Melbourne University Law Review
JULIE STUBBS[*] AND JULIA TOLMIE[†]
[This article examines the Australian use of expert testimony concerning battered woman syndrome (‘BWS’) with particular reference to self-defence as compared with Canadian and United States experience. The authors examine recent developments in North America which move beyond narrow constructions of BWS towards recognition of the need for broader ‘social framework evidence’ to assist juries and the judiciary to better assess the reasonableness of a defendant’s claim to self-defence. They argue that by comparison, although Australian courts have been receptive to evidence concerning BWS, the nature and use of evidence concerning battering and its effects have been narrowly construed. Notwithstanding the recent decision in Osland v The Queen, the authors conclude that Australia still lacks a leading decision which demonstrates clearly why evidence concerning battering and its effects might have value in self-defence cases or other contexts.]
In Australia, Canada and the United States the concept of ‘battered woman syndrome’ (‘BWS’) has been used in the courts as a mechanism through which to introduce evidence about domestic violence and its effects, principally in cases where a woman is charged with killing her abusive partner and seeks to raise the defence of self-defence. However, BWS has been interpreted and applied differently in the three countries. The purpose of this article is to assess recent Australian experience concerning the use of evidence about BWS. In undertaking such an assessment, comparisons are made with developments in Canada and parts of the United States to argue that, although there are indications of change, in some respects the Australian approach has been particularly narrow.
Part II of this article looks generally at the concept of BWS as it has evolved in Australia and in North America. We suggest that Australian courts have interpreted BWS evidence narrowly, and that this may be attributable, at least in part, to a lack of willingness by the Australian courts to acknowledge that there is a need for such evidence because of the historical failings of the criminal law. We argue that Australian practice should follow that in North America in moving beyond the constraints of BWS evidence to introduce ‘social framework evidence’ about battering and its effects. We also document the beginnings of a possible move in the Australian case law towards a broader approach.
Part III of this article catalogues a range of pressing issues which have arisen in the literature concerning BWS and, with a specific focus on self-defence, examines the extent to which recent Australian cases have engaged with those issues. We look at recent case law which has dealt with pre-emptive strikes, previous defensive force, self-defence argued in the absence of expert evidence, pleas for clemency, the issue of violence in same-sex relationships, and the manner in which ethnicity or race may shape social context.
In this section we review the development of BWS in America, Canada and Australia. Activism and legal development over several decades have resulted in a complex and, at times, contradictory picture. Nonetheless, there are several important differences between the North American and the Australian experience. These differences are located around three interrelated issues.
The first issue is what the purpose of introducing such evidence is understood to be.
In the US and Canada, the use of BWS evidence was developed in recognition of the fact that, in the absence of testimony concerning BWS, a battered woman’s experiences might appear to a jury to be incompatible with a claim to self-defence. North American jurisprudence has tended to acknowledge that it was necessary to introduce BWS evidence because the legal doctrine of self-defence historically was gendered in its application. By contrast, the Australian decisions do not start by recognising limitations in the manner in which the law is applied. In particular, they do not recognise the manner in which existing legal defences have been interpreted to exclude the experiences of many women. Instead Australian decisions tend to proceed on the basis that BWS evidence is useful in explaining the allegedly different perceptions of a woman who has been battered and has developed the symptoms of the syndrome. This, in turn, may assist in applying self-defence or another defence to the facts at hand, or may mitigate sentence.
The second point of divergence between the North American and Australian positions flows from the first. This is the question of the nature and range of evidence which might be useful for the purpose identified above.
Recent US and Canadian legal developments have recognised that, where relevant, what is required to assist judges and juries is not so much expert testimony about whether or not a woman might have BWS, but rather broad ‘social framework evidence’ to provide the context within which to understand the issues in a given case. Maguigan expresses the purpose of ‘social framework evidence’ as to provide
the factfinder, usually a jury, with information about the social and psychological context in which contested adjudicative facts occurred. It is presumed that knowledge about the context will help the factfinder interpret the contested adjudicative facts.
Clearly, if one conceives the need for BWS evidence as arising from historical failings in the application of traditional criminal doctrine, for example self-defence, to the kinds of life experiences likely to be faced by women, then expert evidence will be introduced for the purpose of ensuring that the law is accurately applied in those kinds of life circumstances. Such evidence is not so much directed at the questions ‘was the accused a battered woman?’ and ‘did she suffer from learned helplessness?’ but rather, ‘what was the nature of the threat she faced?’ and ‘what lawful protection did she realistically have available to her?’ In contrast to this approach, our reading of many of the Australian decisions suggests that the main use of such testimony has, in fact, been to demonstrate to the court that the defendant had BWS. However, there are some indications of movement away from this narrow approach. These are discussed below.
The third issue around which there has been a divergence between the North American and the Australian developments concerns the question of who delivers the expert testimony in court. The narrow approach arguably taken in Australia to the two preceding issues might explain the preponderance of psychiatrists and psychologists among those who have been accepted as appropriate experts in Australia.
The American jurisprudence concerning BWS evidence has engaged explicitly with the incapacity of traditional understandings of self-defence to accommodate the experiences of women who kill an abuser. Schneider has detailed the feminist strategies which underpinned the development of BWS within the US and has analysed the outcomes of more than two decades of BWS litigation in that country. Schneider argues that the feminist self-defence work on behalf of battered women developed within an equal rights framework which sought to equalise women’s rights to a fair trial within the traditional criminal law framework and not, as some have mistakenly portrayed this work, to invoke special defences or special standards for women. Evidence concerning BWS was offered to provide information and to challenge misconceptions and stereotypes about battering and its effects in order that women’s claims to self-defence might be assessed fairly by the courts.
It has been argued convincingly by some US commentators that it is not so much the law of self-defence that has been inadequate to accommodate the defence claims of battered women but rather misapplications of that law. Such misapplications have included judges keeping evidence of the social context of battering from the jury or failing to instruct the jury adequately on self-defence. One of the criteria which has been found to be crucial in the decision on whether self-defence should go to the jury is the judge’s assessment of the accused’s credibility. This in turn may rely on the judge’s own understandings and conceptions of domestic violence.
Some activists have sought statutory recognition of BWS as a mechanism by which to redress deficiencies in the interpretation of self-defence law. However, there is ongoing debate in the US about the value of statutory recognition of BWS. Twelve US states have statutes recognising BWS evidence and its relevance to self-defence (and in a minority of cases also to other defences). For instance, in South Carolina expert evidence of Battered Spouse Syndrome is permitted to support the defences of self-defence, duress, necessity and defence of another. In addition, lay testimony is permitted concerning a batterer’s prior violent acts as a foundation for the expert testimony. South Carolina has thus codified judicial recognition of the fact that traditional understandings of self-defence may seem to be at odds with the experiences of battered women. Prior case law in South Carolina had relaxed requirements concerning imminence in the defence of self-defence, recognised that an assault might be ongoing even if at the time the woman took defensive action the abuser was not actually engaged in physical abuse (or was asleep), and required that the ‘self-defender’s’ actions be measured according to a standard which acknowledges the circumstances in which the battered woman finds herself.
On the face of it, statutory recognition of BWS may be seen as progressive. However, such statutes are problematic for several reasons, including the implication that issues concerning battering cannot be addressed within the criminal law generally but instead require special measures. The statutes may also introduce new hurdles or tests which must be met for the introduction of expert evidence. One such limit is the requirement in statutes of five states that a case for self-defence must be made out before BWS evidence might be admitted. Such a requirement can be interpreted to exclude the possibility of using BWS to support other defences, or to exclude the use of BWS in other contexts, such as in support of the prosecution of a battering man or in a non-criminal matter. The statutory recognition of BWS may also limit the range of testimony to be admitted, and may hinder the introduction of testimony concerning the effects of battering which arises from research or knowledge developed beyond the context of BWS.
The most important recent development in the US, however, is the widespread acceptance that BWS has been a useful construct as a focus for women’s self-defence work but that it is now outmoded. A major review of BWS evidence and its use in criminal trials, undertaken at the direction of Congress under the federal Violence Against Women Act of 1994, recommended that the term ‘evidence or expert testimony “on battering and its effects”’ should be used instead of BWS to better reflect the state of knowledge and practice. The review, which reported to Congress in 1996, included three components: firstly, an evaluation of medical and psychological testimony on the validity of BWS as a psychological condition; secondly, a study of trends in appeal court cases in which evidence of BWS was offered in criminal trials; and thirdly, an assessment by judges, prosecutors, and defence attorneys of the effects that evidence of BWS may have in criminal trials. The report found that ‘an extensive body of scientific and clinical knowledge strongly supports the validity and relevance of battering as a factor in the reactions and behaviour of victims of domestic violence.’ It therefore affirmed the utility within the trial process, and at various stages of the criminal justice process, of evidence concerning the effects of battering. However, it concluded that the term ‘battered woman syndrome’ is too narrow and has some negative implications. These included the fact that the term no longer reflects the breadth of empirical knowledge now available concerning battering and its effects; the fact that it implies that a single effect or set of effects characterises the responses of all battered women, a position unsupported by the research findings or clinical experience; and the fact that the word ‘syndrome’ may be misleading, either because it carries connotations of pathology or disease or because it creates a false perception that the battered woman ‘suffers from’ a mental defect.
This review has significant implications. By concluding that there is a valid body of knowledge about the effects of battering which is more extensive than that conventionally understood by reference to the term ‘BWS’, and by affirming that this knowledge can be valuable at all stages of the criminal justice system, the review legitimates the use of a broader range of information, so-called ‘social framework evidence’, in criminal cases. The review also gave credence to the concerns of many feminist scholars that BWS evidence risked establishing a new stereotype of ‘the battered woman’ and was being misunderstood to signify a mental dysfunction. However, the move towards a greater focus on the social context of abuse should not be understood to exclude the need for testimony on the psychological sequelae of abuse. As Maguigan argues, such information may continue to be important in some cases in explaining a defendant’s perception of danger. It is difficult to predict what effect, if any, the review might have on subsequent cases within the US but it is likely that the review reflects existing practice at least in many parts of that country.
While the shift to a focus on ‘social framework evidence’ may seem novel, some commentators in the US point out that historically such evidence has always been understood to be significant in dealing with self-defence claims. Cases dealing with bar room brawls, for instance, would seek to understand the self-defence claims of a defendant with reference to what was known about the history of the dispute and the relationship between the parties. It is not the recourse to ‘social framework evidence’ per se which may be novel but rather the recognition that, due to misinformation about battering and its effects and the gendered nature of legal reasoning, courts may need the assistance of expert testimony in order to recognise and interpret the social framework of cases involving women who have been battered.
The conclusions expressed in the review are also consistent with a broader understanding of the nature of the expertise which might be helpful to the court. According to Dutton, one of the leading US scholars on this issue and an author of the review:
[T]he primary purpose of expert testimony is to provide the jury and the judge with both an understanding of general principles of domestic violence and a framework within which to analyze the unique facts of the particular case being heard before the courts.
In order for an expert to do this in any given case, Dutton argues that they must be
informed by four key components supported by the facts in the case:
Dutton suggests that the role of an expert witness will differ depending on the case at hand. The range of issues that an expert might be asked to address is diverse. She offers the following examples:
Will the jury understand that a police officer can also be a battered woman?;
Will a jury hold the view that a woman who is homeless, a prostitute or has a drug and alcohol problem somehow deserves the violence?;
Can the jury understand why a battered woman fought back on this occasion but not on previous occasions?;
Can the jury recognise that a woman who has routinely fought back was nevertheless a battered woman and that the actions against her were illegal irrespective of her responses to them?;
Will a jury recognise the role of immigration status, children, or a debilitating illness in keeping a battered woman with her husband?;
Will the jury assume that a highly educated woman who is battered can simply leave the abusive relationship because she has the economic means to do so?;
Will the jury understand that the risk of serious or lethal violence often increases if she leaves the batterer?; and
Can the jury recognise that physical and mental health effects of domestic violence combined with the effects of culture influence battered women’s responses to abuse?
As the role of the expert differs, so too will the criteria for choosing an appropriate expert. Dutton argues that, depending on what the defence counsel in a given case hopes to achieve by introducing expert testimony, any of the following qualifications may be appropriate: an extensive history of working with battered women (for example, in women’s shelters or advocacy programs), nursing or other health care qualifications or experience, mental health clinical training, or experience in domestic violence research. She is particularly concerned that ‘the expert should understand the phenomenology of battered women’s experience through direct contact with battered women — rather than through academic endeavours alone’. Clearly, this represents a broad understanding of who can qualify as an expert for the purposes of testifying in these types of cases.
Canadian jurisprudence has developed with explicit recognition of the gendered nature of legal doctrine. The first Canadian decision to admit BWS evidence, R v Lavallee, resulted in a powerfully worded decision written by Madam Justice Bertha Wilson which, inter alia, examined the gendered basis of the law of self-defence. The decision also set out clearly the issues which judges should cover in addressing the jury as to the relationship between BWS and self-defence. These included how the evidence might be used in understanding why an abused woman might remain in an abusive relationship, the nature and extent of the violence that may exist in a battering relationship, the accused’s ability to perceive danger from her abuser and thus whether she was responding to a reasonable apprehension, and whether the accused believed on reasonable grounds that she could not otherwise preserve herself from death or grievous bodily harm.
