Ulbrick, Madeleine; Flynn, Asher; Tyson, Danielle --- "The Abolition of Defensive Homicide: A Step Towards Populist Punitivism at the Expense of Mentally Impaired Offenders" [2016] MelbULawRw 16; (2016) 40(1) Melbourne University Law Review 324

THE ABOLITION OF DEFENSIVE HOMICIDE:

A STEP TOWARDS POPULIST PUNITIVISM AT THE EXPENSE OF MENTALLY IMPAIRED OFFENDERS

MADELEINE ULBRICK,^[*] ASHER FLYNN^[†] AND DANIELLE TYSON^[‡]

The offence of defensive homicide was abolished in Victoria in November 2014, following a widely held perception that it was being abused by violent men. While

primarily associated with battered women who killed in response to prolonged family violence — but who were unable to establish their offending as self-defence — a less publicised rationale underpinning the introduction of defensive homicide was

to provide an alternative offence for offenders with cognitive impairments not covered by the mental impairment (formerly the insanity) defence. Cognitive impairments

are complex and varied in their nature and symptomatology. Offenders presenting

with cognitive impairments therefore require an appropriate range of legal responses

to capture the nuances and appropriate moral culpability of their conduct. Drawing

from an analysis of the cases of defensive homicide heard over its 10–year lifespan,

this article contends that the abolition of defensive homicide did not adequately take

into consideration the potential impacts on individuals whose mental conditions are

not typically covered by the restrictive mental impairment defence. We further

argue that the decision to abolish defensive homicide was driven by dominant,

populist voices, without sufficient attention given to the offence’s potential to

achieve the aims underpinning its enactment, including providing an alternative

offence for women who kill in response to prolonged family violence.

CONTENTS

^I INTRODUCTION

In the study of homicide, while mental illness (specifically psychotic illness) has been a common subject of investigation,^[1] this has ‘not produced a thorough understanding of mental incapacity’.^[2] Moreover, the perpetual focus on exculpation at law has ‘marginalized’ the discourse around other ways in which mental impairment can be dealt with before the law.^[3] In turn, offenders with cognitive impairments have received comparatively less academic attention. In our view, a key implication of this is that cognitive impairments are not well understood and consequently the law in this area remains unclear and uncertain.^[4]

Between 2005 and 2014, the offence of defensive homicide

operated in Victoria under s 9AD of the Crimes Act 1958 (Vic),^[5] creating a safety net for an accused person who used lethal violence in circumstances of a mental illness or impairment that did not amount to the restrictive defence of mental impairment.^[6] The defensive homicide offence captured the.

circumstances of a fatality, where a person killed with a genuine belief that they were acting in self-defence, but where that belief was proven to be unreasonable.^[7] Accordingly, this offence sat between murder and manslaughter in terms of legal and moral culpability.

During its almost 10–year operation, 20 offenders who presented evidence of experiencing a history of mental health problems — ranging from formal diagnoses of schizophrenia, bipolar, paranoia and trauma-related mental illness, to cognitive impairments and intellectual disabilities — were convicted of defensive homicide.^[8] Fourteen of these offenders had a guilty plea accepted by the Crown; the remaining six offenders were found guilty of this alternative offence following trial.

This article presents findings from an empirical study of defensive

homicide cases and sentencing judgments from the introduction of the offence in November 2005,^[9] up until 21 September 2015.^[10] Over the 10–year period, we identified 34 defensive homicide convictions. However, due to privacy restrictions on one case, we were only able to access detailed

information on 33 cases.^[11] Of the 33 accessible cases, 23 (70 per cent) involved the Crown accepting an accused’s guilty plea to defensive homicide, while the remaining 10 convictions involved a guilty verdict following trial. Eighty-two per cent (n=27) of the cases involved a male perpetrator and male victim; 15 per cent (n=5) involved a female perpetrator and a male victim; and one case involved a male perpetrator and female victim. In this article, we focus on the 15 cases that involved male perpetrators who presented evidence of a history of mental illness and impairment, which were: (a) acknowledged by the judge in sentencing; and (b) acknowledged by the Crown in accepting a guilty plea, or the jury in reaching a guilty verdict for this alternative offence to murder. While drawing from all 15 cases, our analysis focuses specifically on three cases in which the offenders presented evidence of either intellectual disability or cognitive impairment: R v Trezise;^[12] R v Martin;^[13] and Director of Public Prosecutions (Vic) v Chen.^[14] We selected these cases because the cognitive impairments experienced by the three accuseds do not (and would not at the time have) fit the defence of mental impairment outlined in the Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997 (Vic) (‘CMIA’).^[15] This means that these offenders would (likely) not have been able to access a mental impairment defence to claim a reduction in their moral culpability had defensive homicide not been operating.^[16] In other words, the accuseds may not have been able to plead or be found guilty of a charge less than murder, despite, as our analysis will demonstrate, the evident impact their cognitive impairments had on their levels of moral culpability.

The in-depth analysis of these three cases, coupled with a discussion of the 12 other cases involving an accused with a history of mental illness or cognitive impairment, allows us to test the vocal claims of the dominant abolitionist reformers, who argued that defensive homicide provided an avenue for morally culpable, violent men to ‘get away with murder’,^[17] and that the offence itself was not operating as intended.^[18] We have also selected the cases that involved these specific forms of mental impairment because these conditions have not been considered in the major studies examining defensive homicide in the context of male-on-male lethal violence to date.^[19] Our discussion thus provides new and critical insights into the operation of defensive homicide in the context of mental illness and impairment.

Drawing from the selected case studies, this article sheds light on the complexities surrounding mental impairment and seeks to generate discussion around the absence of mental impairment as a key focal point in the abolition debate. By arguing that these cases did in fact cohere with the intended scope of s 9AD of the Crimes Act 1958 (Vic) — which sought to take into account that people kill in a range of different circumstances and that their culpability may be affected by a range of factors — we challenge the dominant, populist voices that strongly informed the abolition of defensive homicide.

There are multiple complexities and nuances in cases involving a fatality. Our article provides a mere snapshot of some of these by drawing from the personal histories of men convicted of defensive homicide. We argue that these backgrounds are far more complex than the images and focal points used by the media and populist voices to advocate abolition.^[20] Additionally, we seek to demonstrate how defensive homicide did not absolve the legal responsibility of those found guilty (either by plea or trial), nor did it result in accused persons not being punished and facing periods of imprisonment for their actions. Instead, we argue that defensive homicide offered an opportunity for consideration to (rightly) be given to the accused’s level of mental illness and impairment, and its impact on their moral culpability, in a way that would not have been possible without the operation of this offence.

While we believe it is entirely reasonable that homicide is met with legal punishment and social denunciation, we argue that the abolition of defensive homicide was largely premature and insufficient attention was given to the fact that its abolition, combined with the restrictive operation of the CMIA,^[21] would result in situations where individuals with mental conditions insufficient to form the basis of the mental impairment defence would have no defence or appropriate alternative homicide offence available to them in Victorian law.

In light of recent research indicating that 38 per cent of Australian prison entrants have been told they have a mental health disorder^[22] and 42 per cent of Victoria’s prison population have been identified with a psychiatric risk indicating mental health concerns,^[23] we contend that there is a demonstrable need for Victoria to have an appropriate range of legal responses to deal with the nuances and complexities of lethal violence, particularly where the offender suffers from a mental illness or impairment.

Our article commences with a discussion of the link between mental

illness and lethal violence and an overview of the current laws pertaining to mental impairment in Victoria. After considering the findings of recent reviews into mental impairment defences in Australia, we present a

background to the abolition of defensive homicide, with a particular focus on the dominant, populist–abolitionist reformers’ voices that were prioritised in the debate and fuelled the offence’s rapid demise. We then present an in-depth analysis of our three selected case studies to highlight how the offence was working effectively in practice, capturing the very unique and complex circumstances inherent to homicide. The article concludes by showcasing some of the effects of the decision to abolish defensive homicide for accused persons suffering from a mental illness and impairment and summarising why we advocate for a greater range of legal responses to cover the nuance and complexities of lethal violence, including consideration of several of the recommendations arising from the Victorian Law Reform Commission’s (‘VLRC’) review of Victoria’s mental impairment laws.

II THE LINK BETWEEN MENTAL ILLNESS AND LETHAL VIOLENCE

The nature and extent to which those suffering from mental illness are more prone to lethal violence than those without a mental illness is a key question that has informed socio-legal research for decades.^[24] Despite the burgeoning interest, there remains no clear answer, with much of the discussion beset by ‘conflicting opinions and apparently contradictory empirical data’,^[25]

methodological limitations (including unscrutinised, flawed and poorly controlled data sets),^[26] research cathexis and disciplinary silos.^[27] In addition, there exists little reliable statistical information to date concerning a link between homicide and cognitive impairment. The connection between mental illness and impairment and violence is further complicated by the fact that few (if any) longitudinal Australian studies have examined the relationship between non-psychotic mental illness and lethal violence.^[28] Despite this, the most recent Ombudsman investigation into the rehabilitation and reintegration of prisoners in Victoria reported that ‘40% of all Victorian prisoners have been identified as having a mental health condition’ and that ‘prisoners are 10 to 15 times more likely to have a psychotic disorder than someone in the general community’.^[29] This corresponds with the findings of Short et al’s study,^[30] which suggests that schizophrenic mental illness is overrepresented in terms of individuals coming into contact with the criminal justice system. However the true extent of the broad range of mental illnesses (and cognitive impairments) is likely underestimated among those individuals convicted and sentenced.^[31] As a result, the relationship between mental illness and lethal violence remains a highly contentious and contested issue.

While the data in this area is still emerging (making prevalence rates difficult to assess), recent longitudinal research in Victoria has found a correlation between mental illness, specifically schizophrenia, and serious violence — independent of mediating variables such as comorbid substance misuse.^[32] This body of research has found high rates of mental illness among homicide offender cohorts, with offenders who experience mental illness being

significantly more likely to commit homicide than people in the general population without a mental illness. Specifically, the data suggests that rates of

schizophrenia disorder among homicide offenders were ‘thirteen times higher than in the general population comparisons’ and more than 8.7 per cent of homicide offenders had a diagnosis of schizophrenia.^[33] Notwithstanding the accumulating evidence consistent with these findings, there appears to be reluctance (of varying degrees) on the part of some researchers and mental health advocates in accepting this correlation.^[34] As Mullen states, the view that those with mental illness are not more prone to violence than those without a mental illness has been ‘strongly promoted by ... [those wanting] to de-stigmatise mental illness’, particularly following the closure of mental asylums.^[35] Accordingly, it has been argued that the downward spiral of social dislocation, experienced as a consequence of mental illness, has been

consistently offered as the most cogent explanation for any link between mental illness and violent offending.^[36]

It is important to bear in mind that the relationship between mental illness and lethal violence in the majority of cases is not a causal one; rather, a mental illness such as schizophrenia may be a contributory factor.^[37] Based on the available statistics, Short et al hypothesise that, if a causal relationship did truly exist, and all patients with schizophrenia were separated from the community, lethal violence would only decrease modestly.^[38] The link should therefore be considered as one potential risk factor, among others, ‘which, if present, increases the statistical likelihood of an individual committing violent crime when compared to others [without schizophrenia] in the community’.^[39] In light of these arguments, the most recent evidence regarding a link is not about reviving old stigmas, but rather, seeking to accurately portray the risks of lethal violence in order to improve service responses in both the public mental health and legal sectors.^[40] This includes ensuring there are appropriate avenues to respond to mentally impaired offenders who commit lethal violence in a way that captures their legal culpability, while recognising the reduced moral culpability that may exist in these instances.

III MENTAL IMPAIRMENT, LETHAL VIOLENCE AND THE LAW

In Victoria, the current legal defence of mental impairment was introduced in 1997^[41] following the revision of the Daniel M’Naghten’s Case (‘M’Naghten’) ‘insanity’ defence.^[42] To establish the statutory defence of mental impairment under s 20(1) of the CMIA, it must be proved that on the balance

of probabilities:

at the time of engaging in conduct constituting the offence, the person was suffering from a mental impairment that had the effect that —

(a) he or she did not know the nature and quality of the conduct; or

(b) he or she did not know that the conduct was wrong (that is, he or she could not reason with a moderate degree of sense and composure about whether the conduct, as perceived by reasonable people, was wrong).

