‘Am I supposed to be in a prison or a mental hospital?’ The nature and purpose of secure children's homes

Secure children's homes are locked institutions that de-prive children of their liberty. The government are in-vesting significantly in these homes, yet there remains a lack of clarity about their nature and purpose. Drawing on data generated through a substantial ethnography in one secure children's home in England, this paper uses Goffman's (1961) theorising as a conceptual lens to view the institution. It concludes that ambiguity and confusion about what these institutions are, and what they seek to achieve, impacts negatively on the experience of everyday life within. Clarity is needed urgently to improve experiences and to enable the assessment of outcomes


INTRODUCTION
Secure children's homes (also referred to as 'secure care', 'secure accommodation' and 'secure home(s)') form part of the secure estate in England and Wales which also includes Secure Training Centres (STCs) and Young Offender Institutions (YOIs).All three locked institutions accommodate children remanded or sentenced for a criminal offence.Secure children's homes | 3 ANDOW traumas associated with being socially isolated.A decade after Gyateng et al.'s (2013) review of the youth justice estate, the questions raised around cost-effectiveness, meeting children's needs and achieving positive outcomes within secure care remain unanswered.Roe (2022) makes the salient point that until there are agreed aims for secure care, it is impossible to argue whether secure homes are meeting their objectives.Further, there is a gap in data on short-and long-term outcomes for children post-secure care.Without consensus on the nature and purpose of secure accommodation, outcome measures will be inconsistent, if outcome data is captured at all.
Despite this lack of knowledge, the Department for Education is investing millions in refurbishing and increasing secure provision (GOV.UK, 2021).Further, new secure schools are being modelled on secure children's homes (GOV.UK, 2022) despite a lack of clarity around the role of the latter.Secure schools will accommodate children remanded or sentenced for criminal activity within therapeutic environments focussed on education, health and physical activity (Ministry of Justice, 2016).This policy proposal demonstrates that the historic tension between the ideals of welfare and justice persists.At this time of investment, there is unprecedented opportunity to re-imagine secure childcare (Andow et al., 2023).To do so, it is vital to clarify the nature and purpose of secure care, which makes this paper especially timely.
First, this paper illustrates that existing literature has explored ambiguities surrounding secure care without considering any changes in the profiles of the children accommodated.Second, details of the research method make clear the depth of ethnographic data underlying this contribution.Next, Goffman's (1961) theorising on Total Institutions is established as conceptual lens.This lens makes visible how children and staff within a secure home do not share an understanding of the nature of the institution and what it seeks to achieve, and that this confusion has been exacerbated by the inclusion of children with increasingly complex needs.Worryingly, the main consensus seems to be that the secure home acts as a 'storage dump' (Goffman, 1961, p. 73) that achieves nothing at all.

MAKING SENSE OF SECURE CARE
The unique institutional arrangement within secure homes-both in terms of their status as children's homes and because of their accommodation of children on welfare and/or justice placements-has led to questioning surrounding the purpose of these institutions.Various, and often competing, ideas have been put forward, ranging from welfarist notions of care, treatment and education, to more punitive philosophies of control, containment and punishment (Harris & Timms, 1993;Kelly, 1992;Millham et al., 1978;O'Neill, 2001).Kelly reported a 'conceptual conflict… inherent in the ideology of the system' (1992, p. 1) while for Harris and Timms, secure accommodation: constitutes a particular conjunction of the therapeutic and the penal which necessitates rejecting the ideal type explanations of both positivists and retributivists but, because their ideas have been so incorporated into our cultural and intellectual heritage, necessarily containing elements of both, bound inexorably and inextricably together.
(1993, p. 171) around the main purpose of depriving a child of their liberty for welfare reasons and in youth custody (i.e., to punish or rehabilitate) and therefore what the "ideal" system should look like'.She contrasted this with the detention of children in mental health settings where the purpose-to provide treatment-is clear.Ellis (2012) proposed that the contradictions surrounding the purpose of a secure unit can be lessened if the home is conceived of as a socialisation tool.Similarly, Rose (2014, p. 194) suggested the purpose could be considered a 'corrective emotional experience' for all.The most popular objective according to stakeholders interviewed by Bach-Mortensen et al. (2022, p. 20) was 'to safeguard and provide a nurturing and stable environment for vulnerable children', without differentiation by legal status.
