Promoting Aboriginal Women’s Human Rights — Understanding When not to Prosecute Aboriginal Women

Abstract


Introduction
In Australia, most women charged with homicide after killing their abusive partner are convicted following their guilty plea, and a disproportionate number of Aboriginal and Torres Strait Islander women plead guilty (Dioso-Villa and Nash 2024; Dioso-Villa and Nash 2023; Domestic Violence Resource Centre Victoria 2016; Sheehy, Stubbs and Tolmie 2012).These women's cases never come to trial and their plea is in exchange for abandoning any claim to self-defence which would, if successful, have resulted in an acquittal.This paper is about Australian prosecutors' decisions to accept or refuse Aboriginal women's guilty pleas after they are charged with the homicide of their abusive partner.It is argued that prosecutors are misusing their authority by treating 'plea negotiations' as an exercise of discretion even in the absence of evidence capable of proving that the woman charged did not act in self-defence.Acting in the absence of a prima facie case and relying on the immense pressure an Aboriginal woman is under to avoid a trial (often so she can be reunited with her children) amounts to effectively removing Aboriginal women's human right (and criminal law entitlement) to act in self-defence.
The paper sets out the law of self-defence in Australia to explain that an accurate understanding of the violence a person faced is indispensable for the law of self-defence to operate (Tarrant, Tolmie and Giudice 2019).In Section 3, we then examine Indigenous research on violence against Indigenous women and show that taking a social and systemic entrapment approach identifies the kinds of inquiries a prosecutor should undertake to gain an accurate understanding of the violence an individual Legal assessments required by the law of self-defence cannot be made in the absence of an accurate understanding of the violence a person faced before they used force.This article shows how prosecutors are making errors in their decisions to prosecute Aboriginal women who have killed an abusive partner, due to inaccurate understandings of the violence faced by defendants in these circumstances.We argue this error is a fundamental error of law, not one of fact and discretionary authority.It amounts to prosecutors making decisions to prosecute Aboriginal women (and accept guilty pleas) in the absence of a prima facie case.We argue that such decisions by prosecutors can be challenged via an application for stay of proceedings for abuse of process or a 'no case' submission.

The Law of Self-defence
Self-defence is drafted differently across Australia; 1 however, all formulations contain the same primary requirements.The prosecution must prove: • the defendant did not believe her conduct was necessary to defend herself or another person (a subjective inquiry), or • her conduct was not a reasonable response in the circumstances she believed existed (a subjective/objective inquiry). 2  The arguments in this article proceed from a self-evident proposition, that these legal questions required to be determined by self-defence law can only be determined on the basis of an accurate understanding of the violence a defendant believed she faced (Tarrant, Tolmie and Giudice 2019).Put another way, in the absence of an accurate conceptualisation of the threat of violence that a self-defence claim is based on, none of the legal determinations required by the law of self-defence are possible to make.This means there is a legal obligation on prosecutors to understand accurately the violence a charged person faced, including the forms of violence that inhere in intimate partner abuse Aboriginal women face, no less than forms of violence that other defendants face in other contexts.That the law of self-defence relies on an accurate understanding of the violence a claim is based on is recognised in recent Evidence law reform in some Australian states, for example, in Victoria (Jury Directions Act 2015), Queensland (Evidence Act 1977) and Western Australia (Evidence Act 1906).In Western Australia, the reforms were introduced on the release from prison of Jody Gore, an Aboriginal woman who had been subjected to extreme intimate partner violence but was convicted of murder and sentenced to life imprisonment (Hansard 2019).
The interplay of the factors underpinning violence Aboriginal and Torres Strait Islander women experience is, Kyllie Cripps and Hannah McGlade write, 'exceedingly complex'.However, 'to ignore any of these factors and the role they played and continue to play in families is tantamount to not understanding family violence as it occurs in Indigenous communities' (Cripps and McGlade 2008: 242-243).It follows that, where an Indigenous woman was responding to family violence, a prosecutor cannot properly apply the law of self-defence unless these complexities are understood and properly taken account of.

Understanding Violence Against Aboriginal Women so that the Law of Self-defence can be Applied
This section sets out what we suggest is the legal approach required for understanding the violence an Aboriginal woman has faced where family violence is in issue (and, therefore, for making the legal judgments required by self-defence in the prosecution process).This approach is supported by Australian and international research on family violence against Indigenous women.It supports a framing of the facts that reveals the meaning of violence to which an individual Aboriginal woman was subjected, such that her experiences are perceivable under the law.
The approach we argue prosecutors should take in assessing evidence relies on two fields of research and advocacy.The first is Indigenous research and advocacy about violence against Indigenous women in Australia, including in international human rights frameworks.The second is the 'social and systemic entrapment' approach to conceptualising intimate partner violence developed by Julia Tolmie and colleagues in New Zealand, Aotearoa, and within feminist and intersectional theory and advocacy.
The arguments made later in the article about how criminal law principle should be applied by Australian courts concern centralising an understanding of violence as it is experienced by the Aboriginal women who were subjected to it.We take the same approach to formulating how we argue prosecutors should inquire into the violence, by prioritising Indigenous women's understanding of these forms of gender-based violence.Bevan, Lloyd and McGlade (2024) reviewed research from the past 25 years, seeking to understand to what extent Indigenous-led research describes the determinants and responses to missing, murdered and incarcerated (MMI) Indigenous women in Australia .They found that only seven Indigenous-led sources specifically examined MMI Indigenous women (2024: 12).They presented this paucity of Indigenous-led research as one of the findings of the review.Thus, although there is little published research in this field led by Indigenous women, where it exists, we have prioritised it.We have also relied on non-Indigenous and international research.