Substantively, paragraph (b) differs only slightly from the former M’Naghten elements,^[43] which focused on the accused person’s capacity to determine right from wrong without the clarifying reference to the person being unable to reason ‘with a moderate degree of sense and composure about whether the conduct, as perceived by reasonable people, was wrong’.^[44] Pursuant to the CMIA, persons who satisfy either element must be found not guilty because of mental impairment (‘NGMI’).

While there is no specific partial defence for mental impairment in

Victoria, it is important to note that the sentencing of accused people with mental impairment is governed by the R v Verdins (‘Verdins’) principles, which are applicable to sentencing in ‘at least’ six ways.^[45] The Verdins

principles restated the former R v Tsiaras (‘Tsiaras’) principles,^[46] following a landmark appellate decision in 2007, which ‘radically altered the judicial sentencing landscape in Victoria’.^[47] The appellate judges in Verdins^[48] sought to clarify the Tsiaras principles,^[49] which had long been considered ambiguous, due to the lack of clarity around which conditions were included or excluded, thereby making the principles difficult to invoke and apply, particularly in cases not involving psychotic illness.^[50] In this way, the seminal decision in Verdins shifted the focus away from severe psychiatric illness not amounting to insanity,^[51] and liberalised the law to take into account a broad range of mental conditions.^[52] While the Verdins principles are ‘thoughtful and sensible ... they are still not well understood by clinicians’.^[53] Furthermore, it is standard practice for courts to apply the same sentencing principles for accused people with cognitive impairments as they do for those with mental illness. This has been described as a ‘strange anomoly’,^[54] with cases such as R v Mailes^[55] representing a ‘failure of both the courts and the legislature to comprehensively address the key differences between offenders with an intellectual disability and those suffering a mental illness’, resulting in a gap in sentencing practice.^[56]

As part of its 2004 final report on defences to homicide, the VLRC

evaluated some of the concerns associated with the mental impairment defence.^[57] While noting ‘almost universal support for leaving the defence of mental impairment unchanged’,^[58] the VLRC advocated for the ‘scope of “mental impairment” to be clarified’ on the basis that:

While the CMIA explicitly abolishes the common law defence of insanity, the tendency by the courts has been to interpret mental impairment restrictively by reference to the common law defence of insanity and the notion of a ‘disease of the mind’.^[59]

Accordingly, the VLRC recommended that ‘a new provision be inserted into the CMIA to make clear [that] mental impairment includes but is not limited to a disease of the mind’.^[60]

In August 2012, then Victorian Attorney-General, Robert Clark, commissioned the VLRC to undertake a comprehensive review of the CMIA.^[61] According to Clark, the opportunity to examine, reflect and make changes to improve and modernise the operation of the CMIA was desirable. This became the first substantial review of the CMIA since its inception 15 years earlier.^[62] The review aimed to assess whether the CMIA was operating ‘justly, effectively and consistently with its underlying principles’ and to ‘examine whether changes ... [were] needed’.^[63]

Following extensive roundtable consultations with stakeholders, including one on the operation of the CMIA in the higher courts, which was attended by legal practitioners (defence and prosecution), staff of advocacy groups, forensic clinicians and academics, the VLRC determined that the CMIA was difficult for accused persons to use due to lack of definition and unclear legal tests.^[64] This was because the legal test was formulated around ‘disease of the mind’, which provided insufficient judicial guidance and flexibility.^[65] As a result, interpretations of a ‘disease of the mind’ were considered to be inadvertently narrow and excluded a range of mental conditions.^[66] The VLRC also found that accused persons were reluctant to rely on the CMIA due to the stigma attached to the NGMI verdict, and the possibility of indefinite detention (associated with the mandatory 25–year nominal term).^[67] In the Supreme Court and County Court, persons found NGMI may be unconditionally released or liable to a custodial or non-custodial supervision order.^[68] In the Magistrates’ Court, an accused must be discharged following a finding of NGMI.^[69] In the context of homicide, the supervision order is indefinite, but contains a 25–year nominal term at which time the supervision order is subject to review. During consultations, the nominal term attracted criticism based on its length, which ‘creates the impression of a sentence and discourages people from raising mental impairment as a defence’,^[70] and because ‘a significant number of people remain on [supervision] orders after the expiration of the 25–year term’.^[71] In many cases, this is longer than they would have been detained had they been found or pleaded guilty to the offending behaviour and sentenced accordingly.^[72]

The narrow ambit of the CMIA’s operation has meant that, in practice, mental impairment can only be argued ‘in very limited circumstances’,^[73] and it is successfully relied on in even fewer cases.^[74] For example, between 2011 and 2012, orders under the CMIA were made in ‘approximately one per cent of the total cases ... [processed] in the higher courts’.^[75] Such infrequent use is not simply due to its restrictive application, which ‘does not properly reflect [contemporary] medical understandings of mental illness’,^[76] but is also reflective of the:

decisions made by accused people who are mentally ill to take a chance that they will be acquitted at trial, or to plead guilty ... and serve a prison term,

rather than to run the risk of being confined in a psychiatric hospital for a

lengthy period.^[77]

The VLRC’s report was tabled in Parliament on 21 August 2014.^[78] Among the 107 recommendations, three are of particular relevance to homicide offences. First, recommendation 84 proposed the introduction of a new system of

five-year progress reviews to replace the 25–year mandatory nominal term.^[79] Second, recommendation 16 proposed ‘[a]dapting the test [for unfitness to stand trial] when the accused wishes to plead guilty’.^[80] And third,

recommendation 24 proposed introducing a flexible statutory definition of mental impairment into s 20 of the CMIA that defines ‘“mental impairment” for the purposes of the defence as a condition that “includes, but is not limited to, mental illness, intellectual disability and cognitive impairment”’.^[81] These proposed changes attempt to clarify the law and improve safeguards for managing vulnerable people with intellectual disabilities and cognitive impairments by ensuring that they are treated equally before the law.^[82] The recommended changes may also encourage accused persons to rely on the mental impairment defence. By making it more accessible to a broader range of impairments, the changes would extend the scope of the mental

impairment defence, thereby providing an appropriate option for offenders suffering from a condition, while still ensuring the definition remains narrow enough to avoid inappropriate use. The change in nominal sentence would also be more likely to encourage a reliance on mental impairment in

appropriate cases, as opposed to providing an incentive to offenders with mental impairments and illnesses to plead guilty for a lower sentence option. Despite these potential outcomes, the VLRC’s recommendations have yet to

be implemented.

IV REVIEW OF THE LAW AND PROCEDURE APPLYING TO PEOPLE WITH MENTAL CONDITIONS IN NEW SOUTH WALES

A review of the laws applicable to people who have mental conditions has also been conducted in New South Wales (‘NSW’).^[83] The NSW Law Reform Commission’s (‘NSWLRC’) report stated that the submissions acknowledged that, although ‘modifications to the M’Naghten rules were suggested ... the defence of mental illness works in practice without significant difficulty and ... the right results are achieved’.^[84]. However, like the VLRC’s 2014 report, stakeholder submissions to the NSWLRC review identified similar issues to the law and procedure applying to people with mental conditions, including that NSW’s mental illness defence contained a problematic definition of the accused’s mental state.^[85] The NSWLRC found that ‘the fit between cognitive impairments and the terminology that presently describes the mental state of the person for the purposes of NGMI is not a good one’.^[86] The submissions argued that the terminology be replaced with ‘contemporary definitions of mental illness and cognitive impairment’.^[87] In particular, they criticised the outdated notions of ‘disease of the mind’ and ‘defect of reason’, which appear to be part of the common law interpretation of the defence.^[88] To remedy this, the NSWLRC recommended revising the defence ‘to include a statutory test for the defence of mental health or cognitive impairment’.^[89] It also recommended that under s 38 of the Mental Health (Forensic Provisions) Act 1990 (NSW), the ‘verdict’ should be articulated as ‘not criminally responsible by reason of mental health or cognitive impairment’.^[90] This amendment would recognise the culpability of the accused, while also simplifying the law to better align with public understanding of this outcome, given that they may ‘find it [the present law] confusing and unpalatable’.^[91]

In addition to considering how to label and define the mental illness

defence, a further objective of the NSWLRC review was to consider arguments surrounding the abolition of substantial impairment.^[92] Substantial impairment is a statutory partial defence available in homicide cases in NSW which operates to reduce liability for murder to manslaughter^[93] where the culpability of the accused was significantly diminished by ‘an underlying condition’.^[94] There have been calls for amendment to the similar mental impairment defence in South Australia based on concerns that it is relied upon as a defence to murder more than in any other Australian or

international jurisdiction.^[95] As Toole argues, the Sentencing Advisory Council of South Australia’s 2014 review was ‘welcome’ because the threshold for the defence is too low.^[96] This view was similarly voiced by the NSW Office of the Director of Public Prosecutions in the NSWLRC’s review.^[97] The NSWLRC summarised the Office of the Director of Public Prosecution’s position (expressing support for abolishing substantial impairment) as follows: substantial impairment ‘confuses issues’ that are more relevant and ‘more appropriately dealt with in sentencing’,^[98] and the defence is ‘overrepresented in court and in pleas’.^[99] This view, however, was not supported by the majority of stakeholders involved in the NSWLRC review,^[100] and is also not reflective of NSW court statistics which showed that between 1998 and 2011, the defence of substantial impairment was raised in 82 cases (constituting an average of only six times per year).^[101] Of these, just over half (58 per cent) were accepted, resulting in the lesser conviction of manslaughter.^[102]

Significantly, the accused persons who successfully relied on substantial impairment ‘generally displayed severe mental health conditions’,^[103] thereby suggesting those who were accessing the defence were, in fact, deserving of doing so.

Ultimately, the NSWLRC recommended the retention of substantial

impairment based on the infrequent use of the defence and overwhelming stakeholder support.^[104] In addition, the report stated that substantial impairment should be retained because ‘the complexity of cognitive and mental health impairments, and their nature and effects, requires an appropriate range of legal responses’.^[105]

This recommendation and the arguments detailed in the NSWLRC report more generally are significant in the context of the abolition of defensive homicide in Victoria. As noted, during its 10–year operation, 15 of the 27 defensive homicide cases with both a male perpetrator and a male victim involved the defence raising evidence of mental health and moral

culpability.^[106] Significantly, the judicial responses in these cases, as

documented in the sentencing judgments, acknowledged that this evidence went some way towards explaining the accused’s unreasonable belief in the need for lethal violence. The abolition of defensive homicide, combined with the restrictive operation of the CMIA,^[107] has culminated in Victorian law now not accommodating those who have been charged with murder and who have a mental condition insufficient to form the basis of the mental impairment defence. Before considering the implications of this, it is important to place the decision to abolish defensive homicide within its appropriate context, including the rationale for the creation of the defence, and how dominant, populist voices helped fuel its rapid demise.

V HOMICIDE LAW REFORM AND THE RAPID DEMISE OF DEFENSIVE HOMICIDE

In 2005, a comprehensive package of ‘trendsetting feminist ... reforms’ was implemented in Victoria through the Crimes (Homicide) Act 2005 (Vic)

(‘2005 Act’).^[108] The 2005 Act abolished the controversial partial defence of

provocation,^[109] and codified self-defence as a defence to murder,^[110] expanding the scope so that it would more adequately accommodate the experiences of women who kill in response to family violence.^[111] The 2005 Act also

introduced a new provision to allow the admission of evidence highlighting the relationship and social context of family violence in cases of homicide where family violence is alleged.^[112] Included in the reform package was the new offence of defensive homicide.^[113] This offence permitted a person to be ‘convicted of defensive homicide (rather than the more serious offence of murder) where they killed with the belief that their actions were necessary in order to defend themselves, or another, but they had no reasonable grounds for that belief’.^[114] While a main rationale for these reforms was to make the law ‘more accessible to people who kill in response to family violence’,^[115] an equally significant — but less publicised — rationale underpinning the reforms was to provide a ‘halfway house’^[116] for offenders with mental illness or impairment not amounting to a defence of mental impairment. This was demonstrated during the second reading speech of the Crimes (Homicide) Bill 2005 (Vic), in which then Attorney-General, Rob Hulls, stated that while ‘[r]elatively few cases are likely to fall into the new defensive homicide category’, the distinction between the components of the ‘belief’ and

‘reasonable grounds’ tests would be most applicable, outside the context of family violence, in situations where ‘the accused person is not suffering a mental impairment within the ... [CMIA] but is suffering from a form of paranoia or distorted perception’.^[117] This suggests that it was expressly contemplated that defensive homicide would be used by individuals with a mental illness or mental impairment falling outside of the CMIA defence.

