Displaying family at a women’s refuge

Abstract

This article explores the family relationships of mothers and children living at a women’s refuge because of intimate partner violence. Theoretically, the article contributes to the sociological literature analysing family relationships in terms of ‘doing’ and ‘displaying’ rather than ‘being’ a family. Empirically, it is based on ethnographic fieldwork and semi-structured interviews with children living at a refuge in Denmark. The article shows that family display at the refuge is conducted by both mothers and children and sometimes in conflicting ways, not least when it comes to the question of how to define the position of the father (who has committed the violence) in the family’s future life. This highlights how display – as an evaluative practice – can be done in different ways in order to protect or reject family relationships, both internally and when addressing external audiences. Furthermore, the article analyses family relationships in a semi-public setting (the refuge) where powerful audiences are active interpreters of the family display enacted, and participants in decisions concerning the families’ futures.

Keywords

displaying family doing family ethnography intimate partner violence mothers and children at a refuge

Introduction

Several mothers and children are outside in the garden of the women’s refuge. Most of the children are playing while the mothers are talking. Sara’s phone rings. She answers it and a male voice is heard: it is Nora’s father, the man who has carried out violence in the family. Sara hands the phone to Nora, who starts talking to her father. ‘Move away a little bit’ Sara says, pointing to the garden. Nora walks away and one of the other mothers asks Sara in a low voice: ‘Is it her father?’ Sara rolls her eyes and nods. Then she says: ‘She is not used to talking to her dad this much [Pause]. So in that sense this [staying at the refuge] has been good. She has never talked this much with him before.’ After a little while, Nora comes over to Sara and hands her the phone. ‘Have you hung up?’ Sara asks. ‘Yes. Dad told me to’, Nora replies, and Sara says: ‘Good. Was it nice talking to dad?’ Nora nods and says: ‘Yes’ before running into the garden again.

According to Morgan (1996), families are constituted by the ‘doing’ of family rather than the ‘being’ of family. In this perspective, family is not a static category or structure but something that is actively produced and reproduced by the people involved. In 2007, Finch extended Morgan’s concept of ‘family practices’ by adding that ‘families need to be “displayed” as well as “done”’ (Finch, 2007, p. 66). She emphasised the changeability of family relationships and that family members have to continuously exhibit to each other (and to external audiences) that they are a family. This brought attention to the distinctly social nature of family practices, and how they need to be both conveyed and understood by relevant others as characteristic of a family. As the extract above shows, the women’s refuge is a place where display is at work in front of different audiences – both family members and people outside the family. It is an unusual place to ‘do family’, as practices normally considered intimate and private are carried out in a semi-public setting. The extract illustrates the multitude of family display that is practised side-by-side at the refuge. From the child and the father affirming a family relationship by talking to each other on the phone, to the mother’s conflicting displays as she rolls her eyes when the father is mentioned and her daughter looks away, but also expresses appreciation for the father’s increased interest in their child.

This article contributes to the research tradition focusing on the doing and displaying of families (Finch, 2007, 2011; Morgan, 1996, 2011) by applying these concepts to family practices enacted in a semi-public setting. We analyse how family relationships are displayed by both mothers and children in varying and sometimes conflicting ways, depending on the audience. Family display at the refuge is a means of maintaining and protecting some relationships while others are exposed and questioned. The relational changes that many families undergo when a mother and her child(ren) move into a refuge happen ‘under the gaze of others’ (Gengler, 2011). For the mothers in particular, this means that they have to adapt their family practices to normative standards of ‘good’ parenting, including standards on how to inform (or not to inform) their children about the violence.

Staying at a refuge represents a unique experience for women and children following intimate partner violence. At a refuge, mothers and children live in an unfamiliar setting with other families and staff members, and sometimes far away from their previous home. However, as shown in a meta-analysis by Robinson et al. (2020), little research has focused on the lived experience of staying at a refuge. Among the few studies addressing this, two focal points can be identified as paramount for our research: one concerning the fact that refuges not only provide safety for the families living there but also expose them to control, and another concerning how children/teenagers experience refuge life.

First, studies have shown how families are subjected to rules and restrictions at refuges, for example, in terms of curfews, mandatory chores and limitations of contacts with people outside the refuge (Gengler, 2012; Glenn & Goodman, 2015; Haj-Yahia & Cohen, 2009). Research demonstrates how refuges may group women into suitable ‘victim’ and ‘non-victim’ categories, and how they develop disciplinary measures to cope with ‘troublesome’ or ‘difficult’ residents (Gengler, 2012; Loseke, 1992). Furthermore, studies have focused on the ideological framing of shelter services. An example is Krane and Davies’s (2007) Canadian study, which described how motherhood was ‘romanticised’ by the staff at the refuge and influenced by ideas of ‘good’ mothering, and how this sometimes conflicted with the women’s needs and interests. Similarly, a US study (Gengler, 2011) found that refuge staff attempted to reform mothers based on ideologies of ‘appropriate’ mothering, which led to an experience among the mothers of being ‘policed’ by the staff – an experience that for some reminded them of the controlling behaviour of their partners. This finding illustrates how family life at refuges needs to be understood in relation to the gendered dynamics of intimate partner violence. During the refuge stay, women’s experiences and practices may be influenced both by past incidents of violence and by post-separation violence that attacks or undermines their mothering (e.g. manipulating tactics or negative talk by the violent partner) (Katz, 2019; Thiara & Humphreys, 2017).