Indigenous Researchers Prioritise Systemic, Colonial Violence Within Frameworks for Understanding Interpersonal Family Violence
Indigenous women's research and advocacy centralises Indigenous women's relationship with the state when conceptualising the interpersonal violence to which they are subjected.The research shows that individualised violence against Aboriginal women by their partners can only be understood through colonial relations, which continue to impact women today.
Bevan, Lloyd and McGlade found that Indigenous-led research 'described an injustice experienced by Indigenous women in Australia through the processes of colonisation and ongoing harm by the state' (2024: 13).For example, in a digital case study of Indigenous femicide, Allas and colleagues show that what we understand in Australian society as intimate partner or family violence against Indigenous women is 'inextricably linked to the state' (Smith andRoss 2004: 1 quoted in Allas et al. 2018).
McGlade also writes, 'Black women know the Australian state was built on such violence and that the instruments of law, the police and courts, can never be trusted to protect black women's bodies' (quoted in Allas et al. 2018: np).

Allas and colleagues (2018) observe that:
Indigenous women die outside the formal custody of the state, on the streets; on the open road; in their own home or at the edge of communities.In these spaces, although outside of the carceral confines, the violence of the settler state is enacted through diverse practices that render Indigenous women's lives unsafe and produce their deaths.
In these ways, it is recognised persistently that 'discrimination against Indigenous women and girls and its effects should be understood in both their individual and collective dimensions' (Committee on the Elimination of Discrimination against Women (CEDAW Committee) 2022: para 17), and that 'collective dimensions' are primary (Allas et al. 2018;Longbottom 2020;McQuire, Sisters Inside and Institute for Collaborative Race Research 2022;Jones et al. 2023).In a submission to the Australian Inquiry into Missing and Murdered Indigenous Women, McQuire and colleagues (2022: 3) write that there must be a 'mov [e] away from the standard focus on Indigenous women's "vulnerability" to make visible the structurally violent and deeply racialized relations between Indigenous people and state agencies in this place'.They observe (2022: 8) that the 'only way we can understand rates of victimisation of Indigenous women, girls and gender diverse people is by examining the structure of violence created by colonisation … and police and state agencies are on the front line of that violence'.Indigenous women do not say that the violence in the racial hierarchy that structures colonisation is a consideration that throws light on what 'actually happened' between an accused and a person killed in a domestic killing.Rather, they say that a domestic killing by an Aboriginal woman is impossible to understand at all unless an inquiry is made into how the racialised world both the accused and deceased lived in was experienced by the accused.This involves inquiry into the parties' experiences of institutional discrimination (such as discrimination in state policing, corrections or health services and policies), as well as racism by individual state actors over time and through generations.This has been documented by numerous Australian inquiries (Australian Law Reform Commission 2017;Queensland Government 2022: Pt 4).
Moreover, colonial ideologies and practices are gendered (Behrendt 1993;McGlade 2012).Aboriginal women bear the racist inequalities of colonisation for themselves.They also bear an accumulated burden from violence against Aboriginal men.In Australia, the national violence prevention agency, Our Watch, reported that a 'high proportion of Aboriginal men who commit violent crimes are themselves suffering the psychological symptoms of cumulative harm and intergenerational trauma as a result of numerous traumatic stressors … [including] institutional violence ' (2018: 53;McGlade 2012).This is also true of Aboriginal women who are incarcerated at increasingly high rates, and who extensively report experiencing interpersonal violence as women and also as children (Wilson et al. 2017;Wilson, Jones and Gilles 2014).The United Nations Special Rapporteur on violence against women, reporting on violence globally against Indigenous women and girls in 2022, recognised that Indigenous women and girls 'bear the gendered consequences of the violence against themselves and their communities disproportionately' (Alsalem 2022: para 8-9).Further, 'targeting victims in a gender-oriented manner destroys the very foundations of the group as a social unit and leaves long-lasting scars within a group's social fabric ' (Allas et al. 2018: np).
Australian research is consistent with international research about violence against Indigenous women and girls and, therefore, what is required of decision-makers.Under the United Nations (UN) Convention on the Elimination of all forms of Discrimination Against Women (CEDAW), determinations about gender-based violence against Indigenous women must be made by integrating a gender perspective and an Indigenous women and girls' perspective (CEDAW Committee 2022).
Thus, national and international research about violence against Indigenous women makes it clear that the starting points for comprehending interpersonal violence against Aboriginal women are the structural racism of colonisation and its institutional violence and discrimination.