Although much of the commentary on Lavallee has focused on the issue of expert testimony concerning BWS, there is evidence that this focus has been unduly narrow. For instance, in the recent Supreme Court of Canada decision R v Malott, L’Heureux-Dubé J wrote of the decision in Lavallee that:
A crucial implication of the admissibility of expert evidence in Lavallee is the legal recognition that historically both the law and society may have treated women in general, and battered women in particular, unfairly. Lavallee accepted that the myths and stereotypes which are the products and the tools of this unfair treatment interfere with the capacity of judges and juries to justly determine a battered woman’s claim of self-defence, and can only be dispelled by expert evidence designed to overcome the stereotypical thinking. The expert evidence is admissible, and necessary, in order to understand the reasonableness of a battered woman’s perceptions ... that she had to act with deadly force in order to preserve herself from death or grievous bodily harm ... But it is wrong to think of this development of the law as merely an example where an objective test ... has been modified to admit evidence of the subjective perceptions of a battered woman. More important, a majority of the Court [in Lavallee] accepted that the perspectives of women, which have historically been ignored, must now equally inform the ‘objective’ standard of the reasonable person in self-defence.
In addition to the concern with ensuring that the objective standards utilised by the standard criminal law defences reflect women’s experiences, Lavallee also demonstrates the importance of a broad range of evidence in informing the consideration of whether self-defence is available on a particular set of facts. Judge Ratushny, who conducted a review of self-defence cases in Canada, wrote that:
The significance of Lavallee for the law of self defence, in my view, is not in the recognition of the condition referred to as the ‘battered woman syndrome’, although it is this aspect of the case that has probably received the most attention. Nor is the case important just for its treatment of the issue of expert evidence. Rather, its real significance for the law of self defence lies in the fact that the Court took a broad view of the evidence that is relevant to the legal elements of that defence. Having done so, the use of expert evidence and, more particularly, expert evidence on battered woman syndrome was obviously relevant to the issues before the jury. The Court recognized that the experiences, background and circumstances of the accused should be taken into account in determining whether she actually believed she was at risk of serious bodily harm or death and had to use force to preserve herself, and the reasonableness of her beliefs. In effect, Lavallee tells us that we must consider a broad range of factors that may influence the accused’s beliefs. Similarly, many of these factors will be relevant in assessing the reasonableness of those beliefs.
Moreover, in Malott L’Heureux-Dubé J commented that an emphasis on any particular accused’s learned helplessness, dependence, victimisation, and low self-esteem should be avoided. She said that:
There are other elements of a woman’s social context which help to explain her inability to leave her abuser, and which do not focus on those characteristics most consistent with traditional stereotypes. As Wilson J herself recognized in Lavallee ... ‘environmental factors may also impair the woman’s ability to leave — lack of job skills, the presence of children to care for, fear of retaliation by the man, etc may each have a role to play in some cases.’ To this list of factors I would add a woman’s need to protect her children from abuse, a fear of losing custody of her children, pressures to keep the family together, weaknesses of social and financial support for battered women, and no guarantee that the violence would cease simply because she left. These considerations necessarily inform the reasonableness of a woman’s beliefs or perceptions of, for instance, her lack of an alternative to the use of deadly force to preserve herself from death or grievous bodily harm.
Thus, although by a different route to that evident in the US, Canadian jurisprudence has also determined that broad ‘social framework’ or ‘social context evidence’ is required in order to assist the judge and jury to make their findings in cases involving self-defence. Additional support for the utility of social framework evidence has been provided by Canadian scholars who have undertaken research using mock juries. They have found that broad ‘social framework evidence’ is extremely effective from a defence perspective.
Consistent with the acceptance of social framework evidence, some Canadian courts have accepted a broad range of expertise as relevant and admissible, including expert evidence from a crisis counsellor and from a lawyer with experience representing battered women. Sheehy has identified this as one of the positive consequences of Lavallee.
In those cases in which a more conventional construction of BWS is relied on, there is Canadian authority to the effect that expert testimony might no longer be necessary. In R v Lalonde the accused was acquitted of social security fraud. Trainor J took into account the implications of the accused suffering from BWS in interpreting the facts of the case, even though there had been no expert testimony led on BWS. Citing a passage from the unreported decision of R v Eagles, Trainor J held that it was open to a trial court to refer to and rely on the scholarly authorities cited by Wilson J in Lavallee, at least to the extent that they were referred to by her:
It would be both absurd and wasteful to require expert witnesses to attend court for the purpose of describing the generic battered wife syndrome. Such experts may not be available in remote and isolated communities.
The judge was fortified in this conclusion by ‘these days of restraint, particularly with respect to legal aid cases.’
Since 1991 expert evidence concerning BWS has been accepted in all Australian states and territories and in a range of contexts. It has been used to support the defences of self-defence, provocation and duress. It has been used extensively in mitigation of sentence, in an appeal finding a miscarriage of justice, and may have been influential in at least one decision not to go ahead with a prosecution. It has not been confined to homicide but has been introduced in a diverse range of matters including social security fraud, shoplifting, armed robbery, and charges of perverting the course of justice, breaching the Companies (Tasmania) Code and dishonestly obtaining financial advantage.
Most Australian decisions have tended to simply admit BWS with little or no debate about its nature or relevance and there have been few appeal decisions. Although this ready acceptance of BWS in Australian courts may have had positive outcomes for some defendants, it has resulted in an impoverished jurisprudence. From the cases available to the authors it is evident that there has been little, if any, attention paid to questions about what the introduction of BWS is intended to achieve, what constitutes BWS testimony, what the relationship of that testimony might be to self-defence or to other issues at trial and very little recognition of the limitations of existing defences to adequately reflect women’s life experiences. Even though the first High Court decision to address the issue of BWS evidence was recently handed down, Australia still lacks a leading judgment, similar to Lavallee in Canada or Wanrow in the US, which addresses these issues at a comprehensive and sophisticated level. The lack of Australian jurisprudence is in stark contrast to the well-developed academic literature debating the subject.
There have also been very few reported decisions involving BWS in Australia despite the fact that the authors have on file at least 25 such cases. As noted by Tolmie elsewhere, the paucity of law reporting in this area has made it difficult for lawyers, judges and academics alike to analyse legal developments and to put information about relevant decisions to use in legal argument.
The absence of a developed set of precedents, or statutes, governing the use of BWS evidence is, of course, somewhat of a mixed blessing. It does avoid an overly prescriptive approach which may introduce additional hurdles which must be met in order for the evidence to be introduced. Unfortunately, the absence of guidance concerning the purpose of such evidence and how best to use it also leaves much open to interpretation by the prosecution, defence and judiciary and can result in inaccurate and inconsistent usage. A clearer purpose and a more settled sense of the legitimacy and range of expert knowledge in the area and how it links to the legal defences could also assist the Director of Public Prosecutions in making more principled decisions not to prosecute in appropriate cases.
One factor which marks out the difference between Australian jurisprudence and that in North America is their different starting points. The recognition in North America that the law is gendered, and thus may not adequately respond to the experiences of battered women, allows a consideration of what additional information might assist the judge and jury to understand and fairly assess the actions of a battered woman on trial. Such additional information need not be restricted to BWS. This approach explicitly admits the possibility of both the admission of ‘social context information’ and the further development of legal doctrine. The Australian cases, however, have tended to start from the notion that battered women have developed different perceptions from other people, and that, with the assistance of evidence concerning BWS, judges and juries may be better able to apply existing law to cases involving battered women. This approach offers little prospect for the admission of broader social context information, except perhaps as an adjunct to expert testimony concerning BWS. It also diminishes the prospect for the further development of legal doctrine. If the problem is not a limitation in the application of the law, but rather the different perceptions of battered women, there is little need for legal reform.
One of the factors which may have contributed to the manner in which Australian law and practice has developed in this regard is the unusual nature of the case in which BWS evidence was first introduced. Unlike the North American developments which dealt with self-defence, BWS was first introduced in Australia in support of the defence of duress. Runjanjic and Kontinnen dealt with charges of false imprisonment and causing grievous bodily harm with intent, and concerned the allegation that the two women defendants had held another woman against her will and assaulted her. The defendants argued that they had acted under duress, that is under threat of violence directed towards them at the hands of a man called Hill, who was the de facto partner of both defendants. The appeal court dealt with whether expert evidence concerning BWS should be admitted, and ruled that BWS may be relevant in order to assist the court to understand
that methodical studies by trained psychologists of situations of domestic violence have revealed typical patterns of behaviour on the part of the male batterer and the female victim, and typical responses on the part of the female victim. It has been revealed, so it appears, that women who have suffered habitual domestic violence are typically affected psychologically to the extent that their reactions and responses differ from those which might be expected by persons who lack the advantage of an acquaintance with the result of those studies.
Repeated acts of violence, alternating very often with phases of kindness and loving behaviour, commonly leave the battered woman in a psychological condition described as ‘learned helplessness’. She cannot predict or control the occurrence of acute outbreaks of violence and often clings to the hope that the kind and loving phases will become the norm. This is often reinforced by financial dependence, children and feelings of guilt. The battered woman rarely seeks outside help because of fear of further violence. It is not uncommon for such women to experience feelings for their mate which they describe as love. There is often an all-pervasive feeling that it is impossible to escape the dominance and violence of the mate. There is a sense of constant fear with a perceived inability to escape the situation.
Another factor which may have contributed to the manner in which Australian law and practice first developed was the need in Runjanjic and Kontinnen to justify the introduction of expert evidence on the basis that it went beyond the ‘behaviour of normal people and of situations which are within the experience of ordinary persons or are capable of being understood by them’. What this meant, of course, is that it was originally easier to justify the admission of psychological testimony if it was directed at explaining a ‘syndrome’ or abnormal mental state. This evidential rule, sometimes known as the ‘common knowledge rule’, has now been abolished under the Uniform Evidence Law which applies in the Australian Capital Territory, Northern Territory and New South Wales.
The recent High Court decision of Osland is the only other Australian appeal case to have discussed the preliminary issue of whether BWS evidence should in principle be introduced. In that case, all five judges based their decision primarily on the application of the legal principles of ancillary liability to the facts of the case. Only four of the judges also considered the issue of BWS evidence, with Gaudron and Gummow JJ producing a joint judgment that otherwise dissented on the issue of ancillary liability.
The most problematic judgment in Osland on the subject of BWS was delivered by Callinan J. His discussion of BWS betrayed a fundamental misconception about the nature and purpose of the evidence. He suggested, contrary to all authority, that the courts in Canada may regard BWS as a separate legal defence in itself. Fortunately, he concluded that ‘[t]here is no such separate defence in Australia.’
Gaudron and Gummow JJ provided a more traditional account of BWS evidence. They described a set of psychological features which ‘are characteristic of battered women, but not necessarily present in all cases’ and concluded that:
Given that the ordinary person is likely to approach the evidence of a battered woman without knowledge of her heightened perception of danger, the impact of fear on her thinking, her fear of telling others her predicament and her belief that she can’t escape from the relationship, it must now be accepted that the battered wife syndrome is a proper matter for expert evidence.
This aspect of their judgment arguably goes no further than the approach of King CJ in Runjanjic and Kontinnen and suggests that juries and judges need BWS evidence solely because it explains the very different perceptions and attitudes that may be typical of battered women.
However, Gaudron and Gummow JJ specifically discussed how BWS evidence might be used in the context of the defences of provocation and self-defence. This is rare in Australian cases. They commented that BWS evidence, specifically the ‘heightened arousal or awareness of danger’ that such a woman might possess ‘may be directly relevant to self-defence, particularly to the question whether the battered woman believed that she was at risk of death or serious bodily harm and that her actions were necessary to avoid that risk.’ Similarly, such evidence might explain why an act of ‘apparently slight significance’ might nonetheless be properly viewed as evidence of provocation or why a woman’s heightened perception of danger might have a bearing on the court’s assessment of the gravity of provocation she was offered. Whilst this analysis is useful, it is nonetheless limited.
What is disturbing about the approach of Gaudron and Gummow JJ is that they appear to relate BWS evidence exclusively to the subjective, or modified subjective, components of the defences of provocation or self-defence, rather than to the objective components as well. For example, they do not suggest that BWS evidence might explain why there could be reasonable grounds for the accused’s perception that she was under life-threatening danger and needed to resort to lethal self-help, only why she might honestly have believed this to be the case. Likewise, they do not suggest that such evidence explains how an ordinary person could have lost self-control in the circumstances for the purposes of the defence of provocation — the only objective limb of the provocation test that is not potentially modified by characteristics peculiar to the accused. This reading of their judgment suggests that BWS evidence assists the court in understanding the personal or idiosyncratic — the ‘subjective’ responses of battered women who suffer from the syndrome — rather than explaining the effect that circumstances of violence might have on the responses of ordinary or reasonable women. If this is so, then their judgment represents a considerable narrowing of the interpretation of BWS offered by King CJ in Runjanjic and Kontinnen. It also represents a departure from the basis on which BWS was first developed and understood. BWS evidence was evolved to explain why a particular accused’s response was ordinary or reasonable given the circumstances of violence they had survived and this is the sense in which King CJ clearly understood the concept.
This aspect of Gaudron and Gummow JJ’s judgment illustrates the importance of acknowledging that expert evidence historically has been necessary because of the inequitable application of the legal defences to women’s life experiences, rather than the different perceptions of women.
Arguably the most forward looking judgment dealing with BWS evidence in Osland was delivered by Kirby J. After canvassing some of the controversies around the use of BWS evidence he nonetheless accepted that it was admissible. Whilst his judgment is difficult to interpret in places, his discussion of BWS evidence appears to be consistent with the recognition in North America of the gendered application of the law on self-defence.
In summarising debate concerning BWS, Kirby J noted that the purpose of introducing BWS evidence is ‘to show how a victim’s actions in taking lethal self-help against the abuser was reasonable in the extraordinary circumstances which the victim faced’. In other words, the purpose of such evidence is not to show that battered women’s perceptions are aberrant or different but that they are reasonable. It follows from this that Kirby J also acknowledged that evidence on BWS is relevant to the objective limb of the test on self-defence. Borrowing from Malott, he commented that such evidence is relevant to a series of questions, including:
[W]hether, in the evidence, the particular accused believed on reasonable grounds that there was no other way to preserve herself or himself from death or grievous bodily harm than by resorting to the conduct giving rise to the charge.
