Care(ful) relationships: Supporting children in secure care

Secure children's homes are used to accommodate children aged 10 – 16 under two main categories; while half are sentenced after committing a serious offence, the other half are placed because there are serious concerns around their safety in the community. Secure children's homes are prized within the secure estate, and they administer complex therapeutic support to ‘ the most vulnerable ’ young people, how-ever little is known about the experiences of those employed to work in such spaces. This paper shares findings from PhD research conducted in one secure children's home over 1 year. Data presented are drawn from sensitive ethnographic fieldwork and in-depth interviews with residential staff and residents in the home. Although young people's views are important, we concentrate here on the perspectives of residential staff to share their reflections of delivering ‘ care ’ and the strategies used to manage successful relationships within a secure setting. We conclude that residential staff tread a fine line between creating emotional closeness while maintaining physical distance and that they are sometimes unable to return the intensity of feeling that residents' direct towards them. We recommend that all residential staff receive regular and detailed supervision to provide opportunity to request support when necessary.


| INTRODUCTION
Children enter secure children's homes under two types of orders, while half are sentenced for committing a serious offence, others are placed by social services under a child welfare order. Though placed under different circumstances, it is understood that both types of children are perceived as 'vulnerable' and in need of therapeutic care.
Although young people's perspectives of secure care have been explored (Ellis, 2018(Ellis, , 2020, there has been little consideration of the everyday experiences of the residential staff working to administer this type of care. Hochschild (1983) claims that those working in 'caring' professions enact a particular persona when carrying out their work to ensure that their inhabitants feel 'cared for' in a 'safe place'. Caring work is hard, and Hochschild reminds us that in order to perform their role successfully, carers must complete their duties while disguising feelings of 'fatigue' or 'irritation' as 'seeming to love the job, becomes part of the job' (Hochschild, 1983, p. 6). This paper considers the experiences of residential staff and explores the strategies they use to provide intensive therapeutic placements for vulnerable children.

| Local authority secure children's homes
Though the age of criminal responsibility varies globally, in England and Wales, children as young as 10 can be held accountable for breaking the law. In 2018, 835 young people in England and Wales served a custodial sentence, with the youngest, and 'most vulnerable' assigned places in one of 15 specialized secure children's homes. As well as accommodating young offenders, secure children's homes also accommodate non-offending young people who have suffered serious abuse and are in need of protection. In 2018, 204 children from England and Wales were placed in secure children's homes: 48% were on remand or sentenced for committing a serious offence, 47% were placed under a child welfare order, and the remaining 5% were secured by their local authority over concerns around their safety (DfE, 2018). Secure children's homes therefore have 'a foot in both camps' and straddle both social care and youth justice sectors (Goldson, 2002). As secure children's homes hold only a small proportion of the secure population, research in this area is currently limited, with a few exceptions (Ellis, 2018(Ellis, , 2020Goldson, 2002;Harris & Timms, 1993;O'Neill, 2001;Rose, 2002). Tasked with responsibility for managing challenging behaviour while simultaneously providing supportive environments for especially vulnerable children, the perspectives of residential staff in secure children's homes have received limited attention in the United Kingdom, though international research with a more psychological focus can be found (Harder, Knorth, & Kalverboer, 2013;Seti, 2008).

| The role of residential carers
Although sometimes called 'Residential Social Workers', in England and Wales, residential workers do not have to be formally registered social workers. Although some are qualified social workers, most are not and instead receive specific, on the job training at their place of work. While registered social workers occupy relatively well-paid professional positions and are held accountable for the decisions that they make, in contrast, residential staff are often poorly paid and lack capacity to make 'big' decisions; instead, they provide day-to-day care and behaviour management for vulnerable young people. While aiming to provide emotionally supportive care to young people with significant needs, residential staff must also manage a veil of suspicion that accompanies their role. The abuse of young people in care was made public following the Utting Report (Utting, Baines, & Stuart, 1997), which disclosed widespread abuse of young people in residential settings. Following which, it was ordered that staff 'blow the whistle' on abusive behaviour and consider colleagues as potential perpetrators. Subsequently, research has shown that residential staff are mindful and risk averse when carrying out everyday duties (Smith, 2008;Steckley, 2012) and research notes that staff sometimes appeared 'more concerned with risk to themselves than to the young people they care for' (Horwath, 2000, p. 189).
Although children in care sometimes assert that social workers wield significant power to determine where they live and with whom (Aubrey & Dahl, 2006), research shows that social workers themselves feel tightly bound by regulation and bureaucratic process (Leigh, 2016). In secure children's homes, visiting social workers have significant influence over placement decisions and act as an advocate in court where placement decisions are made. Boundaries are more fluid in residential settings and staff often establish close links with those for whom they provide intimate care (Leeson, 2010). Hence, while a visiting social workers may act within a 'snapshot' of their clients' lives, the role of a residential staff plays out over a prolonged period, within the context of everyday routines, rituals and shared experience (Winter et al., 2018).

