Support and Autonomy: Social Workers’ Approaches in Dutch Shelters for Female Survivors of Domestic Violence

Abstract

It is not uncommon that women residing in Dutch shelters following domestic violence consider returning to their partners during the course of their stay. Social workers cannot prohibit return due to the importance of the client’s autonomy, as stated in the Code of Ethics. Simultaneously, social workers aim to ensure women’s safety and encourage a positive future, which can lead to tensions in their way of support-giving. Based on thirty-five interviews with social workers, this study explores how they navigate these tensions and the support they give in such cases. Building on Martha Nussbaum’s capabilities approach, we distinguish a hands-off and an interventionist approach to autonomy. While the hands-off approach leads to non-intervention and respecting the woman’s decision, the interventionist approach focuses on providing tools and encouraging safer alternatives. The findings show that the hands-off approach is the common type of support, leading to feelings of powerlessness for social workers and often prompting endeavors to intervene more actively. Simultaneously, we argue that, whilst the Code of Ethics demonstrates an ambivalence towards these different forms of autonomy, it is often interpreted by social workers as promoting only a hands-off approach. Consequently, we argue that social workers would benefit from a greater recognition of an interventionist approach to autonomy and more liberty in support-giving.

1. Introduction

“You try to talk with a woman and give her insights on safety. Coming [to the shelter] is voluntary of course, it’s her own choice. You have women who choose to go back to their partner. And sometimes you also see that then something happens, but at least, the social workers did advise not to go back. But you can’t force someone to stay.”

—Nour

Nour is a social worker working at a Dutch shelter where she meets women from diverse backgrounds who have fled abusive relationships. The quote shows her response to the question of how she would advise women opting to return to their partner, as this is not uncommon (Griffing et al. 2002, 2005; Edwards et al. 2018; Herman 2019). Such matters are frequently discussed with social workers. As expressed in the quote, social workers are limited in the scope of the support they are able to offer. This is because they are bound by a Code of Ethics (BPSW 2021), which provides guiding principles in their professional conduct. For example, the Code states that the client’s self-determination is paramount, and that the client must be respected in her decision-making process.

However, the Code also states that one of the aims of social work is to facilitate and encourage “sufficient development opportunities and a safe living environment” (BPSW 2021, p. 11), and that social workers should seek to “promote emancipation of individuals, groups and communities” (ibid., p. 8), and “support and guide people to achieve their goals and life fulfilment” (ibid., p. 11). Thus, whilst the Code foregrounds a model of client self-determination that is individual in nature, and that social workers often interpret as requiring —what in this article is referred to as—a hands-off approach, this sits alongside a more interventionist approach that supports individuals in seeking, and sustaining, a safe environment and the achievement of fulfilment in a broad sense. Although self-determination is utilized in the Code of Ethics (BPSW 2021), when exploring the tensions, a perspective of autonomy is adopted, as it encompasses both interventionist and hands-off approaches. For instance, self-determination is not synonymous with guidance and support, thus offering a comparatively limited potential for intervention.

With these ambivalences in mind, we argue that, for many social workers, a potential conflict emerges between these concepts in cases where a woman wants to return to a potentially unsafe relationship. On the one hand, social workers are supposed to support the safety of the women and to assist them to achieve a more fulfilling life, which, as these women have left their violent partners, will include a violence-free life. On the other hand, it is also imperative that social workers respect each woman’s decision, even if it leads to going back to an abusive situation.

Consequently, tensions can arise regarding the support social workers provide: how much can, and should, they intervene in women’s decisions? These tensions are exacerbated in contexts where women of a migrant background are navigating decisions to leave, or return to, a violent relationship. For several years, women of first- and second-generation migration backgrounds, often from non-Western cultural backgrounds, have been overrepresented in Dutch shelters (Jonker et al. 2011), due to, among other factors, a less extensive social network in the country leading to a lack of alternative pathways of support (Bartels 2021). This group of women is often burdened with dependent residence permits (Bartels 2021; de Hart et al. 2022; Roegiers Mayeux et al. 2023) and sometimes cultural expectations that reduce their autonomy, adding further complexity to the tensions faced by social workers. For these women, the consequences of leaving a partner could be losing the right to stay in the Netherlands, and facing shame and rejection from their relatives and social networks in cases of separation (Ahmad et al. 2009; Aboulhassan and Brumley 2019; Hulley et al. 2023).

With this article, we explore the influence these tensions have on social workers’ support, and focus on individual autonomy (henceforth autonomy) as a guiding framework to examine the decisions taken by social workers. We draw on Nussbaum’s perspective on autonomy not only because she has addressed domestic violence as a significant issue, but also because her approach offers a means to address the wider complexities of autonomy in relation to domestic violence. In her theory, the social and economic conditions for autonomy feature prominently, and according to her, intervention is warranted when these conditions impede autonomy (Nussbaum 2000). In social work, too, these conditions are an important issue. For instance, the Code of Ethics (BPSW 2021) states that social workers have a role to play in improving people’s living conditions and addressing constraints on their autonomy (BPSW 2021, p. 11). The Code also explicitly states: “A professional thus focuses on the micro level of the individual, the meso level of the immediate social environment and the macro level of the ordering of society” (p. 5). Simultaneously, Nussbaum encourages respecting women’s autonomy and self-determination in decision-making (Nussbaum 2000), which in the Code of Ethics (BPSW 2021) is also a core value. Due to these connections, we have drawn on Nussbaum’s approach to examine the tensions of autonomy in the work of social workers.