Second, a small number of studies have looked at refuge life from the perspective of children/teenagers. An example is Øverlien’s (2012) Norwegian study. She shows how children at refuges perceived themselves and their mothers as being ‘in transit’, waiting for a ‘better and happier life’ in the future, and how they contrasted refuge life (characterised by passivity and isolation) with ‘normal life’ outside of the refuge. In another paper, Øverlien (2011) describes how security measures at refuges (e.g. keeping the location secret) may increase the sense of isolation among the children living there. Furthermore, Chanmugam’s (2011) study of teenagers living at US refuges shows how some of them compared refuge life to staying in a prison. They mentioned rules pertaining to curfew, bed times and constant parental supervision as affecting their everyday life in negative ways. Finally, Bracewell’s (2017) study of teenagers at UK refuges identified security measures as constituting an obstacle for adolescent life, also showing that many teenagers experienced a lack of social and emotional support at the refuges. Bracewell’s conclusion is that children and teenagers living at refuges need to have greater visibility and recognition as service users in their own right (see Selvik & Øverlien, 2015, for a similar argument).

Our study contributes to this tradition of qualitative research on refuges, addressing the lived experiences of women and children in a perspective inspired by family display theory (Finch, 2007, 2011), also taking into consideration that this display happens under the gaze of powerful audiences. The article is based on ethnographic fieldwork and qualitative interviews with children staying at a refuge in Denmark.

Theoretical frame

Finch (2007) describes family display as the process by which people ‘convey to each other and to relevant audiences that certain of their actions constitute “doing family things” and thereby confirm that these relationships are “family” relationships’ (Finch, 2007, p. 67). Displaying family is different from doing family because of its focus on relevant others and audiences, emphasising that family relationships include a process of affirmation. As McIntosh and colleagues (2011, p. 185) comment on the activities involved in this dialogical process of display and affirmation: ‘That means a particular activity is not only done but it is directed at others and a response is in turn sought from others.’ In the perspective of Finch, all family relationships need an element of display to show that the people involved constitute ‘a family’, but there are situations in which the need for display becomes even more necessary, for instance when family circumstances change. In periods of transitions or downright crises – which is the case for the families in the present study – there is a particular need to display that certain family relationships work and (sometimes) that other relationships do not.

Family display can be directed at audiences both within and outside the family (Heaphy, 2011; Walsh et al., 2019). Even though Finch was primarily concerned with how meaning is conveyed to relevant others within the family, other researchers have shown that the concept is usable in relation to audiences outside the family, such as local communities or official bureaucratic organisations (Finch, 2011; Heaphy, 2011; Walsh et al., 2019). Family display should not be seen as a superficial performance, putting relationships on show for outsiders. Rather, it is a combination of what family members try to accomplish for each other – in terms of maintaining strong relationships and consolidating family identities – and practices aiming at the acceptance of on-lookers.

Family display is related to cultural ideas about what it means to be a family, in terms of, for example, parental responsibility and the protection of children. In this sense, family practices are connected to wider representations of how families ought to be and are supposed to act: representations that both facilitate and impose restrictions on family life (Dermott & Seymour, 2011; James & Curtis, 2010; Kehily & Thomson, 2011; McCarthy, 2012; McCarthy et al., 2000). When parents display family, they do so in relation to cultural meaning systems through which relationships may be legitimated but also rejected (Heaphy, 2011; Kehily & Thomson, 2011). In this way, displaying family also involves risks, as display can be misrecognised, misinterpreted and condemned (Gabb, 2011; Heaphy, 2011). In her original argument, Finch (2007) did not discuss these forms of ‘unsuccessful’ family display in depth, nor did she focus on the fact that such display may come at a high cost for those involved (Finch, 2011; Gabb, 2011; Walsh et al., 2019). Parental practices are always weighted against prevailing normative standards for family life, but some parents find themselves in situations where the risk of being defined as an ‘improper’ or ‘failed’ parent is exceptionally high – something that can entail negative consequences such as losing custody of a child.

Parenthood (and family life in general) is an area where people’s moral identities are at work. McCarthy et al. (2000, p. 789) describe an ‘overall unquestionable imperative’ for family life, namely that adults must always put the needs of children first, and let their own interests take second place, if they are to be moral actors. McCarthy et al. (2000) also claim that women in particular have their moral identities at stake when it comes to display of appropriate parenthood, while men may more easily be accepted for combining care for their children with a more individualistic pursuit of other interests. May (2008) even suggests that a proper display of motherhood includes securing that children receive good fathering (p. 473).

