Abstract
This article analyses domestic violence fatality reviews/coronial files for a whole-of-population study of First Nations women killed by male partners across several Australian jurisdictions between 2006 and 2016, alongside yarning/interviews with First Nations domestic violence workers, violence survivors and Elders. Findings show that most women had domestic violence-related police contact before their deaths, and these interactions were frequently harmful. Harms resulted from police inaction, including failures to respond or enforce the law. Harms also resulted from police action, with policing enhancing state surveillance of victims’ families, eroding victims’ autonomy and criminalizing victims. Findings are located in neo-colonial context, emphasizing a policy need for meaningful Indigenous self-determination and reinforcing the importance of inclusive disciplinary and epistemological practices in gender-based violence criminology.
INTRODUCTION
First Nations women across Australia experience violence and assault disproportionately and at high rates compared with non-Indigenous, ‘settler’, women. First Nations women are 32 times more likely to be hospitalized due to family violence, and 11 times more likely to die due to assault than non-Indigenous women (Australian Human Rights Commission 2020: 44). Between 2010–14, despite representing around 3.3% of the Australian female population (Australian Bureau of Statistics 2018), First Nations women represented 22% of domestic violence femicide victims (Australian Domestic and Family Violence Death Review Network 2018: 17).
In accounting for disproportionate rates of domestic violence against First Nations women, First Nations academics have consistently linked domestic and family violence to the historical and ongoing violence of ‘settler’ colonization (Atkinson 1990; Lucashenko 1996; Aboriginal and Torres Strait Islander Women’s Task Force on Violence 1999; Cripps and Davis 2012; Langton 2016). As Atkinson (1990) describes, the importation of patriarchal constructs and values through processes of British colonization and ‘settlement’ significantly altered Indigenous ways of being, living and traditional knowledge systems. The ongoing effects of colonial violence and control have also led to widespread poverty and socio-economic disadvantage for many First Nations people (Atkinson 1990; Bolger 1991: 44). Violence against First Nations women accordingly cannot be understood without recognising the impacts of colonization and the ongoing violence of ‘settler’ occupation. This adds complexity to mainstream ‘settler’ feminist understandings of domestic violence, which cite gender inequality as the primary driver of men’s violence against women (Our Watch 2015). It also raises questions around the ability of responses founded in limited ‘settler’ conceptualizations of domestic violence to effectively respond to domestic violence against First Nations women.
Australia’s National Plan to Reduce Violence Against Women and their Children 2010–22 (the National Plan) and its subsequent action plans explicitly identify the need for governments to support First Nations community-led solutions to domestic and family violence (Council of Australian Governments 2011: 20). Yet, these same plans simultaneously advocate expansions of, and increased investments in, mainstream police powers and laws, from which First Nations people are certainly not exempt. This strong ‘settler’ emphasis on domestic violence criminalization is founded in carceral feminism(s): a range of feminist ontologies that emphasize uses of criminal justice solutions to resolve gender-based violence issues and support victims (Sweet 2016). Findings profiled in this article highlight that despite the recognized need for alternative responses, First Nations women experiencing domestic violence regularly engage with the ‘settler’ response system, most typically with police. This necessitates further exploration of not only First Nations women’s police interactions, but consideration of the broader, structural emphasis on ‘settler’ policing as a key domestic violence response in Australia’s neo-colonial context.
This article draws on two data sources: analysis of coronial files/specialist domestic violence fatality reviews (DVFRs) of a ‘whole of population’ sample of all domestic violence-context homicides of First Nations women who were killed by a male intimate partner in New South Wales (NSW), Queensland, the Northern Territory (NT), South Australia (SA), Victoria and Tasmania between 2006 and 2016 (N = 68); and yarning/interviews with First Nations domestic and family violence specialist workers, Elders and/or survivors of violence. In addition to exploring the nature of women’s police interactions and locating these findings within a broader structural critique, this study’s implications for domestic violence policy and knowledge production within criminology are discussed.
A NOTE REGARDING THIS STUDY’S CRIMINOLOGICAL APPROACH
This article deliberately locates its analysis of domestic violence policing practices within a neo-colonial context. This reflects post-colonial and Indigenous approaches to criminology, which intentionally place the ‘colonial matrix of power’ at the centre of criminological inquiry (Blagg and Anthony 2019: 1). Post-colonial approaches posit that, more than many other disciplines, criminology is implicated in neo-colonial regimes of power and control (Agozino 2003; Cunneen and Tauri 2016; Blagg and Anthony 2019). This control is expressed via the discipline’s preoccupation with individual criminality, penal systems and the idea of ‘risky’ individuals (Agozino 2003; Cunneen and Tauri 2016; Blagg and Anthony 2019). In the Australian context, this also manifests in the discipline’s construction and management of the ‘Indigenous problem’ (Tauri 2018). This study accordingly attempts to exercise a degree of disciplinary ‘disobedience’ (Mignolo 2011; Blagg and Anthony 2019) by examining expressions of state power within responses to domestic violence. In line with post-colonial and Indigenous criminological methodologies (Cunneen and Tauri 2016; Blagg and Anthony 2019) this article also intentionally prioritizes the scholarship of First Nations people, and does not draw equivalence with other, differently situated, minority populations (Behrendt 1993).
TERMINOLOGY
The term family, rather than domestic, violence is considered preferable by many First Nations people as it reflects holistic, extended conceptualizations of intra-familial violence and harm (Blagg 2016; Cripps and Davis 2012). This article adopts the narrower language of domestic violence as intimate partner relationships dominate Australia’s responses to domestic and family violence, as well as the gendered evidence-base that underpins these responses. The terms ‘family’ and ‘domestic’ violence are often used interchangeably in ways that may obscure a dominant system focus on intimate partner violence, as well as conceal the operation of gendered narratives. A specific focus on policing intimate partner violence against First Nations women can help elucidate some of these issues.