| Care and carers
The concept of 'care' has been subject to sustained interrogation, with ongoing debate focused about the interrelationship, and potential mismatch, between care as practical support and care as emotional support (Finch & Groves, 1983;Noddings, 1984). The distinction between 'caring for' and 'caring about' continues to be important.
Though previous focus was placed upon the distinction between active care-givers and passive care-recipients (Milligan & Wiles, 2010), attention has shifted to consider the interdependencies through which care and caring relationships are constituted (Watson, McKie, Hughes, Hopkins, & Gregory, 2004). Moreover, as noted by Milligan and Wiles (2010, p. 738), 'the nature, extent and form of these relationships are affected by where they take place'. To date, care involving children has been considered in a number of institutional contexts including both non-residential settings (Boyer, Reimer, & Irvine, 2013;Colley, 2006) and residential settings (Emond, McIntosh, & Punch, 2013;Steckley & Smith, 2011). In this paper, we attend to secure children's homes, a space that has previously received limited attention as a context for care.
The relationships that young people build with staff are crucial to the success of secure placements (Rose, 2002). In such settings, relationships become the therapy, and staff are tasked with the responsibility of reforming the emotional conduct of their residents while displaying care and concern (Perkin, 1990). Caring work is inevitably emotionally challenging, and often the emotional labour required is unacknowledged and taken for granted (Winter et al., 2018). In the context of the secure children's home, this emotionality is compounded by residents' enforced confinement (Coy, 2009). Thus, secure care constitutes a particularly charged emotional space in which carefully scripted interactions take place between children and the adults who are paid to care for them. While relationships between staff and residents are the lynchpin by which secure children's home are felt to be most effective (Rose, 2002), research tends to neglect the attitudes and experiences of employees who forge and maintain such relationships. This paper shares the views of residential staff, to explore the everyday challenges they face in managing their own emotions, with a view to creating a better understanding of this type of role.

| METHODS
Findings derive from an ethnographic study conducted over 12 months in a local authority secure children's home in England, named here as Hester Lodge. Ethical consent was granted by the university and the local authority responsible for the home. Data were collected using a range of qualitative methods, including participant observations, in-depth interviews and case file analysis. The researcher spent over 300 hr observing daily life, focusing particularly on everyday interactions between staff and young people. Participant observations were written up each day into a fieldnote diary and were complemented by a series of in-depth interviews conducted with five staff and 15 girls.
Although Hester Lodge was a mixed unit, the study focused on female residents to explore notions of vulnerability that are applied to girls in this context (Ellis, 2018). Twelve girls consented for their care files to be included in the study to show how professionals documented their 'progress' before, during and after placements. Consent was re-agreed at the beginning of each contact, and participants were reminded that they could withdraw at any time without sharing their reasons. All participants were informed of the instances in which confidentiality would be broken, such as if they disclosed that they were being harmed by someone or if they were intending to harm someone.
Data have been anonymised to protect the identity of participants and the home where the research took place. Fieldnotes and case file analysis were typed out in full and entered alongside interview transcripts into NVIVO where data were analysed according to a number of themes, including control, rules, identity, relationships, negotiation, feeling connected, care, concern and risk. Following data analysis, we used the theoretical lens of the sociology of emotion to interpret data contained within the themes listed.

| Participants
While the research prioritized the views and experiences of girls in secure care, this paper focuses primarily on the perspectives of residential staff. There were approximately 30 residential staff who ranged in age from early twenties to early sixties. They were predominantly white, with a small number of black employees. There was a roughly even male/female split, with marginally more female care workers than male. Though in-depth interviews were conducted with only a handful of residential staff, the ethnographic approach used ensured that everyday interactions between staff and young people were included, along with staff writing in case files, staff reporting in meetings and everyday informal conversations between staff and the researcher.

