to draw a comprehensive picture of institutional entitlement using archival documents, since the materials needed are so rare. One can presume from this, in relation to routinely strip-searching children, that the entitlement to act was seldom challenged, and that the entitlement to define was not usually needed.

Any number of those subject to these practices may have experienced them as harmful, but as far as the archive is concerned, these experiences are effaced or erased: even the Aldington newspaper report presumes that the experience is humiliating without affording any direct voice in the matter to the boy in question. But as noted above, abuse inquiries have linked the routine use of procedures such as strip-searching with the neutralisation ordinary moral norms, as staff become accustomed to crossing these boundaries on a daily basis. This can result in the dehumanisation of child residents, and an increased risk that some staff then feel less inhibited (or more entitled) to transgress other moral boundaries in their interactions with other children [109] (pp. 38–50).

What of occasions where boundaries were crossed in ways which did prompt questions about the legitimate use of power? There is very limited archival evidence on staff disciplinary procedures, namely a single file [133] which reviews Prison Department policy on what to do if prison officers came under police investigation. All the evidence presented in this paragraph comes from this file. The policy was constructed ‘on the hoof’ in 1980: this was two decades into the period we cover, after Neville Husband’s 1969 arrest and also after many of his offences at Medomsley. A memorandum discussing whether prison officers under police investigation should be suspended from duty proposes ‘establishing the Department’s policy in writing, rather than proceeding case by case on the basis of assumptions as to policy’; such an approach would help to ‘avoid future embarrassments’. It is not stated whether the lack of a written policy had caused past embarrassments, but the file describes a number of previous examples in which suspension from duty had been applied inconsistently. One was from HM Borstal Feltham, where ‘there was prima facie evidence that an officer had burned an inmate with a red hot spoon’. This officer was suspended, but after ‘later evidence showed that the burn was probably accidental’ no criminal charge resulted; whether the officer faced any other action is not recorded. Documents in the file compare this case with others, discussing circumstances in which placing an officer on alternative duties (i.e., with no contact with prisoners) might be a better response than suspension. What is striking in this discussion is the criteria of evaluation that are used. The discussion is animated by the perceived risk of potential collusion and witness tampering by the accused member of staff, such that any police investigation might be impeded. In short, procedural concerns predominate. By contrast, the file offers no evidence that suspending a staff member following allegations of this kind might have been seen as a way to protect prison residents from potential harm. What appears to have been paramount for the officials who wrote these documents was to safeguard organisational priorities, such as the avoidance of embarrassment and the maintenance of procedural propriety for staff. Child (or prisoner) welfare does not appear to have significantly affected the discussion.

The written policy formulated by this 1980 review retained the presumption of innocence—another right to which staff were entitled—as a strong thread. While it defined some kinds of allegations as grounds for automatic suspension pending investigation (and automatic dismissal if upheld), other kinds were not seen as so serious, or were judged marginal. Those which would not have led to automatic suspension/dismissal included sexual offences and some forms of physical violence, provided they were committed off-duty. Offences committed on duty were taken more seriously, and appear to have usually been seen as grounds for dismissal if proven [133].

Taken together, then, these points support the theoretical proposition that the victimisation of children in custody took place in a cultural context of ‘entitlement’, as Kauzlarich et al. suggest [18]. Strip searching certainly illustrates the entitlement to act. It is less clear that staff always and everywhere felt ‘entitled’ to use violence to keep order and achieve other organisational goals, though reporting on the recent Medomsley trials [11] suggests that at some times and in some places they did. However, what the archive suggests is that the entitlement to define was hardly challenged in youth custody, even where the entitlement to act was. Particularly serious outcomes, such as where a child was burned with a hot spoon, were handled as a staff disciplinary issue, suggesting recognition on some level that certain lines could not be crossed. But the point here is that no criminal charges were pursued, presumably on the unlikely basis that the injury had occurred by accident. This, along with other documented examples of physical abuse such as the recent Medomsley convictions [11], certainly suggests an unchallenged entitlement to define and hence an effective moral hegemony. Those in power were hardly attuned to the victim’s experience, even if this was of harm and abuse and was represented as such. A sense of entitlement made it too easy for staff to use unlawful violence in furtherance of legitimate official aims, and (in most instances) to then define this away as something other than the abuse which it was. The permissive, unconstrained environment this created could also be exploited by those pursuing illegitimate ends, such as at Medomsley, where CSA was committed with apparent impunity. It also has clear implications for the recognition of victimhood, as Christie [19] suggests: without an offence, there is no offender, and without an offender, no victim.