Our article contributes to the literature on family display in three ways. First, we focus on family display enacted by both adults and children, which few other studies have done. Second, we show how family display is practised differently depending on who the audience consists of (e.g. children or adults, family members or outsiders). This question has only marginally been touched upon in research on family display (see Seymour, 2011, though). Third, we show how the need for family display is extraordinarily strong under conditions that are not private but semi-public, and in a situation characterised by an ongoing refiguration of the participants’ family life. Before analysing these aspects, we describe the context and methods of the study.

Context and methods

In Denmark, 2,000 women and 1,900 children stayed at a women’s refuge in 2018. This is only a small fraction of the number of women and children who experience intimate partner violence every year. A study by Helweg-Larsen (2012) estimates that around 38,000 women and 33,000 children in Denmark (in a population of 5.8 million) experience intimate partner violence annually. In the present study, 26 mothers and 31 children were part of the fieldwork that formed the basis for the data collection. The mothers were between 25 and 45 years of age. Half of them were Danish-born; the rest came from other countries in Europe, Africa, the Middle East, Asia and South America. Before moving to the refuge, almost all of the mothers (24) had lived with the violent partner. In two-thirds of the cases, the violent partner was the child’s birth father, and in one-third the mother’s new partner. Regarding length of stay at the refuge, two-thirds of the families had lived there for more than six months (and a few as long as two years) while the rest had been there for shorter periods.

For many families, moving to the refuge marked the beginning of a divorce process and custody negotiations. In Denmark, decisions on custody and contact arrangements are made by the parents if they can agree, and if they cannot, by the Agency of Family Law (at the time of the study called the Danish State Administration). In most families involved in this study, the parents did not agree, and therefore, the State Administration had a hand in determining custody and visitation rights. At the time of the study, almost half of the children of violent birth fathers had contact with him – from seeing him every other week to occasional visits – while none of the children of violent stepfathers maintained the contact. In a few cases, the child and birth father did not see each other.

Ethnographic fieldwork was carried out over a period of eight months in 2018. Participant observation provided important insights into how family display was done and in front of which audiences at the refuge. The researcher (the first author of this article) participated in everyday activities such as watching television with the families, playing with the children or sitting in the kitchen when dinner was prepared, as well as having informal conversations with the mothers and children. She positioned herself as ‘different’ from the staff in order to gain insight into the families’ everyday lives rather than focusing on typical resident–staff issues (e.g. related to practical arrangements or contacts with the social services). This position is reminiscent of the ‘non-official adult’ role, described by Mayall (2008) as someone who had a lot of time at her disposal and therefore could engage in lengthy conversations with the mothers and leisure activities with the children in a way that staff members could not. The ‘non-official adult’ position was established by spending as much time as possible with the families and little time in the staff areas, as well as assuring the residents that what was talked about was confidential and would remain so.

In addition to the participant observation, semi-structured interviews were carried out with 10 children (aged 6–12) during their stay at the refuge, and eight of them took part in a follow-up interview 6 to 12 months later. The main themes of these interviews were everyday life at the refuge and the children’s experiences of different relationships (with family members, relatives, friends, schoolmates, etc.). The interviews lasted between 20 and 45 minutes and were conducted in a meeting-room or the family’s own room at the refuge (first round of interviews) and at the mother’s place of residence (second round).

The study followed the ethical standards stipulated by the Danish Council of Independent Research and by Danish universities concerning consent, anonymity, respect for participants, integrity, and safe data storage. Before the fieldwork, the mothers and children were informed about the study through a leaflet distributed at the refuge. At the beginning of fieldwork, all mothers and children were approached by the researcher and the consent of participating families was obtained. During the fieldwork, the researcher relied on the common areas of the refuge (shared kitchens and living rooms) and only entered the rooms of the families when invited to do so. For the interviews with children, consent was achieved from both the children and their mothers, and the children were told that they did not have to answer all questions and could end the interview whenever they wanted. Furthermore, the researcher made it clear for both children and their mothers that the interviews were confidential. Hence, no information from the interviews was passed on to the mothers or other people. In this article, all names are pseudonyms, and any information that could reveal the identities of the participants and the refuge has been removed.

When analysing the data, both authors read the fieldnotes and interviews carefully, specifically focusing on parenting practices and how family relationships were talked about and enacted. We immediately noted the pressure the mothers (and partly also the children) were under, not only because of the violence but also because of the ‘transit’ situation (Øverlien, 2012) they found themselves in. We turned to the literature on ‘doing families’, well aware that family life in our study was conducted under very specific, partly public, circumstances. Finch’s concept of family display was used to open up our data, because of the differences in how the women and children were doing family in relation to various audiences (each other, staff members, etc.). We then read the fieldnotes and interviews again, marking all sequences where the women and children were talking about, or enacting, family relationships. We categorised these sequences with the help of focused, theoretically inspired codes, such as: ‘children protecting family relationships’, ‘women displaying family in front of different audiences’ and ‘women displaying the character of their family relationships (and of the people involved)’. After comparing our notes, we picked the quotes we found most informative and representative of these codes (to be presented in the following sections).