| Caring relationships: staff and young people
The nature of secure children's homes is akin to what Goffman (1961) refers to as a 'total institution'. Doors are locked at all times, and residents are not permitted to leave unless a court deems it so. Residents are contained to spaces where they are supervised at all times. Behaviour is reported upon in a group setting at regular intervals, and good behaviour is rewarded by points which can be exchanged for unit privileges, such as a TV or posters and books. There are strict protocols around most aspects of everyday life and unit rules dictate which clothes are 'appropriate' to wear, the types of language that are 'allowed' and the personal boundaries that are considered 'acceptable'.
Though strict rules were enforced, the atmosphere in the unit was overwhelmingly upbeat and staff endeavoured to create positive relationships with those in their care, as fieldnotes show: Case files confirmed that before entry, most residents had been living what professionals had termed, a 'chaotic life'; they were often reported as missing, did not attend school, used (or abused) drugs and alcohol, and committed crimes (Ellis, 2016). As noted by Bell (2002), residents confirmed that they had previously 'successfully' avoided professionals in their day-to-day lives, though in Hester, they were forced to interact with the same professionals all day and every day.
As a consequence, staff and residents recognized that daily life contained echoes of family life: It can feel really homely and it can feel like we are like a strange unique family. That's a really weird word to use but it is and that's what I like about it, that feeling that you know, you've built up those rapports with those people you know. (Dawn) Residential staff reported that they found their work 'rewarding' and that they often 'cared about' individual residents. However, they were emphatic that the unit was their 'workplace' and despite developing friendly working relationships, staff also maintained strict professional boundaries noted by others working with children in care (O'Leary, Tsui, & Ruch, 2012). While residents felt that they got 'to really know' the adults who were caring for them, staff reported that they maintained a 'guarded' presence that allowed them to show only a certain 'side' of themselves: They don't get to know us. They get to know this kind of, it's not false but they get to know this guarded side and only a little bit, because we don't talk about our lives you know (Jayne)

| 'Caring' for children in secure 'care'
Staff hoped to equip residents with strategies to manage everyday negotiations. To do this, they acted calmly and with unrelenting empathy when facing difficult or heated discussions and later encouraged residents to consider the impacts of their behaviour after flash points had passed. Although effective in de-escalating difficult situations, staff noted that managing a space with such potential for outbursts could be challenging. They explained that positive environments were maintained through the successful use of what Svensson (2002) terms 'coercive power'. Dawn explains: They could choose not to get out of bed, yet they know there are consequences for not getting out of bed. But then they know that we can't yank them out of bed either. So a lot of the whole day is built around the fact that they will be … I hate this wordcompliant. It wouldn't work if they all just said one day 'Oh none of us are getting out bed' … the whole day, we are banking on the majority doing what we tell them (Dawn) Sometimes, residents displayed extreme emotions, which acted as a stressor for maintaining group harmony. On one occasion, three girls 'commandeered a corridor' and refused to move. They banged their heads on the wall, threatened staff and threatened suicide.
While staff described such circumstances as distressing, they calmly acted to deescalate behaviour and helped residents to work through their emotions. Alfie recounted his management of a similar situation: Everyone was buzzing around and saying 'he's gone mad' … He was swinging and banging windows, kicking doors, smashing batteries into windows and doors, really aggressive. I looked at him and I said 'I know you don't want to hurt me, and I also know that you are very angry and you are very upset. I know that you need to talk, so what I'm going to suggest is, put the batteries down on the window ledge and we'll talk, you can stand or sit.' And do you know what he did? He just walked down and passed them me and I said 'thank you' and we talked, and it was as easy as that … I am scared, you have to have some fear in you, because you still don't know, you just go in and be calm.
Residential work can be both challenging and rewarding (Leeson, 2010). While research has shown that young people come to see residential staff as primary their carers (Punch & McIntosh, 2014), Hester staff reported that they also formed close relationships with individual residents: Sometimes you can't help it … you can get too involved  Ruch (2014) explains that professionals working with young people are expected to demonstrate 'humane qualities' and be honest, reliable and consistent. While these characteristics were embodied by Hester Lodge employees, they were also maintained alongside an undertone of risk aversion, which often took precedence over the nurturing relationships that residents hoped for (Ellis, 2018). Different opinions around the appropriate balance of nurture and risk aversion created vehement disagreements between staff, which were exacerbated by the dual purpose of the unit. While most described themselves as 'child focussed' and believed themselves employed to care for 'vulnerable young people', others viewed Hester Lodge primarily as a unit for offenders and were critical of staff who were 'over friendly' with residents:

| Maintaining risk averse boundaries
We should hug kids but it is more like a nursery … I know they're children but these are children who've broken the law. It takes a lot for a young person to be secured, to be locked up. These people have been deemed fit not to be allowed into the community, yet Darren and a minority of others appeared to identify their role as 'prison guards' instead of 'carers', which further highlighted the contradiction in the unit's function, since prison guards are needed to discipline (Foucault, 1991) and care staff are needed to 'care' (Hochschild, 1983). Research by Coady (2014) has shown that high profile scandals around child sexual abuse has led to 'significant professional anxiety'. Hence, residential staff have been reported to sometimes shroud their relationships in a 'veil of suspicion' to prevent accusations of abuse (Horwath, 2000;Steckley & Smith, 2011). Subsequently, residential staff were mindful of the boundaries they placed around their relationships with young people: You never ever trust a young person you're looking after. You might feel like you want to, you might be tempted to but you don't … when someone gets suspended from work because a kid says 'He inappropriately touched me' … then [the hugging] will stop.

(Darren)
Although staff acknowledged that they did care for residents, they were sometimes overpowered by the strength of feeling that young people returned to them. In situations where residents got 'too close', other staff stepped in to create distance. Fieldnotes show Jayne asking for help from her team to relieve the tensions around her relationship with Gracie: 'Gracie has a tendency to kick off when I'm on shift, if I buzz, can you put me somewhere else?' Shift manager: 'yes, let's not make an issue of it to Gracie though … Jayne, we'll find a simple way to put you somewhere else' Creating distance in this way helped to provide what Leeson (2010) recognizes as 'an emotional place of safety' for staff. Although these instances created safe spaces for staff, they had the opposite effect on residents who were mostly reluctant to share personal information with professionals. Instances like this were fundamental in reiterating to girls that they had been right not to trust professionals with their secrets (Ellis, 2016). Though distancing strategies used by staff were challenging for residents, they were felt to be vital in ensuring that residential staff did not experience emotional burnout and acted to pre-empt absence from work due to sickness or stress.

