Homicide law reform and gender: Configuring violence

Introduction

This article charts aspects of the engagement by formal law reform agencies with feminist ideas in the context of homicide law reform. This requires, of course, a concentration on violence against women. My article uses law reform work on the provocation defence to map the ways in which violence against women was an apparent driver of reform. It has Victorian law reform as its focus, and concentrates on the various manifestations of independent law reform agencies in Victoria – the Victorian Law Reform Commission (VLRC), and its two predecessor agencies, the Law Reform Commission of Victoria (LRCV) and the Law Reform Commissioner. I explore the thesis that when law reform, at least homicide law reform, is driven by the social context in which the legal phenomenon of interest occurs, one is more likely to get progressive legal change than where reform is driven by legal categories (see Graycar and Morgan, 2005). However, that social context had itself to be configured, and the paper briefly traces the identification of ‘domestic violence’ as a social phenomenon.

The partial defence of provocation

The partial defence of provocation has provided a perennial focus of modern law reform. Provocation, at common law, reduces murder to manslaughter. Broadly, the defence is available where the accused was provoked by the actions of the victim to lose his (or her) self-control and where the ‘ordinary person’ could have also lost self-control and acted in the way the accused did (see Stingel (1990)). The defence has, at least in the last 30 years, been controversial for a series of reasons. Should there be a defence where the perpetrator has formed the intention to kill (see Parker (1964); Johnson (1976))? Is it ‘ordinary’ to respond to provocative conduct with homicidal violence? What should be recognized as causing a perpetrator’s ‘blood to boil’, so that they are ‘out of control’? Are words alone enough (Morgan, 1997)? Did the accused have to act suddenly or could provocation accumulate over time (Parker (1964); Moffa (1977))? Is there a need for a triggering incident (R v R (1981))? What characteristics, if any, should be attributed to the ordinary person (Stingel (1990); Masciantonio (1995))? When the ordinary person test was separated into two aspects concerned with the gravity of the provocation, and the response of the ordinary person, could the jury understand the test? If the test had become so complex that juries could not understand it, should it even remain a defence, at least where the mandatory sentence for murder had been removed? These are all intensely interesting questions for the law reformer. But my interest is whether and how the defence configured violence against women. More particularly, my focus in this article is whether and how formal law reform bodies constructed violence against women as a relevant consideration in their deliberations around reform of defences to homicide. I will touch on many of the perennial questions, but my main focus will be the social context in which spousal homicide occurs, and the extent to which that context constrained, governed, informed or was ignored by law reformers.

Where is the battering?

Catharine MacKinnon commented on the development of the legal claim for sexual harassment, ‘[s]exual harassment, the event, is not new to women. It is the law of injuries that it is new to’ (MacKinnon, 1987: 103). Similarly, women have always been subject to domestic violence, and indeed have been the victim and occasionally, though less often, the perpetrators of domestic homicide, but it took some time for that violence to be recognized as a social phenomenon, and even longer as a problematic social phenomenon apparently relevant to law reform.

First, violence against women – or domestic violence – had to be recognized as something more than an episodic individual failing. Australia’s first women’s refuge, Elsie, was established in Sydney in November 1973 ‘for women and children who have nowhere to sleep’, and by January 1974 it was advertised more specifically for any woman ‘needing to escape a violent home, needing advice and friends when faced with the legal, welfare, health system’ (Lake, 1999: 229). Helen Rhoades et al. have suggested that the first time the ‘prevalence of violence against women was … publicly revealed’ was with the publication of the Royal Commission on Human Relationships in 1977 (Rhoades et al., 2010: 304). The Commission drew on UK research, and conducted two studies of its own – the first of women who had sought refuge at Elsie towards the end of 1975 and early 1976, and the second a ‘phone-in’ the commission undertook in February 1976. ¹ The Commission paints a picture of domestic violence that crosses all classes, and is related to male and female socialization, and more broadly to inequality between women and men, though that language was not expressly used (Royal Commission, 1977: 138–9).