This article uses the term First Nations as it encapsulates the many, varied first peoples and sovereign nations of Australia (Australian Institute of Aboriginal and Torres Strait Islander Studies 2021). Groupings of Australia’s diverse first peoples, particularly under the term ‘Indigenous’, may essentialize plural nations and peoples in problematic and harmful ways. The term ‘Indigenous’ is used where unavoidable. The word ‘settler’ is also used in inverted commas to acknowledge that Australia was invaded and First Nations sovereignty has never been ceded (Commonwealth of Australia 2017: 17). ‘Settler colonialism’ is also understood as a persistent structure and organising principle of the Australian state (Wolfe 1999: 2). Australia is not post-colonial, ‘settler’ colonial structures remain in place and ‘settlers’ benefit from the ongoing dispossession of First Nations people (Haebich 2015/2016).
THE COLONIAL CONTEXT OF AUSTRALIAN POLICING
The colonial relationship between First Nations communities and the police is foundational to understanding domestic violence policing of First Nations women. In 1788 the British invaded Australia without agreement or treaty-making with the Indigenous peoples of the continent. Police were central to racialized assertions of state power from this point forwards—the ‘spokesmen of the settler and his rule of oppression’ (Fanon 1961/2001: 29)—and have been described as ‘the most consistent point of Aboriginal contact with colonial power’ (Johnston 1991). Since the first organized forces were established during the 1830s, Australian police have held an expansive and ongoing role in regulating and administering the lives of First Nations people on behalf of individual colonial governments, and after 1901, the Commonwealth (Finnane 1994: 111; Libesman 2019; Porter and Cunneen 2020). Under racialized protectionist policies, police enforced state/territory legislation which sought to regulate and control First Nations populations by forcing First Nations people from country and onto reserves and missions, regulating rations, controlling access to medical care and removing First Nations children from their families (Cunneen and Libesman 1995: 33). Under subsequent assimilation policies, police carried out large-scale removals of First Nations children on behalf of the government; seeking to ensure that so-called ‘full descent’ First Nations people would die out, while so-called ‘mixed descent’ First Nations people would be assimilated into the ‘settler’ population (Human Rights and Equal Opportunity Commission 1997).
While government policies eventually aborted the language of ‘protection’ and ‘assimilation’—underpinned by notions of extinguishment and extinction (Australian Law Reform Commission 1986)—there is considerable evidence that the policing of First Nations people still operates as a contiguous site of racism, state power and violence (Cunneen 2001; Porter and Cunneen, 2020). The legacy of police complicity in the state’s internal colonization of First Nations peoples has been the ‘disorganization in Aboriginal communities; institutionalized behaviour; and deep mistrust/hatred of police’ (Jennett 1999: 4). This legacy has also shaped the contours of Australian policing institutions, which are today centrally implicated in the over-criminalization of First Nations people (Cunneen 2001: 8; Australian Law Reform Commission 2018; Porter and Cunneen 2020).
DOMESTIC VIOLENCE POLICING IN AUSTRALIA
Police are the central pillar of Australia’s domestic violence response. Since the 1970s mainstream feminist advocacy has focused on recognising and treating domestic violence as a crime (Murray and Powell 2011; Goodmark 2018: 1–2) with anti-violence advocates promoting increased law enforcement, police prosecutions and escalating criminal penalties to protect victims (Goodmark 2018: 13–5). The anti-violence movement keenly embraced this carceral feminist ‘criminalization agenda’ (Goodmark 2018: 15), and largely due to the influence of anti-violence groups, across Australia today most domestic violence behaviours are criminalized. Police enforce criminal domestic violence laws as well as civil protection orders (DVOs)—hybrid civil/criminal court orders where a breach constitutes a criminal offence (Douglas and Fitzgerald 2018). Police are also gatekeepers of not only the criminal justice (Blagg et al. 2018), but frequently the social service, system (Dowling et al. 2018: viii). In Australia, domestic violence laws and enforcement mechanisms are administered at a state and territory, rather than national, level. Separate state/territory police forces accordingly enforce different jurisdictional laws (Commonwealth of Australia 2015), however, there are often considerable cross-jurisdictional similarities in both laws and enforcement.
FIRST NATIONS WOMEN’S EXPERIENCES OF DOMESTIC VIOLENCE POLICING
Positioning police as key domestic violence responders raises complex issues for First Nations women. Colonial policing practices and police violence (Human Rights and Equal Opportunity Commission 1991) have gendered impacts on First Nations women (Behrendt 1993; Behrendt 2000). First Nations women have long reported concerns that police fail to administer domestic violence laws fairly or equitably (Atkinson 1990; Bolger 1991; Johnston 1991; Aboriginal and Torres Strait Islander Women’s Task Force on Violence 1999; The Special Taskforce on Domestic and Family Violence in Queensland 2015). Domestic violence laws may also negatively impact First Nations women, particularly where the law’s content or administration may be paternalistic, or it may lack sensitivity to First Nations women’s racialized realities (see Bolger 1991: 73; Nancarrow 2019). Given issues of over-policing and high levels of incarceration of First Nations people (Johnston 1991; Mitchell 2019; Porter and Cunneen 2020), domestic violence policing may also produce criminalizing outcomes for victims of violence, for instance where police arrest First Nations women for unrelated criminal matters when responding to domestic violence callouts (Blagg and Anthony 2019: 212–3).