Research inside secure homes demonstrates that staff differ as to the purpose according to their role within the home and/or the mix of legal placements.In Scotland, Kelly (1992) found that staff working directly with children perceived more of a custodial ethos than those in management and that staff felt managers did not share a definition of the purpose of the home.Hart and La Valle's (2016) research with senior managers in local authorities in England supports Kelly's finding that those in management do not define secure care in terms of punishment.Senior managers identified the purpose of secure care as keeping a child safe, offering a period of stabilisation and the opportunity to engage with adults wanting to work with them, creating a space for assessment of a child's needs and the ability to plan forward and reduce further risks (Hart and La Valle 2016).There is a further similarity between the views of these managers and those surveyed by Brogi and Bagley (1998) almost 20 years earlier.Brogi and Bagley (1998, p. 323) described managers giving idealistic responses as to the purpose of secure accommodation, quoting one as writing, 'The emphasis is upon enabling children and young people to begin to deal with the circumstances, problems and difficulties that have led to their liberty being restricted'.
In O'Neill's (2001, p. 121) research, perceptions of purpose of a secure children's home differed according to the legal placement of the child: to keep the public safe from 'offenders' or provide children placed on welfare grounds an 'escape or sanctuary from risks in the community'.O'Neill found that the staff struggled to evaluate the effectiveness of their work because of confusion as to what they were trying to achieve.This was less problematic in relation to the justice placements where the focus was on offending behaviours, and, in contrast to the perceptions of managers discussed above, most staff considered secure accommodation as primarily a facility for offenders and were confused as to the appropriateness of the institution for those with complex welfare needs.O'Neill (2001, p. 124) reported that the staff she interviewed failed to identify a 'theoretical framework which informed the regime and their practice within the unit'.
Children with direct experience report understanding the purpose of secure care as punitive (Barry & Moodie, 2008;Ellis, 2012;Kelly, 1992;O'Neill, 2001).Children on welfare placements have described a disconnect between the stated purpose of their placements as care, and their experience of their placements as punishment (Ellis, 2012;O'Neill, 2001). However, Goldson (2002) and Ellis (2012) both found that children's perceptions of secure care changed over time; initially, they viewed the home as prison-like because of the fixtures and facilities, but over time, they came to view it as a therapeutic facility.Neither study, however, explored whether perceptions differed according to the child's legal status.
The problem of defining the purpose of secure care has been compounded by the increasing number of children with mental health needs accommodated.Data from the Secure Welfare Coordination Unit show that 68% of the 215 children referred for a welfare placement in 2021 had a diagnosed mental health condition, compared to 48% of 254 children in 2017 (Bach-Mortensen et al., 2022).Children entering secure care are often referred to as being more 'complex' than they  ANDOW were previously, defined as having greater and wider-ranging needs, as well as being involved in more serious types of crime such as rape and murder (Bach-Mortensen et al., 2022).The use of welfare beds in secure children's homes for children with severe mental health problems when there is no available hospital bed was highlighted in 2006 (Jane Held Consulting Ltd, 2006) but academic research has not explored how children and staff experience secure children's homes with this issue in mind.
In summary, the purpose of secure care has been explored academically, but much of the research is outdated and/or limited to one group of stakeholders or another.Direct insights from more recent research on children (and staff) in secure care in other jurisdictions such as Sweden (Vogel, 2018) or Denmark (Henriksen & Refsgaard, 2020) cannot be drawn because of differences in institutional arrangements between jurisdictions (Souverein et al., 2022).Moreover, how children are constructed in popular discourse will have an impact on how secure children's homes are viewed, and these constructions are historically and culturally specific.As Daniel (2010, p. 231) indicated, '…in the context of child welfare and protection, the explicit and implicit definitions of "adversity," "risk" and "vulnerability" can have profound effects upon understandings about the most appropriate policy and practice response'.Muncie (2021) describes youth in Britain as being constructed in media discourse as dangerous, 'yobbish' (if white and working class) troubled and troublesome.It is against this backdrop that this paper explores understandings of the nature and purpose of a secure children's home in England.
This paper makes a new contribution by exploring the perspectives of staff in different roles, and children on both types of legal placement, using data generated through an in-depth ethnography.It asks, 'How is the nature and purpose of secure care experienced by children and staff within a secure children's home?' and thereby goes beyond the preoccupation with purpose in existing literature.Importantly, it also pays close attention to perceptions about the changing profile of children.