The Severity of Abuse Against Indigenous Women
In addition to centralising Indigenous women's relationship with the state, research also shows that physical family violence against Indigenous women is more commonly severe and more likely to be life-threatening.In Australia: • Hospitalisation rates from family violence are 32 times higher for Indigenous women compared to other women in Australia, and three times higher than that of Aboriginal and Torres Strait Islander men (Australian Human Rights Commission 2022: 44).These statistics are likely to be underestimates because it is recognised that gender-based violence against Aboriginal and Torres Strait Islander women and girls is 'drastically underreported' (Australian National Research Organisation for Women's Safety (ANROWS) 2020; Alsalem 2022: para 8-9).• Aboriginal and Torres Strait Islander women are almost 11 times more likely to die due to assault (ANROWS 2020).
• In Western Australia, Aboriginal mothers are 17.5 times more likely to die from homicide than non-Aboriginal mothers (Uibu 2016).This rate of mortality of Aboriginal mothers 'in turn carries a range of grave implications for Aboriginal children's health and wellbeing' (Allas et al. 2018: np).• Intimate partner violence contributes 10.9% of the burden of disease for Indigenous women aged between 18 and 44, which is higher than any other health risk factor, including alcohol or tobacco use and obesity (Webster 2016: 4).
These statistics are well known and can be presumed to be within the knowledge of state prosecutors.They are relevant to all Aboriginal women who experience violence, whether they die by violence or defend themselves (Australian Domestic and Family Violence Death Review Network and ANROWS 2022: 34-36, 48-50). 3Aboriginal women also know personally the severity of violence and their risk of dying; they more often have family members who have been murdered.
Research into lethality risks based on forms or severity of violence by an intimate partner is an important, growing field.It is relevant to non-Indigenous and Indigenous women victimised by intimate partner violence.However, the increased risk of high levels of physical violence-including maiming and lethal violence-that an Indigenous woman accused of using lethal force may have lived with is not incidental to or accidentally different from the experience of a non-Indigenous woman.The risk is itself an indicium of colonisation (Atkinson 2001 [The] framing of Indigenous women as sexually useful yet dispensable provides an explanation for the lack of care or concern demonstrated in the public sphere when Indigenous women experience violent assaults and death.Our status as gendered subjects has always been differentiated in the public domain from that of non-Indigenous women to the degree that Indigenous women's bodies invoke a unique and brutal form of misogyny that incurs little interest or judgement in the broader public domain.(2022: 86-87) It follows that an Aboriginal woman's experiences of violence cannot be comprehended without ensuring that specific inquiry is made into the forms and severity of violence to which she was subjected.This will prompt inquiry into research about, for example, the significance and underreporting of brain injuries caused by intimate partner violence (King et al. 2023;Meyer, Jammula and Arnett 2022), the lethality risk in strangulation (Douglas and Fitzgerald 2022), and sexual violence.In turn, this will determine the relevance of facts that can otherwise be overlooked and ignored (McGlade and Tarrant 2021: 109).
Thus, there is a paucity of research about violence against Aboriginal women from Aboriginal women's perspectives (Bevan, Lloyd and McGlade 2024).However, available research shows that the racialised violence of colonisation-and, within that framework, the incidence of severe and life-threatening physical violence-are central and distinct experiences of intimate partner violence to which Aboriginal women are subjected.

The 'Social and Systemic Entrapment' Framework for Understanding Intimate Partner Violence
There is a growing general awareness in Australia, the United Kingdom, New Zealand, the United States and other jurisdictions, of intimate partner violence as 'coercive control'.This is the idea that intimate partner violence cannot be understood accurately as only physical assault but is better conceptualised as patterns of abusive behaviours over time, motivated by a desire to dominate, using fear (and usually physical violence) instrumentally to police another person's behaviour (usually a woman's Recently, Tolmie, Smith and Wilson have developed the framework to emphasise the structural and intersectional dimensions of intimate partner violence by reversing the order of the sets of inquiries and formulating the framework as follows.Inquiries are required into: 1.The operation of intersectional inequities and state-sanctioned violence in the victim-survivor's and their relational support network's lives (within an infrastructure of colonial violence in settler states).This includes understanding whether and how these exacerbate their abusive partner's ability to coercively control them and impact the safety responses that are available to them; 2. The realities of the family violence safety responses available to the victim-survivor -this includes the responses of the immediate community surrounding the victim-survivor and their abusive partner and of those agencies charged with assisting them.The question here is whether the victim-survivor, and those that matter to them, have access to equitable and dignifying systemic safety responses that are effective in addressing the coercive controlling behaviours of their abusive partner; and 3.The abusive partner's coercive and controlling behaviour and the impact of that behaviour in closing down the victimsurvivor's space for action.(Tolmie, Smith and Wilson 2024: 61-62)

A Prosecutor's Inquiry Into Violence Against Aboriginal Women to Make Determinations About Self-Defence
We argue that without conducting detailed inquiries consistent with the research discussed above, a prosecutor is not in a position (Rubio-Marin and Sandoval 2011) to make considered and lawful determinations.We centralise Aboriginal women's relationship with the state in understanding the interpersonal violence they experience and adopt the social and systemic entrapment approach as an organising framework.We also include specific inquiry into the severity of abuse to which an Aboriginal woman has been subjected.The social and systemic entrapment framework offers an approach designed to be useful for all women (Tolmie, Smith and Wilson 2024: 55).In the context of this article, we suggest a focussed inquiry into the forms and extremity of physical violence is relevant.As discussed, the normalisation of extreme physical violence against Aboriginal women is not incidental to or accidentally different from the experiences of non-Aboriginal women.Thus, we suggest the following levels of inquiry are required by a prosecutor before s/he can evaluate whether an Aboriginal woman acted to defend herself: • The role of state actors and state institutions in producing, facilitating and/or exacerbating the individualised abuse to which she was subjected; • The responses of those around her/her and her partner to the individualised abusive behaviour and to her help-seeking actions-including responses, or absence of response, by state institutions-and whether realistically available, effective safety options existed; • The severity of abuse to which she was subjected, including the presence of forms or extremity of abuse that indicate the risk of lethality for her as an Aboriginal woman; and • The ongoing patterns of individualised abuse to which she was subjected and the ways she managed and resisted the abuse.