Major criticism of defensive homicide began to emerge in 2010 in the wake of R v Middendorp (‘Middendorp’),^[118] which was the first and remained the only case involving a male perpetrator convicted of defensive homicide for killing his former female partner. In this case, the jury found Middendorp not guilty of murder, but guilty of the lesser offence of defensive homicide after stabbing his former female partner four times in the back (by reaching over her shoulder), after she ‘came at [him] with a raised knife in her right hand’.^[119] Moments after stabbing her, witnesses reported hearing Middendorp say ‘words to the effect that she got what she deserved and that she was a filthy slut’.^[120] There was also evidence presented that at the time of his offending, Middendorp was in breach of bail conditions, an intervention order and ‘almost certainly ... the good behaviour bond’.^[121]

In response to the public criticism of this perceived unjust outcome,^[122] the Victorian Department of Justice (‘DOJ’) published a discussion paper calling for submissions on a range of issues, the most important of which for our purposes was ‘whether the defensive homicide “safety net” is still required, and if it is required, whether the offence of defensive homicide should be limited in some way’.^[123]

At the time of the discussion paper’s release, there had been only two cases in which women had killed in response to family violence since the 2005 reforms.^[124] The DOJ interpreted the outcomes in these two cases as a sign that the 2005 Act had ‘introduced significant improvements to the criminal justice system in dealing with situations in which a woman kills in response to

long-term family violence’.^[125] The DOJ’s assessment of how defensive homicide had been operating for male perpetrators was based on 13 cases, 12 of which involved offenders and victims who were male.^[126] The DOJ found that all 10 cases where a guilty plea was entered involved ‘young men in one-off violent confrontations’,^[127] with only one case including a history of

family violence.^[128]

The decision to abolish defensive homicide was already being

contemplated by the time the DOJ released its 2013 consultation paper.^[129] In assessing the value of retaining the offence, the DOJ noted that it was ‘difficult to conclude that this defence clearly works to the benefit of women who kill in response to family violence’,^[130] given the small number of women who had been convicted of defensive homicide at the time (n=3).^[131] In its consideration of ‘defensive homicide both in policy terms as well as how it applies when men kill’,^[132] the DOJ drew on the long-standing criticisms of provocation, suggesting ‘the very existence of defensive homicide inappropriately condones or excuses male violence’^[133] and ‘supports a culture of blaming the victim’.^[134] Using a rationale that was based largely on the public outcry from the guilty verdict in cases such as Middendorp,^[135] the DOJ further argued that ‘[a]bolishing defensive homicide should reduce victim blaming as it will no longer partially excuse male violence’.^[136] This was material to the DOJ’s proposal that ‘defensive homicide be abolished’.^[137] In November 2014, defensive homicide was abolished through the Crimes Amendment (Abolition of Defensive Homicide) Act 2014 (Vic) (‘ADHA’).^[138]

VI THE POLITICS OF HOMICIDE

The rapid demise of defensive homicide after Middendorp is reflective of how crime policy has more generally shifted towards increased punitive measures, which has not only resulted in a focus on punishment and stricter social controls, but has problematically shifted the ways in which law reform is implemented, to the point that there is often fast-paced, legislative change that fails to consider the broader environmental, structural, social, political, economic and physiological contexts in which crime occurs.^[139] As Garland argues, in contemporary criminal justice, ‘[p]olicy development appears highly volatile, with an unprecedented amount of legislative activity ...

No one is quite sure what is radical and what is reactionary’.^[140]

Clear examples of this shift in homicide-related crime policy can be seen in government responses to fatalities resulting from public ‘one-punch’ incidents.^[141] In NSW, legislative changes introducing a mandatory minimum sentence of eight years’ imprisonment for anyone who, while under the influence of drugs or alcohol, fatally punches someone^[142] were implemented with ‘alarming speed ... passed by both houses without substantial

amendment and on the same day they were introduced’.^[143] Quilter writes:

within the space of just over a week, without a public consultation process and without any apparent input from the NSW Law Reform Commission ... or other expert groups, the Government moved from the announcement of a

16–point plan to tackle alcohol-related violence to fully operational legislation which had exceptional features: invoking for only the second time in recent NSW history the policy of mandatory sentencing and constituting the first additional offence to the law of ‘homicide’ since 1951 ...^[144]

As Flynn, Halsey and Lee observe:

Such law was enacted with little (if any) understanding of the

phenomenological dimensions of one-punch events ... [Instead] [o]ne-punch fatalities have not only been presumed to be redeemable through law (by

proscribing [sic] hefty sentences for perpetrators) but preventable through such as well (by general deterrence of would-be perpetrators).^[145]

The Victorian government responded similarly in September 2014 by

introducing a new offence into s 4A of the Crimes Act 1958 (Vic) (manslaughter by single punch or strike)^[146] and enacting ss 9A and 9C of the Sentencing Act 1991 (Vic)^[147] to make this offence subject to a statutory mandatory minimum sentence of 10 years’ imprisonment.^[148] The new offence allows for an individual to be found guilty of manslaughter by single punch or strike,^[149] regardless of whether the punch or strike directly caused the death (for example, if death occurred as a result of hitting the ground),^[150] and it is irrelevant whether the death occurred from one punch or strike or a series of punches or strikes.^[151]

To put the frequency of these one-punch offences in context, Pilgrim, Gerostamoulos and Drummer’s study identified 90 recorded one-punch fatalities across Australia between 2000 and 2012.^[152] During this 12–year period, there were 24 one-punch fatalities in Victoria — an average of two per year.^[153] In contrast, during the 2013–14 financial year, 167 homicide offences were recorded by Victoria Police.^[154] Bearing these statistics in mind, we might question why the one-punch offence — of which there was an average of 7.5 per year over 12 years across all Australian jurisdictions^[155] — was responded to so quickly and with such severity by multiple governments. In a similar vein to Garland,^[156] we suggest that such practices are reflective of a shift in law reform practices away from nuanced inquiry and evidence-based research, towards quick fix, populist responses:

The policy-making process has become profoundly politicized and populist.

Policy measures are constructed in ways that appear to value political advantage and public opinion over the views of experts and the evidence of research. ... The dominant voice of crime policy is no longer the expert or even the practitioner ... A few decades ago public opinion functioned as an occasional brake on policy initiatives: now it operates as a privileged source. The importance of research and criminological knowledge is downgraded and in its place is a new deference to the voice of ‘experience’, of ‘common sense’, of ‘what

everyone knows’.^[157]

In the context of defensive homicide, the dominant public arguments

supporting its abolition focused on its failure to act as a viable option for female offenders who killed in response to ongoing family violence (commonly and inaccurately cited during the public debates as the only reason for the offence’s introduction).^[158] The dominant voices also invoked a populist, punitive framework, claiming that the current laws were akin to violent men ‘getting away with murder’,^[159] leading Fitz-Gibbon to state, ‘let’s call a spade a spade. Committing lethal violence with an intention to kill is murder and the operation of defensive homicide in Victoria has unjustly blurred the

distinction’.^[160] Such claims were made despite the fact that 70 per cent (n=23) of defensive homicide convictions were resolved by the Crown accepting a guilty plea to defensive homicide, meaning the Crown determined that there was not a viable likelihood of the accused being found guilty of murder at trial. As the Victorian Director of Public Prosecutions has continually stressed,^[161] the ‘structured and accountable’^[162] decisions to accept guilty pleas are made by ‘experts in the field of court craft, advocacy, and the law and procedures relating to the conducting of the state’s most serious criminal trials’,^[163] and take into account a careful consideration of all available evidence, the public’s interests, ‘any probable defences; the views of the victims and the informant; the accused’s criminal history; and the likely length of a trial’.^[164]

Given the number of cases involving Crown-accepted guilty pleas and the public interest and evidence thresholds prosecutors are bound by in Victoria,^[165] it can be assumed that, had the offence of defensive homicide not been available, a significant portion of these guilty plea outcomes may either have become manslaughter convictions, or possibly no conviction at all — had the matter proceeded to trial with an evidence base the Crown deemed unlikely to result in a murder conviction (an outcome much more reflective of the populist abolitionist reformers’ concerns, however unjust, that violent men were ‘getting away with murder’).^[166]

With a powerful combination of claims the system was allegedly failing vulnerable, abused women and a focus on ‘violent men who’ve been able to get away with murder ... [and] escape responsibility’,^[167] defensive homicide, or more specifically, the abolition of defensive homicide, took on a strident political character. Debates around potential reforms rapidly shifted to calls for abolition, which reflected what Garland called, with respect to punitive responses to crime control in Britain in 1996, ‘no negotiation, no question of whether or not [this response] might “work”’.^[168] Concerns raised by family violence stakeholders that ‘[a]bolishing defensive homicide would be a backward step in legal responses to victims of family violence’^[169] — including a submission to the DOJ endorsed by 17 community and family violence organisations, women’s services and academics^[170] — were downplayed by the dominant, punitive voices: ‘the concerns of domestic violence stakeholders are unsurprising. However, these concerns are potentially unwarranted’.^[171] At the same time, no attention was given to the significant number of mentally ill and impaired offenders convicted of defensive homicide in the media, in many abolitionist reformists’ public critiques,^[172] or in the two major studies of male-on-male defensive homicide cases,^[173] despite the fact this offence was, in part, introduced with such offenders in mind.^[174]

Constructing the abolition of defensive homicide in the context of preventing murderers from getting away with their crime (‘an avenue away from murder for men who kill’)^[175] and drawing the discussion (however

misguided) into the realm of protecting vulnerable, abused women (‘[a]bolishing defensive homicide will benefit female victims and offenders’),^[176] presents a powerful narrative, and one in which the current political climate focused on violence against women is very easy to accept. However, this narrative is not reflective of the reality of the cases that were being heard, or the evidence-based research suggesting that defensive homicide had the potential to benefit vulnerable and abused female offenders.^[177] Indeed, recent research from Tyson et al has concluded that there remains a need for a ‘safety net’ in the form of defensive homicide for women who kill abusive partners.^[178] This finding was based on an analysis of cases in which defensive homicide was working for some women perpetrators, such as Angela Williams,^[179] who Tyson et al argued may have been convicted of murder had defensive

homicide been abolished at the time.^[180] The findings of Tyson et al’s study also suggest that, while there have been some improvements in legal

understandings of family violence, the way forward lies in ‘the need for comprehensive, consistent and ongoing [family violence] training for

prosecuting and defence counsel, judges, expert witnesses, and other legal professionals’,^[181] rather than the abolition of defensive homicide.

The dominant views heralding defensive homicide’s abolition offered a superficial presentation of events that negated the complexities of crime, the individual circumstances of offences and the causes of offending behaviour.^[182] This reflects Garland’s view that a populist storm may encourage all

perpetrators to be treated as ‘violent individuals for whom we can have no sympathy and for whom there is no effective help’,^[183] with ‘[t]he only practical and rational response ... [being] to have them “taken out of circulation” for the protection of the public’.^[184] This is particularly evident in Fitz-Gibbon’s arguments which drew on populist notions of governments needing to be ‘tough on crime’ and respond punitively, to support an abolitionist agenda:

the operation of defensive homicide has also served to minimise the seriousness of male lethal violence perpetrated with knifes [sic]. Convictions for

defensive homicide in this context undoubtedly conflict with the government’s expressed intention to show a ‘tough on crime’ approach to knife crime, which aimed to curb the knife culture amongst young Victorian males. A conviction for defensive homicide in cases where a knife was not only used but was brought to the scene of the crime certainly trivialises the seriousness of this form of lethal violence. What is needed is clear. ... [T]he government should abolish defensive homicide. ... It has no place in our legal system.^[185]

Voices such as this effectively worked to shut down nuanced inquiry into the immediate and extended effects of abolition, which in the context of this article, is particularly concerning in light of its impact on mentally

impaired offenders.

While we believe it is entirely reasonable that homicide is met with legal sanction and moral censure, in the scramble to abolish defensive homicide, we contend that much more unsettling questions about how this change would impact on offenders suffering from a mental illness and impairment have been permitted to be ignored. As recognised by the VLRC with respect to diminished responsibility,^[186] and articulately noted in the NSWLRC’s recommendation to retain substantive impairment as a defence: ‘it is

inappropriate [and unjust] to apply the label “murderer” to a person whose [mental] capacity ... was substantially impaired’.^[187]

The section that follows highlights some of these unconsidered impacts through an analysis of three of the 15 defensive homicide cases where evidence around mental health and moral culpability were successfully raised.

