'Neither a professional nor a friend': the liminal spaces of parents and volunteers in family support

Home-Start is a family support charity whose delivery model is a national and global example of how targeted volunteer support can benefit parents, carers and children experiencing difficult times, in both domestic and other spaces. Parenting support continues to be a key policy area for the current UK government and other policy-makers across the Global North. In this article we draw on qualitative findings from an ethnography of a Home-Start organisation in a city in the north of England. The theoretical framework of liminality, a space between social structures, allows for an appreciation of the ambiguous nature of supporting parents in the private domestic spaces, and the ways in which this support enables parents and families to move forward. The article has broader implications for global social care and social work practice, specifically demonstrating the importance of the relationships between parents and volunteers in the every day, and contributes to the literature on liminality.


Introduction
Since 2010 there have been austerity measures in the UK and across the Global North, instigated by the national and global financial crises (Clarke and Newman, 2012), and 'politically reframed from an economic issue to a problem that can be blamed on the welfare state and its dependents' (Fisher et al, 2014: 39). Austerity continues to be a dominant feature in shaping social care services in the UK (Slay and Penny, 2013), and increasing numbers of families are experiencing difficulties in everyday life (Orton, 2015). Central to the austerity narrative are widespread public sector welfare article funding cuts across social care, housing, criminal justice and health, and these are not exclusive to the UK (Wiggan, 2012). At the same time, current UK government policy continues to place parenting at the centre of children's outcomes, including a focus since 2011 on both troubled (DCLG, 2017) and hard-working families (Crossley, 2015;Runswick-Cole et al, 2016).
Previously, family support has been conceptualised in a number of ways including social capital (Strange et al, 2016), social support (Sheppard, 2004;Wisso and Plantin, 2016) and empowerment (Vuorenmaa et al, 2016 ). For this article we deliberately chose to focus on liminality as a lens to understand families' perspectives of support. This was influenced by the thematic analysis (Braun and Clarke, 2006) that highlighted the in-between nature of Home-Start volunteers. In this article we draw on qualitative research undertaken between 2013 and 2014 in a city in the north of England with a Home-Start organisation that provides voluntary support for families experiencing difficulties with children aged five and under.
First, we explore how policies positioned families within the austerity rhetoric of the UK government, and provide some context to family support in the UK and globally, including a brief outline of the role of Home-Start. Second, we consider the theoretical framework of liminality (Turner and Turner, 1982), a space between social structures, and a distinct time and space, and argue that it provides a useful framework for understanding the effectiveness of the practices of a voluntary organisation in supporting families. Third, we locate our critical realist, theoretical perspective methodology, prior to a discussion of the ethnographic findings in relation both to liminality and two key themes identified through inductive analysis: the coping experiences of the mothers and the relationships between the volunteers and the mothers. Finally, we contribute to the health and social care literature through our focus on liminal spaces of parenting support, and argue that the concept of liminality allows for an understanding of everyday experiences of family support in domestic spaces provided by a voluntary organisation.

Family shaped through policy
The role of government in family and parenting has a long history within the UK and in other European countries (Cornford et al, 2012). Family is a contested and contestable concept, and while space precludes a detailed exploration of what is a 'family', Morris et al (2017) provide a useful discussion of contemporary theories of family. For the purpose of this article, a family can be identified as a parent(s) living with a child(ren) in the same household. It is acknowledged that parents may be of the same sex, or where there are two adults in the household, one adult may not be the parent of all the children or any, while families may also be constituted through legal proceedings like adoption. Parenting, families and family support programmes continue to be the focus of research across academic disciplines including psychology, sociology, social policy and public health. However, there is limited research concerning the role of voluntary sector family support organisations that work with families assessed as requiring support. There are some notable exceptions to this, including research by the Family and Childcare Trust (2013). The increased interest in 'family' for policymakers and implementation in policy and programmes can be traced back to the UK New Labour government (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010) with the introduction of the Childcare Act 2006, Sure Start and the Every Family Matters agenda (Cornford et al, 2012). This policy focus on the family was continued by the UK Coalition government with the Troubled Families programme, a deficit-focused model with a lack of regard for the structural inequalities that families can experience (Bunting et al, 2017), and Crossley (2015) provides an excellent exploration of this. For the previous UK Coalition government , and the UK Conservative government at the time of writing, policies, including those focused on family (and for family, read parent; see Murray and Barnes, 2010), have been underpinned by a neoliberal ideology, framed by minimum intervention from the government and an emphasis on individuals to take personal responsibility for their family within society (Hartras, 2014). Parents are expected to be responsible and resourceful citizens (Jensen and Tyler, 2012;Runswick-Cole et al, 2016), able to adapt and manage within a fragmented austerity-based social care system (Jensen and Tyler, 2012), and able to assume blame and stigma for a myriad of social issues (De Benedictis, 2012). De Benedictis (2012: 2) states that policies place 'parents as responsible for their offspring, the economy, the locality and the prosperity of society overall.' Turning to austerity measures, anxiety and insecurity continue to have an impact on family life (Orton, 2015), with negative impacts (Family and Childcare Trust, 2013) for those on low incomes and increasingly the 'squeezed middle' (Orton, 2015: 22).