Following Nussbaum’s approach, we differentiate between two approaches to autonomy: an interventionist and a hands-off approach. While Nussbaum does not employ these terms explicitly, there are compelling reasons for doing so. The rationale underpinning their use is that they accurately reflect a paradigm that fosters a safe environment, while simultaneously demonstrating respect for the autonomy of women in decision-making. An interventionist approach to autonomy aligns with Nussbaum’s conceptualization of autonomy as a capacity. This means that interventions are aimed at supporting women in strengthening their capacities, like the ability to critically reflect on their choices and to realize their life plans, including having a safe and violence-free life. In contrast, a hands-off approach aims to respect women’s choices, arguing that autonomy is a matter of free choice and self-determination, regardless of the outcome. Consequently, with these distinctions in mind, this paper asks: How do social workers say they navigate the contradictory approaches to autonomy and support when working with women considering returning to an abusive relationship?

This article is divided into two sections. The first focuses on the recognition of the existence of multiple tensions written in the Code of Ethics (BPSW 2021), guiding social workers. The second presents an analysis of how these tensions can be addressed through support from a hands-off and interventionist perspective on autonomy. Is one of the two paramount for social workers, and if so, how does that influence their support-giving? Both sections commence with an examination of the experiences of social workers with women they work with, regardless of the latter’s cultural background and migration history. We then proceed to examine whether additional complexities emerge when considering the intersection of cultural background and migration history.

Thirty-five social workers of diverse positions and locations based in the Netherlands were interviewed for this research. In the next sections, we discuss the current literature on domestic violence, social work, and autonomy, before describing the methodology and examining the results and implications of the research.

2. State of the Art

2.1. Staying with, Leaving, or Returning to a (Violent) Partner

The Netherlands was among the first European countries to implement a policy on domestic violence in 1984 (Roggeband 2002, 2012). Nevertheless, non-funded shelters had been in operation since before the formal policy was established, providing assistance to survivors of violence (Roggeband 2002). Shelters provide temporary, emergency accommodation on a country-wide basis, offering a safe environment for those in need. The permitted duration of stay for survivors varies by city, with a maximum period of nine months (Bargellini 2021). During their stay in the shelter, women are offered various forms of assistance, similar to other countries (Cattaneo and Chapman 2010; Gregory et al. 2021; Robinson et al. 2020). The objective is to reduce negative emotions such as stress (Jonker et al. 2011) and to empower women to make their own decisions and reclaim their autonomy (Robinson et al. 2020).

Internationally, scholars have considered multiple factors that influence women’s decisions to stay in, or to leave, an abusive relationship (Rhodes et al. 2010; Clark 2016; Heron et al. 2022). For instance, staying can be perceived as safer than (trying) to escape. This is especially the case for women who have experiences with intimate terror as a mode of domestic violence. Research indicates that intimate terror is characterized by patterns of control and coercion that can lead to violence and femicide. The fear that a partner may leave a perpetrator significantly increases the risk of violence and harm (Corradi et al. 2016; Epstein and Goodman 2018; Monckton-Smith 2021). In such contexts, women may decide to stay with their partner as leaving may be life-threatening. Other factors which have been shown to impact decisions to remain in, or leave, an abusive relationship are the presence of children (Rhodes et al. 2010), financial dependency (Jeltsen 2014, Sep 12), belief that the abuse was normal (Clark 2016), and the fear of not being believed when reporting violence (Heron et al. 2022). Additionally, survivors of domestic violence do not always regard professionals as trustworthy, in particular fearing that their children will be taken away if they share their experiences of abuse (Salmon et al. 2015; Heron et al. 2022).

A closer examination of women of diverse cultural backgrounds reveals that they encounter the same obstacles mentioned above, but with a range of additional factors. For example, cultural expectations may co-exist with stigma and the exclusion of the woman and her family members by their community, which could ultimately result in the woman losing (part of) her social network (Ahmad et al. 2009; Aboulhassan and Brumley 2019; Hulley et al. 2023). Furthermore, residence status is a factor that deserves attention, especially for women of first-generation non-European migration background who often arrive on a dependent residence permit (Okeke-Ihejirika et al. 2020; Elbelassy et al. 2023). For example, women with an insecure residence status, such as refugees and marriage migrants, are subject to a constant fear of removal from the country. For instance, during the first five years, marriage migrants’ right to stay depends on the residence permit of their spouse. In cases of separation, the spouse can ask the state to terminate the residence permit, which may result in the marriage migrant being required to return to their country of origin—as she can only ask for an independent residence permit after five years. If domestic violence occurs, it is possible to apply for an independent permit; however, it is difficult to obtain the evidence required to acquire this. As a result, some women tolerate violence due to their fear of losing the right to stay (Voolma 2018; de Hart et al. 2022; Roegiers Mayeux et al. 2023).

While there are plenty of reasons why women choose to stay in a relationship, there is also past research exploring what makes women leave an abusive situation (see, for example, Ulrich 1991; Yamawaki et al. 2012; Heron et al. 2022). Alongside wanting to end the abuse (Ulrich 1991), another common concern is the protection of children (Heron et al. 2022). In the research of both Yamawaki et al. (2012) and Heron et al. (2022), participants noted that leaving became less difficult when receiving formal (e.g., professionals) and informal (e.g., relatives) support.

At the same time, it is not uncommon that women return to their partner after leaving them, often at a period of fraught consideration (Griffing et al. 2002, 2005; Edwards et al. 2018; Herman 2019). For example, Herman (2019) examined a range of existing literature on returning to an abusive relationship and found that often women go back to their partner due to emotional attachment and love, citing the willingness to forgive their partner. Alongside this, financial dependency, such as the need to support yourself and, if applicable, your children, was also a factor seen to affect decisions to return. Building on such work, Monckton-Smith (2021) argues that a fear of femicide is a further powerful reason driving some women to return to an abusive partner.

These factors are all significant as they affect and may constrain the choices made by a woman. To explore this further, we next consider what autonomy means in such contexts, and focus on both hands-off and interventionist approaches to autonomy.