He also cited approvingly from Crocker to the effect that ‘[t]he defendant introduces testimony to offer the jury an explanation of reasonableness that is an alternative to the prosecution’s stereotypic explanations.’
It is possible to read this aspect of Kirby J’s judgment as acknowledging that expert evidence is necessary because of the historical failings of the application of the law, rather than the allegedly unique perceptions of battered women. There is implicit recognition of the need to modify the manner in which the concept of ‘reasonableness’ has been applied in this context in respect of self-defence. This reading of Kirby J’s judgment is reinforced by his discussion of the issue of ‘imminence’.
Nonetheless, as we have mentioned, Kirby J’s judgment is difficult to interpret. He rejected the nomenclature ‘Battered Woman Syndrome’ and was even more uncomfortable with the notion of a ‘Battered Wife Syndrome’. Whilst he acknowledged that domestic violence is overwhelmingly an issue which affects women, and is likely to remain so, he commented that, unlike conception and child birth it is not inherently specific to women. Given this fact, he was not in favour of creating a sex-specific category or construct. However, considering the social context of the offender in a detailed and realistic fashion when applying the law on self-defence is not to create a sex-specific construct or stereotype, even though that social context will be strongly influenced by the gender of the offender concerned. To this degree, it is unfortunate that Kirby J did not go so far as to explicitly acknowledge that the law has historically had a gendered effect. Without this explicit acknowledgment one may fail to really appreciate the value of ‘social context evidence’ in matters concerning abuse.
A unique recognition of the gendered nature of the law on both self-defence and provocation in the Australian context is to be found in the judgment of Gleeson CJ in Muy Ky Chhay, a case in which BWS was not argued. Gleeson CJ endorsed the criticism in relation to provocation that ‘the law’s concession to human frailty was very much, in its practical application, a concession to male frailty.’ He cited from Lavallee in support of his comment that the ‘orientation of the law towards relief of the plight of males, rather than females, was also noted in the area of self-defence.’
The marked inconsistencies in the manner in which BWS has been constructed by expert witnesses and interpreted by the prosecution, the defence and the judiciary in the Australian context are not surprising given the lack of an authoritative and accessible jurisprudence. However, it is probably fair to say that in the majority of Australian cases in which BWS has been introduced it appears to have been narrowly construed and directed primarily towards explaining the psychology of the particular accused or of battered women in general. This is in marked contrast to the literature indicating that what is likely to be of greatest utility to courts is broader evidence concerning battering and its effects.
Some expert witnesses have characterised BWS in such a way as to suggest that battered women in general, or a particular accused, are not reasonable. This has the capacity to undermine claims to self-defence. A classic example of this approach is found in the expert testimony in J v The Queen. The trial judge summarised the expert’s account of BWS as follows:
There was a well recognised psychological condition known as the battered woman syndrome; it was part of or closely akin to a post-traumatic stress syndrome; the feature of it was a learned helplessness, a difficulty in breaking away from an assaultive or destructive relationship; if the complainant’s narrative was accurate ... that narrative was consistent with the ... pathology of the condition.
The construction of BWS as pathology in this manner has, of course, been widely criticised in the international literature. Other expert witnesses have added their own psychological embellishments to the more usual features associated with BWS. For example, the expert in Secretary also talked about regressions to a childlike space and childhood abuse experienced by the accused. Such evidence can have the effect of suggesting that the accused killed the deceased in a state of mind that was far from that of the reasonable adult.
In some Australian cases expert witnesses called to testify concerning BWS have also attempted to discuss social factors which might be relevant to understanding the perceptions and behaviour of an accused person. Evidence of social circumstances may support the notion that the accused’s circumstances were objectively life-threatening and may have necessitated the kind of defensive response she in fact engaged in. However, the introduction of evidence concerning social circumstances has tended to be secondary to a discussion of the psychology of the accused, with a particular focus on establishing whether the accused has characteristics (or symptoms) consistent with BWS.
By contrast to the more usual expert testimony concerning the psychology of the accused, there have been a few recent instances where an expert witness providing testimony concerning BWS has focused primarily on the context in which the accused found herself, rather than her psychology. One example is contained in R v Lyons where the expert witness’ account focused very heavily on the violence and psychology of the perpetrator rather than the accused. Whilst he listed 12 characteristics commonly seen in women with BWS, seven of these characteristics are typical of the violence such women are likely to face and only four describe states of mind on the part of the woman. These latter states of mind are not presented by the expert as pathological but rather reasonable or normal in the circumstances. For example, number 12 is the simple realisation on the part of the woman that the person she is in a love relationship with is quite capable of killing her.
Also at issue is how judges interpret the nature of BWS evidence. There are obvious examples where expert witnesses have added ‘social context evidence’ to their psychological testimony to suggest that the accused’s actions may have been a reasonable response to her circumstances, but the judge has nonetheless construed the evidence given in the narrowest possible way. In Tassone, an expert witness testified that in some cases a woman’s perception that she is trapped in an abusive relationship might be reinforced by reality. She said that:
[S]ometimes ... that sense of no escape[,] of being trapped, helpless in the situation is reinforced by the reality of the situation. They can be caught up with that partner because of children, because of nowhere else to go to, no money to get there, no means of support. Sometimes those are definite realities sometimes they are just the result or perception of the individual because they have such a sense of worthlessness and hopelessness about themselves.
She also made the point that ‘the actual studies show that in a significant number of cases the abuse does end in the death of the woman’ and that it is unusual for women to kill in these circumstances — they are more likely to be victims. In spite of this testimony Gray J, instructing the jury in Tassone, said that the expert’s evidence on BWS
seeks to explain why a woman suffering this kind of ill-treatment feels trapped within such a relationship, suffers a massive loss of self-esteem, is likely to exaggerate in her mind the immediacy of threats of further violence and generally feels overwhelmed by the endless hopeless future she sees before her, until very often a violent reaction occurs.
One of the most powerful aspects of Kirby J’s judgment in Osland, and consistent with his treatment of the need for BWS evidence, was the manner in which he conceptualised the expert evidence. When he first discussed the admissibility of such evidence he did not use the phrase ‘Battered Woman Syndrome’ but, instead, ‘expert testimony about the general dynamics of abusive relationships’. At a subsequent point he spoke of admitting ‘expert evidence of BWS, or an analogous condition’. There is also a point in his judgment where he appeared to invite defence counsel to think more imaginatively about presenting such evidence. After having canvassed a number of the criticisms made of BWS, he commented that:
[I]t was the appellant, at trial, who raised BWS. She called expert evidence about it and argued that it was relevant, in her case, to self-defence and provocation. It is therefore unsurprising that this was the way in which, until now, the issue has been addressed in these proceedings. It is too late in this case to adopt a change of course.
While Kirby J was vague about the kind of evidence which might be admissible, and at times rather problematically appeared to have in mind evidence that was limited to the psychological, he was clearly anticipating the significance of evidence going beyond BWS. Whilst we have documented a number of recent expert accounts of BWS that represent quite broad and contextual understandings of the nature and purpose of such evidence, Kirby J’s judgment is the first judicial acknowledgment of the need for broader testimony in these types of cases. In this respect his judgment is the first in Australia to open the way for hearing broader evidence which is more consistent with the North American developments which have been outlined above.
Since Australian cases largely have interpreted BWS by requiring that the psychology of the person who has experienced battering needs to be explained to the court, it follows that expertise is considered to be both necessary and to lie in the domain of psychology or psychiatry. As outlined above, this is a narrow approach which is contrary to the findings of a major US review of these issues. In most Australian cases of which the authors are aware, the expert testimony has been given by a psychiatrist or psychologist. However, there are some interesting exceptions.
In R v Gadd, the accused was acquitted of murder after stabbing her husband. It appears that she was acquitted on the basis of self-defence. One of the most interesting features of this case is that the expert testifying about the material normally understood in terms of BWS was a social worker rather than a psychologist. The social worker had extensive experience working with battered women and amongst many other qualifications, had worked as the coordinator of a women’s health centre, a domestic violence resource centre, and a women’s refuge, as well as doing counselling or crises intervention work with over 700 women who had experienced domestic violence. She testified about domestic violence generally, including the cycle of violence. She explained the difficulty battered women might have in leaving violent relationships and their tendency to hide the abuse in terms of the violence they experience. She did not testify about the concept of learned helplessness nor any other psychological characteristics of the woman concerned.
Osland was one of the first Australian decisions to attempt to spell out explicitly the connections in principle between BWS evidence and the legal defences of provocation and self-defence. We have argued above that in Osland the approach of Kirby J is to be preferred to the approach of Gaudron and Gummow JJ.
One of the grounds of appeal in Osland was that the trial judge had failed to clarify the connection between the evidence of BWS and the law of provocation and self-defence. Three of the judges in the High Court firmly placed the onus on defence counsel either to make these links apparent or to actively seek specific directions from the judge. Gaudron and Gummow JJ said that:
It need hardly be said that there is an obligation on counsel to make clear to the jury and the trial judge the precise manner in which they seek to rely on expert evidence of battered wife syndrome and to relate it to the other evidence and the issues in the case. In circumstances where evidence of battered wife syndrome is given in general terms, is not directly linked to the other evidence in the case or the issues and no application is made for any specific direction with respect to that evidence, it cannot be concluded that the trial judge erred in not giving precise directions as to the use to which that evidence might be put.
Kirby J concurred:
[A]ny imperfections in the charge of the trial judge in this case in relating the evidence of BWS to the issues of provocation and self-defence can properly be ascribed to the general terms in which [the expert] gave his evidence about the typical features of BWS and the absence of clear evidence linking such testimony to the appellant’s own conduct. In the circumstances, the judge’s directions were adequate.
It is unfortunate that the judges in Osland who dealt with this issue chose to place the onus on defence counsel rather than the trial judge. This approach was taken in spite of severe cutbacks in legal aid that may well affect the quality of legal representation that many accused experience. It also can be contrasted to the position taken by the Supreme Court of Canada in Malott. In that case the majority held that a trial judge instructing the jury in cases involving BWS and self-defence must communicate certain principles set out in Lavallee. That is, the judge must inform the jury that BWS evidence may be of use in explaining:
1. Why an abused woman might remain in an abusive relationship.
2. The nature and extent of the violence that may exist in a battering relationship.
3. The accused’s ability to perceive danger from her abuser. [In particular whether, for the purposes of self-defence, she has a reasonable apprehension of death or grievous bodily harm.]
4. Whether the accused believed on reasonable grounds that she could not otherwise preserve herself from death or grievous bodily harm.
In Part II of this article we addressed general understandings of the need for, and nature of, BWS evidence, as well as who is qualified to testify about it in court. We documented best practice in North America and argued that the judiciary in Australia needs to broaden its understandings of the nature of expert evidence to include evidence directed at ‘social context’ rather than merely BWS. We also documented hints of a sea change along these lines in some of the most recent Australian authorities.
In this section we discuss the extent to which recent Australian cases have engaged with a diverse range of more specific legal issues currently arising in cases concerning battering, especially those involving self-defence. Firstly, we describe modifications to self-defence doctrine that make it more accessible to battered women who engage in defensive pre-emptive strikes. Secondly, we discuss the problems that women who have used previous defensive force in respect of relationship violence might face in arguing self-defence. Thirdly, we document a number of recent cases in which self-defence was successful even though it was argued without expert evidence concerning BWS to support it. Fourthly, we describe a case involving a successful clemency plea. We argue that Australian developments around all of these issues are very positive. Less positive is the way in which Australian courts have dealt with the issue of violence in the context of same-sex relationships, or treated ethnicity or race as factors which may shape social context.
A number of commentators have pointed out that requiring that an attack be ‘imminent’ before defensive force is legally justified presents a major obstacle for battered women seeking to raise self-defence in respect of homicide charges. For obvious reasons such women are unlikely to take an abuser on in hand-to-hand combat. Therefore it is not uncommon for them to defend themselves by means of pre-emptive strikes.
Since Zecevic v Director of Public Prosecutions reformulated the law on self-defence in Australia, imminence is no longer a technical requirement. In addition, at least four Australian cases have explicitly allowed battered women to successfully raise self-defence in respect of their pre-emptive defensive actions, thus establishing that an imminent attack is not essential if an accused is to succeed in a self-defence claim in this context. Of course it is still essential for an accused to demonstrate that their defensive action was necessary in the circumstances they faced.
Two recent cases which expressly looked at the issue of pre-emptive strikes involved sleeping aggressors and both were decided under the Northern Territory Criminal Code 1983. The Northern Territory Code differs a little from the common law in the manner in which the issue of immediacy arises. In the Code, a person cannot raise self-defence unless there is a serious ‘assault being defended’. An assault can be ‘threatened’, as opposed to an actual application of force provided that the person threatening assault has ‘an actual or apparent present ability to effect his purpose and the purpose is evidenced by bodily movement or threatening words’.
In Tassone, the Northern Territory Supreme Court left the question of self-defence to the jury. The accused was charged with attempted unlawful killing. She shot her violent husband (who survived) whilst he was sleeping and after he had assaulted and raped her. Her evidence was that she was terrified of his extreme and unpredictable violence, that she had unsuccessfully tried to leave him on a number of occasions and now believed that there was no escape from him, and that the rape had ‘upped the ante’ in the sense that it demonstrated a new level of violence towards her. Although her husband had not verbally threatened her before he fell asleep, the general and ongoing threat that he presented to her, which was demonstrated by his past behaviour towards her, was obviously satisfactory to the jury in terms of the Code. She was acquitted on the basis of self-defence.