Family display in a semi-public setting

At the women’s refuge, family practices were done in a setting comprised of overlapping private and public arenas and in front of multiple audiences. The social and institutional setting influenced how display was done and what was displayed or left un-displayed. The first section below analyses how children displayed family (especially concerning the father) and how this was conditioned on the frames available to them. The next section explores the mothers’ displays of family relationships – in a context that could be influenced by post-separation violence in the form of coercive control attempts – and how the acts of display varied, and were expected to vary, depending on the audience. The final section analyses how women displayed motherhood at the refuge, relating their practices to cultural ideals of parenthood and (in custody cases) to the demands of the State Administration.

Children’s display of family relationships

Display of family relationships is dependent on how agentic and autonomous the ‘displayers’ are (Heaphy, 2011; Walsh et al., 2019). In the case at hand, children’s doing and displaying of family was conditional on the frames set up for them by adults (their parents, staff at the refuge, etc.). One such frame was related to the children’s knowledge of the character of their parents’ present relationship, and of the violence that had brought them to the refuge. In most cases, and especially among young children, this knowledge was limited. One example was Will (7 years old), first interviewed while he was living at the refuge. Will started out by saying that he could not remember when and why he and his mother and sister had moved into the refuge, continuing: ‘My mum at first said that dad had hit her. But then dad said that . . . it was an insect, it wasn’t a bee [searching for the right word] . . . It was a wasp. Dad said that a wasp stung mum, and that she was badly hurt by the wasp.’ Avoiding further discussion of this, the researcher then asked Will about the family’s present living arrangements. Will said that he and his sister Ida (6 years old) alternated between staying at the refuge with their mother and living with their father in the family’s house. Neither Will nor Ida (who was also interviewed) seemed to know why their mother did not want to move back to the family house. Ida also told the researcher that their father sometimes talked negatively about their mother and that he cried because his wife did not ‘want them to be a family anymore’. This illustrates how post-separation violence can become part of children’s contact arrangements, in this case with a father enacting his own family display by placing the blame for the family’s situation on the mother (cf. Thiara & Humphreys, 2017).

As pointed out by Finch (2007, p. 70), family display is an ‘evaluative’ definition of who belongs to the family and who does not. Most of the children at the refuge included their birth father as a central figure in their family. For them, it was the father who represented stability and the well-known, while the mother – for reasons they did not understand – had made them leave their home (and their friends, neighbourhood and school) and move into an unfamiliar and sometimes stressful form of collective accommodation. The following quotes from observations and interviews show examples of how children talked about their fathers, and how their mothers responded to this.

Two mothers, Maria and Nicole, are in the shared kitchen with their toddlers. Maria shows the researcher some pictures of her apartment, telling her that she would like to have new locks on her door in order to keep her husband out. Oscar (Maria’s son) is babbling, and the researcher jokingly asks the other child, three-year-old Ben, what Oscar said. Ben replies: ‘He said he wants to go home . . . home to his dad.’ Ben’s mother Nicole turns around, asking: ‘What did you say?’ Ben repeats: ‘That he wants to go home to his dad.’ Nicole turns around facing the sink, muttering ‘Mmmmm’ after which the dialogue ends.

Another example of the children’s ways of including their father in their family display, and the mothers’ tendency to react with silence, was the following incident. Ida (6 years old) had just been interviewed by the researcher who was making an appointment with her mother about a follow-up interview, when she said: ‘Yes [you can come] when we have moved back to dad.’ Ida looked at her mother who did not respond but said something to the researcher instead. Ida insisted: ‘We have to move back to dad!’ Her mother still did not respond but gently stroked her hair and gave her a smile. After this, Ida went into their room.

A third example of the children’s ways of including their father in their family display comes from the second interview with Will (7 years old). At the time of the follow-up interview, Will had not seen his father for a year. Towards the end of the interview, the researcher asked him if there was anything he would like to say that could help other children living at refuges. Will answered: ‘It would be good to see one’s dad.’ When the researcher asked him if he had mentioned this to his mother or anyone else, he said: ‘No.’ In this interview, as in the first, he described himself as being close with his father. By stressing his wish to see him again, he displayed the importance of their relationship, but only to the researcher.

