Children in domestic violence shelters: Does the feminist perspective collapse?

SUMMARY: Even though an extensive body of literature on children has swept the field of domestic violence in the last 30 years, little is known about how domestic violence shelter workers understand children's situations and how they intervene with them. This article seeks to address this gap in the literature, and presents the results of a study conducted with 48 advocates in the province of Québec (Canada).
FINDINGS: The data suggest that most of the participants adopt a child-centred perspective and consider the children in their own right during their stay. The accounts of the participants' practices also reveal that they perceive children as being vulnerable and at-risk. With a moderate emphasis on vulnerability and risk, the participants tend to support the children alongside their mothers, while associating potential risks with the behaviour of the perpetrator of domestic violence. However, with a strong focus on vulnerability and risk, participants tend to cast aside the perpetrators' behaviour and monitor the women-as-mothers during their stay while associating potential risk with their [in]actions under the circumstances. This can lead to mother-blaming, surveillance and more authoritarian interactions. APPLICATIONS: The understanding of children living with domestic violence needs to remain rooted in a feminist analysis of violence against women in order to avoid some of the issues highlighted in the article. Furthermore, studies that seek to shed light on best social work practices when working with children in alliance with their mothers from a feminist perspective are crucially needed.

Introduction

A vast network of domestic violence shelters was established in North America and in the United Kingdom in the 1970s, allowing abused women and their children to escape violent and controlling men (Dobash & Dobash, 1992; Walker, 1990). Initially, interventions were mostly designed to support women and protect them and their children. However, shelter workers quickly realized that the children also needed help in their unique context (Côté, 2016). Accordingly, practices have been developed over the years to meet their needs.

Although several studies investigating children living with domestic violence have been conducted, very little has been written about shelter workers’ practices with children. The current article will seek to fill this gap in the literature by providing a critical analysis of social work practices with children and women-as-mothers in domestic violence shelters, drawing upon data from a doctoral thesis. The purpose of the thesis was to analyse the evolution of practices in domestic violence shelters in the province of Québec (Canada) since their development in the mid-1970s (Côté, 2016). During the interviews, the development of practices with children was identified as a critical component of the evolution of these shelters, but the participants’ perception of children and women-as-mothers raised a number of issues that will be discussed in the article.

This article is divided into four sections. The first section draws upon existing literature to discuss domestic violence shelters and the development of feminist practices within them, the recognition of children living with domestic violence, and the underlying tensions of working with children and women in shelters. The second section presents the research methodology. The third section focuses on the results and addresses three main themes. The implications of the findings for research and social work practice are discussed in the last section.

Literature review

Domestic violence shelters and the development of feminist practices

Feminists have highlighted the social and political dimensions of domestic violence and its roots in the patriarchy (Dobash & Dobash, 1979). This social issue can be understood on a continuum, along with various manifestations of male violence against women (Kelly, 1988). This perspective has never reached a unanimous agreement amongst researchers and in the general population. Many have challenged it, including researchers of the ‘family violence perspective’ who understand domestic violence as the exacerbation of conflicts in the family (see Johnson, 2008) and fathers’ rights groups who aggressively argue that women are just as violent (if not more) than men in intimate relationships (Dragiewicz, 2008). Some tensions have also occurred between shelter workers regarding this theoretical perspective (see Mann, 2002). However, despite opposition and tensions, the feminist perspective on domestic violence has been at the forefront of the shelter movement in Canada (Côté, 2016; Goodhand, 2017).

In the early 1980s, in the wake of extensive feminist work on the issue, feminist practices have emerged as an alternative perspective to working with women (Corbeil et al., 1983; Walker, 2002). According to Corbeil et al. (1983), feminist practitioners have strongly opposed traditional therapies recognized for their sexist and paternalistic approaches to working with women. The same authors have pointed out that the fundamental premise of feminist practices lies in its willingness to raise women’s consciousness of the political dimension of issues formerly thought of as private matters, and to empower women by deconstructing female socialization and gendered social roles. Overall, feminist practices focus on strengths with the ultimate goal of empowering women (Walker, 2002).