VII WHAT DOES THIS MEAN FOR MENTALLY IMPAIRED OFFENDERS?

Table 1: Summary of the Dataset

* Denotes case where evidence of mental illness and/or cognitive impairment was successfully raised.

A total of 20 cases involved evidence of either mental illness or cognitive impairment. Of the 15 cases (involving a male perpetrator and male victim) where evidence of mental illness or cognitive impairment was presented, four involved a jury verdict, 11 were resolved by the Crown’s acceptance of a guilty plea, and the sentences imposed varied in length from a minimum of four years, to a maximum of 17 years’ imprisonment. In the majority of these cases, the offenders had a dual diagnosis of either mental illness or cognitive impairment, and polysubstance dependence, as well as a history of trauma such as family violence, kidnapping and torture, being forcefully removed from family as a child and detained and being grievously attacked by a gang. In addition, 14 of the cases involved an offender from an environment of chronic disadvantage such as homelessness, transient, dysfunctional or unstable accommodation and a cycle of foster care and state ward placements. Forty per cent of the cases involved evidence of all these factors.

As noted, for the purpose of this article we have selected three of the 15 cases to examine in detail. The three cases all involved offenders with cognitive impairment (including intellectual disability and traumatic brain injury). The decision to focus on these impairments was made because such conditions in the context of homicide offending have been scarcely examined within the socio-legal sphere, and are often misunderstood and not adequately accommodated for in the law. In presenting the analysis, we seek to highlight the complexities surrounding cognitive impairment and thus demonstrate the significance of this discussion not being adequately considered in the abolition debate. In doing so, we provide evidence that these cases did in fact cohere with the intended scope of s 9AD of the Crimes (Homicide) Act 2005 (Vic) — the counter claim to which was levied as a key reason to abolish defensive homicide. We also highlight the need to maintain an appropriate range of legal responses to deal with the nuance and complexity of these types of cases, thereby rendering the abolition of defensive

homicide problematic.

A R v Trezise

Trezise and the deceased had spent some time together consuming alcohol when an argument developed.^[188] The argument culminated in Trezise inflicting 36 penetrative stab wounds on the deceased.^[189] Trezise pleaded guilty to defensive homicide and the Crown contended that (as summarised by Coghlan J), ‘where there is a claim to self-defence based upon your belief, which cannot be rebutted beyond reasonable doubt, it is reasonable and appropriate to accept a plea to defensive homicide’.^[190] The evidence collected from the crime scene suggested that ‘there had been a substantial disturbance inside the house’^[191] and that Trezise had ‘suffered some injuries, which were comparatively minor, although ... required sutures’.^[192]

In its opening statement, the Crown noted that Trezise had made a number of inconsistent statements — ‘all patently false’ — to police and medical staff.^[193] It further noted that none of these statements provided an adequate overview of the events that transpired.^[194] While using these comments as an aggravating component of the offence, such outcomes are indicative of the unique challenges accused persons with intellectual disabilities face when coming in contact with the criminal justice system.^[195] In the context of police interviews specifically, ‘modes of communication are often not meaningfully altered to accommodate [the communicative abilities of] people with

intellectual disabilities’; thus the offender may have difficulties in engaging, interpreting and responding to the questions posed by police.^[196]

In sentencing Trezise, Coghlan J accepted that Trezise’s intellectual

disability played a role in his actions, stating: ‘I am satisfied that you have very little, if any, proper recollection of what happened on the night ... taking into consideration the amount of alcohol ... consumed together with your underlying intellectual disability’.^[197] Trezise’s intellectual disability was also acknowledged by the Court in the summation of his personal background:

Your background and upbringing have been extremely difficult. ... Your

mother was Aboriginal. ... [S]he was an alcoholic. That fact and other birth

difficulties have contributed to your intellectual disability. On recent

examination you were found to have a full-scale IQ of 76, which places you ... in the bottom six per cent of the community. ... Because of your intellectual disability, your schooling was difficult. ... Your schooling having been all the more difficult because of the itinerant lifestyle that your family had observed for many years ...^[198]

The impacts of Trezise’s disability were further noted in that he had been ‘placed on a disability pension and since then [had] never been in paid employment’.^[199] Coghlan J also stated that Trezise ‘continue[d] to have difficulties comprehending and organising [his] time ... [and had] plain communication difficulties’.^[200]

A further consideration relevant to the sentencing was Trezise’s exposure to family violence:

Your parents had separated when you were around five years old ... Your father formed another relationship ... That relationship was very difficult and marked by domestic violence. ... The Department of Human Services and other

agencies intervened to protect you on a number of occasions ...^[201]

Trezise’s legal representative relied on expert evidence from a forensic psychologist who tendered a report which diagnosed Trezise ‘as suffering from a chronic adjustment disorder with mixed disturbance of emotions and conduct’.^[202] In responding to this evidence, Coghlan J remarked:

Given your upbringing, such a diagnosis is hardly surprising. That, together with your intellectual disability form another relevant factor in your sentencing. The application of the [Verdins] principles of both general and specific

deterrence will be necessarily moderated because of those matters.^[203]

Trezise was sentenced to eight years’ imprisonment with a non-parole period of four years.^[204]

B R v Martin

Martin went to the home of the deceased where they ‘drank some beer and watched tennis on the television’.^[205] Later that night, the deceased allegedly ‘made sexual advances’ towards the accused,^[206] in which he asked Martin to remove his clothes and ‘[s]it with [him] in the nude’,^[207] and then started touching the accused ‘down between the legs’.^[208] The accused also told police that the deceased ‘was flashing himself, rubbing himself against me. I didn’t like it and then I went to the toilet, he followed me and ... pushed [me] into his bedroom and “[t]hen I was on the bed and he tried [to rape me]”’.^[209] Curtain J held that, in accepting Martin’s guilty plea to defensive homicide:

The Crown accepts that when the injuries were inflicted, you believed that what you were doing was necessary to defend yourself from really serious injury, in that you believed the deceased was trying to rape you. The Crown considers that your intellectual disability was at least partly responsible for you holding this belief. By your plea to defensive homicide, you have acknowledged that you had no reasonable grounds for that belief.^[210]

In sentencing Martin, Curtain J accepted that the offending was driven by a perceived need for self-defence, on the basis of the threat of rape.^[211] Curtain J remarked that Martin’s answers in this way were ‘supported, to a degree, by [DNA spermatozoa] evidence’.^[212] Furthermore, Curtain J found that the unsolicited statement to police by another party, referred to as ‘Mr Y’, ‘alleges that the deceased sexually abused him from the ages of 10 to 14’.^[213] Although this allegation was ‘untested’, it ‘may be said to support the credibility of [Martin’s] answers’.^[214]

In commenting on this case, Toole argues that ‘Martin’s claim to defensive homicide is critically undermined by his admission of being angry and having “lost it’’’.^[215] However, this argument fails to consider Martin’s intellectual capacity appropriately, which, as accepted by the sentencing judge,

significantly diminished his ‘moral culpability’ and capacity to control anger.^[216] Martin was assessed as ‘having a full scale IQ of 59’ whereby his ‘intellectual functioning was described as in the extremely low range’, and he was noted as having ‘marginal cognitive abilities’.^[217] According to the Diagnostic and Statistical Manual of Mental Disorders: DSM-5, persons assessed within this IQ range may present with profound deficits in adaptive behaviour and intellectual functioning (such as conceptual, social and practical intelligence) required for everyday life.^[218] Furthermore, it was tendered that Martin had periodically been diagnosed with depression, which was likely a long-term sequela of other physical debilitations he suffers.^[219] Accepting this evidence, Curtain J remarked: ‘I am satisfied that the principles of Verdins ... are here applicable and operate to reduce your moral culpability by reason of your intellectual disability’.^[220] Martin was sentenced to eight years’ imprisonment with a non-parole period of five years.^[221]

C Director of Public Prosecutions (Vic) v Chen

In this case, Chen had responded to ‘an internet advertisement for the sale of a motor vehicle by the deceased’ approximately one week before the homicide occurred.^[222] It was acknowledged that Chen ‘did not really want the vehicle but, under pressure from [the deceased], ended up buying it for $1500’.^[223] Following this monetary exchange, Chen fatally stabbed the deceased at a train station — the offence was committed during the day in front of

witnesses and captured on CCTV.^[224] After a trial lasting two weeks, a jury found Chen not guilty of murder, but ‘guilty of the alternative charge of defensive homicide’.^[225] In sentencing the accused, Bell J accepted that the deceased ‘took advantage’ of Chen’s obvious vulnerability and ‘intimidated ... and threatened’ him, ensuring that Chen ‘saw a knife which he carried on his person and in his vehicle’, allowing him to extort ‘a further payment of $600’.^[226] In describing Chen’s state of mind at the time of committing the offence, Bell J stated, ‘[y]ou feared death or really serious injury at [the deceased’s] hands’.^[227] Bell J was guided by the expert testimony of the clinical neuropsychologist which established ‘a causal connection between [Chen’s] neuropsychological condition and the crime which [he] committed’.^[228] Accepting this, Bell J remarked:

There was a causal connection between that impairment and your offending. Consequently, your moral responsibility for the offending conduct ... is

reduced. ... The sentencing consideration of general deterrence must be sensibly moderated because you are not an appropriate vehicle for making an example to others. Specific deterrence must also be so moderated.^[229]

The Court also recognised that Chen ‘will find ... jail harder than most’ on account of cultural and linguistic barriers and the fact that Chen was a ‘young man with no family support in Australia ... [who speaks] practically no English and [has] a range of health issues, including ... hepatitis B’.^[230] These vulnerabilities were a significant consideration for Bell J, who further

remarked that Chen will ‘really struggle with the prison system’.^[231] Chen was sentenced to a period of eight years’ imprisonment, with a non-parole period of five years.^[232]

D The Case Analysis

These three cases provide a basis for arguing that defensive homicide was operating as a viable alternate offence for offenders with a mental illness and impairment. They also highlight the broad spectrum of mental illnesses and impairments that range on a continuum of severity, which is not always covered under the mental impairment legislation. Significantly, these cases show that despite the populist voices supporting abolition on the basis that it was not operating as intended,^[233] the offence was in fact operating in

accordance with one of its primary intentions — offering a halfway house for offenders with mental illness or impairment not amounting to a defence of mental impairment under the CMIA.^[234]

As evidenced by these cases, defensive homicide did not negate the legal responsibility of these accused persons, nor did it mean they did not face punishment for their actions. In all three cases, the accused were sentenced to a term of imprisonment — with the minimum non-parole period sitting at four years. What defensive homicide did offer in these cases was an

opportunity for the Crown, in cases of guilty pleas,^[235] or the jury, at trial,^[236] to determine the extent to which the accuseds were morally culpable for their conduct. In this way, defensive homicide allowed for the lower moral

culpability of these offenders to be taken into consideration in both the type of conviction recorded and then also in regards to the sentence ultimately imposed. Importantly, having an alternative offence that can capture these circumstances does not mean that all accused persons will have a guilty plea to defensive homicide accepted by the Crown, or be found guilty of this offence by the jury. Take, for example, R v Romero, in which the accused had an intellectual disability and killed in similar circumstances to Trezise.^[237] On plea, the accused did not seek to rely on the principles of Verdins owing to the fact that the psychological report tendered inadequately responded to the criteria required to enliven the principles.^[238] Instead, counsel for Romero invited the jury to convict the accused of defensive homicide rather than murder in recognition of his reduced culpability.^[239] However the jury returned a guilty verdict to murder. Romero was sentenced to 18 years’ imprisonment with a non-parole period of 15 years.^[240] This case supports the need for maintaining the flexibility to accommodate a range of conviction options and indicates that defensive homicide was not simply operating as a way for

males to ‘get away with murder’,^[241] but instead, offered an opportunity to capture the reduced moral culpability of some offenders, where it was

deemed appropriate.