However, a close reading of the judge’s comments throughout the trial transcript indicate that Tassone is not a strong authority for the proposition that pre-emptive strikes will satisfy the self-defence test. It is clear from comments made during the trial, although not from his instructions to the jury, that Gray J did not have much confidence that a sleeping aggressor could present a ‘threatened assault’ against which a woman could be defending herself. He also expressed doubts about the veracity of BWS evidence but felt bound by authority to admit it. In the end he left the question of self-defence to the jury on the basis that a trial judge needed to be extremely careful about withdrawing issues of fact from the jury. He commented that the case before him
[o]n any conceivable view...[is] on the outer limit of self-defence cases, [it] will be very likely to be beyond the outer limit, but I say my inclination is to leave it [to the jury].
In Secretary, the Northern Territory Court of Criminal Appeal re-examined the issue. In this case, the accused had shot and killed her sleeping aggressor who had fallen asleep after he had terrorised her, assaulted her, and told her that further violence would commence when he awoke. In the Supreme Court the trial judge held that self-defence was unavailable to the accused. His reasoning was twofold. Firstly, he held that the danger to which the accused was responding had to be ‘imminent’, a requirement which was obviously not satisfied in the case of a sleeping aggressor. Secondly, he held that such an aggressor did not have a ‘present and apparent ability to effect his purpose’ in terms of s 187(b) of the Code. On appeal the Court of Criminal Appeal held, by a majority of 2 to 1, that the Code did not require an ‘imminent’ attack. A threat of future harm would suffice provided that the threat was ‘current’, ‘continuing’ or ‘not completed’ — a question of fact for the jury. Mildren J observed, in obiter dictum, that the ‘common law has moved away from the requirement of immediacy, favouring a more flexible approach to the law relating to ... self-defence’. Obviously the question of whether the threat necessitated the accused’s defensive force must still be answered. Secondly, the majority held that a sleeping aggressor could have an ‘actual or apparent present ability to effect his purpose’. Mildren J said that the ‘reference in the section to “present ability” means in this context, an ability, based on the known facts as present at the time of the making of the threat, to effect a purpose at the time the purpose is to be put into effect.’ A retrial was ordered and the accused was ultimately acquitted.
One of the features of Secretary which differs significantly from Tassone is that all of the judges dealt with the matter as though the assault against which the accused was defending herself were the words uttered by the deceased before he went to sleep, rather than the general threat he represented in the relationship with her.
In R v Stjernqvist a Queensland jury acquitted the accused on the basis of self-defence after only 15 minutes of deliberations. She had shot her husband in the back as he walked away from her. In this case, according to the trial judge’s instructions to the jury, the assault that the accused was defending herself from was to be found in the general nature of the relationship and particularly in the threats the deceased had made to the accused over a period of years, rather than in any specific action he had taken on the day in question. We discuss this case further below.
Kirby J in Osland has provided additional support for the proposition that in the context of a battering relationship (as in other contexts) the law does not insist that an accused be facing an imminent attack before she can legally use force to defend herself. The determining factor is not the imminence of the attack but the necessity of the defensive force used. In his words, ‘[t]he significance of the perception of danger is not its imminence. It is that it renders the defensive force used really necessary and justifies the defender’s belief that “he or she had no alternative but to take the attacker’s life.’’’
There has been some suggestion that women who are not able to construct themselves as dependent and helpless, for example because they have previously used force to defend themselves against the ongoing violence they have experienced, will have difficulties in raising self-defence. This may be the case for a number of women because, as Besley points out, the notion that battered women tend to respond to the abuse they experience in a passive way is a myth. The Canadian courts have recognised this problem. In Malott, for example, it was said that:
It is possible that those women who are unable to fit themselves within the stereotype of a victimized, passive, helpless, dependent, battered woman will not have their claims to self-defence fairly decided. For instance, women who have demonstrated too much strength or initiative ... or women who might have fought back against their abusers on previous occasions, should not be penalized for failing to accord with the stereotypical image of the archetypal battered woman.
Women who have fought back may also run the risk of activating traditional misconceptions of domestic violence as ‘marital discord’ or a situation for which both parties are mutually responsible.
There has been some interesting scholarship around this issue. A study undertaken by Schuller and Hastings attempted to test whether ‘portrayal of the woman’s behavior toward the batterer, specifically her degree of “passivity,” would influence the decision process.’ This study used simulated juries and court materials. Half of the ‘juries’ were presented with material in which the accused was portrayed as an active agent, including information showing that on previous occasions she had fought back when attacked or threatened. For the other half this information was omitted. Whilst commenting that they may not have made the woman in the first set of materials ‘assertive enough’, Schuller and Hastings found that:
Overall, the woman’s prior response to the violence in her relationship (passive vs active) had little impact on the decision process. It was also not the case that the battered woman syndrome evidence was detrimental in the case in which the woman was portrayed as having previously fought back against her abuser.
Besley conducted a similar study also using simulated court materials and juries. Her findings provide stronger support for the development of forms of expert testimony alternative to BWS evidence in cases involving women who have a history of using defensive force. In one fact situation the accused was portrayed as very passive. In this scenario she had failed to seek help, was from a lower socio-economic status and adhered to traditional female gender roles. In the other fact situation the accused was financially independent, did not conform to traditional gender roles, actively fought back (had broken the perpetrators arm and pulled a gun on him), had sought legal help and had fled the home on a number of occasions. Besley found that if the accused did not conform to the stereotype of a ‘battered woman’ then expert testimony on BWS would actually decrease the number of not guilty verdicts, just as it increased the number of not guilty verdicts if she did conform.
The case of R v Lock illustrates some of the potential difficulties women who fight back have in raising self-defence quite aside from issues of BWS. It may be just such difficulties which account for the fact that evidence of BWS was not offered in that case. The fact that the accused in Lock previously had fought back against abuse by her partner may have been the reason why Hunt CJ at CL characterised the relationship as a ‘“love/hate” relationship’, rather than as a relationship characterised by violence. It also may have been the reason for his gratuitous comment that:
[The accused and the deceased] had earlier that evening been involved in a long argument and the accused had acted at that time in a way which was almost guaranteed to provoke any person prone to violence in the privacy of the home to be angry with her and to attack her later in the home.
Lock allegedly had stabbed her husband and was charged with his murder. She argued that she was defending herself and that, although the deceased had not physically assaulted her on the night in question, she knew from past experience that he was going to be violent and had therefore taken up a defensive position with a knife. To support a claim to self-defence the accused led extensive evidence of the deceased’s past violence, backed up by police records of attendances at her home and hospital and ambulance records of her admissions into hospital. After a trial which ended in a hung jury, the accused was retried and found not guilty of both manslaughter and murder. Although the basis for the jury decision is not clear there are indications that she was acquitted on the grounds of self-defence.
The central legal issue in Lock was the admissibility of evidence about past defensive efforts by the accused. The Crown sought to tender evidence of three instances where the accused had stabbed the deceased with a knife. The Crown unsuccessfully sought to tender this evidence on the basis that it showed that the accused had the tendency to deliberately stab the deceased when arguing with him. With three exceptions the Crown was successful in admitting this evidence on the basis that it showed the true nature of the relationship between the parties — that is that they had a relationship of mutual violence. The accused had failed to establish under s 137 of the Evidence Act 1995 (Cth) that the probative value of the evidence was outweighed by the danger of unfair prejudice to her. It was held that the evidence was highly probative because it was
relevant to rebut self-defence, which was opened to the jury by counsel for the accused as the significant issue in the case, one which in turn depended strongly upon the general relationship between the accused and the deceased. The true nature of that relationship was therefore of great importance in the case, and the probative value of this evidence upon the accused’s state of mind as to the necessity to do this act in self-defence was correspondingly high.
Lock contains a strong and explicit acknowledgment that a relationship of violence is a highly relevant context for the assessment of an accused’s claim to have acted in self-defence. This is a positive step. It has been complained in the past that courts often decontextualise the accused’s behaviour in fact situations when there has been a history of violence by placing primary emphasis on the immediate facts surrounding the killing. Obviously, an acknowledgment of the significance of relationship violence means that the accused’s own behaviour in the context of the relationship can be used by the Crown to cast doubt on her credibility. It is possible to imagine cases where expert evidence might be used to provide the court with a realistic context in which to judge the accused’s past defensive force.
Recent cases demonstrate that it is now possible to succeed with a claim to self-defence arising in the context of domestic violence without offering evidence of BWS. The authors are now aware of three Australian cases in which self-defence has been successfully run by women who have been the target of domestic violence without the use of supporting BWS evidence. Such outcomes may reflect a growing awareness by the judiciary and the community of the incidence and nature of domestic violence. This is a development which seems to have been contemplated by Callinan J in Osland, when he envisions a point in time at which expert testimony might be redundant in cases involving women who are the targets of violence. He says that with ‘growing community awareness’ of ‘these matters’ (it is not clear whether he is speaking of the phenomenon of domestic violence or of BWS) they might become ‘matters for a jury to decide with proper assistance from the trial judge’ as opposed to matters for expertise. The other judges in Osland did not specifically address this issue, although Kirby J did comment in passing that expert testimony ‘about the general dynamics of abusive relationships is admissible if ... proved by a qualified expert.’
It is to be hoped that Callinan J’s predictions in this regard are correct. BWS, and other expert evidence concerning battering and its effects, were originally developed in order to extend traditional legal doctrine to the kinds of life experiences likely to be faced by women. It was not, as we have indicated above, intended to develop a special defence or pleading for battered women. It is to be hoped that there will come a time when community and judicial understandings of domestic violence as a phenomenon are sufficiently informed and sophisticated that expert testimony is not necessary for the finder of fact to realistically understand the context in which women’s defensive force might take place.
Stjernqvist is an interesting case in which an acquittal was achieved without expert testimony concerning BWS. The case involved a ‘non-traditional’ self-defence fact situation in which the accused was using pre-emptive force. Derrington J delivered a very traditional summing up on the defences of provocation and self-defence. However, the judgment also reveals a sophisticated understanding of the phenomenon of domestic violence. Instead of analysing the violence which the accused faced as a series of discrete instances with periods of calm in between, he analysed it in terms of a general overall threat that the accused lived with:
[W]hat emerges is necessarily a sad picture of serious violence — not violence that has caused any great physical harm at any particular time, but violence of such a nature that, you might think, would be virtually intolerable, particularly if one had the view that it was going to be never ending. To live in an atmosphere where there is a constant threat of violence, you might think, is a very hard thing and must be very emotionally wearing. And, of course, after a while it becomes a case where not only is there physical violence, but the mere endurance of the threat of violence also becomes a form of psychological violence as well.
Derrington J also had a clear grasp of the concept which Mahoney has labelled ‘separation assault’, assisted perhaps by evidence of the deceased man’s repeated threats to track down and kill the accused should she leave him. Derrington J interpreted these threats as a ‘continuing assault’ in terms of the Queensland Criminal Code Act 1899:
[I]f she were living under the threat of his killing her if she left him, then that was a very substantial threat. It was not immediate, but certainly there, and very real if she did leave him.
Derrington J discussed the reasonableness of the accused woman’s belief that defensive force was necessary in her circumstances with reference to both the general threat from remaining in the relationship and the serious threat from leaving it:
Then you would have to consider whether or not in those circumstances the situation was so intolerable that she could not stay, having regard to his refusal to let her have other women around and that type of thing, which means she had the impossible situation of remaining there in those circumstances, or leaving and then being subject to the threat of being killed by him.
It is apparent that Derrington J did not minimise the violence faced by the accused but stressed the serious nature of such violence and the social reality that ‘there are irresponsible people with guns who go and blast their wives and children, their wife’s parents, when their wives are trying to hide from them.’
For cases like Lock and Stjernqvist to be successful without the introduction of expert opinion evidence it was obviously necessary for defence counsel to educate judges and juries about the factual context of the accused’s actions. In both cases extensive and detailed evidence was led from a number of different sources about the deceased’s violence. It is also possible to lead extensive factual evidence about the accused’s options for dealing with that violence. Although Gilbert was also a case in which BWS evidence was introduced, it does illustrate the kinds of broader framework evidence that might be used. In Gilbert the accused was an Aboriginal woman from a small rural community. An Aboriginal elder from the community testified that traditional mechanisms within the Aboriginal community that should have protected this woman had broken down. He testified that it was well known within the community that the accused was being bashed but that no one was willing to intervene because they were terrified of the deceased. An Aboriginal Police Liaison Officer gave evidence that the police did not offer the accused protection from the deceased and he described one occasion on which he had seen two police officers walk past the accused when she was lying injured by the side of the road.
The value of expert witnesses is not confined to the trial. Defence counsel may in fact usefully employ experts during pre-trial preparation. For example, Maguigan comments that:
[D]efense lawyers rely too much on expert testimony at trial (often overlooking the necessity of convincing judges that the cases require the application of existing substantive, procedural, and evidentiary law) and utilize experts too little at the stage of trial preparation.
It is worthy of note that in at least two of the three Australian cases that the authors are aware of which resulted in acquittals in the absence of BWS evidence, counsel engaged experts in the pre-trial preparation. The same social worker contributed to both cases.
Defence counsel do need to be careful about the degree to which expert psychological testimony has the capacity to undercut factual ‘social framework evidence’. For example, the outcome in Gilbert was a conviction for manslaughter on the basis of provocation. It is possible that the psychological evidence of BWS, as it was presented by the expert in that case, may have undercut the accused’s strong circumstantial self-defence case. Dutton also points to circumstances in which expert testimony may be unnecessary, or counterproductive. She argues that:
[E]xplanations that rely too heavily on psychological constructs to explain a battered woman’s behaviour or state of mind may take the emphasis off the violent and coercive behaviour of the perpetrator as a basis for understanding the battered woman’s behaviour.