Initially, feminist shelter workers focused their energy on listening to women’s stories, which provided insights into the cycle of violence (Walker, 1979) and into various manifestations of power and control in their relationships (Pence, 1987), but they had a broader mission. In fact, and unlike faith-based or humanist-oriented shelters, feminist shelters believed that structural changes were necessary to tackle the issue of male violence against women, and thus, that they had to get involved in collective actions for long-term social change. Individual interventions alone would have resulted in what Walker (1990) calls a ‘band-aid approach’. In other words, while it was necessary to support each woman coming to the shelter and taking into account her needs and personal history, social and collective actions were just as important as they allowed feminists to fight domestic violence by tackling its pillar: the patriarchal social order (O’Neil, 1998). Male domination in the family was thus understood as a snapshot of existing power relations between men, women and children in society. Children living with domestic violence had to have their needs addressed; however, little was initially known about them and their experiences.

The recognition of children living with domestic violence

An extensive body of literature on children has swept the field of domestic violence in the last 30 years, raising awareness about the scope of the phenomenon, types of exposure and consequences on the development and well-being of children (see Holt et al., 2008). In 1998, a substantial review by Appel and Holden raised the issue of the co-occurrence of spouse and physical child abuse in homes where men abuse their female partners. These studies have corroborated the observations of shelter workers, who have always expressed concern for the children in their services and have outlined the importance of ensuring their safety inside and outside of shelters (Lapierre, 2010; Mullender et al., 1998; Peled, 1993; Walker, 1979).

In the late 1990s, several countries made changes to policies and protocols to support and protect children living with domestic violence (Nixon et al., 2007). Legislative changes at the beginning of the 2000s also recognized children living with domestic violence to ensure their safety (Edleson, 2004). In the province of Québec (Canada), this issue was formally recognized in the 1995 Policy on Intervention in Conjugal Violence and, a few years later, framed as a form of psychological maltreatment in the 2006 revision of the Youth Protection Act. In their conceptualization of the problem,

psychological ill-treatment refers to a situation in which a child is seriously or repeatedly subjected to behaviour on the part of the child’s parents or another person that could cause harm to the child, and the child’s parents fail to take the necessary steps to put an end to the situation. Such behaviour includes [ . . . ] exposure to conjugal or domestic violence.

Although domestic violence shelter workers had been in favour of the recognition of exposure to domestic violence in the Youth Protection Act, frustration and disappointment grew over the ubiquity of mother-blaming and punitive practices in the child protection system (Lapierre & Côté, 2011a). The same phenomenon has been observed in other provinces and other countries (Douglas & Walsh, 2010; Johnson & Sullivan, 2008; Nixon et al., 2007; Nixon & Tutty, 2009).

Working with children and mothers in shelters: A significant source of tension

The growing recognition of the phenomenon of children living with domestic violence has led to changes in shelter practices, which have mainly focused on women’s needs (see Peled, 1997), and has increasingly developed child-focused interventions. In Québec, shelters have developed ‘youth’ or ‘mother–child’ components to better meet the needs of children, with a mandate ‘focused on the consequences of violence in the psychosocial development of children’ (Chayer & Smith, 2012, p. 10 – loose translation).

Mandates focused on children and women-as-mothers in domestic violence shelters have raised concerns in the literature. One of the main criticisms is that while it is of crucial importance to work with women in their role as mothers, this work is not always framed within a feminist perspective (Krane & Davis, 2002). Women’s parenting skills and the way they interact with their children are more often assessed rather than supported during their stay, at a time in their lives where being a mother is highly stressful (Gengler, 2011; Krane & Davis, 2007; Peled & Dekel, 2010).

Moreover, an idealized view of mothering, which reproduces the dominant ideology, emerges from the discourse and practices in a shelter (Gengler, 2011; Krane & Carlton, 2012; Krane & Davis, 2007). In this regard, Krane and Davis (2007) argue that women find themselves in ‘difficult’ and ‘unusual’ conditions for mothering. Thus, while shelters should remain a place of solidarity and support, women may become reluctant to verbalize difficulties related to mothering, not only because they fear the judgment of other residents and workers, but more importantly, to minimize the risk of being reported to child protection services (Gengler, 2011).