2.2. A Hands-Off and Interventionist Approach to Autonomy

In its classical understanding, individual autonomy means to live a self-directed life. To be able to exercise autonomy, certain conditions must be met. For instance, people need to be free from coercion or manipulation, and their life choices must be respected. This, thus, requires non-interference, even where people make choices that may not appear to be in their best interest. In this sense, autonomy maintains that people have a right to make their own mistakes (Friedman 2003).

For Nussbaum (2000), this freedom to live according to one’s conception of the good requires that other people respect a person’s choices in life and do not interfere with those choices. She emphasizes that respecting people’s decisions is an essential aspect of recognizing their dignity. Therefore, Nussbaum encourages respecting people’s decisions, even if they might not fully align with her vision of autonomy. This is because respecting a person’s decision is part of recognizing their inherent worth and dignity.

Yet, Nussbaum argues that non-interference might not be sufficient to assure women’s autonomy in their capacity to lead lives that are of significance to them. She states that “the central human capabilities, then, are those that allow individuals to lead lives that they have reason to value. Women’s equality, on this view, is not just a matter of legal rights, but a matter of actual freedom to pursue these central capabilities, and to make choices about how to use them in ways that are meaningful to the individual” (Nussbaum 2000, p. 78).

Consequently, this demands that the structural inequalities that constrain women’s autonomy and capabilities need to be addressed. These inequalities frequently manifest themselves as economic dependence, social subordination, sexual violence, and discriminatory legislation. Nussbaum thus argues that to have autonomy, “…people must have the ability to make real choices. This requires not only the absence of legal impediments to their freedom but also the presence of the necessary social and economic conditions, including adequate education, healthcare, and opportunities for employment. For women, this means that autonomy cannot be achieved unless these basic capabilities are accessible, and the social structures that limit their access to these goods are dismantled” (Nussbaum 2000, p. 82).

In the context of domestic violence, a range of factors ensures that these conditions for autonomy might not be realized. For example, dependent residency permits, a lack of income, or a lack of housing options all constrain the possibility of autonomous decision-making following incidents of domestic violence. In the context of social workers’ support, from this perspective, intervening in a woman’s situation can be morally justified if this entails addressing the social conditions that impede her autonomy and that help the woman in making choices that reflect her deeper wants and values. For instance, in the case of domestic violence, this would include a violence-free life as a condition for well-being.

In building on this work, we argue that autonomy can require both non-interference in an individual’s decision and intervention to capacitate a person to exercise autonomy. First, we name a hands-off approach, and second, an interventionist approach to autonomy. The hands-off approach to autonomy is characterized by the principle that interventions in women’s decisions should be discouraged, with the fundamental concern being respect for the woman’s decision and self-determination. An interventionist approach demands active intervention in order to realize the conditions for living a life that they have reason to value. As social workers are strongly encouraged to adhere to the Dutch Code of Ethics (BPSW 2021), the following section will examine how autonomy is interpreted within this framework.

2.3. The Dutch Code of Ethics and Domestic Violence

The aim of social work is to help individuals and communities reach well-being. To help Dutch professionals achieve this aim, a Code of Ethics was developed based upon a set of core values that are regarded as significant within the field of social work internationally as well as in the Netherlands (BPSW 2021; International Federation of Social Workers 2025). The Code of Ethics is for social workers in general and thus not specifically focused on women’s shelters and domestic violence.

The Dutch Code does not refer explicitly to autonomy as a term, but instead refers to human dignity and self-determination as core values of social work. About the latter, the Code states: “respecting the self-determination of the individual, address the specific qualities, aspirations and limitations of the individual and promote emancipation of individuals, groups and communities” (BPSW 2021, p. 8). Therefore, the initial aspect of “respecting the self-determination of individuals” we have interpreted as the Code’s view on autonomy, entailing, according to the Code (BPSW 2021, p. 8), that social workers should respect their clients’ capacity to make decisions, without compromising the process through which these decisions are made, and avoiding paternalism. This could mean, as it is explained, that they should help women to make choices, without forcing their perspective on them. Considering the theoretical foundation of autonomy mentioned in the previous section, we suggest that this represents a hands-off approach to autonomy. However, this does not imply that social workers cannot express disagreement with their clients’ decisions. The idea behind this is to stimulate reflection on the part of the client, without putting too much focus on the results or consequences.

However, the Code also offers scope for an interventionist approach to autonomy as it argues that “sufficient development opportunities and a safe living environment” (BPSW 2021, p. 11) should be encouraged, and that the promotion of the emancipation of clients also should be taken into account. Social workers are there to “support and guide people to achieve their goals and life fulfilment” (BPSW 2021, p. 11). Social workers must ensure that the client’s interests are clearly identified and that decisions about their lives are made in line with those interests, while empowering them in achieving such fulfilment. In the context of a violent relationship, this could mean encouraging the individual to first bring an end to the abuse. In these terms, we suggest that there is an ambivalence at the heart of the Code of Ethics (BPSW 2021) as this can be read as promoting both interventionist and hands-off approaches to autonomy.

In this context, it is challenging for social workers to assess what is in their client’s interest and how they should be guided in their support-giving. Social workers are trusted with their experience and have a degree of “professional autonomy” (ibid 2021), implying “the freedom and responsibility that professionals have to do justice to people, groups and communities according to their professional judgement. They act within frameworks that determine their scope of action, also called “discretionary space” (BPSW 2021, p. 28). Yet, social workers are not alone in making decisions regarding their interventions. Not only do they have to follow the Dutch regulations (Roggeband 2002, 2012), but they are also part of a professional network of various actors working on domestic violence. For instance, the Code of Ethics states that “Consulting with immediate colleagues, experts and other colleagues helps in applying the abstract standards in everyday work” (BPSW 2021, p. 7). Therefore, consideration and reflection about the severity of the situation, often between colleagues, is of importance.