Kauzlarich et al. [18] suggest that such instances are particularly significant when state crimes are patterned (i.e., replicated in similar ways by different individuals in different locations at different times). The enormous scale of historic allegations in recent years makes this an obvious possibility, one which can only reflect badly on the state: either it has routinely recruited unethical people into positions of power which they abuse to carry out immoral behaviour; or the positions of power turn its employees to unethical paths from which they are not corrected. ‘In the best case, the organisation itself has a problem screening out immoral/unethical decision-makers. In the worst case, the organisational climate itself fosters, facilitates, or encourages such behaviour’ [18] (p. 189).

4.2. Legitimation, Delegitimation and Blame

Thus far, we have argued that power in youth prisons lay predominantly with staff, and that the archive contains little evidence of outside perspectives. This is not the same as saying that there were no ’outsiders’ in youth prisons; in fact, groups of local volunteers were appointed in all of them to monitor custodial conditions and the treatment of inmates [134] (S.6). In this section, we use archival sources to show how these ‘independent’ outsiders interpreted signs of vulnerability among children in custody as evidence of weakness and inadequacy. We also argue that this perception was made likely because the prison’s moral hegemony dulled sensitivities, so that they were also less likely to perceive the risk of victimisation. Having used archival sources to establish their ‘risk-blindness’, we use contemporary ethnographic research to document the kinds of abuse that they might have missed. We show that power disparities endowed even the ‘independent’ observers of custody with an entitlement to define away apparent problems, something which was particularly relevant where they experienced divided loyalties.

Archival records created by custodial monitors contain subtle traces of victim-blaming. For example, the Board of Visitors at HMDC Haslar reported ‘a number of attempted and feigned attempted suicides by swallowing foreign bodies’ during 1966 [119] (emphasis added). The term ‘feigned attempted suicide’ suggests some such behaviour was interpreted as evidence of manipulation, a belief as widespread among custodial monitors as it was among staff [13] (pp. 28–32). Self-harm was therefore interpreted principally as a consequence of individual (rather than institutional) failure. The same attitude is clear elsewhere in the Haslar report, which ‘call[s] attention to the poor standard of lads being committed to Detention’, adding that ‘boys wholly unsuitable by virtue of physical or mental deficiencies, continue to be committed’. Incidences of attempted suicide and self-harm, argue the Board, ‘endorse the absolute necessity for careful selection when sending lads to Detention […] in one case a lad was transferred to Grendon Psychiatric Prison [an adult prison] for necessary investigation and treatment’ [119] (emphasis added in both quotes). This general attitude—that children in custody must ‘measure up’ to institutional requirements is extremely common in archival documents, and can be traced particularly through prevailing norms about gender and disability. So too is the attitude that the establishment is not responsible for meeting the needs of those who are not ‘carefully selected’.

It appears that volunteer monitors took their cues in such matters from staff, who also read the vulnerability of non-conforming children as evidence of a ‘poor standard’ of child, rather than as evidence of unmet (and perhaps complex or entrenched) needs. Reports to the Home Office by DC wardens frequently referred to the problem of ‘unsuitable’ or ‘inadequate’ boys and girls sent by the courts, and similar terms appear in documents on educational provision in DCs, borstals and prisons (e.g., [135]), where the same labels designate children held to be educationally ‘backward’ or ‘subnormal’ [136] and who were presumably seen as beyond help, especially given the constraints of a short DC sentence. Meanwhile, the label ‘inadequate’ was often applied to children who struggled (because of ill health or physical disability) to meet the physical demands of borstal and DC regimes, which included extensive sport, drill and physical training, and sometimes referred to physiological factors such as weight and fitness as measures of the regime’s success (e.g., [83,136]). The wardens of male DCs often asserted that ‘unsuitable’ children belonged elsewhere in the system [13] (pp. 60–63), and sometimes succeeded in having them reallocated [136,137]; as the Haslar case above suggests, this often meant sending vulnerable children to adult facilities [13] (pp. 58–64). Underlying all of this was the unquestioned assumption that the regime itself was unproblematic. The implied role of staff was to deliver it, rather than to identify or meet children’s needs. That of children was to comply with an ostensibly rehabilitative regime: those who did not were ‘unsuitable’, implicitly problematic and faulty [13] (p. 27).