United States advocacy on behalf of battered women has also included clemency campaigns for women incarcerated for killing an abuser. Some of the clemency campaigners have emphasised BWS, while others have chosen not to, wary of the risks of having individual women assessed to determine if they conform to some set of prescribed symptoms. There have been some significant outcomes from such campaigns, including reviews in a number of states of the sentences of women convicted of homicide offences. At least 106 battered women from 23 states have received clemency since 1978, including 26 women who had been imprisoned for killing or assaulting violent partners who were granted clemency in 1990 by the Governor of Ohio and eight women who were granted clemency by the Governor of Maryland in 1991. Despite these successes, concerns have been expressed about arbitrary or politically expedient uses of clemency. It also has been suggested that clemency may be less politically acceptable in the current political climate which favours a tough approach on law and order.
In 1995 the Solicitor-General of Canada and the Minister of Justice established a review of the cases of women under sentence for homicide ‘in circumstances in which the killing allegedly took place to prevent the deceased from inflicting serious bodily harm or death’. The review considered the cases of 98 women, including both pre- and post-Lavallee convictions, and made recommendations in relation to seven women.
The authors are aware of only one successful clemency plea in Australia. Although Eileen Waugh was in an unusual situation in that she was in the onerous position of being caught between jurisdictions, the case does present some hope that battered women who have been incarcerated for killing an abuser in more general circumstances might be able to take advantage of a similar process.
Waugh, who had separated from her violent husband, shot him during a meeting she had agreed to in the hope that she could defuse the situation between them. Prior to the meeting he was persistently phoning her house and threatening to kill her and her boarder. At the time she was living in Queensland, having fled there from New South Wales during one of her many attempts to escape the violence.
The Director of Public Prosecutions offered to drop murder charges if she agreed to plead guilty to manslaughter. She refused and her defence was conducted on the basis that she had accidentally shot her husband whilst struggling with her boarder who was holding a gun (purchased to protect them from the deceased) and arguing with the deceased. At the beginning of 1989 she was found guilty of murder and given a mandatory life sentence under s 305 of the Criminal Code Act 1899 (Qld).
Waugh successfully applied to be transferred to New South Wales, where her family and friends resided, to serve her sentence. At the end of 1994 she made an application to the Governor of New South Wales to exercise the Royal prerogative of mercy, grant her a conditional pardon and immediately release her. This was the only basis on which Waugh could get an early release as the provisions in the Sentencing Act 1989 (NSW) allowing for early release did not legally apply to interstate transfer prisoners. The Queensland Corrections Board would not consider her release whilst she remained a prisoner in New South Wales.
The main ground for her request was that there was new evidence which had not been made available to the courts in the original trial and appeal. In particular the long history of domestic violence that she suffered was not ‘ventilated in any meaningful way in her trial as it did not fall within any of the “defences” available to a charge of murder.’ In particular her counsel submitted that:
The defences of accident, self-defence and provocation, in their present construction, cannot and do not adequately allow for her experience, and her state of mind at the time, to be taken into account. Until there is a real change in the laws so that the reasons why women kill their husbands may be adequately explored, women like Ms Waugh will continue to be convicted of murder. Ms Waugh’s case presents an ideal opportunity to correct a grave injustice and provide that the legal system can and will work in the interests of justice.
The phenomenon of violence within same-sex relationships raises a series of difficult issues, most of which are beyond the scope of this paper. What is important to note in this context is that ‘social framework evidence’, of the kind we have argued above should be introduced in Australia, may be of great importance in realistically understanding and applying the criminal defences to homicide where an accused has used lethal defensive force against an abusive partner of the same sex. For example, it is difficult to realistically assess the range of options a gay man or lesbian might have to escape or defuse domestic violence without understanding what Simone has called ‘discrimination which isolates’, faced by such targets of violence.
The authors are only aware of one Australian case of this kind. McEwen was a Western Australian decision in which the accused was charged with murdering his male partner of 14 years. The jury failed to reach a unanimous verdict and, before retrial, he pleaded guilty to manslaughter on the basis of provocation and received a sentence of five years. Murder charges were dropped. Expert testimony on ‘battered spouse syndrome’ was accepted in support of both his provocation claim and in mitigation of sentence.
The decision in McEwen is problematic. The court did not contextualise the issue of same-sex violence nor focus on the particular isolation of victims of gay and lesbian violence, or the difficulties specific to gay men. For instance, it didn’t explore the accused’s fear of leaving or of publicising a gay relationship in the face of a homophobic society, the inability of the accused to use women’s refuges for short-term shelter and accommodation or otherwise access mainstream support services, his reluctance to hand his partner over to the police or even admit to police the nature of the relationship, mainstream social acceptance of male to male aggression as constituting a conflict of equals, or the social stigma of raising issues of violence within the gay and lesbian community. Instead of realistically exploring the context in which the accused was attempting to deal with the violence he faced in the intimate recesses of his daily life the court ‘feminised’ the accused himself. He was portrayed as passive and childlike, with all the personal and psychological inadequacies implied in suffering from the renamed ‘battered spouse syndrome’. In the words of his defence counsel, ‘his responses [were] different to the responses which might be expected from an ordinary person who had not been in that relationship, or that type of relationship.’ A more expansive understanding of battering and its effects might have assisted the court to better assess the circumstances to which the accused was responding.
It is noteworthy that in Osland Kirby J was clearly influenced by the phenomenon of violence in same-sex relationships in seeking to avoid sex-specific categories in the construction of BWS evidence. He commented that, whilst domestic violence is overwhelmingly a phenomenon that affects women, victimisation is not inherent to women: ‘[i]nstances exist where the reverse is the case, including in some same-sex relationships of analogous dependence and prolonged abuse.’
Elsewhere we have investigated the intersection of race or ethnicity and gender in respect of the experiences of women who kill their violent spouses and seek to raise defences to murder. Our argument is that it may be impossible to accurately interpret the accused’s behaviour for the purpose of applying the various criminal defences (and attaching labels of ‘reasonable’ or ‘ordinary’ to that behaviour) without attempting to understand the context in which the accused was operating. This context may be powerfully informed by the accused’s position as a woman with a particular ethnic or racial identity. Whilst there have been numerous cases involving women from ethnic, racial or cultural backgrounds to which such an analysis might be of critical importance in the conduct of their self-defence cases, there have been disappointingly few Australian cases recognising this fact.
Research has demonstrated that both the incidence of homicide and patterns of homicide vary substantially by race/ethnicity and gender. This issue is particularly important in Australia where empirical evidence indicates that Aboriginal people are significantly over-represented as victims and offenders in homicide generally and in intimate homicides. Strang reported homicide rates for Aboriginal people as compared with the overall rate for all Australians. She found that Aboriginal victimisation rates were 10 times higher for women and eight times higher for men, and that Aboriginal homicide offence rates were 14 times higher for men and 13 times higher for women. In most homicides, victims and offenders shared the same racial backgrounds. Strang also found that 51 percent of all Aboriginal homicide victims were killed by a spouse as compared with only 21 percent for ‘white’ Australians.
Aboriginal women are likely to face significant obstacles in their attempts to deal with violence directed towards themselves and their children. They are more likely than other Australian women to live in remote and poorly resourced locations and to have limited access to information and material resources. They also face serious racism which may limit their recourse to assistance to deal with domestic violence. Among the Australian cases reviewed for this article the authors found a markedly disproportionate number involving Aboriginal women. In a number of these cases the women had been living in remote areas (Gilbert, Secretary) or in very difficult life circumstances on the urban fringes (R v Gadd, R v Hickey, R v Kina) and typically the levels of violence they had endured were extreme. In each of these cases, there was evidence of prior serious violence by the deceased man and of the women’s attempts to seek assistance from family, health services and sometimes the criminal justice system.
Race and ethnicity may mark out different life circumstances, experiences and options for women. Race and ethnicity may also be associated with particular stereotyped understandings of women. Allard analyses the manner in which race plays a significant role in ‘cultural notions of who are “good” women and “bad” women’. She also highlights the manner in which stereotyped understandings of ‘the battered woman’ and of ‘battered woman syndrome’ have developed with reference to ‘limited societal constructs of appropriate behaviour for white women’ to the exclusion of the experiences of non-white women.
The limited research available concerning race as a factor in trials of battered women charged with killing an abuser suggests marked discrepancies in outcomes for ‘black’ and ‘white’ defendants. For instance, in a US study of 96 cases in which she had testified on behalf of battered women who had killed their partner in self-defence, Walker found that women of colour were twice as likely to be convicted and typically received longer prison sentences than other women. However, the relatively small number of Australian cases limits the prospects for similar comparative research in Australia.
Kirby J in Osland is the first Australian judge to acknowledge the relevance of race and ethnicity in the context of defending battered women who kill their abusers, although his comments are made in obiter dicta. He cites Malott and Beri to support his opinion that:
As a construct, BWS may misrepresent many women’s experiences of violence. It is based largely on the experiences of caucasian women of a particular social background. Their ‘passive’ responses may be different from those of women with different economic or ethnic backgrounds.
Although this is a step in the right direction, we would argue that Kirby J’s comment is limited by a stereotyped understanding of battered women as passive and also does not go far enough. First, on one reading, it remains embedded in the notion that it is the psyche of the women concerned that is the important thing rather than their social context. Whilst we have raised the possibility that women from different cultural backgrounds might respond to violence in different fashions, of much greater significance is understanding how the intersection of their gender and ethnic identity might impact on the range of social options available to them in coping with the violence, and might also affect the way in which they are constructed and their behaviour is understood by the legal system. Secondly, Kirby J’s comment is confined to the operation of BWS evidence. He fails to acknowledge that the law, particularly the law on self-defence, whilst maintaining the illusion of race blindness, has a highly differential impact on racialised subjects.
Some positive outcomes have been achieved for some women in battered women’s defence cases in Australia. These include a small number of acquittals, some non-custodial sentences and clemency in one case. They also include developments in the doctrine of self-defence that make the defence more accessible to such women and some indications in recent case law that the courts might be open to taking a more expansive approach to the issue of expert testimony in this context. However, there is a need to go further.
As in the US and Canada the recognition that there exists a valid and credible body of information concerning the impact of battering, and that such information is vital to decisions made at all stages of the criminal justice process, would be a first step. Once it is recognised that such information has a useful role to play within the legal system, then the questions of what should be the nature of the information to be admitted in the courts and what rules should govern its use can be considered. In the US such a review has found BWS to be inadequate and inappropriate. To date the Australian courts typically have forgone the opportunity to benefit from ‘social framework evidence’ concerning battering, its effects and the context in which it occurs. This is to the detriment of defendants or complainants who come before the courts in contexts arising from domestic violence. It is also to the detriment of Australian jurisprudence.
It is also the case in Australia that we don’t yet have a leading judgment which expounds in a clear and informed way the manner in which self-defence operates in the context under consideration, including the relationship between the legal principles and any expert testimony which might be introduced. Until we have an authority like Wanrow in the US or Lavallee in Canada then, in spite of the best efforts of individual counsel, experts who are informed about the issues and some enlightened trial judges, we will continue with a great deal of inconsistency and confusion.
[*] BA (Hons) (Wollongong), MA (Criminology) (Toronto); Associate Professor, Faculty of Law, University of Sydney. Julie Stubbs would like to thank Sue Osthoff and the staff of the National Clearinghouse for the Defense of Battered Women, Philadelphia, (‘NCDBW’) for their wonderful hospitality, advice and access to resources during her visit. Any errors in the commentary on US developments are Julie’s responsibility alone. This article does not necessarily reflect the views of NCDBW. The authors would also like to thank Zoe Rathus and Cathy Miller for their assistance with research materials and other information.
[†] LLB (Hons) (Auckland), LLM (Harvard); Senior Lecturer in Law, Faculty of Law, Auckland University.
 We have reviewed the early decisions concerning BWS in Australia elsewhere: Julie Stubbs and Julia Tolmie, ‘Battered Woman Syndrome in Australia: A Challenge to Gender Bias in the Law?’ in Julie Stubbs (ed), Women, Male Violence and the Law (1994) 192; Julie Stubbs and Julia Tolmie, ‘Race, Gender and the Battered Woman Syndrome: An Australian Case Study’ (1995) 8 Canadian Journal of Women and the Law 122; Elizabeth Sheehy, Julie Stubbs and Julia Tolmie, ‘Defending Battered Women on Trial: The Battered Woman Syndrome and Its Limitations’ (1992) 16 Criminal Law Journal 369.
 We are, of course, focusing in our argument on the best practice in the US and Canada. This is not to say that the North American experience is uniform, unproblematic or ideal. A number of scholars have noted Canadian decisions in which the principles in R v Lavallee (see discussion of these principles below Part II(C)(4)) have not been applied sympathetically: see, eg, Elizabeth Sheehy, ‘Battered Woman Syndrome: Developments in Canadian Law after R v Lavallee’ in Julie Stubbs (ed), Women, Male Violence and the Law (1994) 174; Martha Shaffer, ‘The Battered Woman Syndrome Revisited: Some Complicating Thoughts Five Years after R v Lavallee’ (1997) 47 University of Toronto Law Journal 1. Shaffer has found that ‘Lavallee does not appear to have led to a dramatic increase in successful self-defense claims by women’: at 17. In the US, some states continue to have a restrictive view of when testimony concerning the effects of battering might be admissible. For example, 40 percent of states have required that a woman must be offering a defence of self-defence before such testimony can be admitted (although note that in 29 percent of states such evidence has been admitted in ‘sleeping aggressor cases’ and in 20 percent of states the evidence has been admitted in ‘hire to kill’ cases: Janet Parrish, ‘Trend Analysis: Expert Testimony on Battering and Its Effects in Criminal Cases’ in US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence Concerning Battering and Its Effects in Criminal Trials: Report Responding to Section 40507 of the Violence Against Women Act (1996) pt II, 19–20.) In addition, it has been estimated that in the US between 72 and 80 percent of women accused of killing their batterer are convicted or accept a plea: Elizabeth Schneider, ‘Resistance to Equality’ (1996) 57 University of Pittsburgh Law Review 477, 520, citing statistics from the NCDBW; and a review of appellate decisions found that 63 percent of the convictions were upheld even though expert testimony on battering and its effects was admitted in 71 percent of the cases: Janet Parrish, ‘Trend Analysis: Expert Testimony on Battering and Its Effects in Criminal Cases’ in US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence Concerning Battering and Its Effects in Criminal Trials: Report Responding to Section 40507 of the Violence Against Women Act (1996) pt II, 7.