Mothers’ display of family relationships

Family display is not just a matter of defining who belongs to the family and who does not; it is also a question of conveying the qualitative character of the relationships (Finch, 2007). In situations of change and unscheduled transitions, relationships need to be renegotiated and redefined – internally among the involved parties as well as externally in relation to significant others (Finch, 2007; Morgan, 2011). Family display must be adapted to particular audiences, willing (and when it comes to children: able) to receive and validate it (Walsh et al., 2019). When family display happens in a public, or in our case semi-public setting, the question of audiences becomes particularly important. Hence, at the refuge, there were social restrictions on what types of family display were appropriate in which situations. In the following, we use two mothers, Lisbeth and Julia, to show how family display needs to be adapted to the audiences witnessing it, and how family display is a question of defining the character of relationships – and the character of the individuals involved. The first extract concerns a situation when Lisbeth, a mother with an 18-month-old daughter, did not adjust her family display to the audience:

Lisbeth is sitting in the shared kitchen with her daughter Ella when the researcher enters. She starts telling the researcher about a meeting she had with her lawyer, being very upset about the possibility that her husband will receive shared custody. One of the staff members enters the kitchen and sits down at the table while Lisbeth continues to talk about her experiences with the State Administration. After the conversation, the staff member pulls the researcher aside, telling her that it is important to try to ‘shut down’ conversations like this one. The researcher responds that she had not in any way asked Lisbeth to talk about her problems but merely listened, to which the staff member says: ‘Yeah, but merely listening may also turn a situation into something it shouldn’t be.’

The staff member’s concern was primarily related to the fact that Lisbeth’s daughter was present. However, other conversations with staff members revealed their ambivalent attitudes towards women talking about violence with each other – and with the researcher. As one of the staff members put it: ‘It’s very important to open up about things, but if you say everything to anybody you may regret it afterwards.’ Statements like this were related to worries about the women’s vulnerability and, in some cases, safety at the refuge and to the opinion that the families’ experiences with violence should preferably be handled by staff members and other professionals.

Family display is, as mentioned above, a question of conveying the character of relationships and of the people involved in them. It is also, as Morgan (2011) stresses, a matter of defining what kind of family one’s family is, seen in relation to normative standards of proper vs improper and functional vs dysfunctional family relationships. Several women described their relationships with their husbands as altogether dysfunctional, and their husbands not only as violent partners, but also as poor fathers. One of them was Julia, who in conversations with the other women and the researcher described her husband as a ‘monster’, explaining: ‘Robert hit me badly, also when I was pregnant [. . .] He never wanted our child. And he took my savings while I was away visiting my family [Julia came from a foreign country, her husband was Danish] and then he said “you know, you never had that money, I am so worried about you, you should see a psychologist”.’ She repeatedly mentioned her fear of a future where Robert would get custody: ‘Doesn’t it mean anything that his ex-wife and son have been beaten up too, and I have pictures of red marks on Oscar’s body after he has been with his dad? But they need more proof, they say. Now, what is proof, does he have to kill me?’

Furthermore, Julia was worried about the way Robert presented himself to different audiences (the State Administration, lawyers, psychologists and others) as a caring and loving father, although she said he had never shown an interest in their son before. She also worried about the way he presented her to other people, focusing on and undermining her ability as a mother, a tactic which is described by Heward-Belle (2017) as a way for violent men to assert power and control in families (see also Morris, 2009). When the family’s case was finally settled by the State Administration, Julia said the following to the researcher:

Robert has told everybody I am a poor mum [. . .] But he didn’t get custody, he only got visitation. He will see Oscar for 1½ hours every second week and a psychologist will be supervising [. . .] But you know, I don’t see the point of the psychologist being there. Robert is so smart. It’s not like he’s going to do anything when she sits there. He is going to be the best dad. When we were with other people, he would be the perfect dad [. . .] The best husband, the best dad [. . .] The problem was when we came home and we were alone.

As observed by Finch (2007, p. 68), an overriding need among separated, co-parenting couples is to display ‘good parenthood’. Confirming this, we can add that in couples characterised by severe conflicts (and in our case, violence), there is also a need among the partners to assess and negotiate each other’s parenting capabilities. Although there were variations in how the women described their partners’ fathering practices, there was a general awareness that the State Administration’s decisions depended on how successful their partners were in displaying that they were ‘good fathers, although bad boyfriends’, as one of the women put it.

Displaying motherhood

So far, we have primarily focused on family display concerning the relationship to the partner/father who had committed the violence. However, the refuge and the transitional state that most of the families were in also provided a specific context for displaying motherhood. As previous research has demonstrated, living at a refuge can entail an experience of being controlled and scrutinised as a mother (Gengler, 2011; Glenn & Goodman, 2015). In the present study as well, many women seemed to feel that they were overseen by external audiences, or in Gengler’s (2011) words, that they mothered ‘under the gaze of others’.

One example of this was Charlotte. She had moved into the refuge because of her partner’s violence, together with her two children from an earlier marriage. Because of her own substance use problems (and the violence) during the past two years, her children had been staying with their father most of the time, but Charlotte was now in addiction treatment and was struggling for shared custody. The following observation shows how frustrations sometimes resulted in conflicts, audible (and in some cases visible) to all due to the families’ living conditions. Furthermore, the example illustrates how some women felt observed, by staff members and others, in their mothering practices.