The last two decades have seen a shift from local authority preventative services for families with young children to those centred around community and localities (Cornford et al, 2012), within a landscape of unequal service provision (Needham, 2015). Here, we recognise that Munro (2011) identified the importance of providing 'early help' services and differentiate between help and intervention in her review. We use 'support' here rather than 'help' as this implies a more co-produced approach, and we avoid the term 'intervention', concurring with Featherstone et al (2014Featherstone et al ( : 1742 that early intervention is: ...a future-oriented project building on elements of social investment and moral underclass discourses. It incorporates an unforgiving approach to time and to parents -improve quickly or within the set time limits. Within the landscape of the UK New Labour government (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010), support for families was framed in partnership working and the increasing role of the third sector. The Big Society introduced in 2010 by the UK Conservative-led Coalition government was a central aspect of this policy (Taylor, 2011), and provided a strategy for social action and encouraging volunteering, with a re-imagined and reduced role for the state. Civil society organisations were encouraged to meet the gaps in a rolled back welfare state, particularly faith-based bodies and local charities . Pressure has increased on families to be resourceful in the face of welfare cuts, with a shift to indvidualised support within domestic settings (Kraftl et al, 2012;Needham, 2015). Support services are unequally distributed across the UK, often dependent on the voluntary and community sector (Morris et al, 2017). Previously available preventative community-based services such as Sure Start, an area-based universal programme that provided services to families with young children, particularly in disadvantaged localities, were closed down within the context of austerity (Kraftl et al, 2013).
As part of the austerity measures the local authorities with reduced funding have increasingly moved to focus on statutory and high-risk families, reducing the funding for preventative work. This has resulted in organisations like Home-Start becoming co-opted to deliver on local authority agendas and less able to be develop innovations to connect that enhance social capital in communities. Amidst ever increasing pressure on social workers, volunteer-led organisations are progressively delivering support to families who do not meet the increased social work thresholds yet are experiencing difficulties (Family and Childcare Trust, 2013). This support often takes place in less visible domestic spaces, provided by volunteers, and is about working alongside families within neighbourhoods (Featherstone et al, 2014). We now consider the Home-Start model of support.

The Home-Start model of support
Supporting families with young children in their own homes is an extensively used approach that is 'attractive to professionals and policymakers because of its low-costs and accessibility' (Hermanns et al, 2013: 678). The concept of Home-Start has spread globally, and there are volunteers supporting families in over 22 countries, including Japan, Germany and in Africa. Home-Start UK (2016) describes itself as: ...one of the leading family support charities in the UK. Home-Start volunteers help families with young children deal with the challenges they face. We support parents as they learn to cope, improve their confidence and build better lives for their children.
