“Risk and abolition: perspectives of front-line domestic violence workers supporting refugee and migrant women on safety and policing”

ABSTRACT

Anti-violence advocates continue to criminalise new gender-based violence offences, despite the evidence of unintended consequences for marginalised groups. Using an intersectional framework of risk, this article aims to understand what drives front-line domestic and family violence workers to engage the law for the protection of the women they support. The data draws from a study examining responses in domestic violence policy and service interventions to refugees in Southeast Queensland, Australia. Thirty-one interviews were conducted with front-line workers supporting refugee women experiencing domestic violence. Two-thirds of the interview participants are from refugee and migrant backgrounds themselves. Interview data analysis shows a combination of governing and managing actuarial risk across three themes: 1) Why a preference for the law? 2) Autonomy or discourses of responsibility? and 3) The mutual constitution of risk and inequalities. Highlighting the differences between those who address domestic and family violence on the front-line, the findings reveal that front-line workers are both constrained by and contribute to various assemblages of risk in the carceral nation state. I suggest key steps towards engaging front-line workers in an abolitionist agenda that centre on knowledge translation and research partnerships.

KEYWORDS:

• Gender-based violence
• abolition
• risk
• carceral feminism
• refugees and migrants
• front-line workers

Introduction

In scholarship, practice, and community, debates on abolition continue to build. Abolitionists aim to bring an end to the prison industrial complex and advocate alternatives to laws that build the capacity of communities to respond to social problems, (INCITE, 2017; Crenshaw, 2011; Davis, 2011). Where gender-based violence (GBV) is concerned, laws and criminalistic interventions are considered to be informed by ‘carceral feminism’ (Gruber, 2020). Carceral feminism is often associated with Bernstein’s (2010) definition of feminist advocacy as an overreliance on the justice system by anti-violence and gender equality advocates to protect women. As Kim (2020) points out, the rise of carceral feminism was built on good intentions of addressing violence against women but has become enmeshed in a punitive system that can have ill-fated outcomes for those most marginalised. Despite evidence of the consequences of legal interventions for marginalised groups, however, policymakers and front-line workers who support women experiencing GBV continue to advocate for the criminalisation of offences (Kim, 2019). The last decade in Australia, for example, has seen the criminalisation of new GBV offences – non-fatal strangulation (Douglas & Fitzgerald, 2021) and coercive control (Sandra & Fitz-Gibbon, 2021) – and recent reforms to sexual violence and consent laws (Featherstone et al., 2023; Fileborn & Loney-Howes, 2019). Advocates from non-governmental organisations (NGOs) who implement policy interventions addressing GBV on the front-line played a significant role in advancing these new laws and reforms. This despite cautions and evidence-based arguments presented by academics, survivors of GBV, and First Nations women (see Wangmann, 2022).

Using a theoretical framework of ‘risk’ this article examines what drives front-line¹ workers addressing domestic and family violence (DFV) to engage the law for the protection of women they support. The data are drawn from a study exploring how front-line workers (n = 31) conceptualise the issue of DFV and refugee communities in Australia. Two-thirds (n = 20) of the participants are from refugee and migrant backgrounds themselves. While it is now widely accepted that the criminal justice system acts as a form of social control (Coyle & Schept, 2017), and much attention has been paid to prison abolition (Carlton, 2018; Coyle & Scott, 2021; Seoighe & Rachel, 2021), a growing body of scholarship is turning its attention to the alignment of professions such as social work with the carceral state (Delgado, 2020; Mehrotra et al., 2016; Richie et al., 2021). The literature has problematised policies and front-line interventions addressing GBV for their role in upholding the carceral logics of the state that continue to oppress marginalised groups (Bergen & Abji, 2020; Jacobs et al., 2021; Valenzuela-Vela & Alcázar-Campos, 2020; Whalley & Hackett, 2017), while the depoliticising role of the NGO ‘erode distinctions between state and non-state intervention’ (Musto, 2019, p. 48). Scholars have also theorised the implications of implementing abolition in front-line conditions constrained by funding, limited resources and accountability, paying particular attention to the gap that remains when legal interventions addressing women’s immediate safety are taken away (Brockbank & Greene, 2022; Goodkind et al., 2021; Richie & Martensen, 2020). However, there remains limited examination of the workers who enact such policies and interventions.

The research presented in this article adds a unique perspective from front-line workers themselves on safety and policing. This analysis contributes to literature addressing what Olofsson et al. (2014) describe as a ‘conceptual vacuum’ that has been left by predominant risk theories. Beck’s theorisation of the ‘risk society’, as a fear of the unknown and uncertain futures, has been critiqued for failing to consider social stratification and power relations. Addressing multiple sights of power and inequalities, governmentality refers to the management of social problems and citizenship as aiming to reflect the agendas of governments and institutions. While studies of governmentality identify ‘at risk’ or ‘risky’ populations, and the state’s role in the reproduction of inequalities, how sociocultural determinants of inequalities intersect and change over time has not been sufficiently interrogated. Olofsson et al. (2014) suggest intersectionality as a critical tool for understanding the construction of risk and social inequalities, for its focus on ‘categories of difference in individual lives, social practices, institutional arrangements, and cultural ideologies, and the outcome of these interactions in terms of power’ (K. Davis, 2008, p.68).