Further to the three cases discussed, mental illness and impairment were highly prevalent among those convicted of defensive homicide, including formal diagnoses of schizophrenia,^[242] including paranoid schizophrenia,^[243] Dysthmic disorder,^[244] paranoia and anxiety^[245] and trauma-related mental illness.^[246] In addition to the complex mental conditions identified in the cases of R v Svetina,^[247] R v Monks,^[248] R v Vazquez,^[249] DPP (Vic) v McEwan^[250] and R v Koltuniewicz,^[251] the offenders each had a history of trauma, including chronic disadvantage, a long and significant history of family violence^[252] and kidnapping and torture.^[253] Furthermore, Dambitis, an accused in DPP (Vic) v McEwan, was (as a consequence of the Soviet regime) forcefully removed from his mother’s custody during childhood and involuntarily detained in a psychiatric hospital, which was geographically isolated from his family, and in which he experienced inhumane treatment.^[254]

In R v Monks,^[255] these factors went some way towards explaining Monks’s unreasonable belief in the need for lethal violence. With respect to an expert psychologist’s evidence, the sentencing judge remarked:

[Monks’s] offending could best be explained as a function of chronic factors, being family violence and the trauma it induced and [his] borderline

personality traits which resulted from it.^[256]

Monks was sentenced to eight years’ imprisonment with a non-parole period of five years.^[257] Similar concerns were identified by Kaye J in sentencing Dambitis in DPP (Vic) v McEwan^[258] to 11 years’ imprisonment, with a

non-parole period of eight years:

there are mitigating circumstances in your case. In particular, as a result of your extraordinary experiences during your childhood ... you have suffered a long standing and major depressive condition, and you also have sustained a

post-traumatic stress disorder. ... I accept that, to a limited degree, those conditions, and your traumatic younger years ... mitigate the seriousness of

your offending.^[259]

In R v Ball, the accused had been formally diagnosed with ‘paranoid

schizophrenia with longstanding fixed delusions and chronic [florid]

psychosis’.^[260] Ball was also dually diagnosed as suffering from anxiety and noted as having very low intelligence, with a full-scale IQ of 77, placing him just outside the borderline intellectually disabled range.^[261] It was accepted by the sentencing judge that, according to the evidence tendered by the consultant psychiatrist, at the time of committing the offence, Ball was ‘acutely psychotic’ and ‘it was probable that [Ball] had the defence of mental

impairment available’ to him.^[262] However, perhaps for the reasons we have outlined regarding the limitations of the CMIA, Ball did not rely on the mental impairment defence and instead pleaded guilty to two counts of defensive homicide.^[263] He received a maximum sentence of 20 years’

imprisonment (for two charges of defensive homicide, combined with one charge of recklessly causing serious injury), with a minimum non-parole period of 17 years.^[264]

Without a halfway house between murder and manslaughter, it is likely that the moral culpability of offenders suffering from mental illness and impairment cannot be adequately recognised in the type of conviction or sentence imposed. Further to this, there are additional implications that arise post-sentence, within the prison context. The findings from the Victorian Ombudsman’s investigation into deaths and harms in custody found that almost half the prison population ‘have a psychiatric risk rating indicating mental health concerns’,^[265] but due to severe overcrowding, only the ‘sickest of the sick’ (those with active psychotic symptoms) are able to access psychiatric treatment, ‘leaving the prison system to manage a large number of acutely unwell prisoners’.^[266] While problematic for multiple reasons, in the context of defensive homicide’s abolition, this means that accused persons who would have been eligible for a defensive homicide conviction (and who would now be likely to receive a longer prison sentence for a more culpable conviction) will likely have difficulties accessing effective and sustained treatment while

in prison.

VIII CONCLUSION

Defensive homicide was introduced to take account of the fact that people kill in a range of different circumstances, and their culpability may be affected by a range of factors. While it may not have been deemed completely effective in relation to the number of women able to access the offence prior to its abolition (although the benefits of retaining the offence for women were, and continue to be, strongly advocated by family violence stakeholder groups),^[267] our analysis of the cases suggests it was working for some men.

While this article focuses on male-on-male violence, our arguments also have relevance to female perpetrators. This is because, outside of the mental impairment defence, Victorian law now only accommodates murder or manslaughter convictions, and as we have argued, the law does not deal well with impairments such as intellectual disability. In their study of female homicide offenders in Victoria, Bennett et al found the participants in their sample to be 43.17 times more likely to have schizophrenia compared to the general community comparisons.^[268] Such findings suggest that even despite the relatively low number of female homicide offenders in Victoria, women with a mental illness and impairments who kill (whether or not in the context of family violence) will also be disadvantaged by the abolition of defensive homicide. It is just likely (given the smaller number of female offenders) that the full impacts of its abolition on female accused persons will take longer

to see.

The changes introduced alongside the abolition of defensive homicide through the ADHA included reformulating the test for self-defence,^[269] an amendment to the Jury Directions Act 2013 (Vic) on the relevance of family violence to the defences of self-defence and duress,^[270] and an amendment to the Evidence Act 2008 (Vic)^[271] to allow for the exclusion of evidence, ‘if its probative value is substantially outweighed by the danger that [it] might ... unnecessarily demean the deceased in a criminal proceeding for a homicide offence’.^[272] While we welcome these reforms, as Kirkwood et al have argued, ‘[t]here is no certainty that, in the absence of defensive homicide’, women who kill their abusive partners will proceed to trial and rely on self-defence in the hope of an acquittal.^[273] In fact, they may plead or be encouraged to plead ‘guilty to murder in order to receive a discounted sentence, rather than risk a murder conviction and longer sentence at trial’.^[274] As we have argued

extensively throughout this article, recent comprehensive research into defensive homicide cases has unequivocally shown that the fear of abusive male partners ‘[g]etting away with murder’^[275] was not reflective of the reality of defensive homicide in practice; and defensive homicide was in fact operating (and would likely have continued to operate) effectively for female offenders who required a halfway house between self-defence and murder.^[276] Thus despite claims to the contrary,^[277] the abolition of defensive homicide is unlikely to benefit female offenders (or others) who kill in response to prolonged family violence.

As evidenced by the cases discussed, there are significant complexities and nuances in cases involving a fatality. The personal histories of the men who relied on defensive homicide are far more complex than the image of the ‘violent’ thug depicted in the media and by populist voices in the lead-up to the abolition of defensive homicide.^[278] Accordingly, we believe there is justification for an offence such as defensive homicide, and in line with many others,^[279] we contend that its abolition was premature.

The challenge now is to move towards a new offence or the introduction of a defence that adequately recognises the harms caused to the victim, but also takes into account the impact of mental health on one’s offending behaviour in a way that aligns with community values and expectations. We believe that benefits would flow from the implementation of the VLRC’s

recommendations regarding the CMIA, insofar as the changes have the potential to provide an alternative legal response that would be appropriate for offenders, particularly if the new definition is introduced and the new

five-year review timescale,^[280] as opposed to the 25–year nominal period, is implemented. However, without the safety net of an alternative defence or charge to murder,^[281] Victoria’s homicide laws will be out of touch with modern understandings of mental illness and impairment, and the law will remain inflexible and inaccessible for some of the most vulnerable, disadvantaged and marginalised people in our community.

^[*] BA (Hons) (Monash); Doctoral Candidate and Teaching Associate, School of Social Sciences, Faculty of Arts, Monash University. This article is based on research undertaken by the

corresponding author for the degree of Bachelor of Arts (Hons). We thank the anonymous

reviewers for their thoughtful comments on an earlier version of this article. We also thank the Editors for their generous and incisive feedback on this article.

^[†] BA (Hons), PhD (Monash); Senior Lecturer, School of Social Sciences, Faculty of Arts, Monash University.

^[‡] BA (Hons) (LaTrobe), PhD (Melb); Senior Lecturer, School of Humanities and Social Sciences, Faculty of Arts and Education, Deakin University.

^[1] See, eg, Debra Bennett, An Investigation of 435 Sequential Homicides in Victoria: The Implication of Psychosis, Motive for Offending, Substance Abuse and Gender (DPsych Thesis, Monash University, 2010); Seena Fazel and Martin Grann, ‘The Population Impact of Severe Mental Illness on Violent Crime’ (2006) 163 American Journal of Psychiatry 1397; Arlie Loughnan, Manifest Madness: Mental Incapacity in Criminal Law (Oxford University Press, 2012); Pamela J Taylor and John Gunn, ‘Violence and Psychosis: I — Risk of Violence among Psychotic Men’ (1984) 288 British Medical Journal 1945; Cameron Wallace et al, ‘Serious Criminal Offending and Mental Disorder: Case Linkage Study’ (1998) 172 British Journal of Psychiatry 477.

^[2] Loughnan, above n 1, 16.

^[3] Ibid. For approaches which do not predominantly focus on exculpation see, eg, Bernadette McSherry and Kay Wilson, ‘The Concept of Capacity in Australian Mental Health Law Reform: Going in the Wrong Direction?’ (2015) 40 International Journal of Law and Psychiatry 60; Claire Spivakovsky, ‘Making Risk and Dangerousness Intelligible in Intellectual Disability’ (2014) 23 Griffith Law Review 389.

^[4] Cognitive impairment is the broad term comprising a range of disabilities such as, but not limited to, intellectual disability, acquired/traumatic brain injury, foetal alcohol spectrum disorder, neurological disorders, autism spectrum disorder and dementia. This umbrella term will be used throughout the article, except when referring to specific cases and individuals, whereby the specific disability of the individual is noted (for example, intellectual disability). In this article, we consider mental illness (eg bipolar, schizophrenia, depression) and

cognitive impairment to be distinct from each other, with disability (forming a part of the personhood) considered separate to an illness (typically episodic) treatable with medication. This distinction is important to make because mental illnesses and cognitive impairments have different implications in terms of service provision. For people with cognitive impairments, often their comorbid mental illness(es) can be overshadowed by their lifelong disabilities and can remain undiagnosed/untreated. There are also associated poorer treatment outcomes due to this lack of understanding. For a more extensive discussion of how mental illness can be overshadowed by an existing disability see, eg, Jonathan Mason and Katrina Scior, ‘“Diagnostic Overshadowing” amongst Clinicians Working with People with

Intellectual Disabilities in the UK’ (2004) 17 Journal of Applied Research in Intellectual

Disabilities 85.

^[5] As repealed by Crimes Amendment (Abolition of Defensive Homicide) Act 2014 (Vic) s 3(3).

^[6] Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997 (Vic) pt 4. See Victorian Law Reform Commission, Defences to Homicide, Final Report (2004) xxxviii.

^[7] Crimes Act 1958 (Vic) s 9AD, as repealed by ADHA s 3(3).

^[8] See below Table 1 in Part VII for case details.

^[9] Crimes (Homicide) Act 2005 (Vic) s 6, inserting Crimes Act 1958 (Vic) s 9AD.

^[10] This date range captures all cases involving defensive homicide convictions in Victoria prior to its abolition on 1 November 2014: see Governor (Vic), ‘Crimes Amendment (Abolition of Defensive Homicide) Act 2014 (Vic) — Proclamation of Commencement’ in Victoria,

Victorian Government Gazette: Special, No S 350, 7 October 2014, 1. The sentencing decisions were accessed using the Australian Legal Information Institute (‘AustLII’) database.

^[11] This case involved a child offender. Thus, the sentencing transcript is not publicly available and has been excluded from the analysis.

^[12] [2009] VSC 520 (31 August 2009) (‘Trezise’).

^[13] [2011] VSC 217 (20 May 2011) (‘Martin’).

^[14] [2013] VSC 296 (11 June 2013) (‘Chen’).

^[15] See s 20(1). The Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997 (Vic) will also be referred to as the ‘mental impairment defence’ throughout this article.

^[16] This view is demonstrated by a cursory review of the data available on AustLII over a 15–year period (January 2000–October 2015) which reveals the CMIA was pleaded only once by an accused person with a cognitive impairment in a case involving an intellectual disability. It has otherwise been used exclusively in cases involving psychotic illness: see, eg, DPP (Vic) v Whelan [2006] VSC 319; (2006) 177 A Crim R 449; R v Trucano [2010] VSC 271 (17 June 2010); R v Lloga [2011] VSC 615 (25 November 2011); R v Cavallari [2013] VSC 84 (1 March 2013); R v Konidaris [2014] VSC 89 (27 February 2014).

^[17] ‘Defensive Homicide Law to Be Dumped in Victoria after Violent Men “Allowed to Get Away with Murder”’, ABC News (online), 22 June 2014, quoting Robert Clark <http://www.abc.net.au/news/2014-06-22/vic-dumps-law-that-allowed-men-to-27get-away-with-murder27/5541670> . See also Kate Fitz-Gibbon, ‘Defensive Homicide Law Akin to Getting Away with Murder’, The Australian (Sydney), 3 March 2012, 21.