Moreover, regarding domestic violence, there is a reporting code to higher authorities such as Veilig Thuis1. The reporting code states that the right to report without consent and cooperation is justified only when it has been established that the violence is structural and that the provision of assistance is no longer effective. Here, again, the word “right” is of importance. Looking at it from a hands-off approach to autonomy, if social workers report without the client’s consent, they could be violating their autonomy. In cases involving children and violence, it is imperative to consider the least harmful course of action for the child in question (BPSW 2021; Ministerie van Algemene Zaken 2024). Therefore, social workers do not work alone when it comes to supporting women, as they can ask for advice and guidance from other actors with knowledge and expertise, such as Veilig Thuis, and, although we did not disclose information about it in this article, in case of honor with domestic violence, a center (LEC-EEG2) specialized in the phenomenon.

Thus, the Code of Ethics includes multiple guidelines that are open to varied interpretation, with tensions between the implications of these different readings. On the one hand, social workers must respect a client’s decision [hands-off approach], while on the other hand, they must strive for a violence-free future for the client, which demands intervention [interventionist approach]. In practice, this tension can be found in the support social workers feel able to offer. We turn to these tensions after a brief discussion of the methodology.

3. Methodology

3.1. The Broader Study

3.1.1. Research Topic

This study forms part of a broader study analyzing the intersections between domestic violence, gender, migration, cultural background, and social work from a feminist perspective. While the existing literature on domestic violence and social work (Keeling and Van Wormer 2012; Van Gelder et al. 2021) is extensive, little research has explored the intersection of these phenomena. In this context, the broader study seeks to examine potential gaps in support for women within social work and proposes practical solutions to address such gaps.

The wider study is divided into two parts regarding the insights of the women and social workers. One part of the study explores the experiences of women before arriving in the shelter, diving deeper into the factors that made it more, or less, difficult to leave the abusive relationship. Three main factors have been identified: family dynamics, institutional factors, and cultural expectations. The second part delves deeper into the interventions of the social workers towards the women they work with. Next to this research, we examined the role of the social worker’s identity and positionality and how these factors influenced the assistance they were able to offer women.

3.1.2. Data Collection Strategies

Two main methods have been used: participant observation and interviews. Highlighted are the voices of, on the one hand, women who experienced domestic violence, and on the other, social workers and team leaders working with the former. In total, 47 participants were incorporated: 10 women and 37 social workers, including leading positions.

3.1.3. Respondents

All 47 respondents identified as women. Of the survivors, 5 of them already had (acquired) Dutch citizenship. The others stayed in the Netherlands on a dependent residence permit of marriage migration, and had the nationality Moroccan (n = 2), Iranian (n = 1), Pakistani (n = 1), and Surinamese (n = 1). All the social workers, including the team leaders, were Dutch citizens.

3.1.4. Data Collection

All interviews were recorded through audio. Before the interview, the respondents received a consent form and an information letter in French, Dutch, or English, depending on the language they preferred. The letters were read together with the first author. Afterwards, but also during the interview, the participants could ask questions and request more information. All respondents provided written consent.

3.1.5. Data Analysis

ATLAS.ti 22 was used to code the interviews, using axial coding. The first author coded the interviews following patterns and the recurrence of answers. Afterwards, similar topics were divided into categories and themes. All non-English quotes were translated from French and Dutch.

3.2. This Study

3.2.1. Research Design

The study employed a qualitative approach to focus on understanding the experiences of the respondents, drawing on a narrative research framing. By making use of this framing, it was aimed to make an in-depth and detailed analysis of the way in which stories are interconnected, as well as the way in which they are socially produced and contextualized (Creswell et al. 2007).

3.2.2. Data Collection Strategies

For this study specifically, interviews were utilized and supplemented with observations from fieldwork in one shelter. Interviews entail the examination and interpretation of respondents’ experiences, with the objective of comprehending social realities from the perspective of the participant (Boeije and Bleijenbergh 2019). Furthermore, this approach enables the examination of depth and detail. While the vast majority of the data on which this paper is based are drawn from these interviews, several participant observations are also employed to offer critical context. In this paper, we draw on two central observations from this period of fieldwork, notably the prevalence of women returning to their partner after leaving the shelter, and the way social professionals discussed the challenges of supporting women upon leaving the shelter during informal moments.

3.2.3. Sample of Respondents

From the forty-seven interviews, in this article only the ones with the social workers (n = 35) are used, as they worked directly with the women. Some only worked with women and their children, others with their (previous) partner as well. The shelters where the social workers worked were located in the provinces of Utrecht, Gelderland, North Brabant, South Holland, North Holland, and Friesland. All respondents had a minimum of five months of working experience with women who were exposed to abuse. Twenty-six had degrees in social work. Other degrees were in psychology and pedagogy. Therefore, all respondents had a degree in a social field similar to social work. In the Netherlands, shelters are institutions that fall under the umbrella of social work, therefore, professionals working in and with shelters are considered social workers. Those interviewed came from a range of cultural backgrounds. When asked for their self-chosen cultural background, most respondents identified as Dutch. Nevertheless, Cape Verdean, Surinamese, Moroccan, Hindu, Kurdish, Turkish, Maluku, Tunisian, Brazilian, Indonesian, or a combination with Dutch were also mentioned.

Among the social workers, we identified four functions. Most of the respondents were residential assistants focused on supporting women in the shelter by being present every day, providing activities, and helping with daily life. Secondly, some were personal assistants accompanying women on a legal, social, and psychological basis. Women who arrive in the shelter generally receive a residential and personal social worker to discuss their problems, cases, and other concerns. Thirdly, there were also ambulatory assistants who help and support women in their own homes. Lastly, and less commonly interviewed, there were child(ren)–mother relationship specialists working more closely with mothers and their children. These social workers had already worked as personal or residential assistants before. At the time of the interviews, ages ranged between 22 and 65 years.

Table 1. Respondents’ characteristics.