Misunderstanding Violence Against Aboriginal Women and its Effects on the Law of Self-defence
An Aboriginal woman's claim of self-defence against ongoing and extreme intimate partner abuse is that: her partner would continue to attack and injure her in the future in the same extreme and increasingly serious ways he had done in the past, resulting in further grievous bodily harm or death, and that she had no reasonably available safety option that could have stopped his violence.She claims that her assessment that neither the criminal justice system nor other social services did, or would, stop his violence against her was within reason.This is the claim a prosecutor has the obligation to assess.We have argued this requires taking the deeply contextual approach described in the previous section.
We have written elsewhere (McGlade and Tarrant 2021; Tarrant, Tolmie and Giudice 2019) about the ways in which prosecutors' case theories at trial fail to account for the ongoing violence on which Indigenous and non-Indigenous women's claims of self-defence are based.This is not, primarily, due to the exclusion of evidence of past assaults at trial.It is largely because prosecutors' theories confine the form of violence on which a woman's claim of self-defence is assumed to be based to the moments before she responded with lethal force.Conceptualising the violence in this restricted, and inaccurate, way necessarily renders the violence understood through a social and systemic entrapment framework irrelevant.Further, prosecutors do not consider themselves obliged to answer a claim of self-defence based on that violence.By framing the violence as consisting of no more than that contained in the few moments before she acted, an Aboriginal woman's action is easily misconstrued as disproportionate (Douglas et al. 2020 Without a trial, where an Aboriginal woman pleads guilty, it is not possible to know what a prosecutor's case theory is.It is necessarily the case though that the prosecutor has determined for her/himself that the evidence is capable of proving that the woman did not act in self-defence.Beyond this, some indication can be gleaned from the sentencing remarks, as illustrated by the recent case of R v Walley (Supreme Court of South Australia NO.SCCRM-21-41, 2 December 2019).In this case, an Aboriginal woman, Monica Walley, was charged with the murder of her abusive husband and convicted of manslaughter after she pleaded guilty.It is clear from the sentencing remarks that Monica was subjected to extreme family violence over many years by her partner, Kunmanara. 4The prosecutor must have drawn a conclusion that Monica did not act in self-defence and that decision was never put before a court.In Australia, most women charged with a homicide offence after killing their abusive partner are convicted of manslaughter and most of those convictions follow a guilty plea (usually in exchange for a murder prosecution being discontinued) (Bradfield 2002; Dioso-Villa and Nash 2023; Domestic Violence Resource Centre Victoria 2016; Sheehy, Stubbs and Tolmie 2012).Aboriginal women are more likely to plead guilty than non-Aboriginal women (Dioso-Villa and Nash 2024; Dioso-Villa and Nash 2023; Sheehy, Stubbs and Tolmie 2012).Because trials are not held, comprehensive data is difficult to obtain.However, a study by Dioso-Villa and Nash in 2023 showed that between 2010 and 2020 no Aboriginal woman was acquitted of a homicide charge after killing her abusive partner, whereas 26.5% of non-Aboriginal women were acquitted.Monica's case, therefore, exemplifies the experience of Aboriginal women who have killed an abusive partner.
In R v Walley, the prosecutor evidently confined their case to the moments before and during the stabbing and constructed those moments as containing a mutual 'argument' about 'jealousy'.For example, the sentencing judge said, 'Kunmanara was accusing you of cheating on him and you were accusing Kunmanara of cheating on you' (Sentencing Remarks: 6).Being instantaneous and mutual, the facts conceived of as relevant by the prosecutor contained no relevant threat to Monica, rendering her claim that she was defending herself nonsensical.And Monica's 'misconceptions' about how dangerous Kunmanara really was to her were attributed to alcohol: 'You were both drunk' (Sentencing Remarks: 6).It is important to recognise that, in limiting its case in this way, the prosecution rendered the uncontested ongoing violence to which Monica was subjected, entirely irrelevant to the legal question of Monica's guilt.Put another way-and to stress-the violence she was forced to continue to endure played no part in her criminal conviction.

The Nature of the Prosecutor's Error when the Violence Against an Aboriginal Woman is Inaccurately Understood
By failing to take an approach that would provide an accurate understanding of the violence an Aboriginal woman faced (and therefore her claim of self-defence), a prosecutor is not in a position to evaluate the sufficiency of evidence to prove the offence with which she is charged.
In this section, we set out the law that governs prosecutors' decisions to prosecute.We present the fundamental error of law relevant to the threshold legal determination a prosecutor must make: whether there is a prima facie case on which a prosecution can be commenced.This determination must be made prior to any question about whether a prosecutor should refuse or accept a guilty plea.
From outside the legal field, the impression of prosecutors' decisions is often that they are discretionary.It is believed that, once police charges have been laid, 'prosecutorial discretion' is exercised to decide who is prosecuted and whether a prosecution should be continued to trial.Indeed, within the legal field, how prosecutors' discretion should be exercised is largely the focus of discussion.The concern is that discretionary power 'involves a considerable latitude of individual choice of a conclusion' (Russo v Russo (1953) VLR 57: 62 quoted in Jago v District Court of New South Wales (1989) 87 ALR 577: 615) and that, therefore, there is 'scope for subjectivity and hence for arbitrariness' (Binham quoted in Cowdery 2020: xxv). 5This reflects the primary purpose in creating the prosecution guidelines that operate in all Australian jurisdictions: to remove prosecutorial decision-making from political influence while setting a framework to ensure impartiality and professionalism (Hanlon 2008;Plater and Royan 2012;Rozenes 1996).Discussing the role of the public prosecutor, the former New South Wales Director of Public Prosecutions, Nicholas Cowdery, writes: In common law and some other jurisdictions, prosecutorial discretion is exercised in accordance with the 'principle of opportunity' (prosecuting a crime only if it is opportune to do so), independently of any inappropriate influence by government, politics, the media, the police or any individual or sectional interest in the community, including victims of crime.(2020: 54) This is the authority the prosecutor exercised in R v Walley in eventually accepting Monica's guilty plea to the lesser offence of manslaughter in satisfaction of the charge of murder.
However, the problem we are concerned with is more fundamental than the exercise of prosecutorial discretion.It is worth outlining the legal framework for a prosecutor's decision-making authority to show that by the time a prosecutor enters a plea negotiation, the failure to provide justice to many Aboriginal women in Monica's circumstances is already embedded in the negotiations.