METHOD
This research was undertaken in 2014 as part of a substantial ethnographic case study of a secure children's home in England funded by the Economic and Social Research Council.In the intervening years, there have been fluctuations in the demand and supply of beds, as has historically been the case (Deloitte, 2008;Mooney et al., 2012), but the institutional structures have not changed.The case was selected because it accommodated both welfare and justice placements so the lived experiences of this mix could be explored.
Ethically the project presented challenges, especially as a doctoral student wanting to engage with children considered vulnerable because of their age, likely adverse childhood experiences and detained status.Like negotiating access, gaining ethical approval was a drawn-out process, involving applications to both my university and the relevant local authority, including indicative interview schedules and observation activities: something that would be alien to early ethnographers like Whyte (1993).Generally, children (under 16 years) cannot legally consent to participate in research -their key workers acted on their behalf-but I offered them agency in terms of the extent (or not) of their participation.For all participants, I emphasised the principles of voluntary participation, confidentiality-limited to my perception of risk of harm-and anonymity, which went beyond names to include any identifiable details of the home.I did not offer incentives for participation because even a small reward can be highly valued by a (disadvantaged) child, encouraging them to take part, divulge more information than they would otherwise, or say what they think a researcher wants to hear (Alderson & Morrow, 2011).I did not treat ethics as a one-off process, but rather as a constant negotiation, frequently reminding the participants about my research and their choice to participate, especially at times when my presence felt less welcomed (see Henriksen & Schliehe, 2020).
The initial period of fieldwork was intense to develop familiarity and rapport quickly.I was present at various times of day, weekday or weekend, to build a comprehensive picture of the setting.I spent time variously across the home, conscious not to carve out one 'inhabitable niche' (Hammersley & Atkinson, 2007, pp. 36, 92).I helped staff with menial tasks, attended meetings, played sports, made crafts, shared meals and simply sat in communal areas.I was actively aware of constantly moving along the participant observer continuum-never pure participant, never complete observer (Mason, 1996).
From the outset, I was intent on developing a non-staff identity so that the participants (child or adult) felt able to share their perceptions and experiences with me without consequence.Like Ellis (2012), I requested to not hold keys, to avoid positioning myself in proximity to staff (Henriksen & Schliehe, 2020) but the staff were so overstretched I had to.Instead, I found a symbolic way to communicate my position, by being 'atypical' (Corsaro, 2005;Henriksen & Schliehe, 2020).I noticed that 'on the units' the staff wore their shoes, but the children wore slippers.I wore only socks which encouraged much attention from the children and staff and allowed me repeated opportunities to communicate my non-staff, researcher position.I also carried my notebook around to further distinguish myself and, pragmatically, to record observations immediately.I adopted Ellis's (2012) technique of drawing pictures rather than writing notes.Maintaining this 'liminal position' (Henriksen & Schliehe, 2020, p. 845) felt isolating and required constant effort-especially, as I became a familiar face in the field and accustomed to the way things worked.However, my sustained immersion helped me to develop rapport and trust, among the participants which I argue increases the validity of the data generated (Andow, 2020b).
Overall, I spent 396 h over 7 months in the field.In total, 56 staff and 14 children consented to participate and I interviewed 24 staff and 14 children.Interview topics included the nature and purpose of the institution, peer and staff relationships and daily experiences.These schedules were adapted according to the interviewee.The schedules developed over time as the interviews informed, and were informed by, my observations; a continuous iterative process.
To make sense of the data, I use thematic coding, as espoused by Corbin and Strauss (2008).Moving from concrete to higher-level more conceptual codes enabled me to look for themes and relationships between codes.Exploring these codes in relation to existing literature, Goffman's (1961) theorising and my research questions enabled deep data analysis.

THEORETICAL LENS
Goffman's 1961 ideal-typical concept of a 'total institution' is used here as conceptual lens for seeing.Goffman (1961, p.11) defined a total institution as: a place of residence and work where a large number of like-situated individuals, cut off from the wider society for an appreciable period of time, together lead an enclosed, formally administered round of life.