^[18] See, eg, Oliver Milman, ‘Victoria Will Scrap “Defensive Homicide” and Offer Simpler Test for Self-Defence’, The Guardian (online), 22 June 2014, citing Robert Clark <http://www.theguardian.com/world/2014/jun/22/victoria-will-scrap-defensive-homicide-and-offer-simpler-test-for-self-defence> Kate Fitz-Gibbon, ‘Abolishing Defensive Homicide Offence Essential’, The Age (online), 4 October 2013 <http://www.theage.com.au/comment/abolishing-defensive-homicide-offence-essential-20131004-2uyya.html> Kate Fitz-Gibbon, ‘The Offence of Defensive Homicide: Lessons Learned from Failed Law Reform’ in Kate Fitz-Gibbon and Arie Freiberg (eds), Homicide Law Reform in Victoria: Retrospect and Prospects (Federation Press, 2015) 128.

^[19] See, eg, Kate Fitz-Gibbon, ‘The Victorian Operation of Defensive Homicide: Examining the Delegitimisation of Victims in the Criminal Court System’ (2012) 21 Griffith Law Review 555; Kellie Toole, ‘Defensive Homicide on Trial in Victoria’ [2013] MonashULawRw 17; (2013) 39 Monash University Law Review 473. See also Asher Flynn and Kate Fitz-Gibbon, ‘Bargaining with Defensive Homicide: Examining Victoria’s Secretive Plea Bargaining System Post-Law Reform’ [2011] MelbULawRw 31; (2011) 35 Melbourne University Law Review 905, 911 in which the authors acknowledge that most defensive homicide cases involve male offenders with a mental illness.

^[20] See, eg, Elissa Hunt, ‘Defensive Homicide Law Hijacked by Career Criminals, Drug Addicts and Drunks’, The Herald Sun (online), 4 October 2013 <http://www.heraldsun.com.au/news/law-order/defensive-homicide-law-hijacked-by-career-criminals-drug-addicts-and-drunks/story-fni0ffnk-1226732569460> .

^[21] Victorian Law Reform Commission, Defences to Homicide Final Report, above n 6, xxxviii.

^[22] Australian Institute of Health and Welfare, The Health of Australia’s Prisoners 2012 (2013)

35–6.

^[23] Ombudsman (Vic), Investigation into Deaths and Harm in Custody (2014) 6. The report also noted that 55 per cent of Victoria’s prison population has an identified risk of suicide or self-harm: at 6.

^[24] See, eg, Bennett, An Investigation of 435 Sequential Homicides Thesis, above n 1; Debra Bennett et al, ‘A Study of Psychotic Disorders among Female Homicide Offenders’ (2012) 18 Psychology, Crime and Law 231; Fazel and Grann, above n 1; Loughnan, above n 1; Jenny Mouzos, Trends and Issues in Crime and Criminal Justice: Mental Disorder and Homicide in Australia, Research Paper No 133 (1999); Marnie E Rice and Grant T Harris, ‘Psychopathy, Schizophrenia, Alcohol Abuse, and Violent Recidivism’ (1995) 18 International Journal of Law and Psychiatry 333; Taylor and Gunn, above n 1; Wallace et al, above n 1.

^[25] For further discussion of these ‘conflicting opinions and apparently contradictory empirical data’ see D J Bennett et al, ‘Schizophrenia Disorders, Substance Abuse and Prior Offending in a Sequential Series of 435 Homicides’ (2011) 124 Acta Psychiatrica Scandinavica 226, 227.

^[26] For a discussion of these limitations see Bennett, An Investigation of 435 Sequential Homicides Thesis, above n 1, 72.

^[27] See, eg, Claire Spivakovsky, ‘From Punishment to Protection: Containing and Controlling the Lives of People with Disabilities in Human Rights’ (2014) 16 Punishment and Society: The International Journal of Penology 560; Elizabeth Walsh, Alec Buchanan and Thomas Fahy, ‘Violence and Schizophrenia: Examining the Evidence’ (2002) 180 British Journal of

Psychiatry 490.

^[28] In May 2015, Queensland Health Minister Cameron Dick announced a Queensland-wide clinical review of homicides and attempted homicides where the victim or offender is a person with a mental illness. The purpose of the review, led by Professor James R P Ogloff, was cited as being ‘to make recommendations to improve Queensland’s health system and treatment of people suffering mental illness’: see Tony Moore, ‘Review of Mental Health Role in Murders and Major Crimes’, Brisbane Times (online), 7 May 2015 <http://www.brisbanetimes.com.au/queensland/review-of-mental-health-role-in-murders-and-major-crimes-20150507-ggwpwv.htm l> .

^[29] Ombudsman (Vic), Investigation into the Rehabilitation and Reintegration of Prisoners in Victoria (2015) 32.

^[30] See T Short et al, ‘Comparing Violence in Schizophrenia Patients with and without Comorbid Substance-Use Disorders to Community Controls’ (2013) 128 Acta Psychiatrica Scandinavica 306.

^[31] See also James R P Ogloff et al, Trends and Issues in Crime and Criminal Justice: The Identification of Mental Disorders in the Criminal Justice System, Research Paper No 334 (2007). For a discussion of criminal justice workers’ lack of awareness of the nature of Asperger’s Syndrome see Ann Browning and Laura Caulfield, ‘The Prevalence and Treatment of People with Asperger’s Syndrome in the Criminal Justice System’ (2011) 11 Criminology and Criminal Justice 165.

^[32] See, eg, Bennett, An Investigation of 435 Sequential Homicides Thesis, above n 1, ch 5; Bennett et al, ‘A Study of Psychotic Disorders’, above n 24. See generally Ogloff et al, above n 31; Short et al, above n 30.

^[33] Bennett et al, ‘Schizophrenia Disorders, Substance Abuse and Prior Offending’, above n 25, 230.

^[34] See, eg, John Junginger et al, ‘Effects of Serious Mental Illness and Substance Abuse on Criminal Offenses’ (2006) 57 Psychiatric Services 879, 882. See generally Browning and Caulfield, above n 31, 172.

^[35] Richard Guilliatt, ‘Mentally Ill “More Prone to Violence”’, The Australian (Sydney), 11 May 2013, 1, quoting Paul Mullen.

^[36] See James Ogloff, ‘The Relationship between Mental Illness and Homicide’ (Speech delivered at the Australian Institute of Criminology, Homicide: Precursors and Prevention Conference, Brisbane, 24–5 March 2014) <https://www.youtube.com/watch?v=y9TFzWu3YHQ>. See generally Paul E Mullen, ‘Mental Disorder and Dangerousness’ (1984) 18 Australian and New Zealand Journal of Psychiatry 8.

^[37] See Short et al, above n 30, 312.

^[38] Ibid.

^[39] Ibid.

^[40] See Guilliatt, above n 35, 8.

^[41] See CMIA s 20(1).

^[42] [1843] EngR 875; (1843) 10 Cl & Fin 200; 8 ER 718. This defence required that it be established beyond reasonable doubt that at the time of the alleged act, ‘the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong’: at 210; 722 (Tindal CJ).

^[43] Ibid.

^[44] VLRC, Defences to Homicide, Options Paper (2003) 178 [5.30].

^[45] [2007] VSCA 102; (2007) 16 VR 269, 276 [32]: the principle can be used where the ‘[i]mpaired mental functioning’: (1) may reduce moral culpability as distinct from legal responsibility; (2) may impact on the type of sentence imposed and the conditions in which it should be served; (3) may moderate or eliminate the need for general deterrence; (4) may moderate or eliminate the need for specific deterrence; (5) ‘may mean that a given sentence will weigh more heavily on the offender’ than a person in normal health (based on the existence of the condition at the date of sentencing, or its foreseeable recurrence); and (6) ‘[w]here there is a serious risk of imprisonment having a significant adverse effect on the offender’s mental health’. The principle has recently been held in DPP (Vic) v O’Neill [2015] VSCA 325 (2 December 2015) not to include personality disorders, despite being frequently enlivened in homicide cases involving (trauma-related) borderline personality disorder: see, eg, R v Cook [2015] VSC 406 (19 August 2015); R v Stensholt [2014] VSC 668 (24 March 2014). See also the defensive homicide case, R v Monks [2011] VSC 626 (2 December 2011).

^[46] [1996] VicRp 26; [1996] 1 VR 398.

^[47] Dion G Gee and James R P Ogloff, ‘Sentencing Offenders with Impaired Mental Functioning: R v Verdins, Buckley and Vo [2007] at the Clinical Coalface’ (2014) 21 Psychiatry, Psychology and Law 46, 46.

^[48] [2007] VSCA 102; (2007) 16 VR 269.

^[49] [1996] VicRp 26; [1996] 1 VR 398.

^[50] See, eg, Gee and Ogloff, above n 47, 47.

^[51] [2007] VSCA 102; (2007) 16 VR 269.

^[52] Ian Freckelton, ‘Sentencing Offenders with Impaired Mental Functioning: R v Verdins, Buckley and Vo [2007] VSCA 102; (2007) 169 A Crim R 581 Maxwell P, Vincent and

Buchanan JJ’ (2007) 14 Psychiatry, Psychology and Law 359, 362–3.

^[53] Gee and Ogloff, above n 47, 46.

^[54] Sally Traynor, ‘Sentencing Mentally Disorded Offenders: The Causal Link’ (2002) 23 Sentencing Trends and Issues <http://www.judcom.nsw.gov.au/publications/st/st23/ index.html> .

^[55] [2001] NSWCCA 155; (2001) 53 NSWLR 251.

^[56] Traynor, above n 54. The High Court in Muldrock v The Queen [2011] HCA 39; (2011) 244 CLR 120 held that people with intellectual disability should receive sentences tailored to their unique needs. However, our review of recent homicide cases indicates that this authority has not been applied and that the general Verdins principles continue to be applied for either mental illness or cognitive impairment.

^[57] VLRC, Defences to Homicide Final Report, above n 6, 203–25.

^[58] Ibid xxxvii.

^[59] Ibid xxxviii.

^[60] Ibid.

^[61] Robert Clark, ‘Victorian Law Reform Commission to Review Law on Mental Impairment and Culpability’ (Media Release, 15 August 2012).

^[62] Ibid. However, the meaning of ‘mental impairment’ in the CMIA was considered in the context of broader reform to homicide defences: see, eg, VLRC, Defences to Homicide Final Report, above n 6.

^[63] VLRC, Review of the Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997, Consultation Paper (2013) 3 [1.5].

^[64] VLRC, Review of the Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997, Final Report (2014) 110–13.

^[65] Ibid 113.

^[66] Ibid.

^[67] See VLRC, Defences to Homicide Final Report, above n 6, 239 [5.114]; VLRC, Review of the CMIA Consultation Paper, above n 63, 211, 146.

^[68] VLRC, Review of the CMIA Consultation Paper, above n 63, 138–9 [7.13]–[7.16]. While those found NGMI may be unconditionally released, a review of all Victorian homicide cases between 2000 and 2015 reveals that all NGMI cases resulted in a supervision order with a 25–year nominal term, including one case in which the accused person was found unfit to stand trial due to intellectual disability. In this case, it was found that there were no suitable psychiatric services available to accommodate the accused and no practical alternatives to committing the accused to custody. As such, despite being found unfit to stand trial, it

was directed that the accused be detained in prison: R v Coulter [2014] VSC 42

(27 February 2014)

^[69] See VLRC, Review of the CMIA Consultation Paper, above n 63, 139 [7.17].

^[70] VLRC, Defences to Homicide Final Report, above n 6, 221 [5.56].

^[71] Ibid 221–2 [5.58].

^[72] VLRC, Review of the CMIA Consultation Paper, above n 63, 148–9.

^[73] VLRC, Defences to Homicide Options Paper, above n 44, 171 [5.5].

^[74] Ibid 180 [5.37]–[5.39]. Where it is successfully raised, this nearly always occurs with the consent of the prosecution.

^[75] VLRC, Review of the CMIA Final Report, above n 64, 15 [2.27].

^[76] VLRC, Defences to Homicide Final Report, above n 6, 207 [5.14].

^[77] VLRC, Defences to Homicide Options Paper, above n 44, 180 [5.37].

^[78] VLRC, Review of the CMIA Final Report, above n 64, tabled in Parliament as Victoria, Review of the Crimes (Mental Impairment and Unfitness to Be Tried) Act 1997, Parl Paper No

388 (2014).