The first Australian scholarly discussion of domestic homicide I have been able to find was presented at a conference on ‘family violence’ in 1979, and published in 1980 (Rod, 1980). This was an analysis of all ‘marital murders’ in NSW police files between 1958 and 1967 and 1976 and 1978. Rod points to what is now a familiar pattern: ‘three quarters of spouse murders in NSW were committed by men; yet husbands were the most likely victims of female killers’ (1980: 95). Rod goes on to observe that even though prior violence was likely to be understated in police files, ‘[a] history of assault was reported in approximately one third of the wife murders and nearly half of the husband murders’ (1980: 96). However, the chapter could hardly be described as embodying a feminist analysis as understood in the 1980s and subsequently. For example, the chapter goes on to describe the context in which most of these killings occur in the language of mutuality of conflict and of ‘quarrels’:

It comes as no surprise that the most frequent precipitating cause in the murder of spouses is the marital relationship itself .… Marital quarrels, particularly in cases where there has been a long history of dispute, often cannot be pinned down to one or two specific issues. The couple have been locked into mutual dependancy [sic] and antipathy for some time, so that the most trivial circumstances can unleash animosity directed at points of greatest vulnerability in each partner. (Rod, 1980: 98)

However, Rod did acknowledge the particular circumstances in which women typically kill their partners (see Morgan, 2002):

Most murders of husbands occurred because the wives reacted violently to the hostile and often violent behavior of their husbands. A number of women killed their husbands in self-defence and some others killed them as a ‘protective reaction’ to avoid being assaulted. (Rod, 1980: 99)

And in relation to men who kill, the vulnerability of women even after separation and the level of pre-planning were also recognized (see Morgan, 1997):

An intriguing problem is posed by the number of men who kill their estranged wives. Most of these murders are carefully planned in advance and the women are often tricked into a rendezvous where they are killed. A divorce may not be sufficient protection for women from the murderous intent of their ex-husbands; nor is distance: one man pursued his former de facto wife from Tasmania to Sydney, and lay in wait for her outside her home where he shot her as she came home from work. (Rod, 1980: 102)

This and other feminist work of the time made no apparent impact on formal law reform bodies in Victoria in the 1970s. In 1979, the Victorian Law Reform Commissioner issued a working paper on ‘Provocation as a Defence to Murder’. In this paper, what might be described as a criticism of aspects of the doctrine of provocation from the perspective of its gendered nature is the following:

It seems to have been accepted without question for centuries that a husband coming upon his wife in adultery and killing either the wife or the adulterer can avail himself of the defence. It might be thought that this is an archaic relic of a husband’s right of possession and that instead of forming a separate and legal category of acts it should be relegated to a place amongst the circumstances surrounding the killing. No court has yet decided that a wife is entitled to the same consideration in the case of sudden discovery of her husband in like situation. (1979: 22–23, para 58)

Here we have neither a recognition of the lack of availability of the provocation defence (or indeed self-defence) to women who kill violent men, nor any criticism of the use of the provocation defence by men in the circumstances outlined by Rod (for example, post separation). To the extent gender features at all, the ‘problematic’ issue is the lack of formal equality for women in the context of finding a spouse in adultery – a situation that might be called ‘equality with a vengeance’ (MacKinnon, 1987: 72) – the conferral of formal equality on women where they neither want nor need it. However, it must be recalled that there was not at the time a great deal of material in the Australian environment analyzing the social facts about domestic homicide, nor, as will be seen below, was domestic violence fully mapped.

The emergence of the ‘battered body’

The discursive process whereby ‘the battered body’ was created has been mapped by Anne Genovese (1997), in the context of the NSW campaign around Violet Roberts. At the 1976 trial of Violet Roberts and her son Bruce, who had been charged with the murder of Eric Roberts (Violet’s husband and Bruce’s father), the history of Eric Roberts’ vicious abuse of Violet over many years did not feature at all in the trial. Genovese states:

It makes me angry in hindsight that material available to Violet’s lawyer was never incorporated fully into the defence. However, it must be noted that the preconditions of that material, the personal history, the microhistory, of Violet’s life as a battered woman could not yet be articulated easily by the public, or by many interpretive systems or discourses, let alone the law .… Women and children had been victims of abuse within their own homes for centuries, but the interplay of discursive interpretation of such hidden stories had not yet been identified, named and given a space within the tangle of narratives that constitute a public consciousness. (Genovese, 1997: 95)