It is 'total' in the sense that 'inmates'-the term Goffman used for members of such institutions-'sleep, play and work' in the same place and under the same single authority (1961, p. 17).

ANDOW
For Goffman (1961, p. 17), one of the defining features of the total institution is that 'the various enforced activities are brought together into a single rational plan purportedly designed to fulfil the official aims of the institution'.Goffman (1961) identified five different groups of institutions, categorised according to their purpose and the type of individuals they cater for.Secure homes can be thought to overlap two of Goffman's distinct categories, as they accommodate children considered to pose an unintended threat to society through the danger they pose to themselves (in this case, 'welfare') and children considered intentionally harmful to others ('justice').Goffman's theorising thereby focuses attention on how the mix of children within this type of institution is experienced and understood.
Goffman was writing more than 60 years ago about institutions that excluded people from society entirely for long periods of time and stripped them of their identities.Secure homes are more permeable, and children can maintain contact with their families (in person and online), access services, and even purchase items for delivery.Children are not stripped of their identities in the way that Goffman described.This notwithstanding, Goffman's model-instead of more recent alternatives including Quirk et al.'s (2006) concept of the 'Permeable Institution' (PI) and Scott's (2010) 'Reinventive Institution' (RI)-maintains heuristic value as a way of seeing secure homes because it conceptualises an institution where people enter involuntarily on the grounds of lost liberty.Goffman's (1961) theorising remains of contemporary relevance, even if it does so by making visible how far institutions depart from his ideal type.

PRISON, HOSPITAL OR HOME?
On entry, I had studied academic literature, policy documents and information about the home itself, but I still struggled to develop a sense of the nature and purpose of the institution.Symbols and sounds within the environment-locked internal doors, jangling keys, non-opening windows and toilets without seats-gave a distinct custodial impression.This impression jarred with my knowledge that some of the children within were not detained on youth justice grounds, rather for their own welfare.
In interviews and informal conversations with the children and staff, I found that they too struggled to articulate the nature of the institution, even though they had spent much longersometimes years in the case of staff-inside of it.For most, referring to the home as 'prison-like' is the easiest way to describe the institution to outsiders.However, they do not experience it as a prison; despite commonalities in terms of loss of liberty, the absence of bars marks a crucial symbolic distinction.The following extracts from separate interviews with three residential care staff, and Scott-a boy accommodated on a long justice placement for a grievous crime-capture this.I asked these participants how they describe the home to people who are not familiar with it: Jane: They say 'Is it like a prison?' and I'm like well it's not got bars on the windows but it is secure which means yes, they are not allowed out.Phil: …the best way to say it, is its similar to a prison for young people although it's not got … iron bars and that sort of thing, but that's the best way for someone to understand what secure means… Sinead: …it's like a children's prison, it's probably the easiest thing to describe it as.People don't really understand the terminology, but it is basically a children's home which is like a prison, because they can't get out.But it's not, obviously once you come in here you know it's not a prison because … it's more like a home than it's a prison but it's kind of in the middle… like it's not got like bars on the doors or anything… Scott: There's not like bars, metal doors, stuff like that but other than that I just see it as a mini-prison sort of thing, secure unit.
The focus on the material physicality of the institution as indicative of a prison over any emotional experiences connected with prison life is notable, and links clearly with Goffman's (1961) focus on place and space within his theorising.However, the participants express a sense of uncertainty in describing the institution to others, indicating some ambiguity around its nature.
Staff spoke about being initially shocked about the mix of welfare and justice placements together, which echoes the shock expressed by the children in Shenton and Smith's (2021) research.For example, Natalie (care staff), said 'I didn't think it would be as mixed as it was, where the Welfare kids are living with the Criminal kids'.Theo, another member of the care team but with longer tenure, explained: 'I get asked by friends what I do… Erm, I work with… I don't know, challenging young people?…I've struggled with the justice and the welfare, I struggle with that, the whole concept, I always have done…'.
Making sense of the mix of children-who share broad commonalities but are officially labelled as different-is experienced as problematic.
However, the mix of legal placements became central to how many of the children and staff talked about the home.For example, when asked in his interview how he would describe what secure care is, Trev, a member of the care team said: 'I'd say I'm working in a secure children's home that's split two ways, welfare and criminal'.Some of the children on justice grounds articulated similar definitions, for example, Scott, sentenced for a grievous offence, said: '…it's a secure unit for people that have committed crimes and been remanded or sentenced and a secure unit for people that are here for their own welfare'.In these examples, participants appear to perceive the institution as fundamentally divided based on placement type, even though in practice there is no physical separation.Viewed through Goffman's (1961) lens, the institution is experienced as a secure unit for two different categories of people.