 An extension of the use of BWS testimony to other defences, and/or to non-criminal contexts, followed in many parts of North America.
 Holly Maguigan, ‘A Defense Perspective on Battered Women Charged with Homicide: The Expert’s Role During Preparation for and Conduct of Trials’ (Paper prepared for the National Association of Women Judges, no date) 9, drawing on Laurens Walker and John Monahan, ‘Social Frameworks: A New Use of Social Science in Law’ (1987) 73 Virginia Law Review 559, 560.
 Elsewhere we have suggested that it would be appropriate for defence counsel to pose and for courts to ask some or all of the following questions: What was the nature and extent of the violence she suffered in the relationship? How many times had she called the police and with what result? How had she tried to enlist the protection of the criminal justice system or other agencies and what was the result? How many times had she tried to leave? If she returned, what were the factors that influenced her decision? Did she have a safe and affordable place to go? Was it reasonable to expect her to be the one to leave the family home? How had he responded to her efforts to protect herself in the past? Had he intimated what he might do to her in the future? Was there anything about her cultural circumstances that made it particularly difficult for her to detach from him, negotiate the relationship or seek outside help? See Stubbs and Tolmie, ‘Race, Gender and the Battered Woman Syndrome’, above n 1, 194. See also Sheehy, Stubbs and Tolmie, above n 1.
 State of Washington v Wanrow, 88 Wash 2d 221; 559 P 2d 548 (1977) (‘Wanrow’); Elizabeth Schneider, ‘Equal Rights to Trial for Women: Sex Bias in the Law of Self-Defense’ (1980) 15 Harvard Civil Rights — Civil Liberties Law Review 623.
 Elizabeth Schneider, ‘Describing and Changing: Women’s Self-Defense Work and the Problem of Expert Testimony on Battering’ (1986) 9 Women’s Rights Law Reporter 195. See also Elizabeth Schneider, ‘Particularity and Generality: Challenges of Feminist Theory and Practice in Work on Woman-Abuse’ (1992) 67 New York University Law Review 520.
 Elizabeth Schneider, ‘Resistance to Equality’, above n 2, 485–6.
 Holly Maguigan, ‘Battered Women and Self-Defense: Myths and Misconceptions in Current Reform Proposals’ (1991) 140 University of Pennsylvania Law Review 379, 383; Maguigan, ‘A Defense Perspective on Battered Women Charged with Homicide’, above n 4, 1–2.
 Maguigan, ‘Battered Women and Self-Defense’, above n 9, 432. See also Schneider, ‘Resistance to Equality’, above n 2, 500. Maguigan suggests that in the US ‘there is no single more important fair-trial determinant for any defendant than the rules which govern a judge’s decision on whether to allow the jury to consider the defense asserted’: Maguigan, ‘A Defense Perspective on Battered Women Charged with Homicide’, above n 4, 13.
 Parrish, above n 2, 32.
 SC CODE ANN s 17-23-170 (Law Co-op Supp 1995).
 SC CODE ANN s 17-23-170(c) (Law Co-op Supp 1995).
 See Richard McDowell, ‘Battered Spouse Syndrome: Testing the Traditional Limits of South Carolina Law’ (1997) 48 South Carolina Law Review 673.
 See Parrish, above n 2, 25. The states are Georgia, Louisiana, Ohio, Texas and Wyoming.
 Each of which has occurred in some US states: see Parrish, above n 2, 19–20.
 See also Schneider, ‘Resistance to Equality’, above n 2, 512–14.
 Violence Against Women Act of 1994 Pub L No 103-322, 108 Stat 1902 (1994).
 US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence Concerning Battering and Its Effects in Criminal Trials: Report Responding to Section 40507 of the Violence Against Women Act (1996) vii. As noted in that review:
A general framework for the admissibility of expert testimony in criminal cases is provided in Rule 702 of the congressionally enacted Federal Rules of Evidence: If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.
Mary Ann Dutton, ‘Validity of “Battered Woman Syndrome” in Criminal Cases Involving Battered Women’ in US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence Concerning Battering and Its Effects in Criminal Trials: Report Responding to Section 40507 of the Violence Against Women Act (1996) pt I, 20 (emphasis added). The review notes that ‘[s]pecialized knowledge is that based on experience, training, or skill such as that acquired by a battered women’s advocate or shelter worker, counselor, or law enforcement officer. Specialized knowledge is contrasted with scientific knowledge which is derived from scientific research’: at 30 fn 99.
 Dutton, ‘Validity of “Battered Woman Syndrome” in Criminal Cases’, above n 19.
 Parrish, above n 2.
 Mary Ann Dutton, ‘Impact of Evidence Concerning Battering and Its Effects in Criminal Trials Involving Battered Women’ in US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence Concerning Battering and Its Effects in Criminal Trials: Report Responding to Section 40507 of the Violence Against Women Act (1996) pt III.
 US Department of Justice and US Department of Health and Human Services, The Validity and Use of Evidence, above n 19, vi.
 Ibid vii.
 Ibid xii–xiii.
 Parrish, above n 2, 2.
 Holly Maguigan, ‘Review Essay: It’s Time to Move Beyond “Battered Woman Syndrome”’ (1998) 17 Criminal Justice Ethics 50, 56–7.
 Sue Osthoff, ‘Preface’ in Parrish, above n 2, iii.
 Mary Ann Dutton, Expert Testimony in Criminal Cases Involving Battered Women (A resource monograph prepared for the National Association of Women Judges, 1997) 4–5.
 Mary Ann Dutton, ‘Understanding Women’s Responses to Domestic Violence: A Redefinition of Battered Woman Syndrome’ (1993) 21 Hofstra Law Review 1191, 1202. See also Dutton, Expert Testimony in Criminal Cases, above n 29, 20.
 Dutton, Expert Testimony in Criminal Cases, above n 29, 5–6.
 Dutton, ‘Impact of Evidence Concerning Battering and Its Effects’, above n 22, 7.
  1 SCR 852; 55 CCC (3rd) 97 (‘Lavallee’).
 R v Malott  1 SCR 123, 140–1; 155 DLR (4th) 513, 526–7 (‘Malott’).
 Judge Lynn Ratushny, Self-Defence Review: Final Report (submitted to the Minister of Justice of Canada and to the Solicitor General of Canada, 11 July 1997) 26 (emphasis added) <http://canada.justice.gc.ca/Publications/defence/rtush_en.html> .
 Malott  1 SCR 123, 143–4; 155 DLR (4th) 513, 529–30.
 Regina Schuller and Patricia Hastings, ‘Trials of Battered Women Who Kill: The Impact of Alternative Forms of Expert Testimony’ (1996) 20 Law and Human Behaviour 167.
 Sheehy, ‘Battered Woman Syndrome’, above n 2, 174, 178. Other positive consequences noted by Sheehy include the use of BWS evidence in support of a broad ‘no mens rea’ defence (at 175–6), the use of the evidence in a case where the parties were no longer living together and as such it may have been suggested in the absence of that evidence that the threat was no longer current (at 176), and the use of BWS evidence in mitigation of sentence concerning other offences committed by the defendant (in one case, armed robbery committed after her violent partner’s death; in another case narcotics trafficking; in a third case the manslaughter of a non-abusive partner by a woman who had been previously abused) (at 176–8).
 (1995) 22 OR (3rd) 275.
 (Unreported, Yukon Territories Court, 1991) as cited in R v Lalonde (1995) 22 OR (3rd) 275, 281–2.
 R v Lalonde (1992) 22 OR (3rd) 275, 282.
 Ibid 281.
 Stubbs and Tolmie, ‘Battered Woman Syndrome in Australia’, above n 1, 199–200.
 See, eg, R v Hickey (Unreported, Supreme Court of New South Wales, Slattery AJ, 14 April 1992).
 See, eg, R v Gilbert (Unreported, Supreme Court of Western Australia, Scott J, 4 November 1993) (‘Gilbert’).
 See, eg, R v Runjanjic; R v Kontinnen  SASC 2951; (1991) 56 SASR 114 (‘Runjanjic and Kontinnen’); Webb v The Queen (Unreported, Supreme Court of South Australia, Court of Criminal Appeal, White ACJ, Cox and Mohr JJA, 19 June 1992).
 See, eg, R v Woolsey (Unreported, Supreme Court of New South Wales, Newman J, 19 August 1993); R v Taylor (Unreported, Supreme Court of South Australia, Olsson J, 3 February 1994); R v Gunnarsson Weiner (Unreported, Supreme Court of Tasmania, Zeeman J, 13 August 1992); R v Bradley (Unreported, Supreme Court of Victoria, Coldrey J, 14 December 1994); R v McEwen (Unreported, Supreme Court of Western Australia, Walsh J, 18 March 1996) (‘McEwen’).
 R v Kina (Unreported, Supreme Court of Queensland, Court of Appeal, Fitzgerald P, Davies and McPherson JJA, 29 November 1993).
 See the case of Sherrie Lee Seakins: Gay Alcorn, ‘No Trial after Woman Kills Violent Husband’, The Sydney Morning Herald (Sydney), 14 September 1993, 1.
 See, eg, Winnett v Stephenson (Unreported, Magistrates’ Court of the Australian Capital Territory, Burns M, 19 May 1993). See Marion Frith, ‘Battered Woman Acquitted of Fraud’, Canberra Times (Canberra), 21 May 1993, 1. In the New Zealand context see Ruka v Department of Social Welfare  1 NZLR 154.
 Scott v SA Police  SASC 4038; (1994) 61 SASR 589.
 R v Casotti (1994) 74 A Crim R 294.
 R v Gunnarsson Weiner (Unreported, Supreme Court of Tasmania, Zeeman J, 13 August 1992).
 See, eg, Secretary v The Queen (1996) 5 NTLR 96 (‘Secretary’); R v Kina (Unreported, Supreme Court of Queensland, Court of Appeal, Fitzgerald P, Davies and McPherson JJA, 29 November 1993); Webb v The Queen (Unreported, Supreme Court of South Australia, Court of Criminal Appeal, White ACJ, Cox and Mohr JJA, 19 June 1992); R v Casotti (1994) 74 A Crim R 294; J v The Queen (1994) 75 A Crim R 522; Osland v The Queen  HCA 75; (1998) 159 ALR 170 (‘Osland’).
 Ian Freckelton and Hugh Selby list numerous questions yet to be resolved by Australian courts concerning expert evidence and BWS: Ian Freckelton and Hugh Selby, Expert Evidence, vol 1 (at April 1999) 13 Novel Psychological Evidence, ‘13.30 Battered Woman Syndrome Evidence’ [13.100].
 Osland  HCA 75; (1998) 159 ALR 170.
  1 SCR 852; 55 CCC (3rd) 97.
 88 Wash 2d 221; 559 P 2d 548 (1977).
 See, eg, Rebecca Bradfield, ‘Is Near Enough Good Enough? Why Isn’t Self-Defence Appropriate for the Battered Woman?’ (1998) 5 Psychiatry, Psychology and Law 71; Ian Freckelton, ‘When Plight Makes Right — The Forensic Abuse Syndrome’ (1994) 18 Criminal Law Journal 29; Gail Hubble, ‘Feminism and the Battered Woman: The Limits of Self-Defence in the Context of Domestic Violence’ (1997) 9 Current Issues in Criminal Justice 113; Ian Leader-Elliot, ‘Battered But Not Beaten: Women Who Kill in Self Defence’  SydLawRw 38; (1993) 15 Sydney Law Review 403; Therese McCarthy, ‘“Battered Woman Syndrome”: Some Reflections on the Invisibility of the Battering Man in Legal Discourse, Drawing on R v Raby’ (1994) 4 Australian Feminist Law Journal 141; Stella Tarrant, ‘Provocation and Self Defence: A Feminist Perspective’ (1990) 15 Legal Service Bulletin 147; Patricia Easteal, ‘Battered Woman Syndrome: Misunderstood?’ (1992) 3 Current Issues in Criminal Justice 356; Sheehy, Stubbs and Tolmie, above n 1; Stubbs and Tolmie, ‘Race, Gender and the Battered Woman Syndrome’, above n 1; Julia Tolmie, ‘Pacific-Asian Immigrant and Refugee Women Who Kill Their Batterers: Telling Stories that Illustrate the Significance of Specificity’  SydLawRw 25; (1997) 19 Sydney Law Review 472; Julie Stubbs and Julia Tolmie, ‘Feminisms, Self-Defence, and Battered Women: A Response to Hubble’s “Straw Feminist”’ (1998) 10 Current Issues in Criminal Justice 73; Stanley Yeo, ‘Battered Women: In Between Syndrome and Conviction’ (1992) 4 Current Issues in Criminal Justice 75.
 See, eg, Runjanjic and Kontinnen  SASC 2951; (1991) 56 SASR 114; J v The Queen (1994) 75 A Crim R 522; Secretary (1996) 5 NTLR 96; Muy Ky Chhay v The Queen (1994) 72 A Crim R 1 (‘Muy Ky Chhay’).
 Julia Tolmie, ‘Secretary’ (1996) 20 Criminal Law Journal 223, 227–8.
 For example, Gaudron and Gummow JJ in Osland  HCA 75; (1998) 159 ALR 170, 185 commented that BWS evidence ‘has been received in South Australia, New South Wales, Tasmania [and] the Northern Territory’. In fact it has also been accepted in the Australian Capital Territory, Queensland, Victoria and Western Australia.