To date, there has been limited systematic examination of these issues. There has been little evidence gathered to establish whether such policing patterns are systemic, or anomalous and isolated. Further attention is required to explore the ways policing in this context may also relate to broader ‘settler’ colonial processes. The research set out in this article addresses these gaps.
METHODOLOGY
This study initially involved analysis of a ‘whole-of-population’ sample of all recorded homicides of First Nations women killed by a male intimate partner in NSW, Queensland, the NT, SA, Victoria and Tasmania between 2006 and 2016 (N = 68). Women ranged in age from 19-51 years old—a third of the women were aged between 20 and 30, almost a third were aged between 30 and 40 and over a quarter were aged between 40 and 50. Women were killed by First Nations (N = 57), and non-Indigenous male partners (N = 12).1 All case data was derived from coronial and specialist DVFR files, which include proximal and historical documented service contact histories for domestic violence victims. Case data were derived primarily from review of state records, and accordingly reflect the perspective, interests and values of the Australian state and its agents (Kidd [1997] 2017), while providing insight into the functioning of state institutions and bureaucracies.
Permanent DVFRs currently operate in the coronial jurisdiction in NSW, Queensland, the NT, Victoria and SA, and examine domestic violence-related homicides to identify systemic gaps in service responses and prevent future deaths (Bugeja et al. 2013). In some jurisdictions DVFRs facilitated access to their narratives (NSW, NT), in others, original coronial files were accessed (Tasmania, Victoria), while in SA and Queensland both DVFR narratives (where available) and coronial files (where DVFRs were unavailable) were accessed. Files typically contained extensive proximal and distal service contact histories for victims and their partners, including police, specialist service, courts, child protection and health records. While comparable, not all jurisdictions’ case materials included equivalent detail. For instance, while the NSW DVFR consistently accessed police information systems when preparing its narratives, in other jurisdictions it was not always possible to determine whether police records were complete (for instance, if those records covered the victim’s childhood and all prior relationships). Some of the data provided in this paper may accordingly be an undercount.
Where case materials were not already in narrative form, the author analysed original coronial or DFVR files and recorded details as a narrative at DFVR or coronial offices. All narratives were analysed in NVivo adopting an emergent coding approach, involving content and thematic analysis. Case narratives were, on average, around 10,000 words in length. The majority of homicides occurred in NSW, the NT and Queensland.
Alongside case analysis, this study involved semi-structured face-to-face interviews and group yarning (Power 2004; Dean 2010), conducted between 2018 and 2019, with 22 First Nations participants comprising Elders, domestic and family violence specialist workers and/or survivors of intimate partner violence. Site visits to an Aboriginal community-controlled organization (ACCO) providing domestic and family violence services were also conducted for context. Many study participants fit across multiple categories (for instance, they were both a survivor and an Elder) and were recruited from anti-violence and other networks, and via snowballing. All participants were asked about their views about the overall domestic and family violence response to First Nations people, their views on the police and court response, the availability, accessibility and quality of services, their views on successful community approaches, the adequacy of participatory rights and self-determination and their perspectives about the best ways forward. The study was designed following consultations with First Nations Elders and specialist workers from different communities who were active in anti-violence efforts. The research process was iterative (Mills et al. 2010), with insights and findings arising from yarning/interview informing recursive case analysis, and case findings informing yarning/interview (including via discussing particular case themes and issues). Although participants had lived or worked across all jurisdictions, the majority were living or working in NSW and Queensland when interviewed. Preliminary findings were shared and discussed with participants, and these insights informed the final findings. Findings were then disseminated. The research process is detailed in Figure 1 and the interview table is included in Figure 2. As a non-Indigenous researcher, the author specifically acknowledges the First Nations people whose knowledge and trust was formative in this study, and whose views and perspectives are integral to its findings.
Methodology flowchart.
Interview/yarning sample overview (N = 22).
DOMESTIC VIOLENCE POLICING OF FIRST NATIONS WOMEN: FINDINGS
It is a key study finding that the vast majority of First Nations women who were killed by an intimate partner interacted with police in relation to domestic violence prior to their deaths. This was the case for 60 of the 68 women (88%), with almost all women interacting with police on many occasions. Police contact was by far the most common service interaction women experienced across cases, with far fewer women interacting with ‘settler’-controlled specialist services or ACCOs providing domestic and family violence services. This suggests that notwithstanding a public policy emphasis on community-led solutions for First Nations women (Council of Australian Governments 2011), First Nations women who experience domestic violence are highly likely to interact with ‘settler’ police.
While First Nations women who experience domestic violence have described police responses as sometimes being favourable and sometimes very harmful (Aboriginal and Torres Strait Islander Women’s Task Force on Violence 1999; Cunneen 2010; Blagg et al. 2018) across cases in this study there were very few examples of police interventions that did not produce some identifiable harm. Helpful interactions appeared as an aberration, rather than the norm. Across cases, there were two common ways police responded to domestic violence against First Nations women. Firstly, there were instances where police did not enforce domestic violence laws, even though appropriate laws or remedies were available and/or enforcement was expected under legislation or policy at the time. Secondly, there were instances where police enforced law and policy—or took action—but this nonetheless appeared to produce harmful outcomes for First Nations women, including eroding victims’ agency through paternalistic practice, increasing state surveillance and management of First Nations families and criminalizing victims. These responses are divided within the following sections into two major categories—police inaction and police action. Taken together, these failures and the range of policing practices they encompass, suggest that domestic violence policing in neo-colonial context is harmful to First Nations women.