Bruce and Violet Roberts were imprisoned for murder, having not succeeded in establishing their (partial) defence of diminished responsibility. They each received what was then a mandatory sentence of life imprisonment. In 1980 a large and highly successful campaign to release them from prison was undertaken by Women Behind Bars, Redfern Legal Centre and the Feminist Legal Action Group, well described in Ann Genovese’s work (1997, 1998). The material gathered in this campaign was used with interviews with other women in prison for killing violent partners as tools in a broader campaign to seek justice for these women (see Bacon and Lansdowne, 1982). Remember that this was a period when some lawyers argued that evidence of past violence by the deceased should not be introduced as it provided a motive for the killing rather than contributing to a defence (Bacon and Lansdowne, 1982: 84). The work of Women Behind Bars and others was then used by the NSW Premier’s Task Force on Domestic Violence to recommend reform to the law on provocation, and the sentence for murder. The campaign succeeded and it was the first time avowedly feminist understandings of domestic violence influenced institutional law reform, albeit that the law reform body was a specialist reform body focused on the contextual problem of domestic violence and was not just concerned with legal change (NSW Task Force, 1981: 7–71). ²

That these reforms to the Crimes Act 1900 (NSW) — to s 19 (removing the mandatory life sentence for murder) and s 23 (ameliorating the doctrine of provocation, discussed further below) — had domestic violence as their apparent core driver is obvious not just from the Task Force Report but also from the Attorney-General’s second reading speech:

The bill is intended to give effect to widespread community demand for a modification of the law of homicide, particularly the law affecting what may be called crimes of domestic violence. (NSW Legislative Assembly, 11 March 1982: 2482)

In relation to the removal of the mandatory sentence for murder of life imprisonment, and the substitution of a discretionary sentence, the Attorney-General observed:

On the one extreme, there is the cold and calculated killing performed by a so-called hit man for financial gain, done dispassionately and with premeditation. At the other extreme is the kind of case such as that of Violet and Bruce Roberts … where, in a domestic situation involving a history of brutality by the deceased against his wife and son, he was killed. (NSW Legislative Assembly, 11 March 1982: 2483)

The language of this last statement is interesting: it downplays the agency assigned to Violet and Bruce: he was killed, rather than they killed him, arguably an attempt to make them more sympathetic, but ultimately having the additional effect of rendering them passive.

The changes to the substantive provocation law were clearly driven by the situation of battered women who kill (while the removal of the mandatory life sentence had a number of motives (Weisbrot, 1982)). The legislation removed the requirement that the alleged provocation had to occur immediately before the killing (NSW Task Force, 1981: 70–71). The Attorney-General stated:

The current law of provocation is based on a theory of human behaviour which assumes that all people respond to provocation suddenly – as the present section says, in the heat of passion. This is not true. It is certainly not true for women, and it is also not true for men. (NSW Legislative Assembly, 11 March 1982: 2485)

However, we can see once again the pull of formal equality and gender neutrality. The second reading speech focused on what was described as a Tasmanian case from 1968 where a man had killed another man; the deceased, some 18 months before the killing, had made sexual advances to the accused’s wife, been rejected, and then masturbated in front of her. She told her husband who later assaulted the deceased. The wife subsequently suffered a miscarriage, believed to be because of the stress of the sexual approach. Some five months after the first incident, the deceased again visited, and again made sexual advances which were rejected. The accused husband was told. Five days later the accused killed the deceased. It was held that provocation could not apply. That it is this scenario which is used in the second reading speech again indicates a relentless gender neutrality. Other members of parliament, however, return to the situation of women who kill, citing the Victorian Krope case, R v R (discussed below), and the situation of Bruce and Violet Roberts (for example, NSW Legislative Assembly, 30 March 1982: 2984–5, 2988, 2988–9, 2991).

The battered body disappeared?

The approach of the NSW Taskforce and NSW Parliament, contrast with the contemporaneous approach of the Victorian Law Reform Commissioner (‘the Commissioner’). I have noted above the Commissioner’s Working Paper on Provocation. This was followed by a Final Report in early 1982. There is no mention of the Violet Roberts campaign in the report, nor the NSW Premier’s Taskforce, perhaps a reflection of the very state-based nature of criminal law, and media at the time. ³ The Commissioner was however more interested in a case from another jurisdiction, also a cause célèbre of the time, that of R v R (1981) in South Australia. This case involved a woman who had been the target of her husband’s violence over many years and who had just found out that her husband had been sexually abusing her daughters for many years. A day later, he fondled her and expressed his love for her, and said they were going to be ‘one big happy family’ and there was ‘to be no more talk about the girls leaving home’ (R v R (Full Court) (1981): 324). Soon after, when he had fallen asleep, she got up and got an axe and killed him. An attempt was made to run provocation at trial, detailing the abuse that had been inflicted, but the relevance of the evidence was queried by the trial judge who stated:

I am not sure why [the history of abuse] was emphasized by the defence … if there was this long harsh cruel experience suffered by the family … this is not some kind of defence … You have been urged by [Mr Eames] to do ‘common justice’. Well, your duty is to do justice according to the law and if that is uncommon, so be it. (R v R: 23 quoted in Genovese, 1997: 97)

This passage was not cited by the Commissioner; unsurprisingly he concentrated on the Full Court decision, which, by majority, held that provocation should have been left to the jury, and ordered a new trial (R v R (Full Court) (1981)). There is quite detailed coverage of the facts in this case in the Commissioner’s report – the abuse that the accused and her children suffered, including this quote from one of the judges of the Court of Appeal: ‘His conduct towards his family is about as repulsive as it was possible to imagine’ (Law Reform Commissioner, Victoria, 1982: 12 quoting Zelling J in R v R (Full Court) (1981): 328). The report also includes the extract from the Chief Justice’s judgment where he cautions that the law does not countenance revenge killing or killing to ‘avert … some apprehended harm in the future’ (Law Reform Commissioner, Victoria, 1982: 14 quoting King CJ in R v R (Full Court) (1981): 325). The Commissioner’s report concludes the long discussion of the case with the Chief Justice’s statement:

… the deceased’s actions and words in the presence of the appellant on the fatal night might appear innocuous enough on the face of them. They must, however, be viewed against the background of brutality, sexual assault, intimidation and manipulation .… The implication of … [his] words was therefore that this horror would continue and that the girls would be prevented from leaving by forms of intimidation and manipulation which were only too familiar to the appellant. In this context it was, in my opinion, open to the jury to treat the words themselves and the caressing actions which accompanied them as highly provocative and quite capable of producing in an ordinary mother endowed with the natural instincts of love and protection of her daughters, such a loss of self-control as might lead to killing. (1982: 14 quoting King CJ in R v R (Full Court) (1981): 325)

The Commissioner says only, at the end of his extraction of this case: ‘So there the law stands’ (1982: 15). It is unclear from this epigrammatic statement what the Commissioner had in mind. Certainly, the decision appears to have had no influence on his recommendations for law reform in relation to provocation. (Although it is to be noted that the case arises again in the context of a discussion of the recommended introduction of a defence of diminished responsibility, a defence he said the defendant in R v R (1981) would have been entitled to; he also notes that she was acquitted completely at her retrial (see Law Reform Commissioner, Victoria, 1982: 36–37; Loff, 1982).

I have suggested in my heading to this section that perhaps the battered body disappeared in the Victorian Law Reform Commissioner’s consideration of homicide defences, but it may be more accurate to suggest, as Ann Genovese does in a different context, that ‘the formulation of a public discourse around the battered body was still not complete’ (1997: 97). That is, at the time the Commissioner was writing, there was not a fully developed public discourse about domestic violence. Although the NSW Task Force on Domestic Violence had reported, the NSW lobbying had not yet culminated in the legislative amendment to the law on provocation making it clear that there is no need for a specific triggering incident, and that there can be a delay between the alleged provocation and the lethal violence. The inclusion of a great deal of discussion of the case of R v R (1981), and nothing on the NSW situation, may indicate a traditional preference for ‘real law’ – (appeal) cases or legislation, as opposed to social science, lobbying and media coverage. However, at least violence against women was talked about in a serious manner in a traditional law reform agency document, albeit without a label, and where the ‘natural instincts’ of mothers are extolled. At least the (then) innovative approach to the situation of women who kill violent men of the Court of Criminal Appeal in South Australia was given some currency in Victoria. However, the major focus of the discussion and recommendation was on the ordinary person test: the Commissioner recommended that the test be abolished. Yet the ordinary person test is the most obvious way in which the standards that we should hold men (and women) to can be expressed, and to abolish this objective test would have been a backward step: as put by Leader-Elliott, the ordinary person test ‘give[s] body and substance to the moral imagination and provide[s] a vehicle for the exercise of intuitive judgment’ (1996: 96); see also Morgan, 1997: 260).