^[79] VLRC, Review of the CMIA Final Report, above n 64, 361–9.

^[80] Ibid 80.

^[81] Ibid 115.

^[82] Ibid xxxv.

^[83] NSW Law Reform Commission, People with Cognitive and Mental Health Impairments in the Criminal Justice System: Criminal Responsibility and Consequences, Report No 138 (2013).

^[84] Ibid 46 [3.19].

^[85] Ibid 45 [3.16].

^[86] Ibid 54 [3.51].

^[87] Ibid 45 [3.16].

^[88] Ibid.

^[89] Ibid 50–1 [3.41].

^[90] Ibid 79–80 [3.169] (recommendation 3.6).

^[91] Ibid 79 [3.165], citing Office of the Director of Public Prosecutions (NSW), Submission No MH5 to the NSWLRC, People with Cognitive and Mental Health Impairments in the Criminal Justice System, 2013, 7.

^[92] NSWLRC, above n 83, xv [0.2], xvii [0.17]–[0.18].

^[93] Crimes Act 1900 (NSW) s 23A(5).

^[94] Ibid s 23A(1). See also NSWLRC, above n 83, 85–6 [4.9]–[4.11].

^[95] See, eg, Kellie Toole, ‘Review of Mental Impairment Defence’ (2014) 39 Alternative Law Journal 141. For the relevant provisions see Criminal Law Consolidation Act 1935 (SA) pt 8A.

^[96] Toole, ‘Review of Mental Impairment Defence’, above n 95, 141. See Sentencing Advisory Council, Attorney-General’s Department (SA), Mental Impairment and the Law: A Report on the Operation of Part 8A of the Criminal Law Consolidation Act 1935 (SA) (2014).

^[97] See NSWLRC, above n 83, 90 [4.22]. See also Office of the Director of Public Prosecutions (NSW), ‘Submission to the NSWLRC’, above n 91.

^[98] NSWLRC, above n 83, 90 [4.21].

^[99] Ibid 90 [4.22]. See also Office of the Director of Public Prosecutions (NSW), ‘Submission to the NSWLRC’, above n 91, 11.

^[100] Five out of the six written submissions on this issue recommended retention of substantial impairment: see NSWLRC, above n 83, 90–1 [4.21], [4.23]–[4.25].

^[101] NSWLRC, above n 83, 87 [4.15].

^[102] Ibid.

^[103] Judicial Commission of NSW, Partial Defences to Murder in New South Wales 1990–2004 (2006) 9.

^[104] NSWLRC, above n 83, xvii [0.17]–[0.18], 101 [4.62]–[4.65].

^[105] Ibid xvii [0.17].

^[106] See below Table 1 in Part VII.

^[107] VLRC, Defences to Homicide Final Report, above n 6, xxxviii.

^[108] Carolyn B Ramsey, ‘Provoking Change: Comparative Insights on Feminist Homicide Law Reform’ (2010) 100 Journal of Criminal Law and Criminology 33, 33.

^[109] 2005 Act s 3, inserting Crimes Act 1958 (Vic) s 3B.

^[110] 2005 Act s 6, inserting Crimes Act 1958 (Vic) s 9AC.

^[111] 2005 Act s 6, inserting Crimes Act 1958 (Vic) s 9AH(1).

^[112] 2005 Act s 6, inserting Crimes Act 1958 (Vic) s 9AH(2).

^[113] 2005 Act s 6, inserting Crimes Act 1958 (Vic) s 9AD.

^[114] Flynn and Fitz-Gibbon, ‘Bargaining with Defensive Homicide’, above n 19, 910.

^[115] VLRC, Defences to Homicide Final Report, above n 6, xxix.

^[116] This expression was primarily used in relation to excessive self-defence and family violence. However, the VLRC broadly acknowledged that excessive self-defence ‘may play an important role in providing a “halfway house” for those cases where self-defence is not successful, but where manslaughter is the more appropriate outcome’: ibid 94 [3.91]. The decision to introduce defensive homicide was influenced by these findings.

^[117] Victoria, Parliamentary Debates, Legislative Assembly, 6 October 2005, 1351 (Rob Hulls).

^[118] [2010] VSC 202 (19 May 2010), affd [2012] VSCA 47; (2012) 35 VR 193.

^[119] Middendorp [2010] VSC 202 (19 May 2010) [10] (Byrne J).

^[120] Ibid [9].

^[121] Ibid [20].

^[122] See, eg, Sarah Capper and Mary Crooks, ‘New Homicide Laws Have Proved Indefensible’, The Sunday Age (Melbourne), 23 May 2010, 21; Adrian Howe, ‘Another Name for Murder’, The Age (Melbourne), 24 May 2010, 15; Padraic Murphy, ‘When Luke Middendorp Killed Jade Bownds, He Was Twice Her Size ... and He Said it Was Self-Defence’, Herald Sun

(Melbourne), 12 April 2010, 13.

^[123] Criminal Law — Justice Statement, DOJ (Vic), ‘Defensive Homicide: Review of the Offence of Defensive Homicide’ (Discussion Paper, DOJ (Vic), August 2010) 5.

^[124] Ibid 28–32 [82]–[113]. Neither case proceeded to trial: at 28 [83].

^[125] Ibid 32 [113].

^[126] See ibid 33–40 [114]–[151]. The thirteenth case, Middendorp [2010] VSC 202 (19 May 2010), involved a male offender and female victim.

^[127] Criminal Law — Justice Statement, above n 123, 36 [130].

^[128] R v Spark [2009] VSC 374 (11 September 2009). See ibid 36 [131].

^[129] Criminal Law Review, DOJ (Vic), ‘Defensive Homicide: Proposals for Legislative Reform’ (Consultation Paper, DOJ (Vic), September 2013) 31–5.

^[130] Ibid viii.

^[131] Ibid viii, 8–11. See R v Creamer [2011] VSC 196 (20 April 2011), affd [2012] VSCA 182 (16 August 2012); R v Black [2011] VSC 152 (12 April 2011), affd [2012] VSCA 75 (26 April 2012); R v Edwards [2012] VSC 138 (24 April 2012).

^[132] Criminal Law Review, DOJ (Vic), ‘Defensive Homicide Consultation Paper’, above n 129, viii.

^[133] Ibid 29.

^[134] Ibid 30.

^[135] [2010] VSC 202 (19 May 2010).

^[136] Criminal Law Review, DOJ (Vic), ‘Defensive Homicide Consultation Paper’, above n 129, 30.

^[137] Ibid viii.

^[138] Although the ADHA was enacted in September 2014, s 3 (which abolished defensive homicide) came into effect on 1 November 2014: Governor (Vic), ‘Crimes Amendment (Abolition of Defensive Homicide) Act 2014 (Vic) — Proclamation of Commencement’ in Victoria, Victorian Government Gazette: Special, No S 350, 7 October 2014, 1.

^[139] See Asher Flynn, Mark Halsey and Murray Lee, ‘Emblematic Violence and Aetiological

Cul-De-Sacs: On the Discourse of “One-Punch” (Non)Fatalities’ (2016) 56 British Journal of Criminology 179.

^[140] David Garland, The Culture of Control: Crime and Social Order in Contemporary Society (University of Chicago Press, 2001) 4.

^[141] See Flynn, Halsey and Lee, ‘Emblematic Violence and Aetiological Cul-De-Sacs’, above n 139.

^[142] Crimes Act 1900 (NSW) s 25A(2); see at s 25B(1) for the requirement that courts ‘impose a sentence of imprisonment of not less than 8 years’.

^[143] Julia Quilter, ‘One-Punch Laws, Mandatory Minimums and “Alcohol-Fuelled” as an Aggravating Factor: Implications for NSW Criminal Law’ (2014) 3(1) International Journal for Crime, Justice and Social Democracy 81, 82.

^[144] Ibid 84.

^[145] Flynn, Halsey and Lee, ‘Emblematic Violence and Aetiological Cul-De-Sacs’,

above n 139, 184.

^[146] As inserted by Sentencing Amendment (Coward’s Punch Manslaughter and Other Matters) Act 2014 (Vic) s 3.

^[147] As inserted by ibid s 6.

^[148] Interestingly, given how the abolition of defensive homicide has impacted on attempts to reduce the culpability of mentally impaired offenders, the mandatory minimum sentence applied in Victorian ‘one-punch’ cases is subject to special reason exceptions, including ‘impaired mental functioning’: see Sentencing Act 1991 (Vic) ss 10A(2)(c), 9C(2). This provides scope to address some of the concerns we outline in this article in relation to the abolition of defensive homicide within the context of the one-punch fatality.

^[149] Crimes Act 1958 (Vic) s 4A(2).

^[150] Ibid s 4A(4).

^[151] Ibid s 4A(3).

^[152] Jennifer Lucinda Pilgrim, Dimitri Gerostamoulos and Olaf Heino Drummer, ‘“King Hit” Fatalities in Australia, 2000–2012: The Role of Alcohol and Other Drugs’ (2014) 135 Drug and Alcohol Dependence 119, 120–1.

^[153] Ibid.

^[154] Victoria Police, Annual Report 2013–2014 (2014) 33.

^[155] Pilgrim, Gerostamoulos and Drummer, above n 152, 120–1.

^[156] Garland, Culture of Control, above n 140.

^[157] Ibid 13 (emphasis in original).

^[158] See, eg, ‘Killer Law to Be Scrapped’, The Age (Melbourne), 23 June 2014, 9; Howe,

above n 122.

^[159] Fitz-Gibbon, ‘Getting Away with Murder’, above n 17, 21.

^[160] Fitz-Gibbon, ‘Abolishing Defensive Homicide Offence Essential’, above n 18.

^[161] See John Champion, ‘Plea Offers and Discontinuances: DPP Speech’ (Speech delivered at the Law Institute of Victoria Annual Criminal Law Conference, Melbourne, 27 July 2012) <http://www.opp.vic.gov.au/getattachment/4391b56c-dfce-4569-ba7f-571999aa0388/DPP-speech-to-LIV-annual-criminal-law-conference.aspx> John Champion, ‘The Importance of Plea Agreements: Our Approach and Accountability’ (Speech delivered at the 14^th International Criminal Law Congress, Melbourne, 12 October 2014 <http://www.opp.vic.gov.au/getattachment/8eed0322-7d85-40b0-b593-d8ef97aaf41a/The-Importance-of-Plea-Agreements-Our-Approach-and.aspx> .

^[162] Champion, ‘Plea Offers and Discontinuances’, above n 161.

^[163] Champion, ‘The Importance of Plea Agreements’, above n 161.

^[164] Ibid.

^[165] Director of Public Prosecutions (Vic), Director’s Policy: Prosecutorial Discretion (2014)

2–4 [3]–[7].

^[166] Fitz-Gibbon, ‘Getting Away with Murder’, above n 17, 21.

^[167] ‘Defensive Homicide Law to be Dumped’, above n 17, quoting Robert Clark.

^[168] David Garland, ‘The Limits of the Sovereign State: Strategies of Crime Control in Contemporary Society’ (1996) 36 British Journal of Criminology 445, 461.

^[169] Domestic Violence Resource Centre Victoria, ‘Defensive Homicide an Essential Safety Net for Victims’ on Domestic Violence Resource Centre Victoria, Domestic Violence Resource Centre Victoria (23 July 2014) <http://www.dvrcv.org.au/knowledge-centre/our-blog/defensive-homicide-essential-safety-net-victims> .

^[170] Debbie Kirkwood et al, Submission to DOJ (Vic), Defensive Homicide: Proposals for Legislative Reform — Consultation Paper, 27 November 2013.

^[171] Kate Fitz-Gibbon, ‘Abolishing Defensive Homicide Will Benefit Female Victims and Offenders’, The Conversation (online), 4 October 2013 <https://www.theconversation.com/abolishing-defensive-homicide-will-benefit-female-victims-and-offenders-18484>.

^[172] See, eg, Milman, above n 18, citing Robert Clark; Fitz-Gibbon, ‘Abolishing Defensive Homicide Offence Essential’, above n 18; Fitz-Gibbon, ‘Getting Away with Murder’,

above n 17.

^[173] See Fitz-Gibbon, ‘The Victorian Operation of Defensive Homicide’, above n 19; Toole, ‘Defensive Homicide on Trial in Victoria’, above n 19. But see Flynn and Fitz-Gibbon,

‘Bargaining with Defensive Homicide’, above n 19, in which the authors acknowledge that most defensive homicide cases involve male offenders with a mental illness.

