Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-14T07:46:50.845Z Has data issue: false hasContentIssue false
  • English
  • Français

Women and wrongful convictions: concepts and challenges

Published online by Cambridge University Press:  06 August 2015

Debra Parkes and
Emma Cunliffe
Debra Parkes*
Affiliation:
Associate Dean, Research & Graduate Studies, Faculty of Law, University of Manitoba. Email: debra.parkes@umanitoba.ca.
Emma Cunliffe
Affiliation:
Associate Professor, Faculty of Law, University of British Columbia. Email: cunliffe@law.ubc.ca.
*
debra.parkes@umanitoba.ca
Rights & Permissions [Opens in a new window]

Abstract

This paper draws from the wrongful convictions of women to interrogate the limits of dominant conceptions of wrongful conviction. Most North American innocence projects turn on a conception of demonstrable factual innocence. The paper argues that this focus is problematic as a matter of criminal law principle and presents particular difficulties for women. The paper identifies that family violence forms the primary context for both the conviction of women for violent crimes, and for women's wrongful convictions. Taking two key examples of family violence – child homicide and intimate partner violence – we illustrate that the prevailing focus on demonstrable factual innocence fits awkwardly with identified wrongful convictions in these areas, and argue that this focus may deflect attention from unidentified miscarriages of justice. We suggest that focusing on factual innocence undermines the criminal justice system's proper focus on state responsibilities, including the responsibility to protect women and children from harm, and the asymmetric burden of proof that applies in criminal cases.

Information

Type
Articles
Information
International Journal of Law in Context , Volume 11 , Issue 3 , September 2015, pp. 219 - 244
Copyright
Copyright © Cambridge University Press 2015 

I. Introduction

This paper is an exploratory inquiry into what, normatively and in practice, is included within the concept of ‘wrongful convictions’ and, correspondingly, what kinds of cases do not qualify for that label. We are moved to investigate this question because, as feminists who study the roles and experiences of women in criminal law, we found ourselves asking: Where are the women in the stories and reports of wrongful convictions? In one sense, women are pervasive – they are overwhelmingly likely to be the victims of crimes that are identified as wrongful conviction;Footnote 1 they are jurors, witnesses and official actors; they represent bereaved families and the families of those who are wrongly imprisoned (Milgaard and Edwards, Reference Milgaard and Edwards1999). Women haunt popular narratives of errors made within the criminal justice system as the ghostly victim seeking earthly justice,Footnote 2 or the family member who stands by a wrongly convicted son or brother in the face of condemnation and prejudice (Goldwyn, Reference Goldwyn2010).Footnote 3 Occasionally, the media reports that female family members of victims have supported those who have been wrongly convicted and called for police to focus on catching the real criminal (Hall, Reference Hall2012; Duncan, Reference Duncan2013).

Even when one looks to the wrongly convicted, women are not entirely absent. In Australia, the UK and Canada, a few cases of women wrongly convicted of killing children loom large. The paradigmatic example (in multiple ways) is Lindy Chamberlain, who was convicted of killing her newborn daughter Azaria after she testified that Azaria was taken from a family campsite by a dingo.Footnote 4 Chamberlain was exonerated in 1988,Footnote 5 and a third inquest into Azaria's death was completed in 2012. In England, Sally Clark, Angela Cannings and Donna Anthony were wrongly convicted of killing multiple infants.Footnote 6 Their convictions were based in part on the evidence of paediatrician Roy Meadow, who testified that multiple naturally occurring infant deaths in a single family were unlikely, if not impossible. In Canada, several women were wrongfully convicted of killing children during the 1990s,Footnote 7 partly on the basis of the flawed pathology evidence offered by Dr Charles Smith (Goudge, Reference Goudge2008). Four women have been exonerated in the US in cases involving shaking allegations (Gross and Shaffer, Reference Gross and Shaffer2012, p. 28). In fact, at least nineteen women have been wrongly convicted in these jurisdictions of child homicide since 1980. However, when one reads the websites, books and other mainstream literature on wrongful convictions, the relative absence of stories about wrongly convicted women is striking – even accounting for the fact that women are charged and convicted of crime at far lower rates than men.Footnote 8 This absence from the dominant discourse is particularly stark when one considers the possibility that, taking the low representation of women in the criminal justice system into account (Boyce and Cotter, Reference Boyce and Cotter2013), it may be that women are proportionately overrepresented among the ranks of the identified wrongly convicted.Footnote 9