Police inaction
Study findings highlight that First Nations women routinely and repeatedly experienced police reluctance and inaction following domestic violence. Police failures to act involved both failures to enforce law and policy, and failures to respond to domestic violence callouts in a timely fashion (or sometimes at all). Such failures suggested systemic police apathy towards First Nations women as victims of violence, contrary to the helpful, protective response perhaps anticipated by investment in this system.
Failures to apply law/abide by policy
For three-quarters of the First Nations women who were killed by their partners, there were episodes where police did not enforce criminal laws, nor apply for DVOs, when responding to domestic violence incidents (N = 50). The majority of these episodes involved apparent physical or sexual assaults, which are precluded by law in all jurisdictions. Today domestic violence policing approaches are typically described using language such as proactive (NSW), zero tolerance (NT), pro-investigation (NSW, Queensland), pro-prosecution (NSW), pro-arrest (Victoria), pro-charge (Victoria) and pro-intervention (Tasmania, NSW, Victoria) (The Special Taskforce on Domestic and Family Violence in Queensland 2015; Dowling et al 2018; NSW Police Force 2018; Marks, 2019; Victoria Police, 2019; Tasmania Police, 2021). This language has often been used for many years. Notwithstanding this, across cases officers would frequently ‘write-off’ episodes of domestic violence against First Nations women. By its nature policing involves high levels of discretion (Cunneen 2001: 30; Australian Law Reform Commission 2018) and across cases, it was evident that police frequently exercised discretion to not enforce the law when First Nations women were victims of violence. While jurisdictional police forces all encompass hierarchical quality assurance, and senior or specialist officers may have reviewed responding officers’ discretionary decisions, there were no indications discretionary decisions were ever reversed.
For instance, in a case from NSW, the police attended after a First Nations woman’s partner had punched and kicked her numerous times. She called paramedics, but police intercepted the ambulance to confront her about the assault. Police records describe that she would not ‘make a complaint’ and, despite the victim giving responding officers the offender’s name, and both police and paramedics observing her injuries, officers took no further action.2 In another case from the NT, police records describe that a First Nations woman did not have any ‘visible injuries’ when she reported violence. However, hospital records corroborated that she had presented to Alice Springs Hospital the day before for medical treatment of visible injuries arising from the assault she was now reporting to police. Police records accordingly appeared to minimize the victim’s injuries in order to justify inaction.3
The frequency of police failures to act, as well as the nature of these failures (including officer’s deliberate decisions to refrain from taking further action, and to abandon victims, contrary to law and policy), suggests that police apathy towards First Nations women’s victimhood is systemic and institutionally ingrained. The apparent absence of effective quality assurance through internal police review systems across jurisdictions reinforces the systemic nature of this apathy. First Nations women’s abandonment by police in these circumstances reflects similar findings of the 1991 Inquiry into Racist Violence (Human Rights and Equal Opportunity Commission 1991), suggesting that little may have changed over the past 20–30 years.
Police also routinely failed to enforce DVOs protecting First Nations women. Nearly three-quarters of the First Nations women whose cases were considered in the study had been, at one stage or another, protected under a DVO naming an intimate partner as the defendant (N = 48, 71%). Almost half of those women experienced police failures to enforce those orders in the event of a breach (N = 21, 44%). For instance, in one case from NSW officers not only repeatedly failed to enforce ‘do not approach’ DVO conditions as well as those prohibiting ‘alcohol use’, but also failed to charge the defendant after the victim called police to report that he had assaulted her. The victim even provided police with a video-recorded statement about the assault. According to police records officers nonetheless declined to charge breach or other offences as ‘other witnesses at the party were intoxicated’ and provided varying accounts about what had happened.
Reflecting on this, one specialist worker observed that, in her experience, police inaction was frequently driven by officers’ racialized beliefs around First Nations people. She described that ‘there’s…a real strong common belief out there, a real racist belief that the violence in Aboriginal communities is going to continue—why bother addressing it? It’s an Aboriginal issue. It’s an Aboriginal problem. That’s their way.’ This description had strong continuities with colonial constructions of First Nations women as being unworthy of police protection (Behrendt 1993; Behrendt 2000), including the diminishing of women’s victimization as being part of the ‘Aboriginal problem’ or the diminishing of ‘Aboriginal life’ as ‘violent life’ (Lucashenko 1996: 385). These constructions are founded in ‘settler’ colonial logics (Tauri 2018).
Across cases, police failed to pursue criminal charges when women’s neighbours, family or bystanders reported violence, but also when First Nations women directly contacted police for help. Records highlighted that around half of the women whose cases were considered in the study had reported violence to the police on at least one occasion. However, even when women contacted police directly, police records frequently described that they were ‘un-cooperative’ or ‘unwilling’ to work with police. Almost three-quarters of the 60 women who had police contact in relation to domestic violence had been described by officers using these, or similar, terms (N = 44, 73%). Without victim or family testimony it was often difficult to ascertain from police records whether these descriptions reflected victims rejecting further police involvement, or whether these were racialized labels that officers deployed to excuse or justify inaction. In some cases, however, it clearly appeared to be the latter. For instance, in one case from NSW, a First Nations woman called police to report that her partner was intoxicated and abusive. Police attended but did not enforce the current DVO that he had breached. Officers returned the next day to obtain a statement from the woman, but her partner was there and started yelling at police, accusing them of forcing the victim to make a statement. A crowd began to gather and also started yelling at police. At this stage, the victim told officers she did not want to make a statement. The police left without taking further action, describing in their records that ‘it is apparent from the incident that the victim is willing to call police but not willing to assist police thereafter.’