Using an intersectional lens, scholars have delved into more bottom-up theorisations of risk, highlighting a combination of governing and managing actuarial risk in the prison (Feeley & Simon, 1992; Maurutto & Hannah-Moffat, 2006; Struthers Montford & Hannah-Moffat, 2021). I apply this concept of governing and managing actuarial risk to the case of front-line DFV workers. A scan of the literature on risk and DFV shows that risk is considered as managing the actual safety of victims of interpersonal violence and minimising the likelihood of domestic homicide (Truong et al., 2023). Service provider – and police – responses to risk and DFV often focus on policy- and organisation-specific frameworks alongside the use of tools such as standardised risk assessments (Larance & Kertesz, 2023) to predict and manage the risk of escalating, repeat offending (Meyer & Reeves, 2021). This juncture of managing the actual safety of ‘at risk’ individuals and governing ‘risky’ populations might be considered as lying at the heart of debates concerning abolition and carceral feminism. Abolitionists concerned with the consequences of carcerality for marginalised groups advocate more community-driven solutions outside of the state (Kim, 2020), while carceral feminists, and those seeking to reform the criminal justice system, view the task of protecting women and other vulnerable groups and their rights as citizens to be the role of the state (McGlynn, 2022). Furthering an abolitionist agenda somewhat hinges on the values, beliefs, and subsequent activities of carceral, usually liberal feminists, who are primarily concerned with gender (Sweet, 2016). A focus on gender by those who direct policy and front-line interventions do not adequately consider ‘the effectiveness of criminal law to reconcile the complexities of gender equality, security of the person, and the protection of deeply marginalized poor and racialized women from misogynistic violence’ (Balfour, 2021, p. 70). This article seeks to address the gaps between these two binary positions of carceral and anti-carceral (Terwiel, 2020) by reconstructing how front-line workers understand and manage conditions of risk.

In the following section, I provide further background and definition of terms regarding carceral feminism, the retrenchment of the feminist movement in the neoliberal welfare state, and elaborate on an intersectional approach to risk. I then present information on methodology and the data informing this article, before presenting the findings. The article ends with a brief discussion, outlining implications for GBV policy directing front-line interventions and how these implications might be considered in generating more widespread attention and support for an abolitionist agenda (see Carlton, 2018).

Carceral feminism, NGOs, and risk

The criminalisation of GBV offences was the result of feminist’s efforts to have violence against women taken seriously (Drakopoulou, 2007). Kim notes that ‘progressive critics, however, have argued that many of these gains were made through the anti-violence movement’s collaboration with the increasing U.S. investments in criminalization’ (Kim, 2018, p. 220). While recognising DFV as a crime has been a feminist agenda since the 1800s, the siren call for law reform that came from Western, second wave feminists in the 1970s travelled worldwide to both Western and non-Western countries (Smart, 2002). Bernstein (2012) identified that in attempts to criminalise practices such as forced labour and prostitution, women’s bodies become subject to processes of control by the neoliberal state that both facilitate and reinforce state violence. In many cases, these coercive practices are facilitated by feminists advocating the emancipation and liberation of women. I use neoliberalism in this article to refer to the scaling back of government services that have resulted in individualistic, market-based solutions to social problems aimed at achieving ‘more for less’ (Mehrotra et al., 2016).

The term ‘feminist’ brings into question the current state of feminism in anti-violence work, and its links to carceral intent. Some scholars have been critical of the practice of too frequently associating feminism and carceral politics (Gotell, 2015; Taylor, 2018), suggesting this connection affords feminists a greater role than what they have actually played in the state’s carceral expansion, and its consequences. Bernstein (2012) also identifies problems with linking feminism to carceral intent, but because of the diversity within feminism. Bernstein is concerned with the influence of neoliberalism and the hegemony of liberal feminism. Scholars such as Halley et al. (2018), for example, note the problem of governance feminism, or state feminism, whereby the aims of ‘dominant feminists’ (Halley et al., 2018) are now more in line with state and political agendas, and cannot necessarily be described as ‘feminist.’ This is true of the Australian context where, even though the legacy of second wave feminism remains, strategies addressing GBV such as criminalisation are less protected, or influenced by, feminism (Harris Rimmer & Sawer, 2016; Maturi, 2023a). The case in Australia is one of dominant, often white-led groups defining the terms, norms, and strategies that direct policies and front-line interventions, at the expense of marginalised groups. Rather than suggesting carceral ‘feminism’ as deliberate intent then (Terwiel, 2020), my focus in this article is on the carceral logics that are inherent in front-line work addressing GBV. By carceral logics, I refer not only to front-line workers’ reliance on the law but also ongoing racism and colonial violence in Australia as a settler state (Coyle & Schept, 2017), both of which I address below.

Scholars engaged in anti-violence work often reflect on the transformations of the women’s movement in alignment with criminalisation, professionalisation and neoliberalism (Mehrotra et al., 2016). Bumiller (2008), for example, has articulated the combination of the carceral state and the therapeutic state which entrenched demands on the state to incarcerate dangerous perpetrators and to treat vulnerable women within medical and therapeutic models. As Hardy (2015) notes, the expertise of the therapeutic state was traditionally to diagnose, refer, advise and treat. This expert knowledge began to be questioned, however, as treatments and cures were shown to not necessarily be effective. As a result, welfare professionals have increasingly been allocated new roles of identifying, assessing and managing risk (Maurutto & Hannah-Moffat, 2006; see also Feeley & Simon, 1992). Strategies of engaging and directing victims of DFV through a myriad of systems, with the primary goal being data collection and surveillance, have now become the norm (Delgado, 2020; Jacobs et al., 2021). Scholars have described an overwhelming focus on safety and accountability in DFV policy and front-line organisations, with the aim of protecting vulnerable women (see Stoever, 2019). Safety in the neoliberal welfare state is confined to the violence perpetrated in intimate relationships; however, while violence perpetrated by organisations, institutions and the state is mislaid or overlooked (Abrams, 2019).

Under logics of risk and welfare ‘“social problems,” often associated with particular categories of “vulnerable” people, are redefined, reframed and – ultimately – managed as “risks”’ (Olofsson et al., 2014, p. 417). Regarding GBV, binary constructions of vulnerable victims and violent perpetrators have a dual effect of creating those deserving and undeserving of state support and reinscribe the intersections of racial and gender hierarchies via carceral logics (Mack & McCann, 2018). This combination of risk and carceral logics expand the multiple sights to enact control and punitive measures against marginalised groups (Bergen & Abji, 2020; Valenzuela-Vela & Alcázar-Campos, 2020). In Australian GBV policies and practices, for example, refugee and migrant women are often targeted as ‘at risk’ and ‘vulnerable populations’ (Maturi & Munro, 2023). But research details the multiple ways that Australia’s visa systems fail women on temporary visas who are experiencing violence, who are deemed unwanted by the state and thus underserving of protection (Maher & Segrave, 2018). Other scholarship has discussed the nexus of DFV and child protection, where women experiencing violence in intimate relationships are penalised for failing to protect their children from the perpetrator (Kaur & Atkin, 2018). These mothers then face the difficult (if not impossible) choice of punitive interventions from child safety, including having their children removed, or placing themselves and their children at greater risk of structural violence in the forms of homelessness, poverty (Kuskoff et al., 2023; Maturi, 2023a) and legal systems abuse via the family courts (Douglas, 2018). Whether men face sanctions as perpetrators depend on their own locations at the intersections of race, class, and migration status in the Australian settler state (E. Stanley, 2018).