A Prosecutor's Authority to Commence and Continue a Prosecution
In Australia, a public prosecutor's authority to prosecute is conferred by statute.Prosecutors are governed by statutes that create criminal offences 6  According to this law, there are two fundamentally different determinations a prosecutor must make in deciding to commence a prosecution.The first is whether, within the brief of materials provided by investigating police, there is evidence capable of proving beyond reasonable doubt that the charged person committed the offence they are charged with (or a different offence).
If there is sufficient evidence, the second question is whether the prosecution should be commenced in the public interest.The first question itself contains two processes: a determination of whether material contained in the investigator's brief can prove the offence and a further determination of whether there is a reasonable prospect of conviction.This distinction is between, on the one hand, available material in any form and, on the other hand, material admissible as evidence in a trial which would be sufficiently reliable and strong.For example, material may show a person's guilt but be inadmissible as evidence because it was unlawfully obtained, or witnesses may be unavailable to testify or lack credibility.In some jurisdictions these two legal decision-making processes (contained within the first fundamental determination a prosecutor must make) are described as requirements to determine whether there is a prima facie case and then whether there is a reasonable prospect of conviction. 8 In other jurisdictions it is seen as one requirement that involves both processes: the determination that there is reasonable prospect of conviction. 9Whichever way it is described, this first, fundamental determination in the prosecution process is a determination of law and evidence which, although it involves legal judgement, is not a discretionary authority and does not concern the 'public interest'.There is no discretionary prosecutorial authority to prosecute (or not prosecute) in the public interest, in the absence of a prima facie case.
Thus, prosecutors must make a legal determination on the available evidence (whether there is sufficient evidence) and then, if it is determined there is sufficient evidence, a discretionary determination as to whether it is in the public interest to commence a prosecution.This is the same discretionary authority that is exercised when a prosecutor accepts a guilty plea.The Prosecution Guidelines in each jurisdiction set out the factors that must be taken into account when determining what is in the public interest.
If a prosecutor has failed to engage with an Aboriginal woman's claim of self-defence because they failed to understand the violence by which she was entrapped, this is a problem with the first of the determinations a prosecutor makes: sufficiency of evidence.It is not a question about whether the prosecutor should in her/his discretion, proceed with or discontinue a prosecution, or accept a guilty plea in the public interest.The importance of this distinction is in understanding that when plea 'negotiations' begin, the error in there being no prima facie case is already embedded in those deliberations. 10  Thus, failure to understand the violence an Aboriginal woman faced is compounded by the assumption it produces, that the prosecutor has discretionary authority to accept or refuse a guilty plea.
Two further points need to be made about this erroneous exercise of the powers of the prosecutor.First, they do not operate in a vacuum.Plea negotiations rely on an Aboriginal woman seeking relief from the enormous pressures of proceeding to trial.She avoids a conviction of murder, carrying greater punishment (including mandatory minimum sentences in some Australian jurisdictions) and stigma.She attracts the chance of a mitigation discount for a guilty plea (and in some jurisdictions the chance of avoiding the mandatory minimum sentence for manslaughter).But in addition to these legal implications, she avoids the stress and uncertainty of a trial for herself, her family and community.And, crucially for many Aboriginal women, she can be reunited with her children sooner.This is especially so where separation is worsened because her children are too far away to visit, or are denied any contact with their mother if they could be witnesses at a trial (Wilson, Jones and Gilles 2014: 31).
The second point to make about the erroneous exercise of power by prosecutors is that the ongoing violence against Aboriginal women in these circumstances, which forms the basis of their lawful claim of self-defence, may be transformed into the basis for a prosecutor's acceptance of a guilty plea in the 'public interest'.The factors that determine how a prosecution decision is made in each case are not published or accessible.However, the public interest factors listed in Prosecution Guidelines and sentencing remarks following a conviction will reflect some of the prosecutor's reasoning.Taking R v Walley (2019) again as an example, the 'length and expense of trial' and the likely sentencing outcome of a conviction after trial were probably factors that influenced the prosecutor's exercise of the public interest discretion (Statement of Prosecution Policy and Guidelines 2014 (SA): 6-7).The fact that 'the offence was committed in circumstances of family violence' and Monica's preparedness to plead guilty to manslaughter probably formed the basis of the prosecutor's decision to accept her plea of guilty to the lesser offence.This is reflected in the prosecutor's 'concession' in the sentencing hearing that it was open to the sentencing judge to find that 'exceptional circumstances' (circumstances of family violence and a guilty plea) existed under the Sentencing Act 2017 (SA) (Sentencing Remarks: 8).This was the basis for exempting Monica from the mandatory minimum sentence for manslaughter (Sentencing Act 2017 (SA), s. had defended herself against the life-threatening family violence she had been subjected to over many years) became merely a basis for a charitable reduction in punishment (Tarrant 2018).This is in spite of the fact that neither family violence nor guilty pleas are exceptional for Aboriginal women in these circumstances.
Viewed in this way, the depth of the problem in the inaccurate understandings of Aboriginal women's experiences of ongoing family violence is revealed.A prosecutor's failure to address an Aboriginal woman's claim of self-defence by constructing a case theory that misses the violence on which the claim is based, is contrary to law.Yet, that failure sets the course of the prosecution through the remainder of the process (including plea negotiations and a trial if that occurs). 11Not only is the ongoing violence rendered irrelevant, as the violence she was responding to, but it is used in her conviction (the prosecutor's acceptance of Monica's guilty plea).This also reveals the oppressive bind Aboriginal women in these circumstances are placed in, being required to decide between a trial and a plea of guilty to manslaughter.If her claim to have defended herself does not sound in the prosecutor's process, what chance does she have of it sounding in a trial?(Sheehy et al. 2024)