It was also common for police to disbelieve First Nations women who reported violence, and examples of this were identified in a third of the cases where there was police contact (N = 20). For example, in a case from NSW a 17-year-old First Nations woman attended the police station, crying, to report that her partner had punched her in the face. Despite her presentation, police records subsequently described that she ‘didn’t appear fazed’, and note that ‘the police are unsure of the validity of this report due to the lack of details provided [by the victim] when asked about the incident’. Police took no further action. In another case, from Queensland, a First Nations woman reported an assault to police. When police arrived they overheard the woman’s partner screaming at her, and when the officers came inside she told them that he had slapped her on the side of the head before grabbing her by the throat with both hands. Police records described that the woman’s partner was ‘co-operative’ and he denied using physical violence. Although police records note the victim was ‘credible’ police took no further action due to the ‘lack of witnesses’ and lack of ‘visible injuries’ or ‘property damage’.
Reflecting on the drivers of police inaction, one specialist worker/survivor described that in her experience the police were sexist and racist, and would typically believe the male offender over the victim. She described that ‘you’ll have him there, He’s mister cool, calm and collected, and she is hysterical about what just occurred, you know?’ She remarked that First Nations women may struggle with a range of barriers when engaging police, and that officers would then behave in ways that augmented women’s fears by ‘standing over’ them, failing to show sympathy and behaving in an intimidating manner. She remarked ‘you know [police] do have their cultural safety protocols…but do they practice them?’
Routine portrayals of First Nations women in police records as being undeserving or unable to be believed again evoked neo-colonial ideas of First Nations women as unworthy victims (Behrendt 1993; Kelly 1999; Behrendt 2000). These findings also suggest that police interactions may contribute to the ongoing raced and gendered institutional construction of First Nations women as lacking in credibility, and being unworthy of help.
In two-thirds of cases where police were involved, police records clearly indicated that intoxication (either of the victim, the offender, or witnesses to the assault) contributed to officers’ decisions not to progress charges despite often-considerable evidence that domestic violence offences had occurred (66%, N = 40). While intoxication may impede police officers’ ability to investigate and progress charges, it appeared that intoxication was more commonly used as an excuse not to take action. For instance, in one case from NSW, numerous bystanders and neighbours called police after witnessing a First Nations woman’s partner assaulting her and her relative with a plank of wood. Police records describe that ‘at this stage it is very doubtful if the incident took place due to both of the victims being heavily intoxicated’. Police took no further action. In another case from the NT, a long-term victim of violence reported an incident of domestic violence to the police. The police records management system initially coded the incident as a ‘domestic violence’ callout, however attending officers ultimately recoded and wrote off the episode as ‘drunk person’, taking no further action.
When discussing police interpretations of alcohol and substance use, one specialist worker/survivor described that in her view the drivers of these behaviours for First Nations people related to numbing the pain resulting from colonization—‘having kids removed’ and having ‘a history of such systemic issues in their life’. She described that police would commonly blame First Nations people for racialized disadvantage that has its foundations in colonization, thinking: ‘why don’t they do this? Why is their lives like this?’ She further described ‘…and like I say, it all comes back from systemic colonization. Our people haven’t had a fair go in this country. And that’s for sure.’ These findings echo Atkinson’s (2002) observations around neo-colonial constructions of First Nations trauma and intergenerational trauma as an ‘alcohol problem’, and highlight the extent to which these constructions persist and are continually re-embedded via policing and its associated institutional records.
Finally, in almost a fifth of cases (N = 12) there was evidence that First Nations women had asked officers to take action on their behalf, usually to apply for a DVO or to progress charges against their partner, but police failed to act. For example, in one case from the NT, a First Nations woman attended the police station after an assault, and while police entered her details into the system, they did not take her statement. She left the station after a long wait in the public waiting area in view of passing community members. Police failed to follow up with her, their records describing that officers were dispatched to an ‘urgent’ job. She was killed later that night. In another case from one of the smaller jurisdictions, a First Nations woman attended her neighbour’s property after an assault. When police attended she had several seizures, although both paramedic and police records indicate that responders believed she may have been ‘faking’ those seizures as she ‘seemed intoxicated’. She asked police to arrest her partner for the assault, but officers indicated that they couldn’t ‘understand her’, and did not believe she had been assaulted. Officers left without interviewing any of the witnesses or parties present. She was killed that night.
As Langton et al. (2020) have observed, First Nations women’s help-seeking behaviours are informed by prior police interactions. Findings from this study suggest that not only can ‘under’-policing (Cunneen 2001) expose First Nations women to ongoing, and sometimes fatal, violence, but findings also suggest that these practices are interwoven with neo-colonial constructions of First Nations women as undeserving victims (Behrendt 2000).
Failures to respond
Although police inaction most commonly involved failures to enforce the law, in many cases officers were considerably delayed in responding to callouts, and in almost a fifth of the cases, there was evidence that police never attended when called (N = 10). Many failures to attend occurred on the night of the homicide, and given this is subject to considerable scrutiny in death investigations, such failures may have been more common than these figures suggest. In a case from NSW, a woman’s neighbours called police up to seven times to report that she was experiencing domestic violence, including that they had seen her partner pulling her across the lawn by her hair. Police did not attend for another 24 hours. She had been killed by the time they arrived. In another case from NSW a woman’s family, who lived on a former Aboriginal mission, called police numerous times to report domestic violence between the victim and her partner. Police did not attend for an hour, and when they did officers drove through the community with their windows rolled down ‘listening for loud music or yelling’. Officers closed the job when they could not identify anything untoward from inside their vehicle. Later that night the woman was killed.