It is also worth noting that staying in line with a feminist perspective can be challenging when working with some mothers who, themselves, face several issues that raise concerns about their abilities to care for their children, such as substance abuse, mental health problems, or aggressive behaviours. Furthermore, some women are ‘not so nice to deal with,’ to quote a participant in Mann’s study (2002).

These observations suggest that intervention with women in their role as mothers generates a ‘clash’ of values between feminist principles and the values of workers faced with the realities of shelter work. This has led to a number of contradictions in workers’ practices (Murray, 1988). In fact, workers may feel torn between the desire to empower women and accept that the choices they make are best for them under the circumstances, and having the children’s best interests at heart (Peled & Dekel, 2010). This can lead to pressuring women to end their relationship with the perpetrator in order to their children safer (Krane & Davis, 2002).

Beyond these tensions, additional difficulties have arisen for shelter workers in the province of Québec following changes to the Youth Protection Act, which compels them to report cases where a child’s safety or well-being is compromised. As they now have an obligation to report children who are ‘exposed to domestic violence,’ balancing women’s self-determination and the safety of children can create additional strain on their everyday practices. More importantly, such obligations can create fear or reluctance and even deter women from seeking services for difficulties that could be perceived as ‘failures’ (see Baldwin, 2015).

Methodology

The purpose of the study underlying this article was to analyse the evolution of practices in domestic violence shelters in Québec (Canada), from their development in the 1970s up until today; however, the current article focuses on the participants’ work with children. The study draws upon a critical and feminist epistemological framework and relies on a qualitative methodology.

A total of 48 semi-structured individual interviews were conducted between December 2014 and June 2015. The participants were classified into three categories: pioneers, veterans and workers. Pioneers (n = 8) were women who, between 1975 and 1985, either opened a domestic violence shelter, contributed to the development of the first coalition of shelters, or developed intervention guidelines for domestic violence shelter workers. Veterans (n = 7) were women who, between 1975 and 1985, were involved as employees, volunteers, interns, or activists in a domestic violence shelter but did not ‘fit’ the criteria for the pioneers, as defined above. Given that pioneers and veterans constitute a small group of women in the province of Québec, they were recruited through a snowballing sampling technique. Shelter workers (n = 33) had between 6 and 28 years of experience at the time of the interview and were either doing direct work in a shelter or employed in an umbrella organization. They were recruited on a voluntary basis through the two umbrella organizations and through shelter directors or coordinators. Out of the 33 shelter workers, 11 were in managerial positions, 17 were working mostly with women, and 5 were working mostly with children.

All the interviews (except one) were recorded with the application Smart Recorder, then transcribed into a Microsoft Word document and imported into N’Vivo (QSR International), a software product for qualitative data analysis. Given the amount of empirical material that had to be coded and analysed, an independent researcher/consultant and N’Vivo expert was hired at this stage to work on the data analysis with the main researcher and to provide help with the N’Vivo software.

As part of the coding process, regular meetings were planned with the consultant to develop a tree map (a diagram clustering the main themes) and a guidebook (to ensure the methodical use of the tree map), and to discuss attributes’ values (to make comparisons) for the study. Initially, six substantive interviews were carefully selected, allowing for the emergence of themes. These themes were then transformed into nodes (categories) in N’Vivo to build the tree map and the guidebook. The development of the coding tree, alongside the consultant, required numerous discussions, as well as three inter-rater agreement sessions: two with the consultant and one with the thesis advisor (the second author of this article). After multiple adjustments, all of the interviews were then coded.

Content analysis was then carried out within each node, allowing for a deep understanding of the meaning of the data. Content analysis is a ‘careful, detailed, systematic examination and interpretation of a particular body of material in an effort to identify patterns, themes, biases, and meanings’ (Berg, 2006, pp. 303–304). First- and second-level analyses are too extensive to describe sufficiently here, but can be found in the original study (Côté, 2016).

Results

From the onset of the domestic violence shelter movement, admitting women along with their children was perceived as an essential condition to help them flee their abusive partners. Like their mothers, the children were presumed as being at risk for further violence, and practices were centred on their safety and well-being during their stay. However, very limited child-centred interventions were provided at the beginning, according to the pioneers who took part in the study.