As a social problem that is now largely informed by a public health framework, strategies addressing GBV are a product of the various arms and agendas of a fragmented state that result in different assemblages of risk (Ben-Ishai, 2009; Maurutto & Hannah-Moffat, 2006). Ben-Ishai (2009) describes these different assemblages using the example of CCRs (Community Coordinated Response Systems) in the USA, which bring together sectors such as social work, police and the courts, child safety, women’s shelters, youth and housing programs to provide a multi-agency response to DFV. While Australia has had multi-agency responses for some time (N. Stanley & Humphreys, 2014), High-Risk Teams (HRTs) have recently been implemented, similar to CCRs, or MARACS in the UK (Robinson & Rowlands, 2009). Since 2022 some mainstream organisations have further expanded partnerships with the justice system by embedding police officers in their services. Ben-Ishai (2009) describes that under these different assemblages, an agenda of social control might be less straightforward when it is pitted against logics of care and managing actuarial risk (see also Singh, 2012). Rather the agenda is to foster a sense of autonomy among vulnerable women seeking the state’s protection. On the one hand, front-line workers engaged in multi-agency responses, such as social workers and police, might be subject to their own measures of accountability and success. Policies and laws might encourage coercive practices at odds with the notion of autonomy. Examples include the sharing of information without women’s consent, or police-initiated protection orders.² On the other hand, and despite the risk of structural inequalities addressed above, these interventions might allow an autonomous life free of DFV for women. The mandates and resources of some organisations, for example the police, might enable responses that are not possible for other organisations and sectors who face different constraints (Ben-Ishai, 2009; Hamilton et al., 2021; N. Stanley & Humphreys, 2014).

Although the problem of vulnerable women seeking the protection of masculine institutions remains (Goodmark, 2018), Ben-Ishai (2009) suggests that we need to think differently about risk and governance, and how we conceive of the state towards feminist ends. Ben-Ishai notes the deleterious effects of professionalisation, and the promise she holds for CCRs as a political space has been widely critiqued for failing to affect change through feminist advocacy (Gruber, 2020). However, for the purposes of this article, Ben-Ishai’s understanding of the DFV sector as assemblages of risk, rather than a hegemonic state, does provide an entry point to understanding the conditions which lead front-line workers to rely on the law for the protection of women they support. The point here is to highlight the differences among those an abolitionist agenda seeks to engage.

An intersectional approach to risk

Governance strategies of risk, that juxtapose certain populations as either ‘risks’ or ‘at risk,’ do not take into account the multifaceted ways women experience violence at the intersections of gender, race, class, sexuality, disability and so forth. Nor do they consider the lived experiences of those operating on the front-line who support women experiencing DFV. Intersectionality, in Olofsson et al. (2014) account, focuses on the imbrication of social categories and difference in a context of not only historically mediated inequalities but new structural conditions and power arrangements that come about through ongoing social change (Cho et al., 2013; Collins, 2017; Hankivsky & Jordan-Zachery, 2019). Bringing an intersectional lens to studies of risk can thus adapt to complex social changes and risk phenomena (Olofsson et al., 2014; Struthers Montford & Hannah-Moffat, 2021). Adding an intersectional lens to the management of actuarial risk with the governance of social problems disentangles the inequalities embedded in positivist risk frameworks, for example risk assessment tools and process. An intersectional approach articulates the ways that discourses of gender, race and class framed within logics of risk bound individuals to various subject positions that either foster or inhibit autonomy, as agency or as individualistic responsibility. Taken together, this analysis reveals the ways that risk logics combine with the norms of the carceral, nation state to mutually constitute inequalities for marginalised groups. But in doing so, I argue that the perspectives of front-line workers on safety and policing add valuable insights into opportunities for engagement towards furthering an abolitionist agenda.

Methodology

Methods and data collection

This research used qualitative methods. The research questions explored: i) Workers’ experiences supporting refugee and migrant women experiencing DFV, ii) their perspectives on how factors such as gender, race, ethnicity and class impact on women’s experiences of DFV, and iii) systemic barriers to implementing change. Thirty-one semi-structured interviews were conducted with front-line workers who support refugee women experiencing DFV in Southeast Queensland, Australia – 29 in 2018, one in 2019 and one in 2020. Some organisational documents were analysed, such as policies and assessment tools.

Ethics and sampling

Ethics approval was granted by the University of Queensland’s Human Research Ethics Committee A, approval number 2,018,000,422. Organisations who agreed to participate passed on information about the project to staff as potential participants, who could contact the researcher directly and confidentially. Participants were given information about the project, participation, withdrawal, and written consent was obtained.

About the organisations

Thirteen organisations agreed to participate in this research and included mainstream DFV organisations, refugee resettlement organisations, and migrant specific organisations supporting women experiencing violence. Mainstream organisations usually had multiple programs, including individual case management and DFV programs imbedded in other organisations, such as youth, family- or child-specific. Partnerships with the justice system included having support workers in DFV specialist courts, co-response programs with police, and coordinating the regional High-Risk Team (HRT).³ Mainstream organisations were usually more dominant in size, with upward of 50 staff. The migrant specific services addressing DFV were smaller, some with only 3–5 workers supporting clients. This despite servicing the same region as mainstream organisations (in one case the entire state of Queensland, comparable in size to Mexico or Indonesia). Refugee resettlement organisations were also often larger; however, some smaller organisations or programs took part. Refugee resettlement organisations often focus on integrating individuals and families into the local community, with a heavy focus on training and employment. Only one refugee resettlement organisation (approximately 300 staff) had a DFV program, made up of two part-time workers.