While almost half of the First Nations women who were killed by a partner were living in remote or very remote areas at the time of their death (N = 32), there were many examples of police being delayed or failing to attend callouts in cities and regional areas. In only one case was it specifically identified that there was no police crew available to attend due to the remote location. As a specialist worker remarked, while police responses may be poor in rural or remote communities, ‘the response from Police in like, Western Sydney, is really bad as well…[the clients that I’ve worked with can’t] even get police to respond when they ring to say that an incident’s happening.’ This suggests that police delay and non-response to First Nations people may not be simply a resourcing issue, but a racialized practice that goes to the nature of the police response (Cunneen 2001: 164).
Police action
This article has so far focused on police inaction and failures to respond. This section considers a related, yet seemingly opposing and more complicated problem: the harms that may result when police enforce the law or act in accordance with domestic violence policy. This section suggests that even when police do respond to First Nations women who experience violence this nonetheless produces a range of harmful outcomes. These harms are divided into protective paternalism, state surveillance and criminalization.
Protective paternalism
Paternalistic domestic violence laws, including mandatory or pro-arrest policies and (hard and soft) no-drop prosecution policies, were initially advanced by anti-violence advocates in the United States as a way to remedy police inaction and low rates of domestic violence prosecutions (Goodmark 2018: 14–5). These and related hard-line approaches have been incorporated across Australia at different times and to varying degrees (see, for instance, Murray and Powell 2011). Paternalistic policies and policing strategies are guided by the carceral feminist assumption that a completed prosecution is an optimal outcome following a domestic violence episode (Terwiel 2019). Such approaches render the victim’s wishes irrelevant, replacing these with strong state intervention. As participants observed this may erode First Nations women’s agency, ignore women’s valid reasons for resisting police or court action, and may expose women to further danger. These approaches also clearly reflect paternalistic incursions of state power into the lives of First Nations women, which strongly evokes colonial managerial processes relating to First Nations populations (Human Rights and Equal Opportunity Commission 1997).
Across cases, there was evidence that police repeatedly administered protective domestic violence laws against First Nations women’s wishes. This was most evident in the case of DVOs. In 12 of the cases in the study (18%) there was evidence police applied for a DVO against the victim’s wishes. In a third of those cases, there was evidence that the woman’s partner subsequently breached the DVO; and in two of those cases police then described the woman protected under the DVO as being complicit in, or responsible for, her partner’s breach. Even in cases where it was unclear from the available evidence whether victims supported the orders, it was relatively common for police records to describe that women who were protected under DVOs were responsible when their partners breached those orders (N = 11, 23% of the women previously protected under DVOs). In one case from NSW, a First Nations woman told police that she did not want a DVO, but the police applied for (and the court granted) this order against her wishes. The order included an ouster/exclusion condition,4 even though the woman and her partner remained living together in a relationship at the time. The defendant assaulted the woman in breach of the order, and when police attended their records describe that: ‘police fear the accused will continue to assault the vic[tim] however the vic[tim] is facilitating a breach in the contact condition of the [DVO] by continually allowing the accused to live with her...’ Participants described that paternalistic interventions were harmful as they reduced victim agency, thereby further increasing First Nations women’s distrust in the police and criminal justice system. Participants also described that when women resisted police intervention officers would assume that First Nations women were difficult or unworthy victims.
While protective paternalism is underwritten by the assumption that the best outcome following an episode of domestic violence is a criminal sanction, notwithstanding the victim’s wishes, it is important to note that for almost three-quarters of all of the First Nations women who were killed by their partner (N = 50), the woman’s partner had previously been convicted of domestic violence offences. Prior convictions accordingly did not stop the victim’s partner from using violence, including fatal violence. This obvious failure of a response that is intended to help victims challenges not only the assumption underlying protective paternalistic policing, but also raises questions around the effectiveness of the criminalization agenda more broadly (Goodmark 2018). As one specialist worker described, ‘there’s a tendency to do more of the ‘let’s charge the person—let’s send them to the big house and incarcerate them for a period of time, not provide them any therapeutic intervention while they’re there’—I don’t think that’s being very productive in terms of responding to domestic violence.’
These findings directly challenge the effectiveness of carceral domestic violence response strategies. They also highlight the ways in which carceral and paternalistic approaches mobilized by the state are entwined with neo-colonial regimes of power and control over First Nations women’s lives (Balfour 2021).
State surveillance
In recent years there has been an increasing emphasis on multi-agency work for high-risk victims of violence. Following structured common risk assessment (most often completed by police at a domestic violence callout, and based largely on evidence from the Global North or imported international risk assessment models) victims adjudicated as ‘high-risk’ have their case triaged by a multi-agency committee who undertakes safety planning, typically without victim involvement (McCulloch et al. 2016).
While multi-agency meetings were not a feature of the cases (due to the time period studied), participants described the ways this secondary consequence of police intervention may increase state surveillance of First Nations women and families, and intensify state regulation and management of First Nations peoples. While several participants described that holistic and integrated work practice is important in managing domestic violence victims’ complex needs, they also identified that subjecting First Nations families to state-run multi-agency processes may produce additional negative outcomes for victims and their families beyond interpersonal violence. Reflecting on managerial state practices, one specialist worker also described that ‘families talk about not having a voice [in meetings with State agencies]: cause they don’t have a voice because the people driving the meetings or the conferences, they speak a very different language … they’re [also] seeing you through a particular—“I’m seeing you through this long wrap sheet that you’ve got here’; ‘I’m seeing you through…the latest episode of violence”.’