Some of them discussed the fact that they did interact with the children, but mostly in an informal manner. For instance, one pioneer recalls how they offered to babysit the younger children to give the women some respite: ‘When I arrived at the shelter, the children . . . we babysat them’ (Pioneer 2, former front-line worker).

In this regard, a few pioneers and veterans discussed formal and structured child interventions as an important evolution of practices in domestic violence shelters. On the one hand, they initially assumed that if they helped and supported women, they also helped children. On the other hand, they admitted that they were not fully aware of the impacts of exposure to domestic violence, and thus did not fully grasp the importance of child-specific services. Without denying the severity of the abuse in their homes, the scope of the consequences for children was not known at the time; it was therefore difficult for the workers to assess the situations of the children and develop specific services to meet their needs: ‘We were taking care of the mother and thought that if the mother were better, the children would also be better.’ (Pioneer 3, former shelter founder). ‘We weren’t too aware of the consequences of domestic violence on children. We knew it couldn’t have been very good, but we weren’t really aware of the impacts it had on them’ (Pioneer 2, former front-line worker).

Various factors, including the publication of a considerable body of literature on children living with domestic violence during the 1990s, may have contributed to the recognition of the issue. When provincial funding for shelters was increased in the early 2000s, several shelters hired youth workers whose mandate was to develop interventions specifically addressing the needs of children, to support their relationship with their mothers, and to implement prevention and awareness programmes in schools (Chayer & Smith, 2012). In hindsight, some pioneers have recognized that practices with children are probably more ‘suitable’ and ‘enlightened’ today than they were in the 1970s and 1980s.

Assisting children as individuals in their own right in domestic violence shelters

With regard to the participants’ perspective on the children they receive into their services today, the data are somewhat unclear. Almost all of the shelters accept children, including boys up to 16–18 years old, depending on their policies. Undoubtedly, there is no typical profile of children in shelters; some experience little to no difficulties and appear ‘resilient,’ while others struggle with serious issues (including mental health and behavioural problems). Participants in the current study identified a wide range of needs that are specific to children, mostly with regard to their safety and physical and emotional well-being, and discussed how they address them in their shelters. Broadly speaking, the participants focused on ensuring the safety of the children, educating them and their mothers, helping them identify their emotions, emphasizing children’s rights, and fostering a fun environment during their stay.

All of the participants had the children’s best interests at heart and thought it was important to address their needs in the shelter. One of the crucial components of their work is to focus on the children’s emotions during their stay: monitoring how they feel and how they are managing their transition, addressing their fears, working on their self-esteem, etc. Many participants who discussed their work with children, such as this youth worker, stated that dealing with their emotions is the primary focus of their work: ‘Well, my main goal here when people arrive is to [let them] have the opportunity to talk about what has happened, to vent, to express their emotions regarding the event. That is the first thing to do.’ (Advocate 16, youth worker).

Another important theme emerging from the data on children in shelters relates to enjoyment. Fun is an important feature of childhood. In this regard, the participants discussed a wide range of strategies that aim to make the children’s stay pleasurable: taking them out for sports, activities, city tours or ice cream, playing board games, etc. Even when workers need to address more difficult topics with the children (including domestic violence), they try to incorporate a fun activity to make the discussion easier. This worker describes how she always shows an interest in the children she works with:

Well, I think first of all, you have to spend time with them and listen to them. Spend time with them, but “children time,” as I call it! Bring them to the park, do something they love. For a start, they sure love it when we go to the park. After, it’s really . . . my approach is to be warm and pay special attention to them. Show them that I’m really here for them. (Advocate 25, youth worker)

As well as being a safe environment, shelters should provide children with an opportunity to learn and grow. In this sense, some workers discussed practices that seek to educate children. On the one hand, they want to help children understand domestic violence through their experiences while sending them a clear message that violence is unacceptable. On the other hand, they also take the opportunity to educate children about gender stereotypes to promote equality between girls and boys and women and men. One participant provided an interesting insight into an informal kitchen conversation she had with a young girl in the shelter who believed that it was more important for girls to write and behave well in comparison to boys. She saw this as an opportunity to challenge gender stereotypes with her:

She said to me: “As girls, we need to write and behave well, but it doesn’t matter for boys.” So my job at the time was to show her that no, it’s not normal to think like this. Of course, I didn’t say that to her! But it’s like: girls and boys, everyone must behave well. But I still had a debate with her because she was a bright little girl . . . Oh my god, when you’d get started with her, good luck! So it was a long debate, but I managed to make my point that both boys and girls must have certain standards. It’s not because we are girls that we must behave a certain way and vice versa. (Advocate 23, front-line worker)

Another component of the participants’ work with children is to ensure their rights are respected, within the limits of the shelter’s mandate. This includes respecting their boundaries and confidentiality, and understanding that if they ‘act out,’ it might be a temporary reaction to a difficult transition. One participant referred implicitly to the notion of ‘empowerment,’ a crucial component of feminist practices as it relates to children: ‘I always tell them: you have power. Power to make decisions and not being forced to do things that you don’t want. Be critical, and ask questions.’ (Advocate 10, team leader).

Finally, some participants expressed frustration with various systems that do not sufficiently protect children and that violate their rights, such as the family court. They explained how intensive efforts are sometimes required to ensure the children’s safety in a context where perpetrators are still perceived as ‘good enough’ fathers. Indeed, they feel that society at large is too lenient towards violent men, creating several challenges for their practices. This veteran believes, in hindsight, that society is failing children: ‘Another social issue is the rights of children. I find that when it comes to children’s rights, we are showing them that the strongest win. And the strongest are the violent fathers.’ (Veteran 1, former front-line worker).

Given this context, all of the participants perceived the children as being vulnerable and ‘at-risk’. During the interviews, they reflected on issues that could potentially arise in the children’s current or future lives, and discussed how they attempt to minimize some of this risk. To do so, they usually include the mothers in their interventions. Yet, depending on how they perceive the children’s vulnerability and potential risk, the way the participants interact with the mothers is twofold. With a moderate emphasis on vulnerability and risk, it appears that the participants support the children alongside their mothers while associating potential risk with the perpetrators’ behaviour. With a stronger focus on vulnerability and risk, the participants might instead protect the children by monitoring their mothers during their stay while associating potential risk to the women’s (in)action under the circumstances.

Moderate emphasis on vulnerability and risk: Supporting children alongside their mothers

In a thought-provoking analysis, Parton (2017) argues that risk is a ‘major driver of professional practice’ as it ‘gives the impression of being predictable and scientific and aims to bring the future into the present, so that the future can be controlled and modified’ (p. 6). Avoiding risk is, therefore, central to social work practice, including in domestic violence shelters. In fact, most participants consider children in shelters as vulnerable and ‘at-risk,’ stating that the context of domestic violence impacts them in different areas of their lives. They feel genuine empathy and compassion for the children and their experiences. They see firsthand that these experiences are not always confined to living with domestic violence: they are often also victims of physical, sexual and emotional abuse at the hands of their violent fathers or stepfathers. The participants also understand that mothers can temporarily be less emotionally available for their children as they struggle with the aftermath of domestic violence. They talk about the difficulties the children encounter in their lives, as well as within the shelter after fleeing the perpetrators. As a result, one of the main objectives of the participants’ practices is to ensure the children’s safety, as suggested by this youth worker: ‘Well, in the shelter, we really focus on safety. If they come here, it is to be safe and to get out of a violent environment. That is the main goal.’ (Advocate 16, youth worker).

If the children are perceived as temporarily vulnerable and at-risk due to the perpetrator’s violence, workers are more likely to maintain a positive attitude towards the mothers and express a desire to build on their strengths to help the children. Some participants talked about the importance of strengthening mother–child relationships during their stay, to support women as much as possible in their roles as mothers, to create a space to discuss mothering in the shelter, and to help them work on any issues while there. They noticed that women do their best to protect and care for their children under difficult circumstances and thus, perceive their alliance with the mothers as crucial to the well-being of their children during and after their stay. Mother–child interventions are also offered in some shelters, with the goal of strengthening the dyad and offering them an opportunity to spend quality time together: ‘Every Friday, we organize a mother–child activity to strengthen the mother–child relationship. [ . . . ] We want to strengthen that relationship. It’s about spending quality time with each other.’ (Advocate 24, front-line worker).