About the participants

Thirty-one workers and managers from front-line organisations participated in the interviews. Demographics of participants are detailed in the following table 1.

Table 1. Participant Information.

Cultural Background	Gender	Years of Experience	Qualifications	Work Experience
Anglo-Settler	Female  = 29	2–5 = 8	Social Work (BA) = 5	Mainstream Only = 8
n = 11		5–10 = 7	Social Work (MA) = 7	Refugee Resettlement Only = 5
Migrant Background	Male  = 2	10–15 = 5	Psychology (BA) = 5, (MA) = 1	Migrant Specific Only = 1
n = 8		15–20 = 6	Human Services (BA) = 4	Mainstream & Refugee Resettlement = 7
Refugee Background		20–25 = 1	Counselling (Dip) = 2	Refugee Resettlement & Migrant Specific = 6
n = 12		25–30 = 2	Community Services (Dip) = 1	Mainstream, refugee resettlement and migrant specific = 4
		>30 = 1	Arts/Education (MA) = 1	Management/Supervisory  = 10
		Missing = 1	Arts (BA) = 1
			PhD (unknown discipline) = 2 Missing = 4	Overseas Exp = 6

*Bachelor (BA), Masters (MA), Diploma (Dip)

*Master of Social Work is common in Australia when people credit a former BA degree

*‘Missing’ data – there were some instances that specifics were not provided by participants

* Identifying details have been left vague in the findings to protect confidentiality

Ethnic backgrounds of those who identify as refugee or migrant include Bangladeshi, Brazilian, Chilean/Argentinian, Ethiopian, Fijian, Ghanaian, Indian, Indian/Iraqi/Portuguese/British, Iranian, Italian, Lebanese, Serbian, South Korean, and South Sudanese.

Data analysis

Thematic analysis was used to analyse the interview data, following the method set out by Braun and Clarke (2006). I used an ‘inductive’ process of coding, which Braun and Clarke describe as a bottom-up ‘process of coding the data without trying to fit it into a pre-existing coding frame, or the researcher’s analytic preconceptions’ (Braun & Clarke, 2006, p. 83). Of using semi-structured interviews as a method rather than observation or ethnography, Dingwall gives an account of coding as a rather inaccurate method of introspective research in which ‘the researcher is simply looking for some good quotes to illustrate a previously determined position’ (Dingwall, 1997, p. 3). Sticking to discreet themes or codes can potentially lose the narrative or the context of the extract in question (Braun & Clarke, 2006). I was therefore careful to consider the data as a whole, going between applying theory and literature, engaging in reflexivity through writing and feedback, and refining themes, codes and analysis. Despite there being no direct questions about the law in the interviews, “legal interventions” was a prominent theme that arose in the data. Given the wide reach of the law in DFV work, interview participants’ discussions of the legal system were not always focused on refugee and migrant women per se, but rather their experiences of working within and across the legal system more broadly.⁴

Reflexivity and positionality

I worked in the human services for 15 years, in both refugee resettlement and mainstream DFV organisations. My work experience assisted me with meaning making, providing a greater understanding of the organisations and workers I was interviewing, and the socio-political context in which they operate. This experience also gives me some of the context that Dingwall (1997) says can be missing from using interviews as a method to generate data. However, I also note my role in a system that continues to have many harmful consequences for marginalised groups. As an Anglo-settler woman, I engage a critical approach to theory (Collins, 2019), with the aim of critiquing the structures which I am part of and that centres the lived experiences and knowledge of marginalised groups.

Findings

I begin the following findings by suggesting that front-line workers are guided by a risk logic. Amid a lack of resources, the privatisation of services and the combination of the therapeutic and carceral states, these findings demonstrate further evidence that services and support have been reduced to time-limited interventions focused on managing risk, which ultimately leads to legal remedies. Safety is considered in individualistic terms of being removed from intimate relationships, even though there remain systemic inequalities that make women ‘unsafe’, such as homelessness, poverty, and state violence. For refugee and migrant communities, these structural inequalities might be more pronounced due to colonising practices in Australia that perpetuate racism and sustain exclusionary practices. The findings are presented across three themes: 1) Why a preference for the law? 2) Autonomy or discourses of responsibility, and 3) The mutual constitution of risk and inequalities.

Why a preference for the law?

In the interviews with front-line workers, women’s physical safety from intimate partners appeared to take priority and drive interventions. Bianca, an Anglo-settler woman who had worked in both mainstream and refugee resettlement organisations, described risk and accountability as driving the focus on safety:

I think like the whole risk, safety, accountability. I get it … .Everyone’s worried and doesn’t want a woman to die–who is on their books … … given the focus now on how many women are being murdered a year … … If anything were to happen to a woman that’s come to their service they can say, ‘We did this, we assessed this’ … … .I think that the people that are taking those calls and doing that work genuinely want to help. I think underneath that is this unseen river of expectation that there’s going to be a coroner’s review. That every stone will be overturned.

Participants described workers in NGOs as burdened by a lack of time, resources, and strict policies and reporting structures that governed their responses to clients seeking help. In a climate of competitive funding, both workers and organisations were subject to Key Performance Indicators (KPIs) that determined not only the support they could or should provide but also that set targets for client contacts and outcomes in terms of referrals to services and systems. This work environment was discussed as diminishing the capacity for more reflective practice, worker’s ability to ‘do more’ to help women experiencing violence, or to improve or change responses in policy and practice. Joyce, a young women from an African background who worked for a mainstream organisation, talked about organisations working in silos and needing to conform to funding body requirements to compete for tenders:

I guess every organisation has this core business and what they actually are doing and all of that. So if they actually are making decisions about what tender they actually wanna make, it actually needs to go with the funding department, their capacity as well and it needs to be reflected in their policies and procedures too.