The evidence considered in this section so far suggests, as Kauzlarich et al. propose, that in the context of state crime, victimisation is often enabled by the power of state officials (including volunteer monitors) to represent and define reality, enabling the victims of state crimes—in this case children victimised in custody—to be ‘scapegoat[ed], stereotyp[ed], profil[ed] and typif[ied]’ [18] (p. 184).

These propositions find strong support in a 1975 book by Ericson [86], which is the only contemporary example we found of ethnographic research carried out in one of our establishments of interest. It is valuable because it offers a perspective erased by the archive, from which it is apparent that in the early 1970s the occupational culture of HMDC Whatton—which we argue above was shared by staff and custodial monitors—was very likely to incubate, rationalise and justify abuse. Ericson recorded that staff violence and irregular punishment were routine and that officers routinely delegated power to ‘daddies’ (high-status inmates who kept others in check using ‘knuckle therapy’, kangaroo courts and irregular punishments). He also notes that most inmates preferred these illegitimate and informal arrangements to official procedures and sanctions, because they were less likely to lead to a loss of sentence remission (and therefore more likely to result in early release). ‘Daddies’ also used privileged access to favoured jobs and material goods to exert further power through Whatton’s clandestine economy. Ericson further alleged that officers concealed all of this from Whatton’s liberal Warden, with the tacit approval of a harsh, punitive deputy Warden; we have not been able to establish whether the latter was the same ‘deputy governor’ who was in 2015 accused of historic sexual offences committed around the same time [138].

The potential in these arrangements to foster peer abuse (at least) is clear, and child abuse inquiries on other residential/institutional settings (e.g., pp. 38–50, [109]) have often found that systematic abuses begin with such subversions of legitimate power. Ericson also showed that the victimisation of the vulnerable was clearly a routine feature of life at Whatton. He described how ‘divs’ (low-status or vulnerable inmates who displayed visible weakness or emotion or were otherwise unable to cope) were routinely bullied by peers who subjected them to violence, humiliation, and forms of defilement such as ‘potting’ (i.e., the emptying of a container of excrement over the head). Any one of these could have been experienced as harmful or abusive; some (such as the practice of pulling down divs’ shorts on the sports field to expose their genitals) were apparently carried out in front of officers and with apparent impunity. By crossing boundaries such as nudity, some kinds of peer abuse could have been experienced as sexual abuse. Ericson’s book also made it clear that staff at Whatton expected inmates to ‘toughen up’. Vulnerability was again considered blameworthy and officers felt unabashed about stating such views explicitly in official documents (albeit not ones preserved by the archive): Ericson quotes an official discharge report written by a Senior Officer at Whatton, which stated (of a released detainee), ‘[t]his blubbering giant needed an above average amount of staff support at first … he must be made to face up to reality and be less of a cry baby … A poor prospect indeed’ [86] (p. 96).

The available archival materials offer little insight into the texture of daily life in female institutions but suggest strongly that different cultural expectations applied to their residents. Staff in female establishments were mostly themselves female, potentially altering the power dynamics, and residents in one female borstal were described by a (male) staff member as ‘need[ing] more welfare than men’ [139]. Similar beliefs appear to have influenced the decision to close Moor Court, the only female DC, in 1968. Its militaristic regime was deemed unsuitable for female needs—a striking reversal of the expectation that boys must conform to the institution [81,82,84]. Staff in female establishments also made significant efforts to inquire into girls’ sexual histories, a practice with no obvious equivalent in male establishments. This suggests preoccupation by a different kind of blaming, one which might have seen possible victims of sexual abuse or exploitation blamed for their victimisation [13] (pp. 60–61).