UK Home-Starts have their own board of trustees or management committees, are social franchises, and are supported by Home-Start UK. Each Home-Start scheme can employ a manager/coordinator and administrators. Coordinators are responsible for responding to referrals, recruiting and training volunteers and matching volunteers to families. In the main, families are referred by professionals including health visitors, social workers and teachers, with some self-referrals. Once a volunteer has been matched to a family, the coordinators oversee the support. Volunteers are required to have parenting experience, and undergo eight days pre-volunteering training including safeguarding, attitudes and beliefs, child development and signposting to services. Access is universal with no fixed criteria except that the family must have at least one child under the age of five.
The majority of research about UK Home-Start schemes has been undertaken or commissioned by Home-Start UK, and is evaluation research based on quantitative methods. We recognise that voluntary organisations are increasingly required to demonstrate quantitative evidence of performance outputs at a local and national level (Glasby and Dickinson, 2014) in the context of a general shift towards evidence-based commissioning. Frost et al (2000) undertook a study on families receiving support from Home-Start organisations in the north of England, and identified a positive impact on children's wellbeing that continued for up to three years after the volunteer support stopped. This reflected previous research by Gibbons and Thorpe (1998) that focused on parenting support, early intervention support for families experiencing stress, and the impact of the support on children's behavioural patterns. A Joseph Rowntree Foundation research study (McAuley et al, 2004) found that mothers who had been supported by a Home-Start volunteer valued the service and agreed that it had had a positive impact on their lives as parents. However, the authors concluded that there was no statistical significance to evidence that the Home-Start support was cost-effective or had a significant impact on the mothers or the children's emotional and social development compared with mothers who did not receive support from Home-Start. McAuley et al (2004: 63) also noted that their report should not be read as a mandate to reduce funding, but for Home-Start to reflect on the 'needs of the families it serves' and 'to consider more intensive support'. Barnes et al (2006Barnes et al ( , 2009) also found that Home-Start peer support work reduced parenting stress but had little impact on parental mental health, which they felt was better provided by time-limited professional input. The National Institute for Health and Care Excellence (NICE) evaluated programmes to develop guidance for early interventions with vulnerable children under the age of five. Schrader-McMillan et al (2012) reviewed home visit interventions by professionals, video interaction guidance, and by paraprofessional layworkers. They concluded, based primarily on McAuley et al (2014) and Barnes et al (2006Barnes et al ( , 2009, that professional time-limited support was more effective than Home-Start provision. However, for whatever reason, health visitors continue to be major referrers to Home-Start. In other countries, Asscher et al (2007) concluded that aspects of the Home-Start programme -intensity and length of support -are more effective with lower-economic status families, while Hermanns et al (2013) undertook a randomised control trial and concluded that Home-Start provided long-term changes in parental and child wellbeing as a result of support. The evidence on the effectiveness, and in relation to which outcomes, of Home-Start is thus open to challenge and dispute.

Theorising liminality
Liminality, from the Latin word limen (meaning a threshold), was coined by a French ethnographer, Arnold van Ganeep (1908), and further expanded on by British cultural anthropologist Vincent Turner (1969Turner ( , 1977. Van Ganeep (1908) viewed liminality as concerned with rites of passage and transition between one social status and another within a time and space. Turner (1969) expanded the concept of liminality to cover any period of change or transition (fluidity) in people's lives and applicable to more complex and larger societies than the indigenous tribes studied by van Ganeep (1908). For Turner (1977: 68), the experiences and practices of liminality have the potential to create changes in identity, and was located in 'the meaningfulness of ordinary life' and being 'betwixt and between'. For Turner (1969), the notion of communitas was an important element of liminality -a shared space, inhabited by those in the liminal stage, where connections are formed. Communitas is about shared experiences of liminality (Turner, 1977), commonalities and a sense of belonging between people who are in liminal spaces.
Within health and social care, liminality has been utilised in a number of studies, notably using the ideas of 'betwixt and between' and out of place. Morgan (2012) has also extended the concept to social work training. Roberts et al (2014: 460) identified the 'blurred voluntary/practitioner boundaries' of voluntary community first responders' involvement in rural medical emergencies in Scotland. Mahon- Daly and Andrews (2002) and Dowling and Pontin (2017) applied the concept of liminality to studies of breastfeeding, and drew on communitas as a way of being in-between, on the margins and inferior. Liminality has been applied in research of parenting preterm babies (Watson, 2010), the role of privately funded companion care in a Canadian long-term care facility for older people (Daly et al, 2015), foodbanks in the UK (Cloke et al, 2016), and mental health support work (Warner and Gabe, 2010). While liminality has been applied to experiences of parenting, it has been absent in studies of family support programmes.