Further, a sense of institutional ambiguity and conflicting needs pervaded everyday life inside the home.For example, Reika, accommodated on justice grounds, had the bedroom next door to Shona, a girl on a welfare placement who regularly self-injured.Whenever Shona harmed herself, staff would strip her room entirely, as items such as bedding and clothing could be made into ligatures, and Shona would break apart items or ingest them.Shona would scream loudly and unrelentingly as staff did this because she found it traumatic to have others moving her belongings.I found Reika in the corridor one day, upset about such an event involving Shona the previous night, which had disrupted her sleep.Reika asked me, with tears streaming down her face, 'Am I supposed to be in a prison or a mental hospital?' (RD i 29/01/14).Here she (presumably) unwittingly almost quoted the very title of Harris and Timms' (1993) work, 'Secure Accommodation in Childcare: Between hospital and prison or thereabouts'.This is just one example of how the different needs of children accommodated within the same institution-clash and cause confusion.In this case, the staff faced a dilemma: in trying to keep Shona safe, they negatively impacted Reika's well-being.Consequently, Reika became hostile towards Shona, and disparaging of her needs.
Perceptions of differences impacted how the children identified and related to each other, but not in terms of a direct split between children according to placement type.The label of 'welfare' became associated with the act of self-injury, by the children and staff alike, despite the accommodation on welfare grounds of girls who did not self-injure and one boy over 10990860, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/chso.12828,Wiley Online Library on [15/02/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the course of this research.To illustrate, the welfare placements-as a collective-were described, by Heather, a psychologist, in her interview, as displaying '…a lot more… severe selfinjury'.Similarly, Veronica, a member of care staff with a long tenure at Woodside, described the 'Welfares' accommodated in recent times as more 'prolific and persistent self-harmers' than ever before (RD 27/02/14).And Natalie, also care staff said, '…the type of welfare kids that you do get in are more intense than I ever thought would be in a secure unit.I thought they would be more in a medical hospital place rather than a secure children's unit where you'd have offending behaviours'.The children self-identified and formed relations with each other as 'Criminals' or 'Welfares', but in practice, they were divided according to their engagement-or not-with self-injury, and not their legal status.The children who self-injured were 'othered' by the 'Criminals'; marked out as a distinct and separate social group on the grounds of their self-injury (Schwalbe et al., 2000).
Tensions between the children who (openly) self-injured and those who did not were high because there were perceptions of differential treatment.To return to Goffman's theorising, he spoke about the 'inmates' in a 'total institution' being 'treated alike' and 'required to do the same thing together ' (1961, p. 17).However, in this research, perceived risk of self-injury led to different everyday regulations and disciplining of the children.Staff were reluctant to discipline children who were known to self-injure because of the risk that they would escalate their behaviours.As a result, the children experienced the staff as treating them inconsistently, as explained by Scott, a child on a justice placement: …staff will always, I don't know, jump for them like, if they say jump, staff will say how high?Like that's just how I see it.I think the people that are here on welfare because it's like, if like a kid keeps self-harming that's here on welfare, it's mainly them yeah, they sort of like, because they don't want them to self-harm they'll shut them up with anything they want, but when it's me like I just get, sort of like, 'Oh shut up' like, do you get what I mean?Scott also described a situation where the children's bedding had to be removed because it was not fire retardant.He asked for his bedding back, but was not allowed it, but a child known to self-injure was allowed hers.I knew from my observations that Scott was referring to Beth, and his account of what had happened fitted with Beth's version of events.Beth, who would often harm herself, had told me, in an informal conversation on the day the bedding had been removed, that she would 'kick off' if her bedding was not returned (RD 15.04.14).The next day, while sat with Beth and Shona in the gym, I asked Beth what had happened.Beth delighted in telling me that she had the staff looking for her bedding until 'past midnight' and it was eventually returned to her.Shona then told me her bedding had never been taken in the first place.In response to my surprise, she said 'C'mon, what do you think they think I would do if they went in and changed my bedding?' (RD 16.04.14)These girls seemed very aware of the power the threat of self-injury afforded them in this setting.