Rather than delivering long-term case management and counselling services, most organisations were more likely to engage in lengthy assessments based on managing and identifying risk. Assessments then informed referrals to other services, such as assistance to leave violent relationships via women’s shelters, or to engage the police and the courts for protection. A number of the services who participated in this research used the DASH, a popular risk assessment tool developed in the UK. The DASH, also known as MARAM, has been critiqued for categorising risk based on physical violence from intimate partners, without considering other forms of violence such as emotional or verbal abuse (Robinson et al., 2018). Further critiques of risk assessments, and of DFV services in general, include a lack of consideration of cultural difference (Struthers Montford & Hannah-Moffat, 2021). The DASH examined for this research had two questions tacked on at the end; one was if there were ‘cultural factors’ placing women at greater risk of violence, the second question asked if honour killings were a specific risk. While drawing attention to racialized forms of violence that attribute ‘culture’ to increased risk of violence, DFV services who work from a gender first model aimed at addressing intimate partner violence might not have the tools to conceptualise the nuances of violence in cultural contexts that often differ from the dominant norm.

In the interviews, participants were presented with a case study to elicit responses on how they would respond in practice to a refugee woman experiencing DFV. The case study involves a woman from a Muslim background seeking help for DFV. The case study notes ‘a lot of high-risk factors’ but does not detail what these factors are; there is no mention of whether the violence is physical, emotional, sexual and so on. The woman is hesitant to leave the relationship as she can’t speak English, doesn’t know anybody and can’t work. She also has young children and doesn’t want to break up the family, who have only been in Australia for 6 months. Ines immediately pointed out that it was a ‘high risk’ situation, framing the woman as vulnerable and in need of protection. While she discussed giving the woman ‘options’ and ‘choice’, achieving safety translated into convincing the woman to leave.

I think she’s saying here very clearly she’s worried about what she will do if she separates and then she’s worried about not breaking up the family. So, whether those things are two separate things or whether it’s actually that she just needs information about the fact that you don’t need to speak English, don’t need to know anyone and you don’t need to work to be able to separate. Those are the options I would explain.

Ines identified as Eastern European and was from a refugee background herself. In her interview, she also identified with a liberal feminist ideology underpinning interventions addressing DFV and supports the criminalisation of DFV offences. A focus on the physical safety of the woman overlooks the systemic inequalities identified in the case study that would make it difficult for the woman to leave the relationship, such as a lack of English, social isolation, and socio-economic barriers, which would likely include housing. The wish of the woman to keep the family together, which could be for reasons of religion, a more collectivist sense of family and community, or out of love or loyalty, is also ignored.

A focus on women leaving was a prominent theme in data analysis. While Ines discusses respecting the choices of the woman in the case study, and making sure the woman understands her ‘options and the systems that we’re in’, Ines went on to be a bit more candid about her goals of protecting vulnerable women and children:

I’m not going in there to keep a family together, that’s not my interest. I’m going in there to keep families safe and the safety of the children and women are my priority. It’s my priority, sorry.

Observing this overwhelming focus on risk and safety alongside increasing partnerships with law enforcement, it is not surprising that front-line organisations appear to be increasingly interpreting their work through a legal lens. Alice, a young Anglo-settler woman who had worked for both refugee resettlement and mainstream organisations, including as a support worker in the DFV specialist courts, said of using the law:

… .to me, that’s how I can do my best advocacy, because if there is a book that’s clear and tries to really clearly uphold the rights of people, then I wanna know how to use that … .my really good exposure was doing the DFV court just last year … .that level of advocacy could kick right in knowing that legislation could back us going into court, but also explaining to women what their rights are, really empowering through legislation … .so that’s I guess the law side of things and how to advocate appropriately.

Based on this evidence, in an environment that dictates efficiency and accountability over women’s choice and agency legal remedies offer tangible solutions to front-line workers who are tasked with addressing the safety of individuals they are supporting. Violence articulated in frameworks defined by law and informed by risk assessments, particularly physical violence, can make incidents and acts of violence easier to quantify, reducing risk to measurable and predictable means.

Autonomy or discourses of responsibility?

For front-line workers who are tasked with protecting vulnerable victims, rights-based frameworks under the law inscribe the state as protector, while enforcing the individual responsibility of women to take up their rights. Women are presented with options to address their own safety, with information and a seemingly clear path through the justice system. Informed by risk and carceral logics, it is in these discourses of safety and responsibility that delineations between those deserving and undeserving of state support become apparent.

Blaming ‘culture’

The perspectives of workers in my study on safety and policing emphasised the responsibility of the victim to seek out protections through services and the law, and to emancipate themselves from violent ‘cultural’ practices. Refugee women’s racial and ethnic backgrounds were sometimes to blame for a lack of engagement with laws and institutions believed to protect, rather than the failings of laws and systems themselves.

The police made a DVPO (Domestic Violence Protection Order) against him. This is what we want the police to do … .He blamed her for this DVPO being made … .Her whole community, she had to – She left because of the violence … .essentially she was homeless … … she lost all her support networks … … She was just like, ‘This is actually the worst thing that’s happened. I need to fix this, how can I get rid of this DVPO⁵?’

Jessica is an Anglo-settler woman with 10 years’ experience working for mainstream organisations supporting women. Through Jessica’s quote we start to see some of the unintended consequences of legal interventions, such as homelessness and social isolation. We also see that Jessica, like many other interview participants, is supportive of police taking out DVPOs, despite woman’s wishes. The practice of taking DVPOs out on the victim’s behalf is based on logics of women’s safety; that a woman might be unable to take action against the perpetrator due to fear, a misguided loyalty to the perpetrator or an inability to identify the violence being perpetrated against them.

Jessica actually views problems such as homelessness and social isolation as being caused by the woman’s ethnic community, and the woman not ‘prioritising’ her individual safety from the perpetrator:

Her safety was not her own priority, I suppose, but it was her connectedness to that community which was – Can be the fallout for [the] legal system.