Kauzlarich et al. suggest that victim-blaming in state crimes is often not overt and direct, but instead plays out indirectly through official classification and the inability to recognise victimisation, with other moral beliefs invoked to occlude it. Cohen defines the same subtle dynamics as ‘interpretive denial’ or ‘implicatory denial’, in which the basic facts are not denied (this would be ‘literal denial’) but are instead respectively classified as ‘really something else’ or as ‘justified by something else’ [20]. For Kauzlarich et al., people convicted of crimes are ‘less likely to be treated sympathetically because their assigned master status […] leads to a marginalisation of their human worth, morality, and potential’ [18] (p. 186).

Our research strongly supports this proposition. We have shown how various ‘master statuses’, including those associated with gender and classification as an ‘offender’, made children in custody appear less like Christie’s ‘ideal victim’ [19]. While men and boys were expected to be stoical, self-sufficient and tough, women and girls were seen as vulnerable, sociable and in need of help. For both sexes, such stereotyping could help to marginalise or erase the lived experience of victimisation. A rehabilitative rationale for custody may actually have intensified victim-blaming, by supporting the tacit assumption that a successful detention regime would create non-criminal, healthy, able, ‘normal’ adolescents, and that any ‘growing pains’ incurred along the way were mere side-effects. This could have served as a form of ‘implicatory denial’ [20], by making the individual, not the regime, into the problem. It also allowed non-compliance to be blamed on non-conformity with (for example) gender norms. Rehabilitative aims allowed those delivering abusive regimes to emphasise their creditable intentions while remaining blind to their discreditable outcomes.

4.3. Ineffectual Safeguards

In this section, we build on the analysis of the preceding two sections to show how the entitlement to define and unchallenged moral hegemony of youth custodial institutions combined with their risk-blindness to prevent the recognition of victimisation. We describe the formal safeguards which were supposed to respond to the possibility of custodial abuse and argue that for the most part they were ineffectual. This is relevant to our theoretical frame because (as Kauzlarich et al. suggest) the power disparities inherent to state crimes go ‘well beyond crude asymmetries in the ability to control others’ [18] (p. 183), also acting, through the entitlement to define, to make it difficult for systems of formal accountability to respond to victimisation. This section builds on the argument about ‘risk-blindness’ put forward in Section 4.2, arguing that because of it, the archive supports another Kauzlarich et al. [18] proposition: that the victims of state crime struggle to claim redress because it is from the victimiser that redress must be claimed, making divided loyalties and conflicts of interest a factor.

Across all the different forms of youth custody, formal regulations on staff behaviour were often ineffectual [13] (pp. 28–42). While policymakers issued abundant policy documents and guidance materials to staff, these were haphazardly disseminated and of limited influence. As late as the late 1980s, prison staff were often unaware of the existence or the content of policy documents directly relating to their specialisms, and because prison policy documents were classified as state secrets until the early 1980s and only released piecemeal after that, external scrutiny and independent evaluation were also severely hampered [140,141]. Occasionally, as in the case of suicide prevention, ignorance of policy had fatal consequences [142]. In practice, then, systems of accountability were weak.

During the 1970s and 1980s especially, management too was often weak, industrial relations were often poisonous, and staff unions openly espoused aims which often diverged strongly from official ones, suggesting a kind of ‘malign indifference’ towards prisoners [143] (p. 290). Although individual establishments featured more positive cultures, these were not universally evident [143].

External monitoring responsibilities were divided between external volunteers and an Inspectorate, and archival evidence questions the independence and effectiveness of both. The volunteers (called either Boards of Visitors or, in some establishments, until 1973, Visiting Committees) had the basic duties of ‘satisfy[ing] themselves as to the state of the prison premises, the administration of the prison and the treatment of the prisoners’ [144] (Part IV). They were required to visit regularly, hear complaints, and (somewhat contradictorily) to adjudicate on more serious disciplinary charges against prisoners. It is not clear that these basic duties directed Visitors unambiguously to protect prisoners’ rights, nor that they are free of conflicts of interest. The independence, expertise, and fitness of Visitors to fulfil their duties came under increasing question from the mid-1970s onwards [145,146,147,148,149,150], with their handling of complaints and their tendency to identify with the institution more than the prisoner both frequent points of criticism.