The Law Reform Commissioner was replaced by a free-standing body, the Law Reform Commission of Victoria (LRCV), in 1984. Its first report, focused solely on the sentence for murder, was published in September 1985 (LRCV, 1985). Violence against women is almost completely invisible, and appears to have had little or no effect on the LRCV’s deliberations. In the context of a consideration of the deterrent effect of a discretionary sentence for murder, the LRCV argues that ‘[m]ost murders occur in circumstances making it unlikely that the murderer will consider penalties’, and the footnote to this statement observes that ‘[f]or example, about 70% of murders arise in the context of domestic tensions, sex-rivalry or other disputes … and over one-third involve family members (including de facto spouses and lovers). This may of course indicate that other, more calculated, killings are deterred by the existence of a severe penalty’ (LRCV, 1985: 11). Thus domestic violence, to the extent it is featured at all, is figured as spontaneous and a product of mere tensions in the family.

The social context of violence against women: Now you see it, now you don’t

The battered body had fully arrived by the time the LRCV produced its 1988 Discussion Paper. After a brief introductory chapter outlining the terms of reference and previous law reform efforts, and announcing that it was undertaking an empirical study of all homicide prosecutions in Victoria between 1981 and 1987, the LRCV moved immediately to discuss the ‘Facts About Homicide’ (LRCV, 1988: 8). It drew on the ground-breaking empirical study of homicide in NSW conducted by Alison Wallace which, among other things, highlighted the extent of domestic homicide, and the predominance of prior violence in intimate partner homicides (Wallace, 1986). The LRCV observed that the patterns were likely to be similar in Victoria. It concluded: ‘These patterns must inform policy with respect to homicide. They bear especially on the ability of the law to deter homicides and may have an impact on the assessment of the legal rules’ (LRCV, 1988: 11, emphasis added). In its discussion of the facts of homicide, the LRCV included a heading ‘Domestic Killings’, where it summarized Wallace’s research, noting that in the spouse killings, in close to 50% of cases there was evidence of prior violence. These data are returned to again in a chapter headed ‘Reform’, where the LRCV makes the point that the ‘range of measures designed to alleviate the problems of domestic violence should also contribute to a reduction in domestic homicides’ (LRCV, 1988: 39) drawing attention to the Victorian report, Criminal Assault in the Home (Department of Premier and Cabinet, 1985).

It is difficult to discern the extent to which, if any, the empirical work on who kills whom and why (see Morgan, 2002) informed the recommendations on reform to the laws surrounding homicide. In relation to provocation, the LRCV canvasses some of the arguments for and against the abolition of provocation. It uses three scenarios ‘to provide a focus’ for those arguments, none of which involves the ‘classic’ factual scenario of a man killing his wife because she has left him (Morgan, 2002), or threatened to leave him (LRCV, 1988: 53). However, one involves the killing by a woman of a violent man. By failing to include here the ‘classic’ factual scenario, I wonder about the extent of the commitment to a critique that the defence is misused by men.

The LRCV produced both a lengthy empirical analysis (LRCV, 1991) and a final report on this reference (LRCV, undated). In the final report, after addressing the question of whether the homicide rate is increasing, there is a lengthy chapter entitled ‘Preventing Domestic Homicides’, not directly related to homicide law. Much of the material is directed to policing. Equality for women gets a brief mention (LRCV, undated: 21) in relation to changing traditional role models for men as a prevention strategy.

In relation to provocation, the LRCV did state, in presenting one of the reasons why one might abolish provocation:

that the provocation defence operates in practice predominantly as an excuse for men who kill women .… [T]he vagueness of the criteria allows juries to act upon attitudes which devalue the lives of women victims. Secondly, the law’s emphasis on the suddenness of the defendant’s reaction to provocation means that the defence is consistent with patterns of male response to emotional trauma, but not necessarily with female reaction patterns which may not be so immediate. Under the guise of general rules, the provocation defence operates to excuse aggressive male behavior. (undated: 70)

I am not sure that this makes the focus of concern transparent: is it a concern with the lack of availability of the defence to women, or a concern about the use of the defence by men? It is informative to consider the sources referred to by the LRCV in this context (undated: 70, note 126). The LRCV refers to work by MacKinnon (1982), which does not in fact deal with the killing of women by men, the Victorian report Criminal Assault in the Home, which in relation to homicide is concerned with the lack of availability of defences to women, and a report by the South Australian Domestic Violence Council proposing a complete defence for women who kill after a long history of domestic violence. That is, it seems likely that, in common with previous reports, the focus remained on women who kill, rather than on men who kill and then rely on provocation. In other words, to the extent gender is recognized as an issue, it is in the ‘inequality of access’ to the defence, rather than to the differently gendered social contexts in which women and men kill.