Under logics of risk and colonial beliefs that women are at risk of violence because of their cultural backgrounds, women were often constructed as lacking in agency and having different priorities to those of mainstream services. Workers sometimes discussed refugee women as not knowing or understanding what DFV is. DFV might not be a crime in the countries they had come from, or refugee women might have experienced violence at the hands of authority figures and might be reluctant to seek help from the police in Australia. Presented with the same case study as Ines, about a Muslim woman unsure about leaving her husband, Karen, an Anglo-settler worker, said:

… we might think they’re high risk, she might in her community and her life and her experience, it mightn’t be high risk to her … and having, say the police turn up, that may have created the change or frightened … . I think you’d have to take this gently … .

Karen also suggests that the woman in the case study is not prioritising her safety because she does not consider the DFV perpetrated against her to be high risk. Karen’s use of ‘we’ implicates the othering between mainstream workers such as herself who define the terms of ‘high risk’, and thus how violence is understood, and those who reside in ethnic other communities. Karen at once distances herself from taking an active role in assisting the woman. Following her assertion that ‘you’d have to take this gently,’ Karen went on to discuss referring the woman in the case study to a migrant specific service who could give the woman ‘options,’ such as linking with community via migrant or refugee organisations:

… because they do have connections with the community and she mightn’t even be aware of that, in the Muslim community.

Despite blaming refugee and migrant women’s racial, ethnic and religious backgrounds for violence, there was a tendency by mainstream services to refer to refugee communities for help under the guise that they would recieve some sort of culturally specific support for DFV (see Maturi & Munro, 2023).

The use of law in ‘community’

Responses to DFV in refugee communities were also problematised by participants, however. Refugee communities, who are often struggling with changing gendered norms, racism in Australia, and men’s backlash to gender equality in response to a perceived loss of male privilege, tradition, and ‘culture’ (see Maturi, 2023b) might not have adequate tools and resources to respond to DFV. Maria, who had worked for 30 years in GBV organisations and refugee resettlement, said:

It’s a shame to just talk about [DFV] … it [be]comes ‘our culture’ … like ‘oh we don’t have this in our religion’ … . They want to fix the problem [of DFV] and they don’t want the myth that this our problem, it has to stay in the family, it has to stay in the community because it will bring shame to the community if you go out and you said that you have such a huge problem … .

Participants said that police taking DVPO’s out on women’s behalf could lead to further violence by extended family and community. Sometimes women did not know what the DVPO was for. Amira, a young woman from an African background who worked in both mainstream and refugee/migrant specific organisations, said:

We somehow get women come to our office and they go, ‘the police came to my house the other day. They gave me this piece of paper. But they said I don’t have to go to court’ … .like, why are you telling women not to go to court?

While Amira described taking out DVPOs without women’s consent as problematic, Amira was still supportive of legal interventions. Some workers said that police taking DVPOs out on women’s behalf was a good thing. As Amira said, ‘what if she decides … .I don’t want this relationship anymore and this was my key to get out.’ The problem that participants discussed with police initiated DVPOs was that police often did not explain to the victim or the perpetrator what the DVPO meant. If the perpetrator did not understand broad definitions of DFV, such as emotional abuse or practices like not letting the woman leave the house, then protection orders did not do anything to a) end the violence, or b) end the relationship. Telling women not to go to court took away opportunities for the courts, and potentially DFV workers at the courts, to provide education on what DFV is to both the victim and the perpetrator.

Here, the law appears to provide a useful tool to educate refugee communities, often as new and emerging diasporas, on what DFV is and how women experiencing violence can seek help. Zahra, a Middle-Eastern woman who came to Australia as a refugee, said at the migrant service she worked for they had facilitated a project aiming to educate leaders from refugee communities about DFV. They engaged police to facilitate a number of sessions. Zahra said the sessions helped to disseminate information about DFV to refugee communities, and as a result more women were seeking help from their service for DFV:

… there was more awareness in the community because there’s a number of initiatives that were held for community leaders, run by the police, and a lot of the different agencies that ran these sessions. So, more and more women were starting to come through our door … .

These workers were less concerned with justice or punishment for men, but rather how to affect change in communities and promote the agency and protection of women. Similar to the previous section, however, workers from refugee and migrant specific services said they were limited by the same mainstream policies and funding guidelines that focused on women leaving and legal interventions. Daniel, a South-Asian man who worked for a refugee resettlement organisation, said:

… so the only solution we have here is to approach it from a legal and criminal perspective … so these things are not approached from justice, right … you need to comply with the law … so it’s not because what I’m doing is wrong but, here, if I do, I might get into trouble.

While the law provides a useful tool to educate on DFV and might set a norm for acceptable behaviour, it stops short of fostering an understanding of gender relations and building the capacity of communities to respond to violence. As I go on to show in the following section, not only were there limited other options to address the issue of DFV but the carceral logics evident in Jess and Karen’s responses were also evident in the carceral responses of the state.

The mutual constitution of risk and inequalities

In this section, I highlight the multifaceted ways that women experience violence at the intersections of gender, race and class. I give three examples that demonstrate how interventions managing actuarial risk combine with the governance of ‘risky’ populations under the norms of the carceral nation state to mutually constitute inequalities for marginalised groups: 1. Women victims being charged as perpetrators, 2. The family court system, and 3. Immigration control.

The first example is the mis-identification of women victims as perpetrators. In the interviews for my research, the issue of police taking DVPOs out against women victims came up more than police taking DVPOs out against men. Lisa, an Anglo-Saxon worker from a mainstream organisation, said:

I haven’t noticed it so much in terms of men taking [DVPO] out. Just more police misreading the situation … . I think it does result in women getting labelled as respondents a lot more than they should.

Other research has identified female victims being charged as perpetrators in Australia (Reeves, 2023). Fitzgerald and Douglas (2018) found that Indigenous women were overrepresented in prisons after being charged with DFV offences. A report by the Queensland DFV Death Review and Advisory Board (2017) found that for nearly every Indigenous woman murdered because of DFV, the woman had been identified as the perpetrator at least once. Similarly, it was sometimes suggested by interview participants in my study that police were unable to discern who was the victim and who was the perpetrator. For example, women’s presentation on arrival might be perceived as emotional or heightened while the man was quite passive and calm. Workers described police not using interpreters, police ‘buddying up’ to the man, and police taking DVPOs out against women victims or both parties if there were also injuries on the man.