Legal Responses to Prosecutors' Errors in Failing to Engage with Self-defence
We have argued that prosecutors are making systematic legal errors; failing to understand an Aboriginal woman's experience of violence leads to distorted evaluations of fact and a failure to apply the law of self-defence.In this section, we look at how such errors may be challenged: an application for a stay of proceedings for abuse of process, and a 'no case' submission.

Non-justiciability of Prosecutors' Decisions
Prosecutors' decisions are protected from review under Australian law (Barton v R (1980) 147 CLR 75).In Maxwell v R [1996] HCA 46, the High Court of Australia considered the question whether a trial judge had the power to reject a guilty plea after the plea had been accepted by the Crown in full satisfaction of an indictment.The High Court held that a trial judge has no power to review or reject a guilty plea, even in the absence of evidence that supports proof of the offence, or where an 'artificially restricted view of the facts' is required to make the plea coherent: The decision whether to charge a lesser offence, or to accept a plea of guilty to a lesser offence than that charged, is for the prosecution and does not require the approval of the court.…The role of the prosecution in this respect … 'is such that it cannot be shared with the trial judge without placing in jeopardy the essential independence of that office in the adversary system' (R v Apostilides (1984) 154 CLR 563: 575 quoted in Maxwell v R [1996] HCA 46: para 25). 12  In Maxwell v R (1996) (para 26), the High Court adopted the New South Wales Court of Appeal's observation in R v Brown (1989) 17 NSWLR 472, that 'the most important sanctions governing the proper performance of a prosecuting authority's functions are likely to be political rather than legal'.

Application for a Stay of Proceedings for Abuse of Process
Although a prosecutor's decision cannot be reviewed, a court does have the authority to consider whether a prosecutor has engaged in an abuse of process.Courts make a distinction between reviewing prosecution decisions and protecting the court's own procedural integrity (Jago v District Court of New South Wales (1989) 87 ALR 577;Maxwell v R [1996] HCA 46).In Barton v The Queen, the High Court of Australia held that: It is one thing to say that the filing of an ex officio indictment is not examinable by the courts; it is quite another thing to say the courts are powerless to prevent an abuse of process or the prosecution of a criminal proceeding in a manner which will result in a trial which is unfair when judged by reference to accepted standards of justice.The courts exercise no control over the … decision to commence criminal proceedings, but once [that decision is made], the courts will control those proceedings so as to ensure that the accused receives a fair trial.…There is ample authority for the proposition that the courts possess all the necessary powers to prevent an abuse of process and to ensure a fair trial.((1980) 147 CLR 75: 95-96) Thus, a court can determine the course of a prosecution to protect its own processes.
If an abuse of process is established, a court may grant a stay of proceedings, meaning a prosecution is suspended until the error in the prosecution process is rectified.There is no power in the court to require the prosecutor to re-make their decision to commence or continue a prosecution, only to prevent an abuse of process by staying an indictment (R v Brown (1989) 17 NSWLR 472).Moreover, it has been held that an abuse of process will only be found in 'extreme' circumstances and courts have taken a personalised construction of what is required, focussing on whether an individual prosecutor has acted irresponsibly.In Maxwell v R, it was held that the 'court's power to prevent an abuse of its process … could only arise in this context if the prosecuting authority were seen to be acting in an irresponsible manner', and that, in the Court's view, 'is seldom, if ever, likely to occur' (Maxwell v R [1996] HCA 46: para 26;Jago v District Court of New South Wales (1989) 87 ALR 577: 616).
The argument made here is not that an individual prosecutor has acted aberrantly.It is argued that the offices of Directors of Prosecutions in Australia are making decisions systematically that disregard research about violence against Aboriginal women and that this denies a lawful process for Aboriginal women who respond to such violence.Arguably, this is an extreme circumstance that would warrant a finding that the prosecution process is being abused in this class of case, especially given growing public knowledge of the atrocity that is the femicide of Aboriginal women. 13Were an application to be successful, the remedy an Aboriginal woman would receive is a stay of proceedings only; the charge and possible future prosecution would remain over her head.However, the benefit of such an application would include the opportunity to put the argument before a court, and before the prosecutor, that the prosecutor is acting systematically in a way that is contrary to law.