Inspection of their annual reports to the Home Office suggests that such criticism was merited. Several Boards wrote to local newspapers and to MPs to rebut unfavourable comment on ‘their’ establishment; such comment often related to suicides, self-harm, and alleged staff brutality [114,115,116,117,118]. For example, Annual Reports between 1966 and 1969 by the Visiting Committee at HM Remand Centre Risley (which held both adults and children) [114,115,116] all reported on the Committee’s efforts to counter ‘irresponsible statements’ about the Centre. There had been twelve suicides there during this period, but in 1969 the Home Secretary ordered an inquiry which cleared Risley of any responsibility [151]. Preventive steps were nevertheless taken, including the removal from prison uniforms of possible ligatures. The Committee’s 1969 report spent several paragraphs reflecting on this outcome. It compared the twelve ‘unfortunates’ to Risley’s population turnover of ‘over 50,000’ during the same period; it recorded that ‘[s]ome had expressed thanks for the attention given to them before such a sad ending of life’; it quoted (presumably from a suicide note) to the effect that one prisoner had ‘paid warm tribute to the staff explaining that “he wished to join his wife” whom he had killed’; and it suggested that ‘this our great country must not be condemned by people who understand so little’. By 1971, following two suicide-free years, a new Chair reported that Risley was an ‘undoubted success’ following a difficult first few years for the ‘experimental’ Remand Centre concept.

In the context of the 1969 report, written after Risley had been cleared of responsibility for the suicides, these comments suggest the Committee felt a lasting cognitive dissonance as a result of them. It is obvious that it identified and sympathised with Risley and its staff, and that (like the staff) it was stung by outside criticism. It should be noted that the Committee was not uninterested in prisoner welfare: the same files document its concerted, detailed (and partly successful) lobbying on such matters as overcrowding, staffing shortages, and visits facilities. But nor was it overly interested in prisoners’ voices and experiences, except where these could selectively be quoted to exculpate Risley, its staff, and the Committee itself. Tellingly, the 1969 report closes with the Committee commending itself for its ‘selfless devotion’ and its ‘study, support and constant recognition of appropriate administration’. Although an extreme example, this well illustrates how custodial monitors could lack an independent perspective, identify with the institution and its staff more than with prisoners, and look to official sources for interpretive cues.

In DCs, confusion over the proper aims for custody was common. Medomsley’s files [137] record a particularly sharp example, in that in 1975 open conflict broke out between the Board and the Warden, culminating in all but two members of the Board resigning en masse. They had demanded a more punitive regime and the discontinuation of attempts at a rehabilitative regime, but the Warden had persisted in introducing the latter, presumably because this had been Prison Service policy for some time, and government policy since 1970 [81]. His annual report for the year quotes one resigning Board member as saying that ‘boys are sent here for detention not for training’ [137]. If true, this statement demonstrated ignorance of the change in policy. It appears from this and other files that some DC Boards saw themselves as the guardians of the original, more punitive/deterrent traditions of DC custody; in this, they often formed alliances with the officers’ union, which also espoused a more punitive rationale and had resisted the shift to a more rehabilitative policy when the government consulted on it in 1968–1969 [84,152,153]. By 1976, Medomsley had a new Board comprising the residuum of the original one and new members the Warden himself had recommended [137] (suggesting that the recruitment of these bodies limited their independence). Yet at HMDC Eastwood Park, by contrast, the Board’s 1972 report suggested full and enthusiastic engagement with a more rehabilitative regime: they suggested (apparently with the benefit of members’ professional experience) a number of detailed improvements [120] to the curriculum, as well as pointing to possible sources of local expertise which might help implement them. It is difficult to draw firm general conclusions from these disparate examples, but it is clear that how Boards evaluated the regime (and staff conduct) was more contingent on their own cultural outlook and political sympathies than on actual custodial policy, and that what (if anything) they held staff accountable for varied accordingly.