There are loud noises coming from Charlotte’s room, heavy bumps as if objects were being thrown around. Louise’s (Charlotte’s 10-year-old daughter) shouts can be heard all over the place, and also Charlotte’s voice: ‘You stop now, I don’t want to listen to this anymore.’ Charlotte comes out in the corridor while her daughter is shouting after her: ‘You’re a piece of shit, a real piece of shit.’ Charlotte enters the living room, sees the researcher sitting there, and explains: ‘Louise is angry. If it isn’t her clothes that are wrong it is the roller-skates. I asked her not to bang them on the floor. Right now, she is yelling at me, but I guess that’s normal.’ Charlotte smiles. Louise’s shouts continue from the room, and suddenly a sneaker comes flying through the corridor. Louise then enters the living room, snapping at her mother. Her mother answers: ‘I still love you, sweetheart.’

A few days after this episode the researcher was sitting with Charlotte and one of the staff members in the kitchen. After having talked about other things, Charlotte suddenly asked Helen, the staff member: ‘What did you observe when I had the girls here? Do you think I was a good mother?’ Helen quickly replied: ‘You know, I hate those words, a good mother or a bad mother. It looked as if you had a good time together. Louise put you to a test but otherwise you seemed to have fun.’ Charlotte then said: ‘We have a meeting at the State Administration again in a couple of months. I would really appreciate if this [that things went well] was presented to them.’ Helen looked at her, answering: ‘Let’s make an appointment where we can talk about it.’

Another example showing how some mothers felt that they had to prove themselves as ‘good mothers’ was Klara. One day during the fieldwork, the following took place:

Klara approaches the researcher and tells her: ‘I am stressed. At two o’clock the health visitor will come to see my daughter, and my room is a mess.’ The researcher says: ‘I don’t think she cares. I think she has seen many messy rooms and apartments.’ Klara continues: ‘Yes, but you don’t know how messy my room is.’ The researcher smiles and suggests that Klara can hide the mess, but Klara says she has nowhere to put her things. The researcher tries to reassure her that the health visitor does not care about the mess, but Klara says: ‘Yeah, but you know, in my situation . . . Her dad will do anything. I can’t make any mistakes.’

On another occasion, Klara told the researcher that she needed to stay calm and focused in the whole custody process: ‘My lawyer told me that I shouldn’t get emotional when we go to the State Administration, because then they may think that I am emotionally unstable, and that I can’t take care of my child.’ Klara and Charlotte’s cases illustrate how some mothers felt that their mothering (and behaviour in general) was observed and under the scrutiny of others. They felt that any sign of ‘bad’ mothering, such as a messy room or being ‘too emotional’, could be used against them in their pending custody cases.

As these examples show, there were family displays addressing audiences both at and beyond the refuge. The State Administration is a powerful audience which ‘controls whether and the degree to which an activity is recognised as [an appropriate] display of family’ (Dermott & Seymour, 2011, p. 13). Even though this external audience was absent at the refuge, it played a role, not only as observer of display but as co-constructor of what forms of display were relevant and suitable – if the families’ custody cases were to end in a way that was acceptable for the mothers.

Discussion

In the research tradition inspired by Morgan (1996), family relationships are not regarded as given but as something that needs to be continuously worked on. Finch (2007, 2011) added an important dimension to this approach, claiming that family members need to display their relationships in order to confirm, internally and externally, that they are a family. After Finch coined the term family display, it was quickly taken up and applied by other family researchers, analysing, for example, parenthood in divorced families, sexual minority families, immigrant families and ‘family-like’ display between staff and children in institutional care (Dermott & Seymour, 2011; McIntosh et al., 2011; Seymour & Walsh, 2013). Our study contributes to this research tradition in three ways.

First, in contrast with previous research on family display where the focus has been on adults (an exception being Walsh et al., 2019), we also looked at children’s family display. We showed how children more than the mothers included their father in their family network, and how many of them talked about a future where their parents were living together again. This shows how family display has the function of maintaining and protecting relationships, and how it both reflects images of relationships as they are and images of relationships as the participants want them to be (cf. Wilson et al., 2012, for a similar finding). Research on post-separation families has demonstrated that children can experience loyalty conflicts, and sometimes feelings of being caught in the middle between the parents’ interests (for a review, see Birnbaum & Saini, 2012). In the present study as well, the children’s display can be interpreted as an expression of loyalty felt towards both parents, as their practices seemed to aim at protecting relationships that were otherwise threatened.