We should say at the outset that we are not arguing that there is a single (or simple) ‘woman question’ to be asked of these cases, or that we rely on some essential experience of women criminal defendants as a homogenous group. Volumes of feminist scholarship, particularly critical race feminism and feminist post-structural theory, have problematised such notions in important ways. Aboriginal and Black women are grossly overrepresented within the Canadian prison population, and prison populations in the US, UK and Australia show similar patterns of overincarceration among some groups of racialised women (Balfour and Comack, Reference Balfour and Comack2006, pp. 70–72). Within racialised populations, women's paths to criminalisation show distinct patterns that are linked to poverty, discrimination and victimisation (Pollack, Reference Pollack2000, chapter 4). Clearly, there is no one ‘women's story’ of wrongful conviction. Nonetheless, we think it is important – even imperative – to consider the role of gendered, raced and classed stereotypes in widely shared conceptions of criminal guilt and beliefs about wrongful convictions. Foregrounding some of the research and lived experiences of women who are convicted of crimes, including homicide and other crimes of violence, raises questions about how the concept of wrongful convictions has been framed and circumscribed.

We have come to our shared interest in women and wrongful convictions by somewhat different paths. One of us, Debra Parkes, researches issues of gender inequality in the criminal justice system and has volunteered with the Canadian Association of Elizabeth Fry Societies (CAEFS), a national organisation that advocates for criminalised women and girls.Footnote 10 In doing this work, she has become aware of women whom she thinks have valid claims to ‘wrongful conviction’ status. However, these women tend not to fit into the dominant model of having claimed factual innocence from the beginning. Also, very few seek review of their criminal convictions.Footnote 11 For example, Canada's federal Criminal Conviction Review Group received nine applications from women for post-conviction review between 2002 and 2013 (out of a total of 225 applications received during that period) (Scullion, 2007, 2013).Footnote 12 In the cases we have in mind, women often have taken some form of responsibility for what has happened.Footnote 13 They may have pleaded guilty,Footnote 14 may have refused to testify, or tried to avoid the prospect of their children having to testify or otherwise endure a trial. For many of these women, the ‘wrongfulness’ of their conviction may not stem from the reasons generally cited for the notorious wrongful convictions of men (police tunnel vision, prison informants, faulty eyewitness testimony, and the like), although women are of course also victims of these injustices. Instead, these errors arise from more pervasive social failures, inequities and discrimination – such as the state's failure to provide real and lasting protection and safety for women leaving abusive relationships.

Emma Cunliffe has a long-standing interest in the interplay between gendered and other forms of stereotyping, expert testimony and legal rules in criminal trials. She has studied the transcripts of many of the wrongful child homicide convictions that are identified in the ‘Introduction’ to this paper. The wrongful convictions in the Canadian ‘Smith cases’ and English ‘Meadow cases’ are most often explained by the failures of individual medical experts, and the Lindy Chamberlain conviction is often blamed on the poor quality of forensic science adduced in that trial. Considerable government resources have consequently been directed towards improving the quality of medical investigations into the causes of infant deaths (Goudge, Reference Goudge2008; Kennedy, Reference Kennedy2004; Morling, Reference Morling1987). However, focusing solely on medical errors made by individual experts seems incomplete at best. Such a focus screens out the operation of gendered, raced and classed ideologies in fact determination processes at investigation and trial and, perhaps more insidiously, within expert medical knowledges. It obscures the possibility that internalising these ideologies may affect how women behave when they are grieving, and when they are suspected of or being tried for criminal conduct (Ewick and Silbey, Reference Ewick and Silbey1995, p. 210). By laying the ‘problem’ of wrongful convictions at the feet of individual experts, the prevailing analysis also denies the systemic vulnerabilities that allowed wrongful convictions to occur in child homicide cases in several countries at about the same time. Highlighting the operation and importance of stereotypes within these cases provides a partial corrective to the tendency to quarantine wrongful convictions as anomalous and therefore irrelevant to the main flow of criminal proceedings and penal policy.