Meanwhile, the Prisons Inspectorate (an internal department of the Home Office with no public-facing role before 1982) mainly focused on prisons’ use of resources and their adherence to policy [13] (pp. 30–32); such inspections were of far more limited scope, and far lesser independence, than today. A 1969 Home Office memorandum reviewing the Inspectorate’s functions stated that it ‘must be sure that from the very start [that] all [its] incursions into the life of an establishment are seen to [help] the establishment [to carry] out its functions as effectively as possible […] laudatory comment as well as constructive criticism [must be delivered] in a manner that will enable Governors and staffs to discern that the Inspectorate is well aware of its responsibility to sustain and improve morale’ (emphasis added). It was not until a major inquiry into the prison systems of the UK in 1979 that the need for a more public form of inspectorial independence was acknowledged [148,154]; before this there was certainly no systematic attempt by the Inspectorate to seek prisoners’ views about their treatment. This is very apparent from the small number of full inspection reports appended to Home Office files in the archives [2,113]; both comment briefly on the ‘tone’ or ‘atmosphere’ of the establishment in a general introduction, but then focus principally on technical or administrative matters such as the cleanliness of kitchens or the necessary resourcing of gardening details [13] (pp. 30–32). Prisoners’ voices feature only in passing, in reported conversations.

Complaints systems [13] (pp. 34–39) were therefore the primary means by which child safeguarding problems could be expected to come to light through disclosures. Such systems were weak. They owed much to the traditions of the English legal system: a complainant’s allegations would be heard before a tribunal comprising either a governor (sitting alone or with colleagues) or (in serious cases) members of the Board of Visitors. Adversarial proceedings ensued, with the complainant(s) and the respondent(s) calling witnesses. At the end, a judgment would be given on whether the Rules for the institution had been broken [146,148].

Ultimately, this system for dealing with complaints was only as independent as the outside volunteers (the Boards of Visitors) who were its final arbiters [139,144,145,146,147,148]. From the mid-1970s, it came under increasingly strong criticism from reformers and official sources, and by the mid-1980s, there was a broad consensus for reform, which nevertheless was not enacted until after the prison riots of 1990 [145,146,147,148,149,150]. Among the criticisms were that the system in general was overly weighted towards protecting staff from ‘false and malicious’ allegations; that prisoners lacked access to legal advice and to truly independent consideration of their complaints; and that those responsible for administering the complaints system therefore lacked prisoners’ confidence. Criticism also centred on the perception that Boards of Visitors lacked diversity and were often dominated by the magistracy. This was partly because their members were usually (as at Medomsley) proposed by the Chair or the prison’s governor to the Secretary of State, who usually approved the appointment [147,148,155,156]. In the words of one undated newspaper column appended to a Home Office file, this made some Boards ‘a self-perpetuating oligarchy’ [157]. This may have been a reason why the complaints system was hardly used, especially by children and young people: Home Office research in the 1980s found that the Boards in a DC and a borstal received two and three applications respectively in the sample year, as compared to over 300 a year in some adult prisons [145] (p. 80). On this evidence, children were far less likely to raise complaints.

The archive also shows that between at least 1961 and the late 1970s, the Prison Department pursued a policy which actively and deliberately hindered effective complaints. All prisoner mail was read (circumscribing prisoners’ ability to raise complaints or make disclosures outside the establishment). Policy also required that any prisoner who complained about a staff member should be read a verbal warning that ‘false and malicious allegations’ were a punishable offence. If the complainant decided to proceed, the complaint had to be lodged in writing. While no complaints data were collated centrally (making it impossible today to trace how many complaints there were, or to make any large-scale evaluation of their handling), it is clear that they were usually handled within the prison in which they originated [157].

Between 1956 and 1978, only eight complaints received an ‘external inquiry’, the fullest and most independent form of investigation. Our report [13] (pp. 76–77) summarises these. Nearly all of the eight inquiries appear to have been triggered by adverse publicity, usually stemming from complaints by ex-prisoners to newspapers, the BBC, or their MPs. Four related to staff brutality in institutions which routinely held children; two of these were partially upheld [157]. It appears that even in some cases where staff were cleared of wrongdoing, the investigating body suspected them of collusion and witness intimidation but was unable to establish this beyond reasonable doubt, and therefore retained the presumption of innocence in its findings while also recording its misgivings about the officers in question [158,159]. In none of the eight cases was any prisoner legally represented, while staff were represented in all but one [157]. Documents also record that the Home Office went to some lengths to circumscribe the independence and scope of later external inquiries after the first of the eight [159] angered prison staff by straying from its original terms of reference after witnesses made new disclosures and allegations at its hearings [13,157]. It is also clear that the Home Office acted to head off other potential sources of redress by which prisoners might have complained about their treatment, such as the Race Relations Board [13,160].