Pseudonym	Self-Identified Cultural Background	Age	Role
Aisha	Turkish	25	Residential
Alex	Dutch	37	Ambulatory
Anne	Dutch	30	Personal
Catarina	Dutch	36	Residential/team leader
Cloë	Cape Verdean-Dutch	25	Residential
Ela	Turkish	50	Residential/admission
Eveline	Dutch	58	Personal
Flora	Dutch	42	Personal
Juudje	Brazilian-Dutch	25	Residential
Katharina	Dutch	64	Residential
Lena	Dutch	50	Residential/personal
Lianne	Dutch	35	Personal
Lin	Indonesian-Portuguese-Dutch	41	Residential/ambulatory
Lina	Moroccan-Dutch	33	Residential
Louise	Dutch	61	Ambulatory
Marie	Dutch	47	Ambulatory
Maryam	Moroccan-Dutch	Exact age unknown	Child(ren)–parent
Meyra	Dutch	48	Ambulatory
Mina	Tunisian-Dutch	24	Residential
Mona	Surinamese	56	Residential
Nanda	Maluku-Dutch	49	Personal
Naya	Afghan-Slavic	31	Personal
Niya	Moroccan	31	Residential
Nour	Moroccan-Dutch	44	Ambulatory
Nynke	Dutch	49	Personal
Reile	Dutch	65	Ambulatory
Rolinka	Dutch	23	Residential
Rosa	Hindu-Dutch	26	Personal
Sammie	Dutch	57	Ambulatory
Sanne	Asian-Dutch	22	Residential
Sara	Moroccan-Dutch	38	Ambulatory
Schnreue	Kurdish	51	Ambulatory/HBV-expert
Sofia	Moroccan	31	Child(ren)–parents
Willeke	Surinamese	60	Personal
Yvonne	Indonesian-Surinamese-Dutch	25	Residential

shows respondents’ characteristics.

3.2.4. Data Collection

All the interviews were in Dutch and lasted between forty and ninety minutes. During the interviews, only the respondent and the interviewer were present. Afterwards, all interviews, except for the first one, were sent to a firm to transcribe them. Social workers received their interview once transcribed and could add, remove, or make changes. Information on the data analysis is described in Section 3.1.5.

4. Results

4.1. The Complexity of Autonomy

4.1.1. Tensions Between Safety and Autonomies

As observed during fieldwork, and shared by all the social workers we spoke with, it is not uncommon for women to choose to go back to their partner. In some cases, women leave the shelter overnight without any prior notice. When leaving the shelter earlier than expected, women are required to sign a document indicating their intention to leave. If women think of leaving the shelter, social workers encourage the women to communicate openly if going back is their wish. This is confirmed by Nynke:

“I also say to a woman: you can always go back, but be honest with me; if you say ‘I’d like to go back’, we’ll work on that, we’ll see if we can let your relationship continue in a safe way.”

—Nynke

Social workers generally shared that they do not always oppose the idea of women returning to their partner, depending on the perceived safety of such a move. When social workers assess that the situation is safe enough, and thus that the violence will no longer occur, they do not dissuade a return. Instead, they considered how the relationship could continue safely, as by Aisha:

“I’m not so much concerned with whether they’re going to separate or whether they’re going to move back together, but it’s really about safety for me. If you’re going to choose to separate, how can we make that as safe as possible? And if you’re going to choose to go back, how can we make that as safe as possible?”

—Aisha

Although safe returns do occur, this is by no means a guaranteed outcome. Critically, as several social workers noted, the existence of a trajectory towards a safer return does not signify that the outcome will be safe:

“There was also custody, supervised visitation, all kinds of things. And in the end, they are now back together. If that is really the woman’s wish, then you try to prepare and support her in that and hand it over as best you can, as safely as you can, and let it go. It’s not for me to say: “no, you really shouldn’t do that or that’s not wise”. But it is important to make her aware of the possible risks involved; what are you going to do if it happens again?”

—Lin

Following on this, as Lin notes, the women’s ultimate choice is the one social workers have to respect. This idea is compatible with the hands-off approach to autonomy, where support should be limited as it prioritizes the autonomy of the individual woman in making her choice independently. Nevertheless, Lin does not necessarily endorse the decision as a good one, given the potential for continued risk in the relationship. Consequently, it can be observed that there may be a tension between ensuring the safety and well-being of a woman and respecting her decision, both values shared in the Code of Ethics.

Adding further complexity is the fact that some women also felt pressured by additional constraining factors. Social workers shared that they also work with women who envisaged returning to an abusive situation because of cultural expectations and institutional barriers, such as the coercive nature of residency permits. Social workers noted that these factors were particularly prevalent in affecting the decision-making process of women from first- and second-generation migration backgrounds where non-Western cultural expectations may be present that prioritize staying within a marriage or relationship over other considerations, and where precarious immigration status may serve to constrain the choices women can make.

4.1.2. The Role of Cultural and Institutional Pressure in Going Back to the Partner

Social workers acknowledge that they have worked with women who were faced with cultural expectations that say that they should stay with, or go back to, their partner, even if there was ongoing violence. Social workers recognize cases like this following a familiar pattern: during their relationship, women had good contact with their family. However, leaving the partner might have led to a rejection and rupture. This rupture can be temporary, but sometimes women need to choose between staying separated and losing (part of) their social network, including relatives, or going back to the violent partner and also reconciling with their network. The social workers we worked with noted that this was a particularly difficult choice. While some social workers were more familiar with such cultural expectations due to growing up with similar expectations themselves, for others, such as Lena, this cultural difference is something they constantly need to keep in mind:

“For us [Western perspective] it’s different, but then for this woman it could be a good option to go back because the burden of losing her family as a consequence of a divorce could be worse.”