Second, we have shown how family display was enacted (and expected and directed to happen) in different ways, depending on the audience. Most importantly, the mothers and children were each other’s audience, with the mothers protecting their children from certain information about the fathers, and vice versa, the children keeping to themselves some thoughts and wishes concerning their father. Another audience was the other women at the refuge, where the families lived extraordinarily close, sharing kitchen and living room for months or, in some cases, years. Although there were examples of friendships developing between the women, there were also tensions and disagreements, for example concerning bedtimes, use of shared facilities and the children’s behaviour towards each other. In many cases, the mothers assessed, commented on and intervened in each other’s doing of family. Hence, when the women displayed family, they did so in parallel with and overseen by each other. A third audience was the staff at the refuge, whom the women saw as people offering help and support but whose observations some of them also feared or sought to use in custody cases. One final audience was the researcher, whose role the women did not always understand. Although she repeatedly informed them that she came from the university and was not employed at the refuge, the women tended to forget this and saw her as a trainee or volunteer with a lot of time at her disposal. In her role as a ‘non-official adult’, she was regarded as an ally, one they could engage in ‘grown-up talk’ with and one who could look after their children while they went out for a smoke or to the laundry room. However, the researcher was also (and rightly so) comprehended as an observer, another audience to their family display whose affirmation of them being ‘good mothers’ they actively sought.

Third, we have focused on the extraordinariness of family display enacted in a semi-public context where parenting practices observed by others may have severe social and legal consequences. Finch’s concept of family display has been criticised for focusing too much on individual control and agency and downplaying the fact that different ‘displayers’ have varying degrees of self-determination when doing family (Gabb, 2011; Haynes & Dermott, 2011). This article focused on families that were vulnerable both in terms of their living conditions and in relation to (potentially) important decisions made by the authorities. The family display we analysed belongs to the category of ‘high-intensity’ interaction (Finch, 2007, p. 72), where questions such as ‘who belongs to the family?’, ‘what kind of family is this?’ and ‘what are the characters of the people involved?’ have to be negotiated with internal as well as external audiences. Furthermore, in a semi-public setting like the refuge, the audiences (first and foremost the State Administration but also staff members and other professionals) were not merely spectators of the family display, but active interpreters of what was said and done, and participants in the planning of and decisions concerning the families’ futures.

Finally, some limitations of the study need to be mentioned. The most obvious limitation is that, although we address the display of family relationships, the partners/fathers were not included in the study. Hence, the sparse information we have about them comes from other parties, and especially the mothers, who for obvious reasons focused on specific aspects of their partner’s character and behaviour. The reason why the partners/fathers were not included was ethical. Contacting the partners of women living at a refuge could have jeopardised the women’s and children’s safety and created further conflicts in families already burdened by problems.

Another limitation is the varying degrees of insights the research provided into the lives of the 26 mothers and 31 children who were part of the fieldwork. Some families kept to themselves in their rooms, others used the shared facilities but did not engage in conversations with the other women or the researcher (sometimes because of language difficulties), while others – among them several of the women included in this article – spent a lot of time with each other and the researcher, conveying detailed information about their experiences. As a result, the patterns analysed above are not necessarily representative of all the women living at the refuge at the time of the fieldwork.

Conclusion

This article investigated family display by mothers and children at a women’s refuge, providing insights into what may be at stake for actors in high-intensity family situations. It showed how children’s enactment of their family relationships happened in a context of restricted knowledge, and how their and their mothers’ display revealed different ideas about the father/partner’s position in the family. This illustrated how family display was a question of both protecting and rejecting family relationships. Further, we analysed the women’s doing of family, showing how family display is a way of conveying the character of the relationships as well as the character of the parties involved, and how the women adapted their display to both internal and external audiences. For the mothers involved in custody cases in particular, their stay at the refuge contained experiences of mothering ‘under the gaze of others’, not least that of the State Administration. Significantly then, fleeing from violence at home and seeking safety at a refuge meant that the women opened up their mothering practices to public scrutiny, including from powerful audiences capable of making decisions with far-reaching consequences for the families.

Footnotes

Funding

The first author was funded by a grant from Independent Research Fund Denmark, grant number DFF - 7024-00040.

References

Birnbaum

Saini

(2012). A scoping review of qualitative studies about children experiencing parental separation. Childhood, 20(2), 260–282.

Bracewell

(2017). Teenagers’ experiences of domestic violence refuges [PhD dissertation, University of Central Lancashire].

Chanmugam

(2011). Perspectives on US domestic violence emergency shelters: What do young adolescents and their mothers say? Child Care in Practice, 17(4), 393–415.

Dermott

Seymour

(Eds.). (2011). Displaying families: A new concept for the sociology of family life. Palgrave Macmillan.

Finch

(2007). Displaying families. Sociology, 41(1), 65–81.

Finch

(2011). Exploring the concept of display in family relationships. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 197–205). Palgrave Macmillan.

Gabb

(2011). Troubling displays: The affect of gender, sexuality and class. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 38–57). Palgrave Macmillan.