Participants also raised related concerns about the direct pathway from police involvement to child protection services involvement. Across cases police commonly made mandatory reports to child protection services regarding a child’s exposure to domestic violence between their parents. In several cases, this directly contributed to the victim’s children being removed. Participants described child protection interventions, and the ongoing disproportionate removals of First Nations children (Davis 2019), as a form of state violence against First Nations people. As one specialist worker described as soon as you hear child protection, that incites so much fear. It’s not that far removed from, it might have been them taken as a child. It might have been their cousins. It’s not like that’s something that happened 20, 30, 40 years ago, that’s a distant memory. It’s stuff that happened within their lifetime. [And it happens now].’ This finding, reflected in other studies (for instance, Aboriginal and Torres Strait Islander Women’s Task Force on Violence 1999; Davis 2019; Langton et al. 2020) suggests not only neo-colonial contiguities between historical and contemporary population management practices involving First Nations children (Davis 2019; Human Rights and Equal Opportunity Commission 1997), but highlights the extent to which domestic violence response systems, and especially police, are interwoven with adjacent, broad forms of structural violence against First Nations people.
The criminalization of First Nations women
Across jurisdictions, domestic violence policing regularly contributed to the criminalization of First Nations women. In almost a third of the cases in the study, police had previously identified the victim of violence as a domestic violence perpetrator (N = 21). Across cases, police records typically did not appear to contextualize or understand the victim’s use of physical violence against her abusive partner, or the circumstances surrounding non-physical violence or arguments. When women used physical violence, it was not always possible to accurately ascertain the nature of this violence from available police records, but in many episodes, it was clear that the violence for which women were investigated or arrested involved retaliatory or responsive violence (so-called ‘violent resistance’) or ‘fights’ (Blagg et al. 2018; Nancarrow 2019). This criminalization of First Nations women also had a flow on effects such as women being named as defendants in DVOs (N = 12, 18% of all homicide victims).
In interpreting this, one specialist worker/survivor drew attention to the racialized and sexist components of the police response, describing that when police witness a First Nations woman being upset and angry during a domestic violence callout they react by stereotyping that woman as being ‘a volatile, aggressive, Aboriginal woman’ … an ‘angry black woman’. In her experience, police assumptions around how victims should behave (a response she described as predicated on ‘white women’ victims) would lead to police incorrectly identifying First Nations women as aggressors. For First Nations women with criminal histories, she described that ‘I think women never know who [police officers] are going to blame for the incident’, which may also be a considerable barrier to reporting.
Across cases, police also frequently criminalized First Nations women for unrelated matters when officers were investigating domestic violence incidents. In just under a quarter of cases involving police contact, there was evidence that police investigated, arrested or charged First Nations women for unrelated criminal offences when they attended a domestic violence callout in which the woman was the victim (N = 13, 22%). Several of these episodes involved women being investigated or charged with drunkenness offences, including in some cases being conveyed into protective custody or a diversionary centre to ‘sober up’. In one case police cautioned a woman for making ‘false reports’ of domestic violence, and in another, a woman was charged with offensive language and assault officer offences. For one woman, Queensland police officers helped her to escape from an abusive older male who had kidnapped her, only to arrest her for outstanding warrants once she was in their custody.
The criminalization of domestic violence victims at domestic violence callouts is a very clear example of a response that is intended to help victims mutating quickly into a punitive, penal response. As Cunneen (2001 :163) observes, there are challenges associated with expecting Australia’s ‘settler’ criminal justice system to pivot from seeing First Nations people as criminal offenders (the default position) to seeing First Nations women as victims. The high rates at which domestic violence victims were criminalized during the course of police interactions reinforce Cunneen’s observations.
Locating these findings in a broader policing context, it would appear that First Nations women were frequently coded ‘criminal’ by police, and police systems, even before officers entered the premises to ‘assist’ them as a victim. Of the First Nations women who were killed by their partners around half had previously been identified by police as being ‘criminal offenders’ in either domestic violence or other matters (N = 35, 51%). In NSW, where victims’ criminal histories were complete, the proportion of women who had been identified as criminal offenders increased to just under two-thirds (13 of 20 cases that occurred in that jurisdiction, 65%). In NSW, for some women, the police computerized management system also generated automated warnings that would pop up when police responded to domestic violence callouts. For one First Nations woman an automated warning informed police that she ‘had a history of assaulting police, that she may be violent, that she should be approached with caution and that she makes false allegations against police.’ Such automated warnings appeared likely to negatively influence the police response to those women when they were victims of violence.
Participants described that the over-criminalization of First Nations women is both a cause and a consequence of racism, and is deeply embedded within the colonially constructed relationship between police and First Nations people. As one Elder, a survivor of violence, observed ‘when a cop walks up and goes to see who you are and see criminal offences, that’s all they see. You’re a criminal. Okay? Now we’re going to treat you like a criminal, whether you’re a victim or not, that’s beside the point. You’re a criminal first and foremost, and that’s what we’ll always run to’.
DISCUSSION
Policy implications
Findings from this study suggest that state responses to domestic and family violence against First Nations women are entwined with the persistent structures of ‘settler’ colonialism, and the active colonial project (Blagg and Anthony 2019). Findings indicate that in a high proportion of fatal domestic violence cases, police failed to respond effectively and in some cases, at all, to First Nations victims who experienced abuse. Repeated failures by police to enforce law and policy were highly suggestive of systemic police apathy towards First Nations women. However, findings also suggest that even when police actively responded to First Nations women who experienced violence, applying law and policy, these interactions were frequently harmful. Police interventions were identified as eroding women’s agency through paternalistic practice, exposing First Nations women and their families to increased state surveillance, and contributing to the further enmeshment of First Nations women victims of violence in the criminal justice system as ‘offenders’. Domestic violence policing accordingly appeared to contribute to the ongoing re-construction of First Nations women as colonized subjects through not only persistent, systemic racialized apathy towards women as victims of violence (and the recursive construction of this apathy via police records), but through the mobilization of racialized, neo-colonial state systems to control, surveille and criminalize First Nations women.