The report also emphasizes what was described as the ‘principal argument for abolishing the defence of provocation’ (undated: 69) — that offenders who avail themselves of the provocation defence do have the intention to kill or commit really serious injury. It notes that, once the mandatory sentence for murder had been removed, any factors which reduce culpability could be considered in sentencing. In this report, the LRCV does include as relevant scenarios the basic facts of Moffa and Parker — both involving a sexual jealousy provocation scenario.

The LRCV, in the context of noting arguments against the abolition of provocation, argued: ‘The claim that provocation is gender biased requires empirical evidence to support it’ (undated: 73). Relying on the data in their Prosecutions Study, they noted:

Twenty-two men raised the defence in relation to the killing of a female; 9 females invoked it in relation to a male victim .… It was more likely to be rejected [by a jury] if a male accused killed a female victim (36%) than if a male accused killed a male victim (12%). No woman who raised the provocation defence was convicted of murder. (LRCV, undated: 75)

The LRCV further notes that men who were convicted in cases involving provocation received ‘significantly heavier sentences than women in these cases’ (undated: 75). It continued:

Although juries may sometimes too readily accept provocation defences from men who kill women, this data does not support the conclusion that the provocation defence generally operates in a gender biased way. It refutes the claim made by some commentators that juries routinely accept provocation defences by males who have killed females.

The fact that the provocation defence is used more frequently by men who kill women … than women who kill men, argues in favour of ensuring that any gender bias in the criteria for the defence is eradicated, not in favour of abolishing the defence. (undated: 75–76)

But as Adrian Howe has pointed out, the LRCV not only failed to emphasize the larger raw numbers of men as opposed to women who used provocation, it also failed to interrogate the circumstances that were alleged to amount to provocation for each group (Howe, 1994). As the NSW Law Reform Commission (NSWLRC) observed in a comment on the Commission’s data, echoing Howe:

[]It is … important to be aware of what lies behind these figures. The general pattern that emerges from the cases is that men use the provocation defence when they kill their partners or ex-partners in a jealous rage and that women use it … where they have been the victims of long term domestic abuse. The data treat these situations as commensurate – something which itself should be examined for gender bias. (NSWLRC, 1993: 53)

These very different circumstances in which men and women typically kill, and in which they typically use provocation as a defence to murder, encapsulate the nub of the issue. While this report of the LRCV is undoubtedly aware of, and indeed uses, the data on violence in the domestic context, it remains focused on a truncated understanding of that social context. It also remains committed to resolving issues of inequality between women and men in a formal way, assessing the question of gender bias shorn of context, with a limited count of numbers.

The battered and the battering body?

What was by then the Victorian Law Reform Commission (VLRC) returned to defences to homicide in 2002. In the publications arising from this reference, we see a sustained ‘willing[ness] to grapple with the social context in which homicides take place, as opposed to approaching its reform task at an abstract and decontextualised level’ (Tolmie, 2005: 43, note 78). The reference began with an Occasional Paper which drew together the major qualitative empirical studies of homicide in Australia, especially that of Ken Polk (2004) and Alison Wallace (1986), and drew out the implications of these patterns of homicide for the legal defences (Morgan, 2002). This was the first time that empirical context was placed front and centre in a Victorian law reform body’s deliberations.

Thus paragraph 2 of the VLRC’s Issues Paper commences with a quote from Jenny Mouzos, from the Australian Institute of Criminology’s National Homicide Monitoring Project: ‘violence in general, and homicide in particular, are masculine phenomena’ (VLRC, 2002: 1). This fairly quickly leads to a focus on homicides in the context of sexual intimacy with, in turn, a focus on homicides in a context of jealousy and control, overwhelmingly the circumstances in which men kill their intimate female partners, and on homicides in response to prior violence, overwhelmingly the circumstance in which women kill their intimate male partners (see VLRC, 2002: 19, graph 2).