A 'No Case' Submission (A 'Sufficiency of Evidence' Argument)
The argument made above that prosecutors are achieving convictions without evidence to prove an Aboriginal woman did not act in self-defence against ongoing family violence, is essentially an argument that she has no case to answer.A 'no case' submission is an application to a court for a ruling that the prosecutor has adduced insufficient evidence to prove one or more elements of the offence (Doney v The Queen [1990] HCA 51).In this context, it is an argument that there is insufficient evidence to disprove self-defence, which is required of the state in any homicide prosecution.A no case submission is determined as a question of law (May v O'Sullivan (1955) 92 CLR 654), on all evidence most favourable to the prosecution case.The question asked is, if all that evidence were accepted by a jury, is the offence proved beyond reasonable doubt?If a 'no case' submission is successful, 'it is the judge's duty to direct an acquittal for it is only upon evidence that juries are entitled to convict' (Haw Tua Tau v Public Prosecutor [1982] AC 136: 151;Thomson 1997).A 'no case' submission cannot be made until a trial begins, though it can be made at the start of, or during, the trial.Importantly, however, a no case submission can also be a basis for communications with prosecutors in the pre-trial phase, any time from the point a woman is charged.A 'no case' argument could have been made in the plea negotiations for Monica.
We are not aware of any no case submissions on self-defence that have been successful where a woman has been charged with the homicide of her partner.However, in the remainder of this section, we summarise five cases in which insufficiency of evidence arguments were made in various ways to challenge prosecution decisions.The five cases resulted in acquittal (or in one case, the pardon from sentence after conviction) of a woman charged with killing (or in one case, doing grievous bodily harm to) her abusive partner or ex-partner.None of the acquittals were of Aboriginal women; Jody Gore, a Gidga Jaru woman from Western Australia, had her sentence commuted (The State of Western Australia v Gore [2015] WASC 327).
The arguments in these cases were made at various stages of the process: at the beginning of the trial, at the end of the trial, on appeal or after conviction through public advocacy.In all the cases, arguments were made essentially about the insufficiency of evidence to disprove self-defence.In two cases, formal 'no case' submissions were made.All the cases can be seen as positive progress in relation to how family violence was eventually seen.However, the cases also demonstrate the depth of resistance in the criminal justice system to conceptualising a woman's responses to this form of violence as lawful self-defence (even where she is acquitted or pardoned).Defence counsel made 'no case' submissions in only two of the cases.Neither of the acquittals in these cases was on grounds of self-defence.A woman was acquitted on grounds of self-defence and in only one, other, case: In Western Australia v Gore [2015] WASC 327, Jody Gore was convicted of the murder of her abusive ex-partner.She was sentenced to life imprisonment with a 12-year minimum.As in Monica's case, discussed above, the prosecutor asserted Jody was in no danger at all, that she merely killed in a drunken moment, disregarding Jody's claim that she acted in self-defence against ongoing violence (Douglas et al. 2020) In Silva v R [2016] NSWCCA 284, Jessica Silva was charged with the murder of her abusive partner and convicted by a jury of manslaughter on grounds of 'excessive self-defence'.That is, she was found to have acted subjectively in self-defence, but what she did was unreasonable and disproportionate. 14She appealed her conviction and the New South Wales Court of Appeal held that the jury's verdict was not supported by the evidence; there had been insufficient evidence before the court justifying the jury's conclusion that Jessica had not acted reasonably and proportionately in defence (Tarrant 2023).Jessica was acquitted.
In R v Stephen, 15 Jonda Stephen was charged with the murder of her abusive partner.She reached for a kitchen knife that was at hand while he was attacking her, striking her in the head with a domestic iron.Defence counsel made two 'no case' submissions, first, that the prosecutor had produced insufficient evidence to disprove Jonda acted in self-defence and, later in the trial, that the prosecutor had produced insufficient evidence to prove the intention required for murder.Both these applications were rejected by the trial judge.However, both were part of a defence approach that successfully demonstrated that the prosecutor could not prove all the elements of an 'excessive force' defence (that being the only legal issue remaining at the conclusion of proceedings).The trial was discontinued, and Jonda was acquitted by direction to avoid, in the trial judge's words, the trial descending 'to the level of solemn farce' (R v Stephen (No. 6) [2018] NSWSC 243: para 7; Tarrant 2023).
In the Canadian case of R v Barrett [2019] SKCA 6, a trial judge found the defendant had no case to answer on a charge of causing grievous bodily harm to her abusive ex-partner.The prosecutor had produced insufficient evidence to prove the defendant had not acted reasonably to defend her children from her ex-partner when she feared for their safety and entered his house to retrieve them.The prosecutor argued the conduct was unreasonable because the accused could have sought help from the authorities rather than use force.In acquitting the defendant, the trial judge reasoned that a: practical problem with the Crown's position is that the Crown has provided no evidence … to conclude that going to the proper authorities (whether that be the police, Social Services or others) would reasonably protect the children in the circumstances of the case.(R v Barrett [2019] SKCA 6: para 29) If the Crown 'intended to assert the respondent['s] actions were unreasonable because the matter could have been referred to the authorities, the onus rested on it, not the respondent [(R v Barrett [2019] SKCA 6: para 30).This reasoning and acquittal were upheld on appeal.R v Barrett makes it clear that a mere assertion that a woman could have done other than use force against ongoing family violence does not amount to evidence.
In Western Australia v Bridgewater [2019] WASC 10, Tracey Bridgewater was charged with the murder of her abusive partner after he attacked her and her parents.The first trial ended with a hung jury.The second trial was discontinued, and Tracey was acquitted by direction.Before the end of the trial, and after the trial judge raised the question of sufficiency of evidence, the prosecutor reviewed the evidence.They determined that there was insufficient evidence to prove that Tracey had not acted to defend her home under Section 244 of the Criminal Code of Western Australia.She therefore had no case to answer (The State of Western Australia v Bridgewater [2019] WASC 10, trial transcript: 1062).