Although it is not possible to trace in depth how complaints other than these eight were dealt with, contemporary press coverage (e.g., [161]) supports the archival record in suggesting that when complaints were handled without external investigation, a lack of evidence and a criminal (i.e., ‘beyond reasonable doubt’) standard of proof often made it easy for staff to be cleared. It is unlikely that archival records can be used to investigate the quality of complaints handling for some decades yet to come: where external enquiries were held, data protection regulations preclude further research, and a number of relevant files are also missing, presumed lost [13] (pp. 73–75). It is difficult to avoid three conclusions: that only some children victimised in custody stood any significant chance of claiming redress; that this chance was remote; and that the rules of criminal procedure and the passage of time still constrain it.

Historians working on CSA in other institutional settings (such as schools and children’s homes) have found that when officials were presented with a strong case for regulatory reform, they often placed unjustified faith in the traditions of the English legal model of adversarial proceedings. With its presumption of innocence and its strong protections for the defendant, this system protected the perpetrators of child abuse more than their victims [34] (pp. 428–429). It is hard to avoid the conclusion that the framers of policy in youth and adult prisons invested similar processes with similar confidence, and that their confidence was utterly misplaced. The available safeguards failed to correct power imbalances relating not only to coercive power, but also to skill in self-representation and in the use of systems of redress. Unless they could achieve solidarity or obtain outside support (both of which were difficult, especially in DCs), the child victims of abuse were open to repeat victimisation, and reliant on the state for official redress. As a consequence, staff acted with something approaching impunity.

The above discussion of regulatory frameworks, monitoring arrangements and complaints systems illustrates clearly that children in custody were afforded minimal protection, so that it would have been very difficult for them to obtain recognition of their victimisation. In line with Kauzlarich et al.’s [18] theoretical proposals, they had negligible opportunities to claim redress, because inspection arrangements and complaints systems were systematically biased and procedurally ineffective. In particular, the prison service appeared completely blind to the possibility that adult staff might sexually abuse children in custody, as suggested by its ineffective response to Neville Husband’s 1969 arrest at Portland, and by the alleged failure of his colleagues to act on his subsequent offending at Medomsley [3].

5. Conclusions

Theorisations of the ‘ideal victim’, of the victimology of state crime, and of the dynamics of denial [18,19,20,21] help us to understand how CSA and other abuses went unacknowledged for so long. Children in custody were far from ‘ideal victims’ [19], and they possessed neither the status nor the opportunity to make their voices heard. Kauzlarich et al.’s [18] theoretical framework has particularly drawn our attention to the dynamics of ‘entitlement’. By granting its employees broad discretionary power, the state was able to deflect its responsibility for the harms they perpetrated. Custodial institutions used these entitlements to justify their practices and subdued counter-voices. Given the inherent legitimacy deficit in prisons [99], such moral hegemony is especially likely to veil a denial [20,21] of harm.

Our analysis has wider relevance to the sociology of punishment more generally. It suggests how the aims of and justifications for punishment can affect the visibility (to officials) of the harms it inflicts. For example, in the cases we have described, both punitive and rehabilitative aims may have helped to draw a respectable veil over abusive practices. We have argued that rehabilitative aims generated risk-blindness by emphasising the positive outcomes of custody, suggesting that the pains of imprisonment were ‘growing pains’. Punitive aims, meanwhile, may have given some kinds of abuse a respectable gloss. Policy shifts between 1960 and 1990, especially in DCs, left the purpose of custody contested and unclear, allowing both kinds of neutralising excuse to be deployed against counter-claims of harm and abuse.

Lacking strategic clarity, and amidst operational pressures, staff focussed on order and discipline—internal organisational goals—rather than on the welfare of individual children. Checks on their power were ineffective and failed to guarantee genuinely disinterested scrutiny. In such a context, they could act with impunity, limited only by the degree of inmate resistance—which was, of course, itself punishable. Not only could abusive means be used to secure legitimate organisational outcomes, but the perpetrators of CSA, as Kauzlarich et al. proposed [18] (p. 183), were able to ‘conceal their illegalities and immoralities’. The victims of custodial child abuse can therefore be understood as the victims of state crimes, because the harm they suffered resulted from omissions by state officials (see [162]). This is not denied by the state itself, at least in some cases: compensation has already been paid to the victims of CSA at Medomsley [3], and more may follow the trials relating to physical abuse there which concluded while this paper was under review.