| Moving on
After carefully constructing close and trusting relationships, residential staff were expected to prepare residents to move on afterwards.
Since residents often likened Hester Lodge to 'the best children's home ever', preparing them to leave presented challenges that were difficult for carers to manage: She is refusing to eat and drink. She doesn't want to leave … she says that she wants to die one way or another (Janet) Staff reported feeling powerless and frustrated when external decisions made about young people did not coincide with their own views about what would be best for them moving forward. They want to come back a lot of them, we've had quite a few deliberately committed crime to come back here. (Terri) As special as these relationships were, they were only feasible for staff to maintain while the young person lived in the unit. Although they spoke fondly of those they cared for, staff also admitted that once residents moved on, another soon replaced them. This is illustrated by a conversation with Harry, with whom I shared my first day in Hester Lodge. He was newly employed and we were shown around the unit together. He told me then, 'I'm going to remember these kids'. A year later, in the final stages of data collection, I asked Harry if he knew how Brittany, his first keychild, was faring after leaving the unit. Despite his best attempts Harry could no longer remember her or where she went: There are hundreds of Brittanys. As soon as she leaves we'll get another one in. As bad as it sounds to say, it isn't possible to remember them all. (Harry) 3 | DISCUSSION Milligan and Wiles (2010) argue, the nature of the 'caring' self is constituted through complex, shifting relationships that emerge within, and are influenced by, aspects of place. 'Care' may manifest differently in different spaces, and in association with different forms of waged care work (Boyer, Reimer, & Irvine, 2013). The findings presented in this article contribute to current understandings of care by focusing on the caring practices of residential staff working in secure accommodation, a particular space of care.
The term 'care' has been scrutinized in social work practice, with changes in the U.K. Children Act 1989 meaning that children in care became 'children looked after'. The language of 'care' has also been subject to extensive contestation, with scholars differentiating between professionals who 'care about' children from those who 'care for' children (Noddings, 2002). While those who 'care for' tend to administer everyday nurture, to sustain basic, day-today needs, those who 'care about' tend to safeguard children's overall wellbeing, without performing direct caring duties. For instance, while residential staff ensure that a child brushes their teeth, a visiting social worker would confirm that the child has an adequate placement to live in. Thus, while visiting social workers may be able to distance themselves from the everyday nature of children's experiences (Winter et al., 2018), residential staff form complex relationships with children, built on shared experience and proximity (Punch & McIntosh, 2014). Moreover, residential staff are expected to manipulate their relationships with young people to simultaneously display care and concern, while working unilaterally to influence the development of behaviours that are deemed to be more suitable to young people's participation in mainstream society (Perkin, 1990;Rose, 1999). Such emotion work can be challenging in the long term. Hochschild suggests that waged carers can ease this internal tension by pulling personal 'feeling' and 'display'' closer together either by changing what we feel or by changing what we feign' (Hochschild, 1983, p. 90).
In their description of their relationships with young people, we have shown that residential staff draw, at times, upon narratives of family. The use of family as a metaphor for care resonates with findings from studies of child-care in other contexts (Boyer, Reimer, & Irvine, 2013;Brooker, 2016;Uttal & Tuominen, 1999), reflecting cultural assumptions that the family home is the 'true' site of caring work. However, there are important differences between care that is embedded in familial relationships and care provided in other contexts. These differences are particularly stark in the context of secure children's homes, where paid staff care for residents who are involuntarily detained.
Despite family-like allusions, Meagher (2006) argues that it is unsustainable to expect staff to forge family-like relationships with residents. Indeed, Meagher suggests that outside of the private sphere, care is not underpinned by (the ideal of) family obligation or love. Therefore paid carers have to draw upon alternative resources, which Meagher (2006, p. 35) describes as being 'the moral bonds of contract, of professional duty, and of compassion'. However, such bonds require emotional labour, as Hochschild reminds us, 'to be warm and loving toward a child who kicks, screams, and insults you requires emotion work' (Hochschild, 1983, p. 52). Furthermore, the behaviour experienced by residential staff could be extreme and staff sometimes bore the brunt of young people's frustration. Since they were paid to be there, staff were, as Hochschild suggests of other carers, paid to accept it (Hochschild, 1983, p. 186). While residential staff are clearly contracted to care, their contracts of employment are agreed between the state, or third-party organizations, and individual employees. Secure children's home residents are not, of course, party to such contracts and they have no right to terminate a contract if they are unhappy with the care that they receive. The findings that we have presented in this paper do, however, lend insights into the importance of professional duty and compassion in the constructions of care and the emotional labour that residential staff undertake.
Staff created an environment of measured calm, within which they modelled desired patterns of behaviour management. As they described their care work, staff illustrated the professional norms and expectations that informed their obligation towards residents and the nature of the relationships that they constructed with young people.
Hochschild points out that particular occupational contexts require staff to practice emotion management, to project a certain persona to complement the role of the organization in which they are employed, which requires them to enact a particular 'facial and bodily display' (Hochschild, 1983, p. 7). Residential staff therefore carry out emotion work 'in the classic sense of needing to regulate the emotions of oneself and another' (Boyer, Reimer, & Irvine, 2013, p. 527).
Hester staff acknowledged that they displayed a particular version of themselves to residents and illustrated their mastery of 'face work', described by Bolton (2001) as 'the ability to present a face that is appropriate to the situation at hand' and 'surface acting' (Boyer, Reimer, & Irvine, 2013, p. 527). While such emotion work has been widely commented upon in other contexts (for example, see Osgood, 2012), the reciprocal nature of relationships between staff and residents in the secure children's home inadvertently posed additional challenges to residential carers' universalising professional ideals, as they became intertwined in the intensity of residents' emotional journeys (Steckley & Smith, 2011). 3.1 | Compassion, affective relationships and emotion work Stone (2000) argues that paid carers commonly draw upon contradictory ideals, meaning tensions occur when staff come into conflict with their with their commitment to equality and treating every resident equally comes into conflict with the emotional bonds formed with a specific 'deserving' resident. These conflicts can place staff in a 'moral double bind' (Stone, 2000) and require intensive emotion work, as we have illustrated by Alfie's fierce defence of a young person who was kept in the dark about an imminent placement move.
Our data highlights the significance of compassion as a resource underpinning the provision of care within Hester Lodge. Compassion, Meagher (2006) argues, 'is based on cognitive judgements -judgements about the neediness and the worthiness of the recipient'. Paid care workers' negative views towards cared-for children's families have been reported previously in other contexts (Vincent & Ball, 2006). Staff in our research also illustrated antagonistic views towards residents' families, who were depicted and imagined through case-notes, but infrequently encountered. Hester staff recognized the 'chaotic life' that residents had experienced before being secured and absolved them of-at least some-culpability, as Jennifer demonstrated when noting that Brittany, her keychild, was not at fault: 'she's only copying what she's learned'. Our data suggests that staff experienced the 'moral double bind' (Stone, 2000) as they strove to balance professional ideals with compassion in their day-to-day interactions.
Staff acknowledged that they became emotionally invested in the wellbeing of the young people they were working to reform, and data illustrates real affection for residents, particularly towards individuals for whom they had keywork responsibility. As Boyer et al. (2013, p. 519) have noted, strong affective relationships can develop 'in the context of commoditised care' even though these relationships are embedded within fixed and unequal hierarchies of power.