In addition to helping and supporting the children, the participant’s role is to ensure the women can focus on their own needs – and not just their children’s needs – so that they can get through this difficult and challenging transition in their life. Drawing upon the value of equality, these workers believe that building on the women’s strengths and addressing their flaws or limitations should be done while always considering the domestic violence context that they have fled. This veteran reports a situation where a woman was experiencing difficulties at dinnertime with her baby and suddenly became verbally violent with him. Where her colleagues saw deficiencies, she rather perceived the situation as an opportunity to reflect on the consequences of domestic violence on the mother–child relationship, believing that women need time and support to cope with the aftermath of abuse:

It’s to be in alliance with her, because I want her to regain power in her role as a mother. [ . . . ]. I am not surprised when I see her behaviour: she can’t take it anymore. It must have been something at the dinner table with her husband. [Her behaviour] tell me a lot about what she has been going through. This is the type of women we get in the shelter. We are here to help them regain power over their role as mothers, and over their lives as women. (Veteran 1, former front-line worker)

The participants who believe that addressing the children’s vulnerabilities and potential risk requires working in alliance with their mothers were torn when having to make the decision to call child protection services and collaborate with them. They perceive child protection practices as controlling and feel that collaborating with them often results in pressure to monitor the women in the shelter, thus potentially dissuading them from reaching out to their services. This veteran’s reflection regarding her shelter’s policy is quite interesting in this regard: ‘If we do that [refer every child], women won’t come to the shelter anymore. [ . . . ] We make a referral maybe three times a year.’ (Veteran 3, manager).

If they have to do a referral, the clash of values between building a trusting relationship with the women-as-mothers in the shelter (equality, solidarity) and protecting their children (safety) appears to create a significant source of tension for these workers. For instance, this participant reflects on a recent situation – where she felt she had no choice but to call child protection services – while stressing how heartbroken she felt about it:

Back in February, we had a mother whose four children were taken into care. It’s so, so sad, but at some point, we need to be reasonable and say: it’s better for them. Better than with a mother who is addicted [to drugs], who is dysfunctional, who is crippled with debt, who doesn’t prepare meals, who isn’t there, and who leaves them to take care of themselves. At some point, we have to be reasonable and find them a more suitable environment. But it’s still heartbreaking. (Advocate 21, front-line worker)

When the children’s vulnerability and potential risk remain framed as consequences of men’s violence against women in intimate relationships, workers are more likely to discuss practices that focus on supporting them alongside their mothers and remain consistent with a feminist perspective rooted in the empowerment of women. However, focusing too much on vulnerability and risk is a double-edged sword, as it can lead some workers to focus almost exclusively on the consequences of domestic violence for the children while losing sight of the broader picture. This can result in a shift to a risk-management approach for their practices, away from a feminist analysis of children in domestic violence shelters, and even becoming punitive towards mothers.

A stronger focus on vulnerability and risk: Protecting children by monitoring their mothers

A number of participants discussed some compelling measures that appeared more ‘authoritarian’ when the focus of their practices was centred on the children’s vulnerability and risk. These participants seemed to believe that the risks the children face are related not only to their father’s behaviour, but to a certain extent, to their mother’s behaviour as well. As a result, they emphasize the protection of the children in the shelter by monitoring their mothers based on society’s view of how a mother should and shouldn’t behave (see O’Reilly, 2019). For instance, a number of participants suggested that their work with mothers is mostly education-focused and that they address the women’s parenting skills during their shelter stay. For them, the children’s difficult behaviour in the shelter is related to the women’s parenting skills. This supervisor suggests that part of her duties is to work towards behavioural change with the mothers:

We work on parenting skills with the mom to help her manage crises and bedtime routines. Sometimes, we rehabilitate the parent because of laxity with their children. So we can help her: we give her the tools, and she puts them into practice. (Advocate 2, supervisor)