Conclusion
How prosecutors understand the violence to which an Aboriginal woman was subjected permeates all aspects of the prosecution process.This includes their evaluation of investigators' briefs, and discussions with investigators and with the Aboriginal woman and her counsel.Where a prosecutor fails to conceptualise the violence in a way that allows an accurate understanding, legal determinations about self-defence that follow from those evaluations of facts are distorted.This leads to wrong conclusions of law.Research about violence against Indigenous women from the perspective of Indigenous women and a social and systemic entrapment approach requires inquiry into: the role of state actors and institutions in the individualised abuse to which an Aboriginal woman was subjected; the responses of those around the Aboriginal woman to the abuse, including responses of state institutions; the forms and severity of the violence used against her as an Aboriginal woman; and the ongoing patterns of abuse she experienced and her management of the abuse.This approach, we suggest, provides the grounding for an accurate evaluation of an Aboriginal woman's claim of self-defence.
Where a prosecutor fails to understand the violence an Aboriginal woman faced, s/he decides to prosecute in the absence of evidence capable of disproving self-defence.This is a fundamental error of law.We have argued that although a prosecutor's decision to commence or continue a prosecution cannot be reviewed by a court, that decision may be challenged by an application for a stay of proceedings for abuse of process or a 'no case' submission.
Correspondence: Stella Tarrant, Associate Professor, Faculty of Law, The University of Western Australia, Australia.stella.tarrant@uwa.edu.au.
48(3)(ab)-(b)) -despite the family violence being the foundation of her self-defence claim.Thus, consistent with research (Stubbs and Tolmie 2005; Tyson, Kirkwood and Mckenzie 2017), her claim of self-defence (that she (Bevan, Lloyd and McGlade 2024earch, advocacy and activism about gendered power and control and violence against women was operationalised by the clinical and forensic work of sociologist, Evan Stark.The concept of 'coercive control' was popularised by Stark's Coercive Control: How men entrap women in personal life(2007).However, although 'coercive control' accurately identifies the ongoing nature of this form of harm, and that it is gender-based violence, it is largely limited to the interpersonal relationship(Tolmie et al. 2024.Indigenous researchers and family violence workers do not consider this approach adequate for understanding Aboriginal women's experiences of family violence(Bevan, Lloyd and McGlade 2024; Commissioner for Victims of Crime 2023).Coercive control tends to assume a socially neutral context beyond the individuals directly involved.
(Tolmie, Smith and Wilson 2024ment framework for understanding intimate partner violence has been developed by New Zealand scholars, Julia Tolmie and colleagues, based on a framework originally proposed by James Ptacek(Tolmie et al. 2018).This approach operationalises intersectional theory, which requires consideration of different intersecting dimensions of inequality and disadvantage that affect women's lives, in considering women's experiences of violence.The framework requires inquiry into three dimensions of the facts about intimate partner violence: 1. individual abusive behaviours over time, 2. community and institutional responses to the individual abuse and 3. structural dimensions of inequality and discrimination relevant to the first two dimensions (New Zealand Family Violence Death Review Committee 2016: 39;Tolmie et al. 2018:  185).This approach goes beyond an understanding of intimate partner violence as coercive control(Tolmie, Smith and Wilson 2024:  55).It requires a wider inquiry, at the second level, about the social and institutional responses to those behaviours and conduct.Importantly, this approach assumes the 'violence' in intimate partner violence can only be comprehended in any meaningful way by considering both the individualised coercive and controlling behaviours and family, community and institutional responses to individual behaviours.These actors may, either explicitly or by default, have facilitated the abusive conduct or failed to provide realistic safety options for a victim-survivor(Tolmie et al. 2018).This aspect of a social entrapment approach reflects well-established research which shows that cultural and systems responses play vital roles in the decision-making of victim-survivors and perpetrators and determine the course of the violence (Commissioner for Victims of Crime 2023; Douglas 2018).It also places explicit focus on responses, working against an individualist assumption that a woman could always have acted otherwise to keep herself safe(Tolmie, Smith and Wilson 2024) -an assumption that 'disappears' the state (McQuire, Sisters Inside and Institute for Collaborative Race Research 2022) as a party to the entrapment.
;Tarrant, Tolmie and Giudice 2019).Moreover, her claim that she knew that she was in serious danger (the subjective requirement in self-defence) is easily undermined by evidence that she was drunk or affected by drugs in those moments (Gore v The State of Western Australia [2017] WASCA 163; McGlade and Tarrant 2021; R v Walley Supreme Court of South Australia NO.SCCRM-21-41, 2 December 2019).
Statement of Prosecution Policy and Guidelines 2014 (SA), Prosecution Guidelines 2021 (NSW) and Policy of the Director of Public Prosecutions for Victoria 2023.The Prosecution Guidelines contain the most detailed statements of prosecutors' powers and responsibilities in the prosecution process.
and statutes that create and regulate the Offices of Directors of Public Prosecutions (ODPP) in each jurisdiction (DPP statutes).7Publiclyavailable prosecutorial guidelines have been written for each jurisdiction under the authority of the DPP statutes (Prosecution Guidelines); for example, the Statement of Prosecution Policy and Guidelines 2022 (WA), (McGlade and Tarrant 2021)re than four years of her sentence, she was released pursuant to a royal prerogative of mercy by the Western Australian Attorney General(McGlade 2019).This was following a concerted campaign led by Hannah McGlade involving intensive media exposure and the commencement of work on a legal appeal.Her conviction still stands.The campaign demonstrated the connection between Jody's act of killing her ex-partner and the extreme family violence to which she had been subjected, and the Attorney General introduced family violence amendments to the Evidence Act 1906 (WA) in response.Yet, this connection was not made during the pre-trial prosecution phase or at her trial(McGlade and Tarrant 2021).