—Lena

Moreover, six social workers indicated that, in their experience, women frequently knew that violence was unacceptable. Simultaneously, they also acknowledge that women who are embedded in family structures and subject to stigma frequently aim to prevent disgracing their families. Mone is one of these social workers who has experienced this stigma after leaving her partner herself before becoming a social worker:

“Of course you have women who think: ‘You know what, maybe I’m being too difficult? What are those few slaps? As long as I’m around my family.”

—Mona

On occasion, this can result in a woman being compelled by her environment to decide to go back if she wants to keep in touch with her family. In this case, women are caught in a bind of negative consequences either way, risking losing their relatives, or risking the continuation of violence and harm. In such cases, social workers share that women who have difficulties with making this choice might especially want to talk with social workers of similar cultural backgrounds and experiences. The similarities can help to compare the situations. For example, Ela noted that:

“I am an immigrant, I am a divorced, single mother, and I manage. They can also take me as an example.”

—Ela

In addition, women of first-generation migration background coming from non-European countries may also experience institutional barriers regarding residence permits, sometimes in addition to the cultural expectations and lack of family support. It is not uncommon for a woman to arrive in the Netherlands on a dependent residence permit in the hands of her partner, who may then ask to cancel her residence, resulting in her being required to return to her country of origin. Schrneue recounts:

“[women say:] “I have not been here [the Netherlands] for very long, I don’t have a child”. And chances are very high that the IND3 will then say: you won’t get a residence permit, you have to go back to your country. That could be the motive for someone to say [I’m going back to the partner], and rightly so. This also has to do with a fear that the family would kill her.”

—Schrneue

Women arriving on a dependent residence permit, such as marriage migrants, are thus often in a highly vulnerable situation, and one that can be accentuated by cultural expectations. In some cases, being sent back to their country of origin was feared more than experiencing violence, leading women to endure violence in the Netherlands until they have more safety regarding their residence permit.

In Schrneue’s quote above, we also see that, for some women, leaving their partner does not mean the end of the violence. For instance, family members might be opposed to the decision to leave and may react to such a decision in a negative way. In addition, some partners may be vengeful and pose ongoing risks to women who have left. For example, in one shelter Sara referred to the case of a woman murdered by her former partner after she had left him:

“We recently had a new case, where the woman was actually murdered. A woman who lived with us. […] my colleague did everything she could [to help].”

—Sara

This example shows that, for some women, staying separated from the partner can lead to dangerous consequences and requires considerable support and security. At the same time, the risk of such violence engenders further fear for the woman, serving to act as an additional reason for them to consider returning to their partner. In these contexts, the risk of violence weighs heavily on both moments of return and possibilities of escape.

The most common friction that social workers experience was therefore between having to respect the woman’s choice and worrying about her safety and well-being. Consequently, this prompted the question of how social workers feel about these tensions, and more importantly, how they choose to support women, while keeping in mind their Code of Ethics.

4.2. A Hands-Off or Active Support?

4.2.1. Support and Autonomy

Social workers share that, in contexts where children are not present, women’s freedom in decision-making takes priority. This means that they should not intervene too strongly, adopting a hands-off approach of autonomy. This often resulted in feelings of powerlessness among social workers. For example, as Anne and Rolinka recount:

“Then we say: your life, good luck with it. If you want to get into trouble then do it anyway. At least that is the language often spoken. Ultimately, I don’t decide on someone else’s life; it’s still voluntary assistance. All I can do is make that person aware of the risks of going back […]. Often that’s all you can do, and often we feel extremely powerless.”

—Anne

“We are voluntary social work. We cannot force someone to stay here.”

—Rolinka

Although not specifically naming it a feeling of powerlessness, all social workers—without exception—expressed that they cannot make a decision for someone else, especially as the assistance they offer is voluntary, even if they disagree with the decision being made. Making a woman aware of the consequences of her decisions is the best they can do, so they argued. Therefore, even though raising awareness could be seen as an interventionist action, they predominantly adopt a hands-off approach to autonomy.

Nevertheless, all social workers acknowledged that, in dangerous cases, they endeavor to intervene in a more interventionist manner. In line with the tension between hands-off and interventionist autonomy present in the Code of Ethics (BPSW 2021), social workers also tried to establish an interventionist approach to autonomy; for example, by facilitating an ongoing process of reflection, which is intended to support their clients in making the best decisions. They do this during conversations with women who share that they are considering a return. It might be expressed in a more subtle manner, as stated in the following quote:

“Then I say, “you know, just yesterday we had the conversation about you not wanting it, that you would never go again. So what makes you want to go back now? Did you even talk about it with your partner? I do see risk, because you are here for a reason”.”

—Nanda

However, some social workers acknowledge that they adopt a more direct approach and share their own perspective on the situation, stress the potential implications of returning, and advise the woman to reconsider her decision:

“And then I said to her, or at least advised her: ‘You know, don’t do anything yet, don’t take any steps until you have spoken to your personal assistant and in the meantime you can think about it, because you are here for a reason. You left him for a reason. And so what is the reason for you now that you want to get back with him? It’s, I know you’re in love, but you know, that’s not everything’.”

—Mina

Both examples, in a subtle or direct manner, display a more interventionist approach to autonomy. In these cases, social workers want the woman to rethink her situation because they know that the consequences may not promote her well-being. They verbally intervene and suggest paths that might lead to a safer outcome. Still, they feel limited as the maximum they can do is discuss the situation and outcome.

When children are present in the relationship and are in danger, social workers have the right to inform “Veilig Thuis”. If the woman still chooses to go back, this may lead to an intervention from the state and the removal of the children into custody.

“Suppose a woman says, ‘I’m going back home anyway’, and we don’t support that for the sake of the children, then we have to report it. We have a duty to report that a woman makes a choice that we find unsafe for the children.”

—Flora

So, there is a critical difference in support-giving between women with and without children. Whereas for women without children, a hands-off approach to autonomy is often prioritized in the view of the social workers, and the intervention is limited to discussion, when children are present, an interventionist approach is far more likely to be prioritized. At the same time, when cultural expectations influence women’s decisions, social workers might intervene in different ways.