Some participants raised the issue of women not being aware of the consequences of domestic violence on their children, perceiving that an important part of their responsibility is to increase women’s awareness about the potential consequences for their children. They believe it is up to the mothers to address the consequences that domestic violence has had on their children. For these participants, ensuring that the women understand these consequences is one of the main components of their work, as they educate the mothers and monitor their knowledge and awareness of the issue. This is illustrated in the following quote:

I always ask them if they think domestic violence has had an impact on their children, regardless of their age. Even if they are only pregnant, I ask them the question. After, I talk to them about exposure to domestic violence. [ . . . ], but regardless of their needs, if there is domestic violence, we talk about the child’s exposure; this is non-negotiable. (Advocate 25, youth worker)

This quote provides insight into the notion of ‘responsibility’ regarding potential risk (see Parton, 2017). However, this outlook on children and on women-as-mothers having several flaws seems to set aside the broader context of domestic violence, as the women’s parenting skills and the way they interact with their children shift to the forefront of the workers’ practices. In this sense, some go as far as monitoring mother–child communication in the shelter and ensuring the women listen and talk to their children. However, if the women talk ‘too much’ to their children and discuss their experience of violence, they may be perceived as engaging in parental alienation, thus urging some workers to intervene as illustrated in the following quote:

The mother–child worker works a lot on parental alienation [ . . . ]. She works on: “What happens between the father and you is between you and the father. Children have no role in this. For them, it’s Mom, it’s Dad: it’s the same. They love their mom, they love their dad.” (Advocate 29, front-line worker)

In the same vein, workers who perceive the children as vulnerable and ‘at-risk’ and who have a more negative view of the mothers are less likely to feel torn if they have to call child protection services. They are also less likely to make a decision by reflecting on the risks of reporting and more likely to make multiple referrals per year. In some cases, they agree to provide child protection services with information about the women’s behaviour and mothering skills in the shelter. Perhaps this is related to their view that the central problem lies in the women’s ability to protect their children, and not in the behaviour of the perpetrators. For instance, if the women return to the perpetrators, the logic behind the referral becomes their inability to adequately protect their children. It should be noted that the perpetrators’ violence disappears from their discourse and their concerns are almost entirely focused on the women’s ability to protect rather than the perpetrators’ violence. This is explicitly stated by the following two participants: ‘It could be that Mom, for instance, is not aware of this violence and is not able to take all the necessary measures to protect her children’ (Advocate 24, front-line worker). ‘That’s something we’re going to say to her: “If you go back to him, I have to report because you won’t be protective of your child.”’ (Advocate 4, youth worker).

This shift in the root cause of the problem faced by children in the shelter has important implications, which will be discussed in the following section.

Discussion

One of the main findings of the study from which this article is drawn (Côté, 2016) is that the overwhelming majority of shelter workers who adopt a feminist perspective on domestic violence apply this framework directly to their work with the women in their practices. However, while the feminist perspective provides the lens through which they understand domestic violence, they sometimes draw on other theoretical frameworks to explain some peripheral issues women may be facing, such as substance abuse and mental health problems. The same can be true with regard to the children and women-as-mothers in shelters. This finding is aligned with a theoretical reflection by Lapierre (2010), who argues that the feminist perspective has been marginalized in the development of social work practices related to children living with domestic violence, particularly in the child protection arena. The results of this study point out that, while the difficulties faced by women and children stem from male violence and domination against women and children within the family, the reality of children is not always conceptualized from a feminist perspective, even in domestic violence shelters with clear feminist guidelines. The added notions of risk and responsibility (see Parton, 2017) as they pertain to children can create some challenges for feminist shelter workers.