However, as IICSA has suggested [163], the criminal law is itself an inadequate framework for justice in cases of institutional child abuse. This has also been noted by researchers in other jurisdictions, for example Stanley, who argued that prosecutions fail to reflect the scope and the depth of the harms suffered by victims and survivors. All they achieve is to hold the most blameworthy state agents accountable for the most egregious results of broader crimes of omission by the state. As a result, Stanley argues that it is ‘necessary to move beyond’ individual prosecutions and to ‘[consider] the role of institutional practices as well as structural relations of power and harm’ [164] (p. 1158). Such an analysis might also move beyond clichéd explanatory tropes such as ‘bad apples’, staff who were ‘just following orders’, or ‘inadequate leadership’. It might reflect instead that when organisations are subject to ‘competing or conflicting norms […] leaders and managers respond with structural arrangements that simultaneously absorb those conflicts and become the context for [subsequent] deviance’ [165] (p. 380). Criminal charges alone simply allow the framers of policy to disavow their responsibility and deflect blame onto their subordinates, a process which is also made easier by the passage of time.

We suggest elsewhere [13] (pp. 65–68) that in relation to youth custody in England and Wales, this process may even now be continuing. Despite historical changes in the youth custody estate, there remains great continuity in its underlying conditions: in particular, steep disparities of power remain a structural feature. Past safeguards were believed to address these conditions, but failed to live up to expectations; it should by now be readily apparent that they both ‘absorbed conflicts’ and ‘became the context for deviance’ [165] (p. 380). This alone should lead to continued vigilance and critical attention on secure institutions for children.

Despite gaps in the archive, we believe that historical research offers particular value in thinking about why today’s improved safeguarding regimes still fail (e.g., [166,167]) to prevent custodial child abuse as manifested (and concealed) today. A long-term perspective establishes that flawed safeguards are not only a ‘bug’ of policy design, but also a ‘feature’ of longer-term and more intractable problems: of penal and regulatory change; of political antagonism and contestation regarding the aims and purposes of custody; and of economic change and associated fluctuations in their resourcing. These should prompt policymakers and practitioners to reflect on the kinds of risk distinctive to today’s penal conditions. External perspectives on custody are an indispensable asset here, as the history of regulatory change in the 1990s and 2000s strongly suggests [13] (pp. 44–49).

However, the lessons of the archive are not clear-cut. The archive itself is ‘not merely highly partial, fractured and fragmentary, but also a result of political processes, power imbalances and regimes of governance through which “knowledge” is selected, curated and ordered’ [89] (p. 7). An uncritical reading of its records conveys smooth administration, effective policy and the implementation of clear objectives. A critical reading, privileging a wider range of perspectives, kicks over the traces to reveal contestation, moral ambiguity and the erasure of marginal voices. In making such a reading, we have relied on our professional and research experience in more recent versions of the institutions we read about and have also engaged in dialogue as described by Section 3 above, for example when we used contemporary ethnographic research by Ericson to question and reinterpret archival sources. We would not have been able to make the sense we have of these documents without this experience, or this dialogic method.

It may therefore be that the real value of historical research on child abuse lies in its capacity to provoke reflection and to offer different perspectives. A critical encounter with the archive could reveal to practitioners and policymakers that familiar and taken-for-granted aspects of their work are in fact the historically contingent consequences of political action by their predecessors. Perfectible safeguards are not the logical consequence of this kind of insight. Historical ‘data’ do not (in the strict sense) allow ‘generalisable’ social-scientific ‘findings’, capable of pointing to ‘gaps’ in policy which can then be ‘filled’ so that the possibility of abuse is ruled out in the future.

A more realistic outcome might be a greater willingness on the part of practitioners to think critically about how they wield power, about the intended and unintended harms they can do with it, about the responsibilities they incur in the process, and about the principles that they consider it worth struggling for. At a time when a national politician sees fit to dismiss police investigations of ‘historic offences and all this malarkey’ simply as money ‘spaffed [i.e., ejaculated] up the wall’ [168], we submit that this kind of self-critical attitude might be something to hope for.