This focus on the social context continued throughout the Options Paper (VLRC, 2003), which also contained another detailed empirical study of homicide files, and the Final Report (VLRC, 2004). In that report the VLRC stated:

Defences and/or partial defences to homicide should not be based on abstract philosophical principles, but should reflect the context in which homicides typically occur. In particular, the law should deal fairly with both men and women who kill and defences should be constructed in a way that take account of the fact that they tend to kill in different circumstances. (VLRC, 2004: 15)

The issue of ‘gender bias’ in the provocation defence was covered extensively in the Options Paper and summarized in the Final Report. The VLRC recognized that gender bias does not only arise when women who kill in response to battering spouses are unable to use provocation; it also arises when the typical context in which men use the provocation defence is examined. It cited the NSWLRC’s criticism of the (then) LRCV’s conclusions on (lack of) gender bias that I have noted above (2003: 58). At least two of the four scenarios the VLRC presents are typical circumstances in which provocation is raised and it comments:

This [typical case] often involves a man who has a previous history of violence against his partner, killing her when she attempts to leave him or does not ‘behave’ appropriately. The alleged provocative conduct in such cases … is often a verbal taunt or insult, perhaps accompanied by a trivial physical assault.

… There were no cases [in their empirical study of prosecutions] in which a woman killed her male partner due to his threats to leave her, and no claims of provocation based simply on insults or trivial allegations of violence. (2003: 59)

It noted the preponderance in its sample of men who killed their partners when they tried to leave them, citing Martha Mahoney’s concept of ‘separation assault’ (1991) observing that at least 17 men in their sample, and only one woman, killed in these circumstances (VLRC, 2003: 59). It quotes from Howe, one of the strongest Australian critics of the provocation defence and refers to ‘critics [who] have stated that when these cases are publicized in the media, it often leads women to contact crisis centres, saying that their violent partners have threatened them with comments such as “you’ll be next”’ (VLRC, 2003: 60). Finally, the VLRC tackles the contribution provocation makes to the ‘myth that spousal killings are about a “loss of control”’ (2003: 60).

In some ways, it is not surprising that the issue of men killing women in circumstances of violent sexual jealousy became such a focus of the VLRC’s concerns. In addition to Ken Polk’s empirical work referred to above, and the ‘Who Kills Whom and Why’ paper (2002), Victoria has been the site of tireless advocacy by Phil Cleary, former federal independent MP, whose sister was killed by her estranged husband (see Cleary, 2002). It was also the forum for the publication in 1994 of Blood on Whose Hands?, telling the stories, via an examination of court transcripts, media and interviews with relatives, of Victorian women and children killed by violent partners. This surely was part of establishing a discourse, recognizable to a law reform body, which problematized violence against women in the context of homicidal violence.

A critical scrutiny of the defence of provocation mandates the critical scrutiny of its use by men, a form of scrutiny that crucially requires a move away from formal equality. Historically, formal equality in the provocation context has involved consideration of one or the other of two well-worn narratives. First, formal equality discourses have led to an argument that women who discover their male partners in flagrante delicto and who then commit homicide should also be able to avail themselves of the defence of provocation. Yet this scenario responds to no known social problem. Alternatively, the second of the ‘formal equality’ responses looks at the circumstances in which women typically kill their partners and argues that the traditional provocation rules be reformed by, for example, ensuring the time frame for recognizing provocative conduct is expanded so that a ‘slow-burn’ reaction can be properly recognized. While this latter scenario does have some empirical underpinning, it does not truly question the basic mode of the defence. What the VLRC’s reports do instead (or more accurately, in addition) is to question the very basis on which equality is being claimed: to throw into question the basic standard of the law on provocation, and the common circumstance in which it is used by men when their partners leave them (see also Forell, 2000).