Although for analytical purposes harms of inaction and action were discussed separately in this article, for First Nations women these harms were imbricated. Of the women who had domestic violence-related police contact, 90% experienced harms of police inaction (N = 54), two-thirds experienced harms of police action (N = 40) and over half experienced both kinds of harm (N = 34, 57%). Women often experienced cascading, multiple harms and it is highly likely that women may have experienced far more harms than were reflected in the limited state records accessed.
Findings also highlight that police continue to occupy a priority location in the domestic violence response. While current policy frameworks—such as the National Plan—recognize and purport to accommodate culturally safe, community-led, responses to domestic and family violence against First Nations women within the broader response framework, this study indicates that First Nations women nonetheless frequently interact with ‘settler’ police when they experience violence and abuse. While study participants identified the practical importance of First Nations women retaining the choice to access mainstream services (including police) should they wish to, all expressed significant concerns about this system’s negative effects on First Nations people and considered that there was, at present, a lack of meaningful choice for First Nations women to access alternate culturally safe systems. This lack of meaningful choice was attributed by many to underinvestment in ACCOs and limited service availability, which would likely drive women to access police in the event of a ‘crisis’. One clear policy implication of this study is the need for increased investment in culturally acceptable and safe services for First Nations women (Blagg et al. 2018; Langton et al. 2020), including community-based safety and security measures that disrupt conventional policing paradigms (Porter 2016; Blagg et al. 2018; Porter and Cunneen 2020), as this investment will reduce the likelihood that women will have no choice but to interact with ‘settler’ police.
However, this also begs the question of whether it is simply state underinvestment in community-driven solutions that undermines the system’s ability to deliver safety and security for First Nations women, or whether the underlying issues are more structural and pernicious in character. While frameworks such as the National Plan already purport to ‘make space’ for First Nations justice needs and interests, the persistent and growing emphasis on criminalization within the same framework is telling. First Nations women have for many years directly challenged the suitability of criminalization responses to violence in First Nations communities (Aboriginal and Torres Strait Islander Women’s Task Force on Violence 1999; Nancarrow 2006; Nancarrow 2019), but in recent decades these responses have only expanded. Noting this, and given the negative impact the ‘settler’ criminalization framework appears to have on First Nations women and families (as well as women’s high levels of ongoing contact with these systems notwithstanding these negative effects) it may be observed that the state’s inclusion of First Nations justice solutions into its national plan of action is largely symbolic and performative rather than indicative of true political commitment to alternative responses. Further, it may be argued that the state ‘accommodating’ First Nations women’s justice needs within the existing system may strategically quash First Nations peoples’ stronger claims to differential treatment, effectively preserving the status quo and further consolidating state power over First Nations peoples’ lives. When understood this way, the domestic violence system, with its persistent and growing emphasis on carceral approaches, not only reflects neo-colonial state power over First Nations people, but acts in service of it.
Indigenous self-determination is a strong basis upon which to challenge state responses to domestic and family violence and delimit state power in this area. However, as can be observed from this article, fully recognising and respecting self-determination within domestic violence responses cannot simply involve the ‘settler’ state conceding limited space within existing carceral frameworks that may be (increasingly) harmful to First Nations people. Structural reform is required, including the broad remediation of the fundamental power imbalances that shape social and political relations between First Nations people and the state (Davis 2010; Commonwealth of Australia 2017). Only then will Australia’s First Nations reclaim power to establish a true alternative response system to domestic violence and will domestic and family violence responses serve First Nations women’s rights and needs.
Implications for criminology
This study’s approach and content has significance for processes of knowledge production within criminology, and specifically gender-based violence scholarship. As gender-based violence scholarship, and its influence, continues to grow, it is incumbent on (mainstream) feminist scholars to consider the consequences of criminological research that privileges essentialist gendered narratives that inadequately attend to the localized and broader forces of colonialism and racism. Epistemologies and life experiences outside of the white Western model, including from domains of Indigenous and post-colonial criminology (Cunneen and Tauri 2016), must be central—not marginal—to all gender-based violence research, especially given the intersections this research has with carceralism and state power. Feminist criminologists must accept the call to ‘theorize the relinquishment of power so that feminist practice can contribute to changing the racial order’ (Moreton-Robinson 2000: 186). Only once inclusive epistemological and disciplinary practices are centred, both within gender-based violence scholarship and more broadly, will disciplinary knowledge systems be decoupled from their complicity in reinforcing and reproducing the ‘colonial matrix of power’ (Blagg and Anthony 2019: 227).
Footnotes
One First Nations woman was killed by two non-Indigenous males.
Assault offences were available under the Crimes Act 1900 (NSW) and appropriate arrest powers were available under s99 Law Enforcement (Powers and Responsibilities) Act 2002 (NSW) at this time.
Relevant offences were available under the Criminal Code Act 1983 (NT) and appropriate arrest powers were available under the Police Administration Act 1978 (NT) at this time.
These orders prevent the defendant from living at the same premises as the protected person.
ACKNOWLEDGEMENTS
To all First Nations participants involved in this research—thank you for your generosity and for sharing your considerable knowledge and expertise. Thank you also to Megan Davis, Thalia Anthony, Rachel Condry and Nazila Ghanea for your insightful and invaluable feedback on earlier drafts of this article.
REFERENCES
—— (
—— (
—— (
—— (
—— (