The study
We draw on a critical realist perspective (Bhaskar, 1998;Sayer, 2000) that considers the world as layered and views experiences as outcomes of the interrelatedness of individual and structural causes, effects and tendencies (Sayer, 2000). In our study we explored the experiences of the parents and volunteers of a family support organisation as they described them. The ethnographic research took place with a Home-Start organisation in a city in the north of England, and here we consider the interviews with parents and a focus group with volunteers undertaken in 2013 and 2014. The research was subject to Manchester Metropolitan University's Ethical Review panel prior to undertaking the fieldwork, and research governance processes of the university were followed. Key ethical issues included the need for participantinformed consent, anonymity of participants, data confidentiality and safe storage of data and documents. The Home-Start organisation, within which we located the study, covers a large geographical area to the south of the city that includes a large conurbation that was formerly dominated by local authority housing.

The participants
For the research we interviewed eight parents receiving support from the organisation (see Table 1), and undertook a focus group with active volunteers. We used a purposive sampling strategy that 'works with small samples of people, cases or phenomena nested in particular contexts' (Gray, 2004: 324). Potential participants were first contacted via the organisation's staff. Those who wished to participate were asked to contact the researchers (Mitchell-Smith and O'Neill), and were provided with research information and ethical considerations, and if still interested, an interview was arranged prior to which a consent form was signed. All of the parents and volunteers were female. The participants were reminded at the start of interviews and the focus group that they could withdraw from the research at any time, and permission to audio-record was sought. All of the parents understood that the support from the Home-Start organisation would be unaffected by their participation in the research. Four white British volunteers participated in the focus group and their age ranged from 22 to 75.

Data collection and analysis
Individual interviews lasted between one and two hours and were undertaken at the family homes with the family's agreement, and the focus group with volunteers took place at the Home-Start organisation's office. In the interviews the participants were asked about the length of time they had had a volunteer for, how and why they were referred to Home-Start, and how they felt about the support they had received from the organisation and the volunteer. They were also asked about the type of support the volunteer provided, when they expected the support to end, and how the support made them feel. As with the interviews, the focus group was conversational in style (Morgan, 1997). The volunteers were asked about their motivations for volunteering, their experiences of the training, the impact of the organisation on the families and communities, challenges in their role, and their experiences of being a volunteer.
The interviews and the focus groups were audio-recorded, transcribed verbatim and anonymised. All of the authors of this article undertook thematic analysis inductively   (Braun and Clarke, 2006). We read the transcripts from the interviews and the focus groups individually, continually coding the responses and then allocating them to themes and sub-themes until all data were accounted for (Connolly, 2003). Fisher, Lawthom and McLaughlin discussed the codes, re-coding and co-constructed themes and sub-themes, supported by Mitchell-Smith and O'Neill, who had undertaken the qualitative research and drew on their notes from 'physically being in the field' (Mauthner and Doucet, 2008: 977). This 'being in the field' included attending volunteer training and organisational events. It is important to note that all of the women who spoke to us were positive about their experiences of being supported by the organisation, supporting previous findings by McAuley et al (2014) and Hermanns et al (2013).
The key themes identified were: the mothers' experiences of coping with difficult issues in their lives; the uncertainty surrounding the ending of support from the organisation; and the ways in which the mothers spoke about the support from the volunteers. Within each theme, the participants' descriptions were about a state of 'in between', and this invited the consideration of liminality, the focus of this article.