As many know, the VLRC recommended that provocation should be abolished, a recommendation the government acted on, in the context of extensive media coverage of a particularly controversial case of this nature, the killing of Julie Ramage by her husband (see Kissane, 2006; Cleary, 2005). This is a move I support, as I believe the defence is unreformable (see e.g. Morgan, 1997). And I believe it is more likely that one is led to this view if one keeps the context in which men and women kill front and centre. ⁴ But what is more interesting is that by keeping that context as a central focus of the law reform considerations, I believe one is much more likely to simultaneously address the failings in the doctrine of self-defence for women who kill. (The VLRC notes that a number of submissions urged that the doctrine of provocation should not be abolished unless and until self-defence had been established as a viable option for women who kill violent partners (2004: 99). Hence the VLRC also went on to consider reforms to self-defence, and ways to introduce ‘social context evidence’ where a history of violence is alleged including ‘the history of the relationship, the cumulative effects of violence, and social, cultural and economic factors that may have affected the accused’ (2004: 75). These were also effectively enacted by the state government (see Crimes Act 1958 (Vic) s9AH as inserted by Crimes (Homicide) Act 2005 (Vic) s 6).

A progress narrative?

One reading of what I have presented is of a relentlessly optimistic progress narrative. While I am broadly optimistic, I by no means think the abolition of provocation is the end of the difficulties in this area. At the level of rhetoric, it is not irrelevant to point out that the VLRC said that the principal basis on which it was recommending the abolition of provocation was that ‘an intentional killing only justifies a partial or complete defence to murder in circumstances in which a person honestly believes that his or her actions were necessary to protect himself, herself or another person from injury’ (2004: 55), rather than because of the misogyny apparent in the provocation doctrine. (However, another (secondary) consideration, more fully encompassing the social facts analysis, was said to be ‘that the moral basis of provocation is inconsistent with contemporary community values on what is excusable behaviour’ (VLRC, 2004: 56)). Additionally, I have not factored in the perennial problem of (feminist) law reform which has ‘repeatedly been dogged by what might be termed the “implementation problem” .… [B]efore announcing the successful attainment of feminist goals, it is necessary to determine how the reforms are actually operating in practice’ (Hunter, 2008).

So, for example, there are other things the VLRC recommended, such as the effective reintroduction of excessive self-defence – defensive homicide – which was also implemented by the Victorian Government, which may be proving especially problematic, particularly its use by men (Middendorp [2012]; Tyson, 2011; Fitz-Gibbon and Pickering, 2012).There is the obvious risk that the traditional concerns of the provocation doctrine – spiteful women, sexual jealousy and ‘separation assault’ – will merely be removed to the sentencing stage which is less amenable to appellate and public scrutiny (see Morgan, 1997; Tolmie, 2005). It is probably too early to assess fully the way in which ‘provocation’ cases are being sentenced, but there is at least a research paper from the Sentencing Advisory Council which, while not representing the official views of Council, did present a sustained argument for an approach to sentencing which recognized the equality rights of victims:

Conduct that arises out of the victim exercising his or her right to equality, such as the right to personal autonomy (including the right to form relationships, work and otherwise assert his or her independence), should not provide justification for an offender’s aggrievement. (Stewart and Freiberg, 2008: 94)

The evidence on sentencing in practice seems mixed (see Tyson, 2011). The Victorian courts had at least rejected the notion that a domestic murder is, per se, a ‘less heinous’ category of murder (e.g. Goodall [2000], Azizi [2010]). In Felicite [2011], a recent case where a man pleaded guilty to the murder of his wife, the Court of Appeal reaffirmed this position. In this case, there was evidence of prior violence by the man, including a previous threat to kill (Felicite [2011]: para 2). However, Redlich JA said:

Allowance has sometimes been made for the fact that a murder committed on the spur of the moment in a domestic environment as a consequence of a volatile mixture of emotions, whether or not in response to what was previously recognized as provocation in law, may attract a lesser sentence. (Felicite [2011]: para 19)

Surely the ‘spontaneity’ of the accused’s actions were in doubt, given the prior violence and threat to kill, so one might question the relevance of this statement. However, this observation was followed by:

The taking of a domestic partner’s life undermines the foundations of personal relationships and family trust upon which our society rests. The sentence must reflect both the sanctity of human life and societies’ [sic] abhorrence of violence towards vulnerable and trusting partners …. (Felicite [2011]: para 20)

And the Court of Appeal refused to reduce Felicite’s 19-year sentence (with a non-parole period of 16 years). That is, the decision may be seen as sending some mixed messages.

I do not believe it is easy to displace the cultural story that a woman leaving a man for another, taunting him about his sexuality, or otherwise being ‘provocative’ excuses homicidal rage. However, I do believe it is worthy continuing to try to dislodge it. A focus on the social contexts in which homicides occur may indeed allow us to engage in such a reform exercise in a way that begins such a displacement.