 Parrish, above n 2, 23.
 Ross has argued strongly that many of the public interest factors listed in the Director of Public Prosecution’s guidelines in the Northern Territory (which are similar to those in other jurisdictions in Australia) ‘could lead to a conclusion in certain cases that prosecutors should exercise their discretion and not prosecute’: Stan Ross, ‘Battered Wife Syndrome and the Role of Lawyers’  LawIJV 357; (1998) 72(11) Law Institute Journal 39, 40.
  SASC 2951; (1991) 56 SASR 114.
 Ibid 118 (King CJ).
 Ibid 120.
 See the discussion in R v Singleton (1994) 72 A Crim R 117. See also Ian Freckelton and Hugh Selby, Expert Evidence, above n 56, [13.30]ff.
 Evidence Act 1995 (Cth) s 80. See the discussion in Stephen Odgers, Uniform Evidence Law (1995) 129–32. The authors have discussed the limitations of the construction of BWS arising from Runjanjic and Kontinnen elsewhere: Stubbs and Tolmie, ‘Battered Woman Syndrome in Australia’, above n 1, 205.
  HCA 75; (1998) 159 ALR 170.
 At one juncture he summarised the expert testimony as indicating that BWS was a
‘collection of responses, thoughts, feelings and attitudes’ which compelled the woman who was the subject of the syndrome meekly to comply with the batterer’s wishes without giving any logical thought to her own rights. It entailed a belief on the part of the woman that she could never free herself from the enforced domination of the man, and an induced fear that, if she did not comply with the wishes of the man, it would only serve to worsen her plight. She thus becomes the subject of a ‘learnt helplessness’ which drives her into submission, accompanied by a ‘learnt hopefulness’ that, if she submits, matters will improve.
Ibid 234. At another point he said that the ‘typical features’ of a ‘woman in the position of the appellant’ included ‘emotional and physical submission, reticence in revealing acts of cruelty and threats by a domineering male to a woman, and an ultimate inability to bear any more of these’: at 241–2.
 In support of this extraordinary claim Callinan J cited a passage from Malott  1 SCR 123, 133–4; 155 DLR (4th) 513, 521–2 in which Major J refers to BWS evidence as a ‘defence’ but is clearly contemplating that it be used in the context of a self-defence claim: Osland  HCA 75; (1998) 159 ALR 170, 242–3. However, this misunderstanding of the Canadian position seems to have arisen from a Crown submission arguing this to be the case in Canada, an argument dismissed by Kirby J: at 217.
 ‘The submission for the appellant that this Court should adopt a new and separate defence of battered woman syndrome goes too far for the laws of this country’: Osland  HCA 75; (1998) 159 ALR 170, 243.
 They summarise the expert witness testimony in the following terms:
 Ibid 185.
 Presumably by modifying the ordinary person test for this first limb of its operation. See Stingel v The Queen  HCA 61; (1990) 171 CLR 312.
 Although they did see ‘the history of the particular relationship’ as bearing on the objective question of reasonableness: Osland  HCA 75; (1998) 159 ALR 170, 185.
 Ibid 215 (emphasis added).
 Ibid 217–18.
 Ibid 216, citing Phyllis Crocker, ‘The Meaning of Equality for Battered Women Who Kill Men in Self-Defense’ (1985) 8 Harvard Women’s Law Journal 121, 149.
 See below Part III(A).
 There are reasons for being cautious about taking a gender neutral approach to what is, in fact, a highly gendered phenomenon. Some of these are potentially raised by defence counsel’s argument in R v Manly (Unreported, Supreme Court of South Australia, Nyland J, 17 March 1995). In R v Manly, defence counsel tried to argue that the relationship between the accused and the 19 year-old woman he killed produced a state of mind in him equivalent to BWS. This is because the deceased was unfaithful to him and verbally abusive. Fortunately, the comparison was soundly rejected by the court.
 (1994) 72 A Crim R 1. It might be argued that such a recognition is implicit in the judgment of Mildren J in Secretary (1996) 5 NTLR 96, 105–7, because he cites extensively from Lavallee, particularly those portions of the judgment that discuss the gender implications of a requirement of imminence in relation to self-defence.
 Muy Ky Chhay (1994) 72 A Crim R 1, 11.
 See, eg, R v Bradley (Unreported, Supreme Court of Victoria, Coldrey J, 14 December 1994); R v McEwen (Unreported, Supreme Court of Western Australia, Walsh J, 18 March 1996); R v Tassone (Unreported, Supreme Court of the Northern Territory, Gray J, 26 April 1994) (‘Tassone’); R v Hickey (Unreported, Supreme Court of New South Wales, Slattery AJ, 14 April 1992); Runjanjic and Kontinnen  SASC 2951; (1991) 56 SASR 114; R v Terare (Unreported, Supreme Court of New South Wales, Levine J, 20 April 1995); R v Raby (Unreported, Supreme Court of Victoria, Teague J, 22 November 1994). See also the discussion in R v Singleton (1994) 72 A Crim R 117.
 (1994) 75 A Crim R 522.
 Ibid 530 — although note the unusual use of BWS in that case. BWS was argued in an attempt to rehabilitate the credibility of a victim claiming rape by her father to explain why she did not leave the family home as an adult and why she sent Christmas cards to her father when she was away.
  NTSC 158; (1995) 129 FLR 39 (trial).
 Ibid 64.
 The expert in Secretary also provided some testimony concerning the social circumstances of the accused: ibid 63.
 (Unreported, Supreme Court of New South Wales, Dunford J, 25 August 1995) (‘Lyons’).
 The expert also identified ‘eight types of behaviour which Amnesty International has described as psychological torture’ which can appear in battering relationships. For example, controlling the person’s social contact, exhaustion (sleep and food deprivation), manipulating the way the person perceives reality, threatening to kill a person and then their family, humiliation, administering drugs and alcohol, inducing an altered state of consciousness by disturbing a person’s ability to think clearly, indulging the person for short periods of time: Transcript of Proceedings, Lyons (Supreme Court of New South Wales, Dunford J, commencing 21 August 1995, 25 August 1995) 271.
 These he identifies as: initial surprise (no warning of the violence); that some of the bashing is unpredictable; that the men show an overwhelming and irrational level of jealousy; that the relationship is characterised by types of sexual expression by the man that most people would regard as strange; that certain battering instances stand out in a woman’s memory; that the women characteristically hide the fact that they have been beaten; that almost all of the men characteristically abuse alcohol or drugs; that the men employ extreme psychological abuse; that the threats are not only against the woman but also her family; that the men put an extraordinary degree of terror into the relationship and find a variety of ways to do this; omnipotency (the woman comes to believe that the man has some very unusual power); and, finally, that the women live with an ongoing awareness that the person they are with is quite capable of killing them: ibid 266–70.
 Ibid 270. Later he also described the concepts of learned hopefulness, learned helplessness, and other psychological effects (such as depression and the ‘stop home syndrome’ — a concept that Kirby J in Osland  HCA 75; (1998) 159 ALR 170, 217 labels ‘traumatic bonding’): ibid 273–4.
 (Unreported, Supreme Court of the Northern Territory, Gray J, 26 April 1994).
 Transcript of Proceedings, Tassone (Supreme Court of the Northern Territory, Gray J, commencing 26 April 1994, 28 April 1994) 237.
 Ibid 244.
 Ibid 243.
 Ibid 267 (emphasis added).
 Osland  HCA 75; (1998) 159 ALR 170, 216.
 Ibid 217.
 Ibid 214.
 Such a possibility is also supported by a passage from the New Zealand case of Ruka v Department of Social Welfare  1 NZLR 154, 173–4, which Kirby J cites with approval: Osland  HCA 75; (1998) 159 ALR 170, 214. The first part of this passage explicitly acknowledges the possibility of the further development of current understandings and knowledge:
There is a danger that in being too closely defined, the syndrome will come to be too rigidly applied by the Courts. Moreover, few aspects of any discipline remain static, and further research and experience may well lead to developments and changed or new perceptions in relation to the battering relationship and its effects on the mind and will of women in such relationships.
 Sheehy, Stubbs and Tolmie, above n 1, 178–9; Stubbs and Tolmie, ‘Battered Woman Syndrome in Australia’, above n 1, 197; Stubbs and Tolmie, ‘Race, Gender and the Battered Woman Syndrome’, above n 1, 154. However, as noted by Freckelton and Selby, above n 56, [13.100], ‘[n]ot every psychologist or psychiatrist or sexual assault centre worker could properly claim to have the requisite expertise to offer informed views on battered woman syndrome’.
 In R v Buzzacott (Unreported, Supreme Court of South Australia, Bollen J, 15 July 1993) despite concerns expressed by the Crown, Bollen J accepted a refuge administrator as a ‘practical expert’: at 269. See also Cathy Miller, ‘Domestic Violence — The Issues in Action for the Expert Witness’ (Paper presented at the Challenging the Legal System’s Response to Domestic Violence Conference, Brisbane, 23–26 March 1994). In Winnett v Stephenson (Unreported, Magistrates’ Court of the Australian Capital Territory, Burns M, 19 May 1993), a criminologist was permitted to testify concerning BWS: see Patricia Easteal, Kate Hughes and Jacki Easter, ‘Battered Women and Duress’ (1993) 18 Alternative Law Journal 139, 139.
 (Unreported, Supreme Court of Queensland, Moynihan J, 27 March 1995).
 See Transcript of Proceedings, R v Gadd (Supreme Court of Queensland, Moynihan J, commencing 27 March 1995) 189–98. There was also testimony from a psychiatrist to the effect that Gadd had an unusual physical propensity to an intoxication response from alcohol consumption and had a physical instability of the brain that was exacerbated by alcohol consumption: at 148–51.
 This was backed up in the case by an extensive history of the violence the accused had received from the deceased. The accused herself and a series of doctors and health workers testified as to this history: see Transcript of Proceedings, R v Gadd (Supreme Court of Queensland, Moynihan J, commencing 27 March 1995).
 The same witness had provided affidavit evidence in the final appeal in R v Kina (Unreported, Supreme Court of Queensland, Court of Appeal, Fitzgerald P, Davies and McPherson JJA, 29 November 1993), but the Crown had objected to it on the grounds that as a social worker she was not appropriately qualified because the necessary expertise lies in psychology. The Court chose not to rule on that evidence because the witness had not been cross-examined concerning her qualifications. Nonetheless the Court seemed to rely on her arguments in their final decision: Miller, above n 108, –.
 Osland  HCA 75; (1998) 159 ALR 170, 186.
 Ibid 219.
 Malott  1 SCR 123, 133–4; 155 DLR (4th) 513, 521–2.
 See, eg, Bradfield, above n 60, 76–7; Nan Seuffert, ‘Battered Women and Self-Defence’ (1997) 17 New Zealand Universities Law Review 292, 299–300.
 Stubbs and Tolmie, ‘Battered Woman Syndrome in Australia’, above n 1, 195.
  HCA 26; (1987) 162 CLR 645. In order to raise self-defence the accused must demonstrate that they believed on reasonable grounds that it was necessary in self-defence for them to do what they did.
 In addition to the cases we discuss below, see also R v Kontinnen (Unreported, Supreme Court of South Australia, Legoe J, 30 March 1992), which was a case involving a sleeping perpetrator who was shot. In that case the accused successfully raised self-defence using BWS evidence.
 The relevant provisions are ss 28(f) and 187: Director of Public Prosecutions v Secretary  NTSC 158; (1995) 129 FLR 39, 65.
 In the sense that it causes reasonable apprehension of death or grievous bodily harm.
 Criminal Code 1983 (NT) s 187(b).
 (Unreported, Supreme Court of the Northern Territory, Gray J, 20 April 1994).
 Transcript of Proceedings, Tassone (Supreme Court of the Northern Territory, Gray J, 28 April 1994) 246, 250–3.
 Ibid 171.
 Perhaps Gray J was influenced by the fact that he was the trial judge who was overturned by the High Court in Zecevic v Director of Public Prosecutions  HCA 26; (1987) 162 CLR 645: ibid 172.
 Transcript of Proceedings, Tassone (Supreme Court of the Northern Territory, Gray J, 28 April 1994) 174.
 (1996) 5 NTLR 96.
 Ibid 100.
 Ibid 103.
 Angel and Mildren JJ forming the majority, Martin CJ dissenting.
 Secretary (1996) 5 NTLR 96, 104.
 Ibid 96.
 ‘R v Secretary’ (1997) 18(2) The Legal Reporter: Criminal Law 7. See also Ross, above n 65, 30.
 (Unreported, Cairns Circuit Court, Derrington J, 18 June 1996) (‘Stjernqvist’).
 Transcript of Proceedings, Stjernqvist (Cairns Circuit Court, Derrington J, 18 June 1996) 153, 165, 174.
 Osland  HCA 75; (1998) 159 ALR 170, 221. See also Gleeson CJ in the New South Wales Court of Criminal Appeal in Muy Ky Chhay (1994) 72 A Crim R 1, 11. He commented that:
The orientation of the law towards relief of the plight of males, rather than females, was also noted in the area of self-defence. It was discussed, for example, in the judgment of the Supreme Court of Canada in Lavallee (1990) 76 CR (3d) 329. The leading judgment in that case was written by Wilson J. She observed that the law catered much better for the position of a person against whom another person’s hand was raised in sudden threat or anger, than for a person who, over a lengthy period, has become sensitised to danger from her batterer and who ought not to be required to wait until a knife is uplifted, a gun is pointed, or a fist is clenched, before her apprehension of danger is deemed reasonable.
 Osland  HCA 75; (1998) 159 ALR 170, 221, citing Lavallee  1 SCR 852, 876; 55 CCC (3rd) 97, 115.