In between 'coping' and 'not coping'
Mothers 2 in our study told us that they had been referred to Home-Start as they were feeling unable to cope and manage with caring for their children, life events and everyday activities. For some, they had been coping and then something happenedthey had a new baby, for example -or it was a combination of factors, for example, their partner had to work away from home, they felt isolated and their child became unwell. Referrals to the service were by health visitors, social workers and midwives, when the women reported being at a self-identified 'threshold' point of feeling unable to cope or a professional identified that they weren't coping. Rugina (P7) stated: 'Now I totally understand how people have a nervous breakdown. Before obviously you can read about how it can happen, but when you experience extreme stress, and it's a prolonged thing not just an overnight thing, weeks, months. I totally understand how bad it can get, I was just really upset, it came to a point where I just felt like walking out of the house, with all my kids in the house, just walking out.' Many mothers (and fathers) experience difficulties following the birth of a child, a change in family circumstances, returning to the workplace or a multiple birth. The women in the study remained in a liminal state of feeling like they were not coping, and for some this lasted months or a year. These parents can be seen as in a 'betwixt and between' space as their experiences of parenting are different from those who have managed to cope after a change in circumstances. In the interviews, participants compared themselves with other mothers they knew who were seen to be coping emotionally and practically. Mahon- Daly and Andrews (2002: 63) assert that women experience difficult life events and troubling experiences throughout their lives, including 'adolescence, pregnancy, puberty, childbirth, menopause and death, as well as the more unpredictable, illness and sudden disability. ' Coping, a psychological concept, has been explored in research with mothers previously (Sheppard, 2005), and we did not explicitly seek to explore this. However, it was evident that some of the women accepted they were unable to cope, in that they stated that they felt over-whelmed, had no one to turn to for support, and others had denied that they were unable to cope until they had been referred to Home-Start. In many ways they did cope, and children attended school, meals were made, appointments attended, yet there seemed to be a threshold point where one unexpected event could "push me over the edge" (Rugina, P7). For some, particularly those who were more affluent, there was a feeling that they should be able to cope and that they should not be taking support that may be needed by poorer families; being 'in between' coping and not coping made mothers feel they were not eligible for support, as Samantha (P6) described, 'I was kind of like don't worry I'm sure there's other families that are more maybe in need.' Being in a liminal space, having not reached a point of clearly 'not coping' and therefore not having a clear need, makes it difficult for families to identify if support is 'deserved' or if their state of being in between coping and not coping is part of 'normal' parenting experience: '...you work with people who are struggling, but then when you are experiencing similar kind of things you're very sort of like well is this a problem, or is this normal.... I think there is a bit of a guilt about not being deserving.' (Samantha, P6) Many of the participants talked about feeling isolated, and often felt set apart from others and from communities. For some, this was because their husbands were working on a short-time basis away from home, and they were in a liminal space of not having a partner for weeks at a time. Living with in-laws was also viewed as setting them apart from other peers.
There were significant feelings of disconnection from others, particularly when mothers had no family living locally. Claire (P4) highlighted the importance of this in saying about the volunteer who supported her: 'We don't have a female family member close by, but it feels like she's a grandma to our family, it's lovely, so it's nice to have that maternal kind of experience coming to help us.' When first referred to Home-Start, mothers felt isolated and lacked social support. Elizabeth (P8) explained: 'I was quite isolated I didn't really have any friends, I did slowly start to make them but if you start making friends, you don't really want to start talking about all of your business.' The mothers often developed relationships with other parents during the time they had a Home-Start volunteer, and volunteers supported this, accompanying mothers to parent groups and Sure Start centres. This is illustrative of a liminal phase in which parents lacked communitas, or shared experiences of liminality that Turner (1969) suggests is important for transition. The volunteers aided the development of communitas and, having parenting experience themselves, the feeling that they 'understood' was reported by the families as central to the effectiveness of the support. Having the volunteer to support the mother through the transition was valuable and helped the mothers to move from feeling incompetent to feeling able to manage independently. This included coming to new understandings of coping, reframing issues in relation coping and developing new strategies to cope. Turner (1977) further considers identity change and liminality. Becoming a parent is a significant change in identity, and participants in this study reported feeling a sense of loss of identity or a change in identity from one of coping to not coping. Rugina (P7) and Samantha (P6) both referred to the change from working in careers where they felt in control to becoming a mother and then struggling to cope. They felt they should be able to cope as they had in their working roles. Rugina (P7) made clear the way she felt that 'not coping' was in conflict with her self-identity: 'I knew I really needed the help but I'm the type of person I really want to get on with things and do everything on my own.' A liminal space can be one where a parent is between one identity and another. This was experienced by the participants in our study in relation to 'not coping' as a mother in comparison to coping in previous work roles. May (2008: 471) states: In Western countries, motherhood is part of a powerful nuclear family ideology that permeates all of society and is defined and delineated by strong social norms.