| Limiting the self and maintaining balance
Both professionalism and compassion make strong demands on the character of paid care workers (Boyer, Reimer, & Irvine, 2013;Meagher, 2006), and both are integrally bound up with emotion management (Osgood, 2012), albeit in potentially contradictory ways. Our data illustrate residential staff's concern for, and their strategies to maintain, a professional distance between themselves and the residents in their care-by, for example, limiting one-to-one time with specific residents. Such distancing served to reduce carer stress (Perera & Standen, 2014) and protect staff from becoming too emotionally invested. In contrast, staff described experiences of professional anxiety, when compassion and affective relationships challenged, and could override, professional norms, such as when residential staff felt the need to contest unit rules in order to optimize care for individual residents. They frequently shared their frustration when they felt that other professionals made unwelcome decisions about young people that they cared for, since bureaucracy meant that they lacked authority to override external decisions relating to residents (Leigh, 2016).

| CONCLUSION
This paper explored the complexities of relationships in secure care and examined the emotional and relational dynamics between residential staff and the young people they were paid to care for. It considered the different, normative resources that staff draw upon to constitute 'appropriate' care as well as the associated caring practices and emotion work staff undertook (Colley, 2006). In so doing, we have illustrated the co-existence of, and potential for conflict between, professional and compassionate bonds between staff and residents (Meagher, 2006). By illustrating some of the complexities' of providing intense support within the secure children's home context, the paper supports Boyer et al.'s (2013, p. 519) problematisation of the 'dominant deficit understandings' of paid, institutional care. This paper explored the views of residential staff in their everyday working practices, with the aim of understanding the day-to-day challenges of supporting vulnerable young people in a confined environment. We argue that staff come to care for the young people that they are paid to care for, and that structural boundaries created conflict for those who lacked authority to overturn unwelcome 'care' decisions made outside of the home. Such 'unwelcome' decisions brought feelings of frustration that were difficult to manage and sometimes tempted residential staff to break professional protocol.
Despite the close relationships that staff made with young people, they nevertheless enacted their 'care' with caution, understanding that they were working under 'a veil of suspicion' that meant they should not become 'too close'. For those at risk of becoming 'too close' colleagues acted to create emotional distance, meaning that residential staff were able to maintain their own wellbeing, albeit sometimes at the expense of the young person's wellbeing.
This paper aims to share the voices of residential staff with the intention that those supporting them are better able to understand the tensions they balance and thereby provide appropriate support. To do this we consider the emotion management that is expected in such a role, and the implications for those seeking to support young people in difficult circumstances. While the voices of those experiencing services must be listened to and used to improve services (Ellis, 2016(Ellis, , 2018(Ellis, , 2020, it is important that these voices are heard in conjunction with the professionals and practitioners charged with providing the care that is so crucial to the support of our most vulnerable young people.

This PHD was funded by the Economic and Social Research Council
(grant number ES/F021860/1).

CONFLICT OF INTEREST
None.

DATA AVAILABILITY STATEMENT
Research data are not shared.