^[174] Victoria, Parliamentary Debates, Legislative Assembly, 6 October 2005, 1351 (Rob Hulls).

^[175] Fitz-Gibbon, ‘Abolishing Defensive Homicide Will Benefit Female Victims and Offenders’, above n 171.

^[176] Ibid.

^[177] See Debbie Kirkwood, Mandy McKenzie and Danielle Tyson, ‘Justice or Judgement? The Impact of Victorian Homicide Law Reforms on Responses to Women Who Kill Intimate Partners’ (Discussion Paper No 9, Domestic Violence Resource Centre Victoria, 2013) 49–51; Danielle Tyson et al, ‘The Effects of the 2005 Reforms on Legal Responses to Women Who Kill Intimate Partners’ in Kate Fitz-Gibbon and Arie Freiberg (eds), Homicide Law Reform in Victoria: Retrospect and Prospects (Federation Press, 2015) 76.

^[178] Tyson et al, above n 177, 76.

^[179] DPP (Vic) v Williams [2014] VSC 304 (27 June 2014).

^[180] Tyson et al, above n 177, 89–90.

^[181] Ibid 92.

^[182] See, eg, Kate Fitz-Gibbon, ‘Legitimising Lethal Male Violence: Why Defensive Homicide Needs to Be Abolished’, The Conversation (online), 29 May 2013 <https://www.theconversation.com/legitimising-lethal-male-violence-why-defensive-homicide-needs-to-be-abolished-14578>.

^[183] Garland, Culture of Control, above n 140, 136.

^[184] Ibid.

^[185] Fitz-Gibbon, ‘Legitimising Lethal Male Violence’, above n 182.

^[186] VLRC, Defences to Homicide Final Report, above n 6, 238 [5.110]. Although the VLRC did note this, it ultimately recommended not implementing diminished responsibility due to concerns it may be inappropriately used in the context of family violence. In the context of substantial impairment, the NSWLRC dismissed similar arguments for not being supported by evidence: see NSWLRC, above n 83, 97 [4.48].

^[187] NSWLRC, above n 83, 101 [4.63].

^[188] Trezise [2009] VSC 520 (31 August 2009) [9]–[23] (Coghlan J).

^[189] Ibid [5], [14].

^[190] Ibid [2].

^[191] Ibid [13].

^[192] Ibid [14].

^[193] Ibid [16].

^[194] Ibid.

^[195] See generally Victoria Legal Aid, Submission No 52 to Law Reform Committee, Parliament of Victoria, Inquiry into Access to and Interaction with the Justice System by People with an

Intellectual Disability and their Families and Carers, 2 November 2011.

^[196] Ibid 10. See also Lorana Bartels, ‘Police Interviews with Vulnerable Adult Suspects’ (Research in Practice Report No 21, Australian Institute of Criminology, July 2011) 2–3; Mark Brennan and Roslin Brennan, Cleartalk: Police Responding to Intellectual Disability (1994); Kristine I Ericson and Nitza B Perlman, ‘Knowledge of Legal Terminology and Court Proceedings in Adults with Developmental Disabilities’ (2001) 25 Law and Human Behavior 529; Sarah MacDonald, ‘“Which Way is Justice?” A Practice Manual for Supporting People with an Intellectual Disability in the Criminal Justice System’ (Practice Manual, Community Living Association Inc, 2008); but see B L Spivak and S D M Thomas, ‘Police Contact with People with an Intellectual Disability: The Independent Third Person Perspective’ (2013) 57 Journal of Intellectual Disability Research 635, 635 for the view that police are ‘seen as competent at identifying those with cognitive deficits and seeking appropriate supports for the

person ... in the interview context’.

^[197] Trezise [2009] VSC 520 (31 August 2009) [23].

^[198] Ibid [25]–[27] (Coghlan J).

^[199] Ibid [31].

^[200] Ibid [39].

^[201] Ibid [25]–[26].

^[202] Ibid [42].

^[203] Ibid [43].

^[204] Ibid [49].

^[205] Martin [2011] VSC 217 (20 May 2011) [3] (Curtain J).

^[206] Ibid [4].

^[207] Ibid [5].

^[208] Ibid [6].

^[209] Ibid [5]–[6].

^[210] Ibid [9].

^[211] See ibid [21], [24], [27]–[28].

^[212] Ibid [7].

^[213] Ibid.

^[214] Ibid.

^[215] Toole, ‘Defensive Homicide on Trial in Victoria’, above n 19, 498.

^[216] Martin [2011] VSC 217 (20 May 2011) [28] (Curtain J).

^[217] Ibid [12].

^[218] American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders: DSM-5 (American Psychiatric Publishing, 2013) 33–8. The severity of the deficits will depend on the person’s adaptive functioning as assessed by a clinician. However, Martin’s IQ, well below the intellectual disability range mean of 70 (plus five points as a margin for

measurement error) is an indicator that these deficits exist: at 37.

^[219] Martin [2011] VSC 217 (20 May 2011) [12], [14] (Curtain J). Martin also presented with ‘longstanding evidence of alcohol abuse and likely dependence’: at [14].

^[220] Ibid [28].

^[221] Ibid [34]–[35].

^[222] Chen [2013] VSC 296 (11 June 2013) [16] (Bell J).

^[223] Ibid.

^[224] Ibid [17]–[20].

^[225] Ibid [1].

^[226] Ibid [16].

^[227] Ibid.

^[228] Ibid [28].

^[229] Ibid [29].

^[230] Ibid [36].

^[231] Ibid.

^[232] Ibid [37].

^[233] See, eg, Milman, above n 18, citing Robert Clark; Fitz-Gibbon, ‘Abolishing Defensive Homicide Offence Essential’, above n 18; Fitz-Gibbon, ‘The Offence of Defensive Homicide’, above n 18.

^[234] See Victoria, Parliamentary Debates, Legislative Assembly, 6 October 2005, 1351 (Rob Hulls).

^[235] Trezise [2009] VSC 520 (31 August 2009); Martin [2011] VSC 217 (20 May 2011).

^[236] Chen [2013] VSC 296 (11 June 2013).

^[237] [2009] VSC 376 (3 September 2009) (‘Romero’). In this case, the offender attended a gathering at his cousin’s residence. The offender and the deceased argued over a chair. The deceased sarcastically said, ‘I’ll fight you for it’. The offender said, ‘see that knife there, I’ll put it straight through ya’. The deceased walked towards the offender to shake hands and the offender misinterpreted the gesture and stabbed the deceased three times:

at [16], [25]–[27] (Whelan J). The offender had ‘a full-scale IQ of 72’, placing him squarely within the ‘borderline’ intellectually disabled range: at [37].

^[238] Romero [2009] VSC 376 (3 September 2009) [37], [46] (Whelan J). However, in his appeal, Romero sought revivification of the Verdins argument eschewed on plea, in light of his intellectual disability, which was considered to be at the more profound end of the borderline range. The Court of Appeal stated, ‘[u]pon inquiry from the presiding judge, counsel for the applicant was unable to point to any material before his Honour that established a sufficient causal connection between his intellectual limitations and the offending conduct. The psychologist’s report tendered on the plea simply did not address that issue. The psychologist did not venture any opinion as to whether the applicant’s “history”, which the psychologist opined “indicated a lack of sound judgment, consequential thinking and insight”, was due to his borderline intellectual disability or, if so, to what extent. He did not venture any opinion as to whether the applicant’s intellectual limitations may have contributed to the commission of the offence’: Romero v The Queen [2011] VSCA 45; (2011) 32 VR 486, 490 [14] (Redlich JA).

^[239] Romero [2009] VSC 376 (3 September 2009) [46] (Whelan J).

^[240] Ibid [52].

^[241] ‘Defensive Homicide Law to Be Dumped’, above n 17, quoting Robert Clark. See also

Fitz-Gibbon, ‘Getting Away with Murder’, above n 17.

^[242] R v Smith [2008] VSC 617 (15 October 2008).

^[243] R v Ghazlan [2011] VSC 178 (3 May 2011). The offender had also been ‘variously diagnosed as suffering from endogenous depression’: at [4] (Forrest J).

^[244] R v Svetina [2011] VSC 392 (22 August 2011).

^[245] See, eg, R v Koltuniewicz [2013] VSC 650 (19 December 2013). The offender had also been diagnosed as suffering from depression: at [50] (Hollingworth J).

^[246] R v Monks [2011] VSC 626 (2 December 2011); R v Vazquez [2012] VSC 593

(14 August 2012); DPP (Vic) v McEwan [2012] VSC 417 (13 September 2012).

^[247] [2011] VSC 392 (22 August 2011).

^[248] [2011] VSC 626 (2 December 2011).

^[249] [2012] VSC 593 (14 August 2012).

^[250] [2012] VSC 417 (13 September 2012).

^[251] [2013] VSC 650 (19 December 2013).

^[252] R v Svetina [2011] VSC 392 (22 August 2011); R v Monks [2011] VSC 626 (2 December 2011); R v Koltuniewicz [2013] VSC 650 (19 December 2013); DPP (Vic) v McEwan [2012] VSC 417 (13 September 2012).

^[253] R v Vazquez [2012] VSC 593 (14 August 2012). Forrest J noted that the offender came from ‘an apparently sound family background’: at [14]. The offender’s post-traumatic stress disorder was said to arise largely from the fact that he was ‘kidnapped and tortured by a group of young men’ in 2009: at [19].

^[254] [2012] VSC 417 (13 September 2012) [81]–[83] (Kaye J).

^[255] [2011] VSC 626 (2 December 2011).

^[256] Ibid [22] (Curtain J).

^[257] Ibid [40]–[41].

^[258] [2012] VSC 417 (13 September 2012) [104].

^[259] Ibid [99].

^[260] [2014] VSC 669 (23 April 2014) [64] (Curtain J).

^[261] Ibid [66], [77].

^[262] Ibid [69].

^[263] Ibid [1], [43], [89]. The offender also pleaded guilty to one charge of recklessly causing serious injury.

^[264] Ibid [125]–[126].

^[265] Ombudsman (Vic), Investigation into Deaths and Harm in Custody, above n 23, 6.

^[266] Ibid 121 [556].

^[267] See, eg, Kirkwood et al, above n 170.

^[268] Bennett et al, ‘A Study of Psychotic Disorders’, above n 24, 231, 236–7.

^[269] ADHA s 4, inserting Crimes Act 1958 (Vic) s 322K.

^[270] ADHA s 11, amending Jury Directions Act 2013 (Vic) ss 29–32.

^[271] ADHA s 9, inserting Evidence Act 2008 (Vic) s 135(d).

^[272] Evidence Act 2008 (Vic) s 135(d).

^[273] Kirkwood et al, above n 170, 5.

^[274] Ibid. The submission was endorsed by the following organisations: Domestic Violence Resource Centre Victoria; Domestic Violence Victoria; Victorian Women’s Trust; Human Rights Law Centre; Victorian Women Lawyers; Women’s Domestic Violence Crisis Service; Koori Women Mean Business; inTouch Multicultural Centre Against Family Violence;

Federation of Community Legal Centres; No to Violence; Women’s Health Victoria; Women’s Legal Service Victoria; Victorian Aboriginal Legal Service; Women with Disabilities Victoria; and Peninsula Community Legal Centre.

^[275] Fitz-Gibbon, ‘Getting Away with Murder’, above n 17. See also ‘Defensive Homicide Law to Be Dumped’, above n 17, quoting Robert Clark.

^[276] Tyson et al, above n 177.

^[277] See, eg, Fitz-Gibbon, ‘Abolishing Defensive Homicide Will Benefit Female Victims and Offenders’, above n 171.

^[278] See, eg, Hunt, above n 20.

^[279] See, eg, Thomas Crofts and Danielle Tyson, ‘Homicide Law Reform in Australia: Improving Access to Defences for Women Who Kill Their Abusers’ [2013] MonashULawRw 28; (2013) 39 Monash University Law Review 864, 887–8; Kirkwood et al, above n 170; Tyson et al, above n 177; Nicola Wake,

‘“His Home Is His Castle. And Mine Is a Cage”: A New Partial Defence for Primary Victims Who Kill’ (2015) 66 Northern Ireland Legal Quarterly 151.

^[280] See VLRC, Review of the CMIA Final Report, above n 64, 361–9 (recommendation 84),

80–2 (recommendation 16), 113–15 (recommendation 24).

^[281] Tyson et al, above n 177, 76.