In 'not coping' the mothers in our research felt that they were not meeting these norms. Other cultural expectations also contributed to the feeling of not living up to the required 'mothering' roles, for example, Asma (P2) and Rugina (P7) both spoke about in-laws and cultural expectations to fulfil the role of a 'good wife' and 'good daughter-in-law' having an impact on their ability to provide care for their children. Galam (2016) explored the social significance of the house and conflicts within the house including feeling monitored and having to conform. For the women interviewed for this study who lived with in-laws, this led to feelings of being unable to cope due to difficult relationships, conflicts over housework, cooking and childrearing, and lack of sleep. These mothers had practical help at home from older generations, yet still experienced the liminal stage of feeling unable to cope, and were referred to Home-Start.
In challenging cultural expectations, mothers further moved into a feeling of liminality as they rejected family and community norms, and therefore elements of their own identities. Kenworthy Teather (1999) argues that crossing thresholds is part of moving forward through troubling times. Parenthood, in conjunction with other pressures, had brought the women to a point that was emotionally draining (Vincent et al, 2010). It is important to note, here, that while all parents may experience periods of being between coping and not coping and between identities, the participants in this study had additional issues that contributed to their remaining in a liminal space. These included domestic abuse, multiple births, having multiple children under five, a lack of family support, isolation and children with complex health or developmental needs.
For Shields (2003), liminality enables a transformation from one social status to another, and the space in between is where a person is 'betwixt and between' a stage. Our findings illustrated that the mothers were in a space between feeling like they were good mothers and coping, and being bad mothers for not coping. The nonjudgemental support from the volunteer allowed the mothers to be in this liminal space and accept this as a stage that was not about being a 'bad mother', even if they needed support in coping. Elizabeth (P8) recognised the ways in which her own mental health was having an impact on her children, and thus her feeling of being a 'good enough' mother.
Although we are unable to provide generalisable or quantifiable evidence for this, it is implied that they felt temporarily better when the volunteers visited, and more able to cope. Volunteers told us they would contact the mothers after the support had finished and sometimes the parent had not managed to continue to cope without the volunteer. One volunteer (V2) stated: '...after I'd finished, about a month later I rang her up and said I'd come round and I said, have you been to any Sure Starts, and I could tell she hadn't, and that was sad because that was the whole point, really, of going, was to get her to make friends.' In this theme we have discussed the work of Home-Start as operating in a liminal space where volunteers work with parents who are between coping and not coping. In most instances this appears to prevent 'not coping' to the point where the intervention of statutory services is necessary. The findings from this study suggest that liminality is helpful in highlighting that the mothers experienced positive transitions that were supported by Home-Start volunteers. Although we are unable to provide evidence for this, it is implied in the reports of families, volunteers, staff and trustees and further in the research of Hermanns et al (2013). One volunteer (V3) explains this as follows: 'One family I had, the mother was ill, she had an illness and she was on her own, completely on her own with this little girl who was about a year old, and she wanted to put the little girl into foster care but she didn't, she kept her and after a while she ended up being fine, after a few months her health improved and she was okay but, I'm not saying it's just because of me, but if I hadn't been going in that child could have ended up in care....' It was not always the case, however, that the parents were supported in moving through the liminal space and into a positive transition, and two of the volunteers (V1 and V4) told us about how their support had not prevented children being taken into care.
There is a concern about the provision of individual support that could be seen as identifying the problem lying within the individual and pathologising their inabilities while neglecting the importance of the impact of social class and structure and support for the family. This is a legitimate concern that Home-Start can be seen as part of an individualised service that needs to begin to aggregate its data to challenge conceptions of parents who are 'not coping', and we argue for a wider system of community asset-based approaches to supporting parents (DH, 2017a).