Workers also said that men would use the legal system to take out protection orders against their female partners.⁶ Maria said:

This man … got her to be triggered and to slap him and then called the police and got a protection order against her … because he wanted that to use as an excuse for having to leave the relationship … and trying to make her seem insane.

In the case of protection orders, the masculinity that underlies legal institutions continues to support perceptions of women as emotional, irrational, hysterical and crazy. Perpetrators of DFV therefore draw on representations of what a ‘good’ victim should be, for example vulnerable, quiet and submissive, using the bias of the legal system to perpetrate further abuse and control.

The second example is the convergence of the DFV and family court systems, providing further evidence of what Douglas (2018) refers to as ‘legal systems abuse.’ While both are civil courts, under the Australian constitution federal legislation, which the family court falls under, overrides state legislation, which DVPOs fall under. So while there might be a DVPO in place prohibiting contact between the perpetrator and victim (even when there are criminal charges for alleged DFV or sexual assault), until final orders are made the family court can force contact between the children and the father. The family court process can go on for years. Goeyong, from a migrant service, described the confusion women can experience between being told to leave the DFV perpetrator to protect the child, and then to allow the child to see the perpetrator, as the father, under family law provisions:

They go to DFV court and they all say ‘look, the father is perpetrator.’ Do not see them for five years … and then when family court comes in, all these things goes away and women really confused … why they do not hear me that I’m still in this dangerous situation … why they force me to let this fellow to see my child.

DFV perpetrators using the family court system to commit legal systems abuse against victims has been documented by other research and policy reports and has been the subject of parliamentary enquires (see Commonwealth of Australia, 2021). Atfa, who had approximately 20 years’ experience working in mainstream and migrant organisations in Australia, and overseas in development work, talked about a woman she supported who experienced legal systems abuse from her partner:

So she was really overwhelmed to understanding the mediation and the other party, he was so much controlling … use the legal policies to control her, the legal system.

Atfa went on to describe her client as getting into financial debt over court costs because she did not qualify for legal aid.⁷ The woman’s partner would do things such as set up meetings between the parties, and then not show up to incur costs for the woman and to waste everybody’s time. The legal system was a ‘barrier’ for women, Atfa said.

The third example is the immigration system. Many workers discussed that women on temporary visas were fearful of deportation or having their children taken away if they sought help for DFV. Men being deported from Australia for DFV offences also came up. In 2013, the Australian Border Force Act was introduced, granting government and private contractors special powers to detain and deploy additional surveillance and control measures to ‘protect’ Australia’s borders. This ‘protection’ has extended beyond keeping people out to policing those who are in the community on precarious visas, unwanted by the Australian state because they have ‘failed to thrive’ or are deemed of ‘bad character.’ DFV-related charges, such as physical assault, have been used as a basis for deporting non-citizens (E. Stanley, 2018). Lawrence, an African man who worked for a mainstream organisation, said:

Under the Department of Immigration and Border Protection the policies have changed, if you are on a DVPO for instance and you breach it,⁸ you risk being deported … and I know men that are being deported … there’s one client who’s partner is right now in immigration detention.

Harsh consequences for men, such as deportation, were also discussed as deterring women from seeking help. Sometimes the fear of negative outcomes for perpetrators was not out of women’s concern for men: While deportation could be considered a win for women who could not escape violence any other way, it could also leave women without visa pathways, income, or facing homelessness, single parenthood and social isolation. While their immediate safety from perpetrators might be resolved, refugee and migrant women face additional risks of systemic violence further entrenching gendered inequalities.

Discussion

Adding an intersectional lens to this analysis of the perspectives of front-line workers on safety and policing reveals the multiple inequalities embedded in positivist risk frameworks. Olofsson et al. (2014) say the ways that risk and social difference are constructed can be influenced by emotion, in this case worker’s very real fears over the physical safety of the women they are supporting. Safety was generally considered in individualistic terms of being removed from violent interpersonal relationships, even though there remain systemic inequalities that make women unsafe, such as homelessness, poverty, and racism. State sanctioned violence could also be imposed on men who might occupy subordinated status at the intersections of race and citizenship. Workers were often grappling with a conflict between individualistic, therapeutic practices grounded in supporting women’s choices, agency and broader goals of social justice, and the limited means to address violence through front-line organisations. Under legal definitions of DFV and a shared language of terms such as ‘high risk,’ promoted through the use of standardised risk assessment tools and processes, the law appeared to provide the means to be able to quantify violence for front-line DFV workers, and thus measure and respond to violence.

Attempts at fostering women’s autonomy via legal responses relied on discourses of responsibilisation that inscribed the state as protector of vulnerable women. In a climate of efficiency and accountability to mitigate risk, there were perceptions by front-line workers of women not prioritising their safety. Racialising practices blamed women’s cultural backgrounds for contributing to their experiences of violence. On the one hand, the focus of interventions is directed at identifying and educating those deemed as risky populations, who are unable to articulate what is best for them – the poor, refugees, migrants, mothers – legitimising coercive interventions by organisations and the law. Successful engagements with programs are seen as participants internalising fault for their situation, while any systemic failures are blamed on the participant or intended target of the intervention, rather than the programs themselves (Jarldorn, 2020). On the other hand, the apathy of police and mainstream services to assist refugee women might also be interpreted as failing to recognise refugee women as deserving of the same protections offered to women from more privileged backgrounds, putting refugee women at greater risk of DFV in their communities (Davis et al., 2022).