4.2.2. The Additional Pressure of Cultural Expectations and Dependency Based on Migration History

When cultural expectations affect a woman’s decision-making, social workers, especially personal and ambulatory assistants due to their closeness with their clients and their social network, often undertake extra activities and interventions to help women, provided the latter consent. Some social workers try to support the maintenance of social networks for women who do not wish to return to their partners, but who fear the risk of deportation and subsequent stigmatization by their social network. For instance, in cases of stigma, social workers try to explain the situation to family members or ask religious leaders (e.g., an imam) to mediate. Maryam and Eveline explain this in the following terms:

“We are also open to talking to parents, and sometimes we do that even from a distance, where we actually call abroad to explain. Very often they, especially parents abroad, have an image about the women’s shelter, eh, that it’s a whorehouse and.... well, not with everyone, but with some. So if you then start that conversation and explain a bit about what it’s like here, what kind of assistance, some are then reassured when they know that it’s only women who live here. That gives some space.”

—Maryam

“I would try to look with the client to see how to approach the parents in the best way. Because those parents are actually very important. And the approach doesn’t necessarily have to be us because you also often have mediators. It could also be the imam, or it could be a family member. Are there people who can support her in the process towards the parents?”

—Eveline

If possible, they facilitate a social worker from the same cultural background to converse with the family members, driven in part by commonalities of language but also by a shared understanding of cultural nuances. When the problem of deportation and residence permit comes up, social workers—again, specifically personal assistants—encourage and help women to find proof of the abuse that might help them to be able to stay in the country. This support is aimed at ensuring that women have the capabilities to live a life they value, again adopting an interventionist approach to autonomy.

In some instances, social workers also keep in touch with women even after the latter have left the shelter or are no longer in assistance. For example, Sara told us that she sometimes calls a woman, or the woman calls her, to catch up. If the woman uses a certain phrase, Sara would know that it is not going well and they try to look for a solution:

“Then we had agreed on a code and when I call her we would talk about the bags. And I would ask her: “did the children like the bags?” And she would say: “they all liked them”, then [I know] she could still pull it off. And if she said: “well, the children didn’t like the bags or they weren’t very happy with them”, then I knew, okay, things are really not going well and I have to call the police”

—Sara

This approach is not expected from Sara as a social worker, and might also be unauthorized due to the ideal of client–social worker distance being maintained, a factor that often depends on whether the case is fully closed or not.

We thus see that social workers experience multiple tensions when working with women exposed to domestic violence, particularly in contexts where women may wish to go back to an unsafe situation. Concerns over their safety are the most significant of these, combined with feelings of powerlessness due to the boundaries social workers experience in seeking to support women who have survived domestic violence. This tension also affects their pathways to support, whether that be in a hands-off or an interventionist way. Still, a hands-off approach to autonomy is mostly encouraged in support, leading us to consider whether there might be a fuller role for a more interventionist approach and a more interventionist interpretation of the Code of Ethics.

5. Discussion

5.1. Should More Interventionist Support Be Considered?

The primary objective of this article has been to examine the manner in which social workers intervene when women consider returning to their partners in situations where safety is still a concern. In order to examine these interventions in more detail, we have used the concept of autonomy, based on Martha Nussbaum’s capabilities approach (2000), which we defined as primarily an interventionist approach to autonomy, and contrasted this with a hands-off approach to autonomy. While the hands-off approach limits intervention in women’s decisions, the interventionist approach encourages the provision of tools and conditions to ensure that women make decisions that support their life goals. This led us to ask which approach is more promoted in social work, where interventions are based on guidelines outlined in the Dutch Code of Ethics (BPSW 2021).

In discussing the Code of Ethics, we argue that a tension arises in social workers’ approach to intervention due to the presence of ambiguities within the guidelines when it comes to interpreting and enacting autonomy (BPSW 2021). Within social work, clients’ autonomy is an imperative value, which is necessarily accentuated in the Code as “respecting the self-determination of the individual” (BPSW 2021, p. 8) and is interpreted as a hands-off approach to autonomy. Yet, the Code also encourages having a safe living environment and the promotion of the emancipation of individuals. Thus, the Code recommends both a hands-off and an interventionist approach to autonomy (BPSW 2021). Still, we found that, amongst all social workers, an understanding of autonomy as a hands-off approach is more common: they believe they should not be intervening and that this hesitancy to intervene is a means to respect women’s autonomy. Common belief from social workers is that the most they can do is to explain to women the dangers of the situation and suggest that return—in general—is not recommended.

This restriction prompted feelings of powerlessness, and some social workers endeavored to intervene more actively. When women consider returning because of cultural or institutional pressure, such as marriage migrants, social workers feel that they can intervene more actively as long as the woman consents to the action. Consequently, social workers will do their best to support the woman to prove the violence so that she can receive an independent residence permit (de Hart et al. 2022; Roegiers Mayeux et al. 2023). When a woman opts to unwillingly return due to the influence of her family members, personal and ambulatory assistants might propose to the woman to engage in discussions with the relatives to clarify the circumstances and ease the relatives’ concerns. In adopting this approach, social workers do more than is strictly expected from them, demonstrating cultural sensitivity and utilizing the ambiguity of the Code of Ethics (BPSW 2021) to support a more interventionist approach. Sometimes, against the rules of conduct, social workers choose to maintain contact with women even after they have left the shelter, as is shown through Sara’s example. Likewise, this intervention only occurs when a woman consents. These extra steps illustrate that social workers are highly dedicated and motivated to provide assistance, but feel constrained by the limitations they believe are imposed by legislation that prioritizes the principle of clients’ autonomy.