Neither a professional nor a friend
We found that the volunteers were admitted to the families' homes in a different role than professionals, a liminal space 'betwixt and between' a professional and a friend. While liminality can be positioned as a negative experience (Turner, 1969), here it was mainly a positive experience for both volunteers and the mothers. The regularity of the weekly support facilitated an instant re-entry into the relationship between the volunteer and the mother, with mothers looking forward to the volunteers' visits and valuing that the relationship was distinct from with a professional. They said that volunteers 'chose' to work with them and they felt grateful to them for giving up their time. The volunteers were not viewed as being untrustworthy, as professionals could be: '...other professionals do understand but they have their own agendas, their own thing and so you can't really trust them, at least that's how it feels.' (Elizabeth,P8) Further, the mothers spoke about the importance of volunteers' ability to be in between a professional and a friend, and therefore in a liminal space: '...she understands, but at the same time she's not a friend, I wouldn't want a friend to sit next to me and hear all the personal stuff, but it's nice to have someone there who isn't a professional.' (Elizabeth, P8) The volunteers occupied an in-between space, a role that is different from the conventional relationship between a service user and a professional. Elizabeth (P8) explained that the volunteers were not their friends. Family homes are private spaces, and inviting friends to the domestic space can be problematic for some mothers (Jupp, 2013), and '…it can be a risky strategy of social differences of wealth or cultural capital that were of marginal importance in the original context of the friendships are now made more apparent' (Bowlby, 2011: 616). The volunteers were viewed as significant people in the mothers' lives, and important for their wellbeing. Liminality assumes movement from one state to another, yet the volunteers appeared to be in a state of betweeness that they did not move from, as neither a professional nor a friend: "but really I think it's a friend, somebody going in and doing it because they want to" (volunteer, V3). The liminal role removed the boundaries that clearly define the amount and type of work that might be explicit in professional relationships, and some of the volunteers felt a sense of responsibility to provide support beyond the hours agreed through the Home-Start organisation. However, there were examples of clearly defined boundaries where volunteers were not allowed to provide childcare or support outside of the agreed times.
Recalling Turner's (1977) concept of communitas, where connections are formed and people act in ways that are outside the usual social limits, findings showed that the mothers and volunteers had formed relationships where rules were temporarily revised. Liminality is not always a clearly defined space and blurring of boundaries is a possibility. The meshing of a home space and a formal parental support service led to some suspension of rules. For example, the volunteers were advised not to change nappies, yet in the chaos of everyday life, when a mother was caring for one child, the volunteer would pragmatically change the nappy of another child. The dividing line between what should happen and what did happen was fluid and 'betwixt and between', representative of the liminal role of the volunteers.
One way in which liminality can be seen as problematic in this research was that 'being between a friend and a professional' left both the volunteer and the families unsure about how, if and when the relationship would end. Some of the mothers viewed the volunteer as a friend, "she's like my friend really, she is like a really good friend, we've become quite close in that sense" (Rugina, P7), and it was similar for volunteers: 'I think you get friendship as well. Before I did this I didn't really think I would gain friends from it at all but you do, because you're with somebody for those few hours every week and really you don't see other friends for that long.' (volunteer V2) Although the nature of the liminality of the support was reported to be beneficial in enabling the families to move through their own liminal stage, when support ends there is a move towards a professional relationship in acknowledging that it has to end. For van Ganeep (1908), people leave communitas in the post-liminal stage, but for the families, they understandably became connected to their volunteer and wanted things to remain the same: 'I don't really know how to end it. It's kind of in between a friend and a professional but there aren't really guidelines for how to deal with that, like, do we say goodbye or do we have a cut-off point, kind of, how to manage that relationship, it's a bit strange and I think awkward from both sides.' (Elizabeth, P8) Yet as volunteers they had to be allocated to another family when the mother had developed coping strategies including new support networks. Volunteers found this a difficult transition to manage as they had developed friendships with the mothers, and some continued to be friends unconstrained by the volunteer/user of service dyad. Friendships are complex, and for some, they don't endure for ever (Smart et al, 2012), and these friendships were developed through an organisation and not embryonically. Liminality assumes movement from one state to another, yet the volunteers appeared to be in a static state of betweeness.