The findings show that refugee and migrant communities are at risk of over policing and state violence. However, participants from refugee and migrant backgrounds in this research also want laws due to a lack of alternatives to address GBV in communities. Here, the law is seen as useful to educate new and emerging diasporas on identifying DFV, and where and how to get help. This example highlights the conceptual vacuum that Olofsson et al. (2014) suggest one-dimensional risk frameworks do not attend to regarding the imbrication of social categories and social change. While refugee and migrant participants are speaking from their subjective experiences as workers in professionalised services, they are also often part of refugee and migrant communities themselves. These workers identified issues for refugee and migrant women that were not accounted for in mainstream policies directing front-line interventions, that focus on women leaving violent intimate relationships and legal interventions. Workers from refugee and migrant backgrounds were attempting to build the capacity of communities to respond to violence, while also attempting to address the immediate safety and needs of women experiencing intimate partner violence. As well as a tangible means of quantifying violence, the law becomes a benchmark for norms attempting to address entrenched gendered inequalities, similar to the goals of mainstream feminist movements. Faced with limited resources, smaller migrant services might need to seek legitimacy by adhering to dominant norms in order to compete with other organisations for funding, ensuring their survival and their capacity to support women experiencing DFV (Richie & Martensen, 2020).

These findings reveal the heterogenous views of workers who support women experiencing violence, and how they utilise the law to ends not intended by the state (Ben-Ishai, 2009). Engagements with police, or other sectors that rely on more punitive, carceral interventions, create an environment that encourage reflexivity by those who are largely driven by feminist goals of social justice, equality and client-defined empowerment. Participants in this study are aware of the limits and consequences of law and are challenged by the systemic inequalities facing women, such as legal systems abuse, punitive interventions by child safety and the family courts, visa systems, and the deportation of men for DFV offences. Scholars such as Mehrotra et al. (2016) recommend actions that front-line workers can take to resist the state in perpetuating systemic violence via criminalisation. These actions include to stop criminalising new GBV offences; conceptualise alternatives to individualistic, carceral approaches; expand definitions of safety; and to reimagine funding sources and build coalitions. Diverting funds from policing and investing in community-driven alternatives is critical (Kaba, 2021). The need for practitioners to engage in reflexivity over privilege and the multiple ways their positions and organisations uphold structural inequalities, such as racism, have also been well documented.

These findings provide evidence that front-line workers express the desire to do more but do not have the tools or resources to conceptualise alternative interventions. Viewing the spaces in which front-line workers address DFV as ‘assemblages’ of risk, rather than a mechanism of a hegemonic state, reveal the potential for engagements that create or maintain the political will required to further an abolitionist agenda (Kim, 2019). Strategies aimed at knowledge translation and partnerships with practitioners on the ground would assist to develop frameworks that progress these goals. Partnerships might include:

• building networks between academics and front-line organisations;

• making research accessible through regular exchanges of knowledge, such as symposiums, webinars, or sharing published outputs;

• research partnerships and funding opportunities engaged in participatory action research that centre those most marginalised;

• evaluations of alternative, community-driven programs that document and build an evidence base;

• publishing pursuits that draw on the knowledge and experience of front-line workers (Davis et al., 2022).

Davies et al. (2021) suggest an earlier entry point to knowledge production, by ‘bringing abolition in’ via social work and human service degrees. A core element of reflexivity is reflecting on social location and privilege, and should include historicising how oppressive structures came to be and continue to operate. Speaking to the Australian context, integrating education on race and other inequalities more widely in university degrees such as criminology and the social sciences, rather than being standalone courses, would provide useful tools for those who come from privilege and do not have lived experience. These activities and partnerships might contribute to the development of alternative frameworks and systems of safety that allow for the protection of those experiencing interpersonal violence, while advancing aspirations of abolition as responsive and supportive communities (Richie et al., 2021).

Conclusion

This article sought to understand what drives front-line workers to engage the law for the protection of women they support. Front-line workers in NGOs play an active and often overlooked role in the criminalisation of GBV offences. Front-line workers often feed up into policy and law changes via various consultation mechanisms, as policymakers themselves at different stages of their career, or as activists (Murray & Powell, 2011). An intersectional framework of risk highlights the combination of managing actuarial risk, as the protection of women experiencing violence in intimate relationships, with governing ‘risky’ populations. This research demonstrates that these two dimensions of risk come together to mutually constitute social inequalities and maintain the oppression of marginalised groups (Olofsson et al., 2014). If we understand the state as ‘assemblages’ of risk, however, the differences in how front-line workers use the law towards ends not intended reveal opportunities toward engaging front-line workers in an abolitionist agenda through pursuits aimed at developing alternatives to criminalisation.

Acknowledgments

I would like to thank my PhD supervisors Dr Jenny Munro and Professor Greg Marston for their direction on theory and feedback that informed ealier iterations of this article. Many thanks go to the anonymous reviewers and the editor for their helpful suggestions that informed and improved the current article. I would also like to thank the participants of this research for their generous time and valuable insights, without which papers such as these would not be possible.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This work was supported by a University of Queensland Research Training Program grant.

Notes

1. Professions in front-line organisations might include social work, counselling, psychology and community development.

2. While processes differ by state, the protection order system in Australia has options of both civil and criminal legal proceedings; both can be initiated by either the victim or by police. Australia does not have ‘mandatory arrest’ policies, however police do have powers to take out DVPO’s on the behalf of victims.

3. The HRT was implemented in 2019, and therefore not in operation when most of the interviews were conducted.

4. I focus largely on the DVPO system as this is what arose in the interviews. This focus is possibly due to the crisis-oriented nature of front-line DFV work – workers are often aware of initial interventions by police but might not see processes through to criminal law proceedings.

5. Many workers said that police do not remove police initiated DVPO’s, despite legislation allowing them to. This might be reflective of the particular geographic location of the interviews and the organizations that took part. However, this was also my experience working as a DFV support worker in two different DFV specialist courts in Brisbane in 2019. It was well known amongst lawyers, court staff and support workers that police would not remove DVPO’s, despite women’s wishes or potential harms.

6. ‘Cross orders’ refer to when both parties in a DFV incident take out a protection order. Police can also take out cross orders on both parties when they are unable to discern who is the victim and who is the perpetrator.

7. Legal Aid uses a process called ‘means testing’ to define eligibility, based on income and assets. The threshold is quite low, and varies depending on a number of factors, such as number of children. Women might own a house or business (sometimes with the perpetrator) but still be experiencing financial difficulties, and cannot afford a private lawyer.

8. Breaches of a DVPO can result in criminal charges, bringing other consequences such as immigration detention/deportation.