Thus, in the field of social work, autonomy is normally understood as a right to make your own decisions, a principle that social workers are expected to uphold. However, a secondary interpretation of autonomy is present within the regulatory framework, namely as a strategy for enhancing clients’ autonomy. This approach, though not explicitly delineated in the text, is occasionally implemented by social workers, which social workers do sometimes apply it but without fully feeling entitled (BPSW 2021).

These findings suggest that the balance between hands-off and interventionist intervention in cases of domestic violence needs careful consideration and may need to be reassessed. The current, partly contradictory, Code of Ethics (BPSW 2021), has been interpreted as imposing limits on the ability of social workers to take more interventionist measures. This creates serious safety risks, as the hands-off approach may result in women going back to dangerous, abusive situations, potentially leading to further violence and, in extreme cases, femicide. Therefore, following Nussbaum, we suggest that a priority should be the creation of safe conditions in which women can make a choice. However, it is important that any such intervention is not paternalistic in nature, and this needs to be the situation between an overly interventionist and hands-off approach. For instance, a way of seeing a possible interventionist approach in practice is that women should have access to all the necessary information to be able to make a choice in staying in the shelter or separated, returning to their partner, or even other options (e.g., returning to the country of origin). Therefore, social workers must be aware of all the potential alternatives they could offer women. Next to this, if this is something that women wish, social workers should encourage reflection by offering tools to balance their options. In instances where women do not have proficiency in the Dutch language; for example, recently arrived marriage migrants, this can be through illustrations. For example, Heron et al. (2022) advocate a similar approach in their work examining the reasons for women to leave their partners. Such an approach, we argue, can also be an effective tool in this context. This would lead to them having multiple paths available, each with its risks and benefits. Regardless of the decision they take, it allows the women more agency. While this would require more work for the social workers, it could help the women in making autonomous choices.

If women decide to stay a little longer separated from their partner, or in the shelter, psychotherapy should be available. In case they would still like to return to their partner but first would need some more distance, (gradually) more regular contact in a safe environment can be offered. We do not want to argue that women should never return to their partner; the main criterion for any return should be that violence has ceased and does not recur. The challenges for social workers that we have explored here often center on this removal of violence and the risk of violence as the starting point for rebuilding relationships and sustaining support. Therefore, our approach builds on the perspective of many of the social workers discussed here (for example, Aisha’s narrative), that women should not necessarily stay separated from their partner after experiences of abuse in all cases. Instead, while this is already encouraged, we continue advocating for offering support in ending the violence and reestablishing a healthy relationship for those who wish to return to their partner, and have focused on the varied ways in which social workers navigate these tensions. If such an approach is encouraged, that also means that shelters need additional budgets to provide psychotherapy.

Next to this, it also requires the willing participation of the partner in safe and mediated contact meetings. Parts of the interventionist approach we propose are already endorsed by the city of Rotterdam (Lokale Wet- en Regelgeving 2024), which is focused on the perpetrator of the violence and assists him in becoming and remaining non-violent. Recognizing the dangers of paternalism, this should be under the provisions that these interventions respect the consent of the women, and that they have engaged in reflective exchanges with colleagues about cases where they consider intervening more.

Moreover, regardless of the decision women make, support is needed in all cases: whether a woman decides to return to her partner or maintain a separation from him. Emotional, legal, and financial guidance should be offered in all cases. While this option poses a set of practical challenges, it is based on the principle that support should not be conditional on the woman’s decision. A potential approach includes maintaining contact with specialized social workers after a woman returns, without the need for ongoing ambulatory guidance, in order to maintain support and ensure this remains accessible. As illustrated in Sara’s case, some social workers already keep in touch with women after they have returned to their partner, albeit informally. Taking this approach, and formalizing this contact with specialized support, could reduce the workload for social workers in shelters and contribute to enhancing the safety of the woman. For instance, these social workers could be calling the woman once a month—depending on the situation and the woman’s wish to remain in contact.

Lastly, in case of institutional barriers such as the fear of deportation (Bartels 2021; de Hart et al. 2022; Roegiers Mayeux et al. 2023), we encourage easier access to an independent residence permit for survivors of domestic violence. Thus, it is essential to continue refining and implementing these strategies to create safer, more supportive environments for women experiencing domestic violence.

5.2. Limitations and Further Research

This research has some important limitations. For instance, one limitation of this study is the lack of clarity regarding the influence of culture, nationality, and ethnicity of the social workers on their interventions, and how these self-chosen identities differ, intersect, and interrelate with each other. Although this study acknowledges the importance of respecting the social workers’ self-identified backgrounds, it did not sufficiently explore how their identities shaped their approach to autonomy in practice. Future research could delve deeper into this aspect to provide a more comprehensive understanding of identity and intervention in social work.

Next to this, our focus on the perspective of social workers has necessarily omitted the views of women who went back to their partners or who were seriously considering doing so. Whilst this paper offers no specific insight into the reasons they choose to return to their partner, our focus here has been on the role and decisions of social workers in response to these choices. Similarly, our focus on social workers has omitted the views of family members and the partners themselves. Yet, a fully holistic account of the decision-making process for women who have left violent relationships is beyond the scope of this paper alone and would encompass a broad range of social actors and structural factors, as evidenced in past research (Morrison et al. 2021; Roegiers Mayeux et al. 2023). Instead, we have sought to examine the challenges and dilemmas experienced by social workers as a group at the heart of responding to, and seeking to inform, in often subtle ways, the decision-making of women who have survived domestic violence.

To build on these insights further, interviews with mediators could be conducted to gain insight into their experiences and methods of facilitating reconciliation. Moreover, it would be valuable to examine the elements that could facilitate a survivor’s decision-making process while still preserving the possibility of reconciliation with their partner. This could facilitate a breakthrough in enabling them to make autonomous choices within a safe environment. Finally, once the Rotterdam program for survivors and perpetrators (Lokale Wet- en Regelgeving 2024) has been operational for several years, it would be valuable to more fully consider its effects on couples who wish to reconcile.