In recent years, as discussed earlier, there has been an increasing deprofessionalisation of family support, with social workers and health visitors increasingly being focused on high-risk cases, and preventative work being outsourced to the voluntary and independent sector (albeit limited in this period of austerity). This has resulted in many employees and volunteers from voluntary organisations, like Home-Start, becoming almost like paraprofessionals (Leger and Letorneau, 2015). They work in roles similar to professionals (for example, social workers and health visitors), yet are not professionals, managed by a formal organisation and subject to the rules and regulations of the organisation while valuing the liminal role of being in between a friend and a professional.
A commonly raised theme by the mothers was the uncertainty surrounding when the volunteer's support would end. Generally, the organisation provided support for one or two years. A key issue for the mothers was a lack of a transparent process about how the volunteers would start to withdraw their support, as illustrated by Rugina (P7): '...you don't really know whether you're coming or going with them in terms of how long your volunteer's going to stay with you.... I don't know when the [coordinator] is going to come and assess me and say that's it now.' In this 'betwixt and between' space, the mothers did not want to 'lose' their support. Reviews of the volunteer support take place that are documented by the organisation's coordinator, but there was no clear guidance for the families as to when the support would end. Support from the Home-Start organisation is premised on the fact that families will move on in their lives, and no longer need a volunteer.

Reflections on the study
The richness of this data provides a lens on voluntary organisations that provide family support, and explores mothers' experiences, yet there is a gap in similar research on fathers, particularly around support from voluntary organisations. It is possible that women are more likely to identify the need for help and be primary caregivers, and benefit from sharing an identity with the female volunteers. Fathers did not take part in our study and there were no male volunteers in the organisation at the time of our research. This study, in contrast to previous studies, is qualitative, and the small sample of participants located in one Home Start organisation precludes the generalisation of the findings. We did not include children in our research, unlike many of the previous studies. However, we suggest that it would be appropriate to reconsider parental mental health and children's wellbeing and the cost benefits of Home-Start, and whether the changes made by Home-Start since the previous negative evaluations have made any significant impacts on the experiences of parents and children.
The participants may not be representative of all parents and volunteers and may be influenced by gatekeeper bias (Hammersley and Atkinson, 2007). Further, the volunteers in our study were experienced and not new to the role. However, the focus of this article is the particular use of liminality implicit within the relationships between the volunteers and the mothers, and the mothers' experiences of coping. The particular configurations of family interventions and policy shifts need local interrogation and sense-making, particularly given the continuing policy drive in the UK to integrate health and social care (DH, 2017b) through multidisciplinary community-based teams. Applying liminality to the role of volunteer organisations that provide support for families within a domestic space contributes to an understanding of how professionals and volunteers can renegotiate their roles with patients or service users.

Conclusions
While there are empirical studies that have drawn on the theoretical framework of liminality, there is a gap on the provision and receipt of parenting support in the literature that has foregrounded liminality, and indeed, threshold concepts. We acknowledge that liminal spaces are not neat or clearly defined, yet the betwixt and between nature was positive for our participants in various ways. The liminal space occupied by the volunteers between a professional and friend was significant in building positive and trusting relationships with the mothers that enabled most of them to move from not coping to coping, itself a liminal stage. The application of liminality has enabled us to identify the uncertainty that the families experienced when faced with the withdrawal of the service, and this ambiguity was similar for some of the volunteers who were neither a friend nor a professional. For organisations working in family support, more consideration could be given to managing the endings between volunteer and parent relationships. Conceptually, liminality has proven to be an insightful framework that could be usefully further developed to increase our understanding of volunteers working in health and social care, and the provision of family support. The value of family support and the ways in which it is betwixt and between professional and volunteer support are worthy of further research, including a focus on replicating the McAuley et al (2004) study and Barnes et al (2009) studies, in this period of fiscal austerity, given the reliance on such services.
Notes 1 Corresponding author. 2 Parents use Home-Start support, but it is predominantly mothers who are referred or who self-refer.