As such, when focusing on children as individuals in their own right, feminist and social work values emerge from the participants’ accounts of their practices, including safety, dignity and to a certain extent, self-determination. In this article, we have argued that the more the participants focus on the potential impact of domestic violence and related issues on children’s lives, the more likely they are to shift their attention from the root cause of the problem in the children’s lives (men’s violence against women) to the women’s parenting skills. Participants who moderately emphasized vulnerability and risk believe that their role is to work in alliance with women in their role as mothers. From a feminist social work perspective, this evidently requires a focus on strengths and empowerment. Those who strongly focus on vulnerability and risk might, however, emphasize a risk-management approach and monitor the mothers in the shelter by dealing with their flaws and assessing their behaviours. This confirms research evidence that shelter workers may scrutinize women’s parenting skills (Gengler, 2011; Krane & Carlton, 2012; Krane & Davis, 2002, 2007; Peled & Dekel, 2010) and favour behavioural change rather than empowerment (Gengler, 2011). By doing so, the perpetrator’s behaviour becomes peripheral in their discourse. As a result, they may fail to take into account the broader perspective of the children’s experiences and frame it as a political issue.

Interestingly, children living with domestic violence generally have a positive view of their mothers and understand that even if the perpetrators’ violence does create challenges and difficulties for themselves and their mothers, their relationship still improves over time (Lapierre et al., 2017). Communication is a key component of their relationship, and asking mothers to censor themselves might be more harmful than helpful, for both the well-being of their children and the mother–child relationship. In this sense, the practices of shelter workers – who monitor and interfere in mother–child communication so that the women do not paint a negative image of the fathers – appear at odds with the feminist perspective, in addition to being inconsistent with the views of a number of children living with domestic violence who express a wide range of negative feelings towards their fathers or their mothers’ partners (Lapierre et al., 2015). The assumption that the violent man and the father are two distinct people, that abusive men can still be good fathers, and that children always love their father unconditionally nonetheless requires in-depth reflection.

Limitations of the study

The study has three main limitations. First, all of the participants were recruited on a voluntary basis, either through a snowballing sampling technique (pioneers and veterans) or through shelter directors who agreed to have their staff participate in the study. They may not be representative of all shelter workers across the province where the study was conducted. Second, the data were collected from a sample of French-Canadian participants, and some of the findings may be limited to that particular cultural context. Third, the study from which this article was drawn did not focus specifically on interventions with children in domestic violence shelters; the topic surfaced during the interviews. As these data on children were not the primary focus of the research, they are likely to be limited and lack the depth of a more detailed investigation. Further studies are thus needed to expand on the results presented in this article. More precisely, studies that seek to shed light on best practices when working with children in alliance with their mothers from a feminist perspective are crucially needed. Studies investigating whether too much emphasis on the vulnerability of children could ultimately hinder their safety by making abused women reluctant to seek services (i.e., mandatory reporting to child protection services) would be particularly insightful in this field of research.

Conclusion

Intervention with children constitutes an important stepping stone in the history of domestic violence shelters. In this article, we have argued that, as long as the broader context of domestic violence (i.e., a male perpetrator who is abusing their mother) remains in sight, workers will support children in their own right and in alliance with their mothers. However, if children are perceived as vulnerable and ‘at-risk,’ workers’ practices may move away from a feminist analysis of the experiences of children and women in their role as mothers and become controlling, blaming and even punitive. Paradoxically, violent men do not necessarily seek services for their behaviours and do not face similar monitoring and control of their paternal skills. This provides insight into the robustness and normative pressures of the traditional social institution of motherhood in social work practices and even in feminist shelters.

It seems important to reiterate that the main reason children end up in shelters and may suffer the consequences of domestic violence, and that women can be negatively affected and experience struggles in their role as mothers, remains directly connected to the violent behaviour of men towards women and children in the family. The situation of children and their mothers thus needs to remain rooted in a feminist analysis of violence against women in order to avoid some of the issues discussed in the article.

Finally, if the study from which this article has been drawn took a closer look at the interventions of domestic violence shelters, similar implications would arise with regard to social work practices in other agencies or programmes that provide services to abused women and their children. Carefully balancing risk and safety while maintaining a trusting relationship with abused women in these cases appears to be more helpful than monitoring their behaviours (Lapierre & Côté, 2011b). If women fear services, they may hesitate to seek help for themselves and their children in the future; this can have important consequences for their children’s safety. Engaging with mothers despite challenges is thus crucial if we want to help them, as mothers and as women (Baldwin, 2015), to ultimately support their children. More importantly, perpetrators should be the ones being closely monitored to keep children safe and to send a strong message regarding their responsibility for their behaviour and its consequences on their children.