Gengler

A. M.

(2011). Mothering under others’ gaze: Policing motherhood in a battered women’s shelter. International Journal of Sociology of the Family, 37(1), 131–152.

Gengler

A. M.

(2012). Defying (dis)empowerment in a bettered women’s shelter: Moral rhetorics, intersectionality, and processes of control and resistance. Social Problems, 59(4), 501–521.

10.

Glenn

Goodman

(2015). Living with and within the rules of domestic violence shelters: A qualitative exploration of residents’ experiences. Violence Against Women, 21(12), 1481–1506.

11.

Haj-Yahia

Cohen

H. C.

(2009). On the lived experience of battered women residing in shelters. Journal of Family Violence, 24, 95–109.

12.

Haynes

Dermott

(2011). Displaying mixedness: Differences and family relationships. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 145–159). Palgrave Macmillan.

13.

Heaphy

(2011). Critical relational displays. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 19–37). Palgrave Macmillan.

14.

Helweg-Larsen

(2012). Vold i de nære relationer [Violence in intimate relationships]. National Institute of Public Health, University of Southern Denmark.

15.

Heward-Belle

(2017). Exploiting the ‘good mother’ as a tactic of coercive control: Domestically violent men’s assaults on women as mothers. Affilia, 32(3), 374–389.

16.

James

Curtis

(2010). Family displays and personal lives. Sociology, 44(6), 1163–1180.

17.

Katz

(2019). Coercive control, domestic violence, and a five-factor framework: Five factors that influence closeness, distance, and strain in mother–child relationships. Violence Against Women, 25(15), 1829–1853.

18.

Kehily

M. J.

Thomson

(2011). Displaying motherhood: Representations, visual methods and the materiality of maternal practice. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 61–80). Palgrave Macmillan.

19.

Krane

Davies

(2007). Mothering under difficult circumstances: Challenges to working with battered women. Journal of Women and Social Work, 22(1), 23–38.

20.

Loseke

D. R.

(1992). The battered woman and shelters: The social construction of wife abuse. State University of New York Press.

21.

May

(2008). On being a ‘good’ mother: The moral presentation of self in written stories. Sociology, 42(3), 470–486.

22.

Mayall

(2008). Conversations with children: Working with generational issues. In Christensen

James

(Eds.), Research with children: Perspectives and practices (pp. 109–122). Routledge.

23.

McCarthy

J. R.

(2012). The powerful relational language of ‘family’: Togetherness, belonging and personhood. The Sociological Review, 60(1), 68–90.

24.

McCarthy

J. R.

Edwards

Gillies

(2000). Moral tales of the child and the adult: Narratives of contemporary family lives under changing circumstances. Sociology, 34(4), 785–803.

25.

McIntosh

Dorrer

Punch

Emond

(2011). ‘I know we can’t be a family, but as close as you can get’: Displaying families within an institutional context. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 175–194). Palgrave Macmillan.

26.

Morgan

D. H. J.

(1996). Family connections: An introduction to family studies. Polity Press.

27.

Morgan

D. H. J.

(2011). Rethinking family practices. Palgrave Macmillan.

28.

Morris

(2009). Gendered dynamics of abuse and violence in families: Considering the abusive household gender regime. Child Abuse Review, 18, 414–427.

29.

Øverlien

(2011). Abused women with children or children of abused women? A study of conflicting perspectives at women’s refuges in Norway. Child and Family Social Work, 16(1), 71–80.

30.

Øverlien

(2012). Narrating the good life – children in shelters for abused women talk about the future. Qualitative Social Work, 11(5), 470–485.

31.

Robinson

S. R.

Maxwell

D. R.

Rogers

K. R.

(2020). Living in intimate partner violence shelters: A qualitative interpretive meta-synthesis of women’s experiences. British Journal of Social Work, 50(1), 81–100.

32.

Selvik

Øverlien

(2015). Children with multiple stays at Nordic refuges for abused women: Conclusions, challenges, and causes for concern. Nordic Social Work Research, 5(2), 98–112.

33.

Seymour

(2011). ‘Family hold back’: Displaying families in the single-location home/workplace. In Dermott

Seymour

(Eds.), Displaying families: A new concept for the sociology of family life (pp. 160–174). Palgrave Macmillan.

34.

Seymour

Walsh

(2013). Displaying families, migrant families and community connectedness: The application of an emerging concept in family life. Journal of Comparative Family Studies, 44(6), 689–698.

35.

Thiara

R. K.

Humphreys

(2017). Absent presence: The ongoing impact of men’s violence on the mother–child relationship. Child and Family Social Work, 22(1), 137–145.

36.

Walsh

McNamee

Seymour

(2019). Family display, family type, or community? Limitations in the application of a concept. Families, Relationships and Societies, 9(3), 1–16.

37.

Wilson

Cunningham-Burley

Bancroft

Backett-Milburn

(2012). The consequences of love: Young people and family practices in difficult circumstances. The Sociological Review, 60(1), 110–128.