It was not only the children who perceived the relationship between the care staff and the children at risk of self-injuring in this way.Roger, a senior member of the care time, indicated the same in his interview: Roger: They're treated differently.Certainly, the kids on welfare I think and the ones that are here on youth justice are treated differently, real different expectations.Caroline: Can you give me any examples of that?
Roger: Yeah, if … one of the ones here on criminal justice grounds, if they don't get up for school, don't go to school, they'll be sanctioned.Shona this morning told staff to fuck off, she's not getting up and she won't get a sanction for it.Caroline: Why not?Roger: She should, she should receive a sanction, … maybe she should lose her power or something, but she won't because the risk, if she gets sanctioned, … of her escalating her behaviours and seriously self-harming increases.
Taken together, the data show that accommodating children at risk to themselves alongside children at risk to others led to inconsistencies in regulation and discipline, and tensions among the staff and the children.Notably, this inconsistency does not emanate directly from the mix of legal status.
As part of seeking to find out how staff made sense of the home, I asked them how they describe their role and where they work to others.Many of the staff explained to me that they rarely discuss their role to outsiders, either for reasons of confidentiality (for example, Win, a teacher), or, more commonly, because of the difficulties they encounter in describing where they work.For example, Hugh, a member of the senior care team said in an interview, 'I tend not to talk about it because people don't understand what it's about'.For Sean, the care manager, despite having been in the post for 2 years, he still found it difficult to explain to outsiders the nature of the home because, as he said in his interview, it is a 'hidden world'.In choosing either not to talk about their place of work to outsiders, or drawing on imperfect analogies, staff and children inevitably perpetuate the invisibility and lack of understanding about secure children's homes in popular discourse.

TO PUNISH, CARE OR CONTAIN?
Moving to consider the purpose of the institution, the children were unanimous in their perceptions that the home has a dual purpose; children accommodated on welfare grounds are to be helped, and those placed on justice grounds are to be punished.Naomi, a welfare placement, articulated this dichotomy clearly in her interview when she said: '…Criminals are here to be like taught they shouldn't do that and they've done bad.We're just here to get help'.Echoing Naomi, Beth, another welfare placement said in her first interview, '…I'm here to get help and move on with my life, they're here to be punished for something that they've done'.The children on justice placements had themselves internalised this idea that the purpose of their placements was punishment.For example, Rhianna, nearing the end of her justice placement on the grounds of petty offending, said '…obviously we've been naughty and are supposed to be getting punished…' Jacob, another justice placement with a long history of low-level offending expressed the same view: '… because we come here for punishment'.And, like the welfare placements who saw the justice placements as deserving of punishment, those placed on justice grounds described that children were placed on welfare grounds 'to get help and get better' (Scott).This distinction between the remits of help and punishment maps onto the 'particular conjunction of the therapeutic and the penal' (Harris & Timms, 1993, p.171) that is unique to secure children's homes.
In contrast, the staff were divided as to how far the institution has a single goal-as Goffman (1961) envisaged-or whether there are dual goals of care and punishment.The care staff-tasked with managing the day-to-day behaviour of all the children; 'whatever is thrown at them within a shift' (Harry, psychologist)-made a distinction as to the purpose based on legal status.However, instead of a dichotomy between punishment and care, these staff more often  ANDOW distinguished between changing the behaviour of the 'Criminals' (e.g.Sinead) and keeping the 'Welfare kids' safe (e.g.Natalie).The care staff, like the children cited above, 'label' (Becker, 1963) the children by legal status and identify different purposes in line with these labels.In contrast, staff with more specialised roles identified universal aims, in line with the role they played.The teachers spoke about re-engaging children-regardless of legal status-in education and giving them opportunities to succeed (Hugo, Win and Valerie).The psychologists described enabling each child-again irrespective of placement type-to function in wider society post-release (Harry and Heather).While these groups were internally unified as to the purpose of secure care, and broadly in line with each other, there was not a sense of any official aim(s) as Goffman envisaged in his theorising.