 Rowena Besley, Empirical Study of Battered Woman Syndrome (Honours thesis, University of Sydney, 1996). One US study found that in 54 percent of spouse killings police had previously intervened five times or more in the previous two years: Police Foundation, Washington DC, Domestic Violence and the Police: Studies in Detroit and Kansas City (1996). See also Women’s Coalition Against Family Violence, Blood on Whose Hands? The Killing of Women and Children in Domestic Homicides (1994); Martha Mahoney, ‘Legal Images of Battered Women: Redefining the Issue of Separation’ (1991) 90 Michigan Law Review 1; Alison Wallace, Homicide: The Social Reality (1986).
 Malott  1 SCR 123, 142; 155 DLR (4th) 513, 528 (L’Heureux-Dubé J).
 See Miranda Kaye and Julia Tolmie, ‘“Lollies at a Children’s Party” and Other Myths: Violence, Protection Orders and Fathers’ Rights Groups’ (1998) 10 Current Issues in Criminal Justice 52.
 Schuller and Hastings, above n 38, 172 (citations omitted).
 Ibid 184.
 Besley, above n 140.
 Ibid 13, 19.
 (1997) 91 A Crim R 356 (‘Lock’). See also Pamela Lesmond, ‘I Killed the Love of My Life’, New Idea (Melbourne), 25 May 1997, 16–17.
 Lock (1997) 91 A Crim R 356, 362.
 Ibid. See also Rebecca Dobash and Russell Dobash, Violence Against Wives: A Case Against the Patriarchy (1979). They comment that ‘the idea of provocation is a very powerful tool in justifying the husband’s dominance and control and in removing moral indignation about his resort to force in securing, maintaining and punishing challenges to his authority’: at 136.
 Lock (1997) 91 A Crim R 356. For example, the judge commented that the main thrust of the defence case was self-defence. This is also reported to be the understanding that the accused has of her trial. See Lesmond, above n 148.
 Hunt CJ at CL held that, in accordance with s 101(2) of the Evidence Act 1995 (Cth), the prejudicial effect of the evidence was not substantially outweighed by any probative value it might have had. For this to be so ‘there must be no reasonable view of the evidence available which is consistent with the innocence of the accused’: Lock (1997) 91 A Crim R 356, 363.
 The three exceptions were comments made by the deceased that he was afraid of the accused’s use of knives (excluded under s 137 of the Evidence Act 1995 (Cth) on the basis that its prejudice outweighed its probative value), an incident where a neighbour took a knife from the accused (excluded on the basis that it had no probative value and was irrelevant), and a comment by the deceased that he had been in hospital for a fortnight after the accused had once stabbed him (excluded under s 65 of the Evidence Act 1995 (Cth) on the basis that it was unlikely to be reliable and that it was in fact inaccurate and therefore highly prejudicial).
 Lock (1997) 91 A Crim R 356, 365 (citations omitted) (emphasis in original).
 Lock (1997) 91 A Crim R 256; R v Stephenson (Unreported, Supreme Court of Queensland, August 1992) and Stjernqvist (Unreported, Cairns Circuit Court, Derrington J, 18 June 1996).
 Osland  HCA 75; (1998) 159 ALR 170, 243.
 Ibid 216–17.
 (Unreported, Cairns Circuit Court, Derrington J, 18 June 1996).
 Ibid 162–7.
 Ibid 153.
 Mahoney, above n 140.
 Stjernqvist (Unreported, Cairns Circuit Court, Derrington J, 18 June 1996) 174.
 Ibid 165.
 Ibid 177.
 (Unreported, Supreme Court of Western Australia, Scott J, 4 November 1993). Note also that in the cases of Runjanjic and Kontinnen  SASC 2951; (1991) 56 SASR 114, counsel for the defence did lead evidence from a director of a women’s shelter to explain the facilities and the physical problems associated with the shelter alternative. They were not permitted to lead evidence from the director relating to any behavioural characteristics typical of battered women as she was not an ‘expert’. Counsel also tried to get some women who had lived in violent relationships to testify, but it was difficult to find someone prepared to testify and, in any event, after the judge’s ruling on the shelter worker, counsel realised that they would not be permitted to call this evidence: Letter from Counsel (6 January 1994) (copy on file with authors).
 Maguigan, ‘A Defense Perspective on Battered Women’, above n 4, 2.
 R v Stephenson (Unreported, Supreme Court of Queensland, August 1992); Stjernqvist (Unreported, Cairns Circuit Court, Derrington J, 18 June 1996).
 Dutton, Expert Testimony in Criminal Cases, above n 29, 6. She also cautions that ‘[e]xpert testimony based on dated or stereotypic characterizations of battered women, uninformed by recent contributions to the scholarly literature, may solidify, rather than debunk, misconceptions about battered women’: at 6.
 See Patricia Gagné, Battered Women’s Justice: The Movement for Clemency and the Politics of Self-Defense (1998).
 Schneider, ‘Resistance to Equality’, above n 2, 519, notes that at least 26 states have set up committees to review the cases of incarcerated battered women.
 Note that the term clemency is ‘a generic legal term that includes any executive act that reduces or alleviates a penalty for a crime’: Gagné, above n 169, 29. It includes: pardons, commutation of a death sentence to life imprisonment, and reduction of sentence. Typically women had their sentences reduced, often from life imprisonment, to a level that made them eligible for parole — some were released immediately but not all. Of the 106 battered women known to have received clemency, only one was pardoned and one received a conditional pardon: see NCDBW, Battered Women Who Have Received Clemency 1978–1996 (1997) 1. Note that this information may be imprecise — it relies on matters brought to the attention of the Clearinghouse, and since reasons for clemency decisions are not always provided, it cannot be assumed that the woman’s battering history was a factor, or even relevant, in each of these cases. For instance, at least two of the women whose cases are listed appear to have been released on medical grounds. See also Alison Madden, ‘Clemency for Battered Women Who Kill Their Abusers: Finding a Just Forum’ (1993) 4 Hastings Women’s Law Journal 1.
 The Governor of Ohio signed legislation recognising BWS evidence, instigated a review of cases of women imprisoned for crimes against their batterers, and ensured that the parole board was educated about BWS and the recent changes to the law before they commenced their role in the clemency process. See also Linda Ammons, ‘Discretionary Justice: A Legal and Policy Analysis of a Governor’s Use of the Clemency Power in the Cases of Incarcerated Battered Women’ (1994) 8 Journal of Law and Policy 1, 20–1; and Schneider, ‘Particularity and Generality’, above n 7, 525 fn 25.
 Gagné has argued that ‘[i]n every state, the success or failure of clemency was dependent upon the political-opportunity structure’: Gagné, above n 169, 130. See also Ammons, above n 172; and Madden, above n 171.
 Ratushny, Self Defence Review, above n 36, 5.
 Ibid 62.
 The following discussion is taken from the application for conditional pardon, prepared on behalf of Waugh (December 1994) (copy on file with authors).
 Simon Rice, ‘Additional Submission on Behalf of Ms Waugh’ (January 1995) 1 (copy on file with authors).
 For example, whether it is appropriate to understand the phenomenon of violence within the framework developed for heterosexual couples and, if not, what framework should be used to analyse and interpret it.
 Barbara Hart, ‘Lesbian Battering: An Examination’ in Kerry Lobel (ed), Naming the Violence: Speaking Out about Lesbian Battering (1986) 173; Debra Reid, ‘Am I Being Recognized?’ (1993) 1(2) The Network News [newsletter of the Network for Battered Lesbians, US] 1 (note that Reid was convicted of killing her abusive female partner — she was one of the ‘Framingham Eight’, eight women petitioning for commutation of sentences. She argues that the Board of Pardons was unable to perceive her as a battered woman); Claire Renzetti, Violent Betrayal: Partner Abuse in Lesbian Relationships (1992); Ruthann Robson, ‘Lavender Bruises: Intra-Lesbian Violence, Law and Lesbian Legal Theory’ (1990) 20 Golden Gate Law Review 567. Note also Foster v Commonwealth, 827 SW 2d 670 (1992), certiorari denied 506 US 921 (1992). The appeal court held that the trial court had erred in admitting BWS because although generally accepted, ‘by its own definition’ it does not apply to lesbian relationships: at 683.
 Catharine Simone, ‘“Kill(er) Man Was a Battered Wife” the Application of Battered Woman Syndrome to Homosexual Defendants: The Queen v McEwen’  SydLawRw 13; (1997) 19 Sydney Law Review 230, 234.
 (Unreported, Supreme Court of Western Australia, Murray J, 25 April 1995 (trial); Walsh J, 18 March 1996 (sentencing)).
 Simone, above n 181, 234–7.
 Arguably too the relationship the accused had to the deceased was understood and explained by reference to traditional heterosexual active–passive gender roles. This imports the idea that ‘[h]eterosexuality is posited as the foundation or the original, upon which homosexuality can only mimic or copy’: ibid 239. Jenni Millbank, among others, has argued that this kind of legal reasoning denies the originality and diversity of gay and lesbian relationships: Jenni Millbank, ‘If Australian Law Opened Its Eyes to Lesbian and Gay Families, What Would It See?’ (1998) 12 Australian Journal of Family Law 99.
 Transcript of Proceedings, McEwen (Supreme Court of Western Australia, Murray J, commencing 18 April 1995) 340 as quoted in Simone, above n 181, 233.
 Osland  HCA 75; (1998) 159 ALR 170, 212–13.
 See Stubbs and Tolmie, ‘Race, Gender and the Battered Woman Syndrome’, above n 1; Tolmie, ‘Pacific-Asian Immigrant and Refugee Women Who Kill’, above n 60.
 Valli Kanuha, ‘Women of Color in Battering Relationships’ in Lillian Comas-Diaz and Beverly Greene (eds), Women of Color: Integrating Ethnic and Gender Identities in Psychotherapy (1994) 428.
 See, eg, R v Hickey (Unreported, Supreme Court of New South Wales, Slattery AJ, 14 April 1992); Secretary (1996) 5 NTLR 96; Muy Ky Chhay (1994) 72 A Crim R 1; R v Gadd (Unreported, Supreme Court of Queensland, Moynihan J, 27 March 1995); R v Terare (Unreported, Supreme Court of New South Wales, Levine J, 20 April 1995); R v Simon (Unreported, Supreme Court of New South Wales, Bruce J, 21 July 1995); R v Varagnolo (Unreported, Supreme Court of New South Wales, McInerney J, 21 March 1996); R v Buzzacott (Unreported, Supreme Court of South Australia, Bollen J, 21 July 1993); Gilbert (Unreported, Supreme Court of Western Australia, Scott J, 4 November 1993).
 One of the rare exceptions is R v Kina (Unreported, Supreme Court of Queensland, Court of Appeal, Fitzgerald P, Davies and McPherson JJA, 29 November 1993).
 For a discussion of the relevant US material see Linda Ammons, ‘Mules, Madonnas, Babies, Bathwater, Racial Imagery and Stereotypes: The African-American Woman and the Battered Woman Syndrome’ (1995) 5 Wisconsin Law Review 1003, 1006–7ff.
 However, note some similarities with the patterns for native Canadians: Sharon Moyer, ‘Homicides Involving Adult Suspects — 1962–1984: A Comparison of Natives and Non-Natives’ (Working Paper No 1987–29, Research Group, Solicitor General of Canada, 1987).
 Heather Strang, Homicides in Australia 1990–91 (1992). Note however, that this data is derived from police statistics in which race is determined on the basis of ‘racial appearance’. The data is therefore an imprecise measure of Aboriginality and should be treated with some caution. Nonetheless, the magnitude of the differences reported between Aboriginal rates and overall rates (see below in text) are very large and consistent with other literature: see Wallace, above n 140; Paul Wilson, Black Death, White Hands (1982).
 Strang, above n 193, 25.
 Ibid 33.
 Ibid 26.
 Ibid 37.
 Pam Greer, ‘Aboriginal Women and Domestic Violence in New South Wales’ in Julie Stubbs (ed), Women, Male Violence and the Law (1994) 64.
 Australian Law Reform Commission, Equality Before the Law: Women’s Access to the Legal System, Report No 67 (1994) 31–2.
 See Human Rights and Equal Opportunity Commission, Racist Violence (1991) ch 5.
 These include R v Hickey (Unreported, Supreme Court of New South Wales, Slattery AJ, 14 April 1992); R v Kina (Unreported, Supreme Court of Queensland, Court of Appeal, Fitzgerald P, Davies and McPherson JJA, 29 November 1993); Secretary (1996) 5 NTLR 96; R v Gadd (Unreported, Supreme Court of Queensland, Moynihan J, 27 March 1995); R v Terare (Unreported, Supreme Court of New South Wales, Levine J, 20 April 1995); R v Simon (Unreported, Supreme Court of New South Wales, Bruce J, 21 July 1995); R v Varagnolo (Unreported, Supreme Court of New South Wales, McInerney J, 21 March 1996); R v Buzzacott (Unreported, Supreme Court of South Australia, Bollen J, 21 July 1993); Gilbert (Unreported, Supreme Court of Western Australia, Scott J, 4 November 1993). There may well be other cases known to the authors in which race was not clearly identified in the reporting of the case, and there may be other cases involving Aboriginal women which have not come to the attention of the authors.
 Sharon Allard, ‘Rethinking Battered Woman Syndrome: A Black Feminist Perspective’ (1991) 1 UCLA Women’s Law Journal 191, 194.
 Ibid 194–5.
 Lenore Walker, ‘Legal Self-Defense Issues for Battered Women of Color’ (Paper presented at the Advanced Feminist Therapy Institute, Seattle, 12–15 May 1988).
 Suzanne Beri, ‘Justice for Women Who Kill: A New Way?’ (1997) 8 Australian Feminist Law Journal 113.
 Osland  HCA 75; (1998) 159 ALR 170, 213–14.