Neil, the manager, who rarely interacted directly with the children, suggested that the purpose was to 'enable children to fulfil more rewarding, fulfilling lives'.Neil was keen to dispel the notion that there were separate purposes according to legal status.However, the senior members of the care staff-who occupy a middle ground between care staff and management-were conflicted as to whether the home had multiple aims given the mix of legal placements.For example, for Gary there was a distinction, as captured in this extract from his interview: Caroline: And what is this place designed to achieve with the young people?Gary: I think, that becomes quite murky because you've got that punitive and punishment thing as well because you've got the welfare and then youth justice young people mixed.Now the kids on welfare aren't here as a punishment, they're here to keep them safe and manage their behaviour, where ultimately, young people that have been sentenced to a custodial are here as a punishment of society, it's youth justice, it's justice, so it's murky I think.
In contrast, Nasa, another member of the senior care team, unified these seemingly contrasting aims into a common rehabilitation towards an ideal standard: Yeah, there's multiple aims but in reality all those aims come down to, 'What can I do to help this young person develop some internal controls that are socially acceptable and then go out and maintain that in the community?
As Ellis (2012) cited above, Nasa's perspective suggests that there could be a 'single rational plan to fulfil the official aims of the institution' (Goffman, 1961, p. 17) if this is conceptualised as broad resocialisation.
However, Nasa's optimism about the function of the institution was not experienced in practice by the children and staff.Scott, a justice placement, said in his interview: '…you sit here, do your time and that's it'.Indeed, many of the staff also spoke about the home in a way that echoes Goffman's (1961, p. 73) notion of institutions acting as 'storage dumps for inmates'.For example, Rich, a member of the care team, described the home as 'more of a holding pen than a rehabilitation centre'.Similarly, Sinead, in the same role, described it as 'Just somewhere to hold kids' and said '…we're not really doing anything for them, we're just containing a safe period of time for them here.'Time spent in the field, sat in the (ironically named) 'quiet-area', attests to this, as I watched children who were not engaged in any activity, complain about their boredom, just as Bengtsson (2012) found in her research in a secure care institution in Denmark.Harry, one of the psychologists lamented in his interview, 'Sometimes it is just purely containment and then they move on….' Roger, a senior care worker suggested that there is a significant gap between the public presentation of the institution as rehabilitating children and the impact-or lack thereof-on children in practice: I can't honestly say … I can think of anyone here in the last six months who walked out of here a better person than when they walked in.They've come in here because they've done naughty stuff, and they've continued to do naughty stuff whilst they've been here, and then we just breathe a sigh of relief when they've left because we don't have to deal with that anymore.
Unlike in Goffman's (1961) theorisation however, there does not appear to be a clear presentation within public discourse or even internally as to how these secure children's homes present themselves, and what they claim to achieve.
Neil, the manager, indicated that he was aware that the purpose of the institution needed to be made clearer.He said, …we need to be smarter about…what it is specifically that we do and then, with our statement of purpose… we can then be a bit clearer about, should you place a child here, this is what you can expect at the end of it, for parents, courts.
The fact that this is not already in place demonstrates why there is ambiguity surrounding the purpose of the home.

CONCLUSION
This paper demonstrates that not only the purpose of secure care, but also the nature of it too, is experienced as unclear and contested.The staff and children in this ethnography experienced uncertainties, lack of transparency and confusion around an institution that is widely perceived to be catering for the needs of two different groups.However, unlike existing research, this study found that it is not legal status that marks a crucial distinction between the children.It is rather the act of self-injury, which is exclusively associated with children on welfare placements, even though children are accommodated on welfare grounds with other needs.The association of this act with this label is so ingrained, that the staff and children do not question it.The accommodation of children severely injuring with those who are not creates confusion about the nature and purpose of the institution, tensions within interactions and inconsistencies around regulation and discipline.Arguably, the focus on the mix of legal placements is outdated and contemporary questioning of these institutions must take into account the changing profile of children accommodated.
Looking at the secure children's home through the lens of Goffman's (1961) theorising encouraged questioning of who the institution caters for, how it operates and how it is experienced.Using Goffman's (1961) theorising has made the inherent complexities and ambiguities visible, and in doing so, this paper has shown the continuing heuristic value of his work.Unlike in the institution he envisaged, there is no transparency about what secure care is and what it seeks to achieve.The nature and purpose of secure children's homes must be defined for the well-being of children and staff, and more broadly, so that the impact of this costly form of intervention in children's lives can be evaluated.