This paper suggests that feminist and other critical interrogations of criminal law and procedure demand a broader role for the concept of wrongful convictions. The notion of criminalisation as a social and political practiceFootnote 15 points to a correspondence, rather than a divergence, of interests between ‘wrongly convicted’ and ‘justly convicted’ women. We begin with the focus of most scholarship on wrongful conviction as requiring demonstrable factual innocence, and suggest why this focus is problematic both for women and as a broader matter of criminal law principle. We also briefly note the often-cited ‘common causes’ of wrongful conviction, and prescriptions for avoiding those errors, and suggest that the dominant conceptualisation excludes factors that may be particularly important for women. Next, we comment on the prevalence of women ‘taking responsibility’ (primarily by pleading guilty) in two areas: homicide cases and domestic violence. We conclude by noting some potential implications of expanding our enquiry about wrongful convictions to take seriously the experiences of women.

II. Locating (and limiting) wrongful convictions: the role of factual innocence

The decision to classify a case as a ‘wrongful conviction’ or ‘miscarriage of justice’ is value-laden. A survey of the scholarship on wrongful convictions shows that the paradigmatic wrongly convicted defendant is demonstrably factually innocent – for example, the accused is proven not to have been present at the scene of the crime, or was clearly not involved (Bedau and Radelet, Reference Bedau and Radelet1987; Gross and Shaffer, Reference Gross and Shaffer2012; Rattner, Reference Rattner1988; Risinger, Reference Risinger2007; Caddell, Reference Caddell1999; MacFarlane, Reference Macfarlane2006). In an illustrative passage from a classic study, Hugo Bedau and Michael Radelet explain: ‘If person A is convicted of killing person B, and A did in fact kill B, then A is not a victim of a miscarriage of justice, even though it is later discovered that A was insane, acted in self-defense, or had some other legally valid excuse or justification’ (1987, p. 40). Wrongful convictions that meet a standard of factual innocence have been called ‘wrong person errors’ (Poveda, Reference Poveda2001, p. 691), and are often defined in distinction to ‘mere’ legal errors such as the denial of an available defence or an unremedied breach of constitutional rights within the investigative process (Gross and Shaffer, Reference Gross and Shaffer2012; Raymond, Reference Raymond2001; Leo, Reference Leo2005, p. 205; Medwed, Reference Medwed2008).Footnote 16 Since Bedau and Radelet defined the field, DNA exonerations have reinforced the notion of factual innocence as the gold standard of wrongful conviction. The valorisation of demonstrable factual innocence is illustrated in the discussion provided by the Honourable Antonio Lamer in his inquiry into three Newfoundland cases. One conviction, that of Gregory Parsons, was acknowledged by the prosecution to have been wrongful. A second, that of Randy Druken, was still contested by the prosecution despite the fact that the only direct evidence against Druken was testimony by a prison informant who was later prosecuted for obstructing justice in the case. Lamer characterised the two accused as follows: ‘Gregory Parsons played no part whatsoever in the murder of his mother, Catherine Carroll. He is completely innocent. His innocence was established by DNA evidence’ (Lamer, Reference Lamer2006, p. 70);Footnote 17 and ‘[m]y mandate is not to re-try Randy Druken … Nor am I authorized to determine whether he is factually innocent … Although his factual innocence has not been acknowledged, I have concluded that there was no reliable evidence on which to base his prosecution’ (Lamer, Reference Lamer2006, p. 175; see also Risinger, Reference Risinger2007; Gross and Shaffer, Reference Gross and Shaffer2012). While Parsons was thoroughly exonerated by Lamer's inquiry, the speculative possibility that Druken had some involvement in the crime of which he was convicted (which could not be investigated because police failed to collect complete crime scene evidence after their investigation began to focus on Druken), coupled with Druken's criminal record, led to an equivocal characterisation of the wrongs of this particular wrongful conviction.

Although the question of how best to ascertain factual innocence is occasionally discussed within the existing literature, the relationship between this research emphasis on factual innocence and the ostensible legal requirements of proof is rarely given sustained consideration. Keith Findley provides an exception to this trend in his 2011 paper ‘Defining Innocence’ (Findley, Reference Findley2011; see also Plaxton, Reference Plaxton2002). Findley identifies that ‘innocence’ performs a variety of functions in different legal and political settings. For example, attention is paid to evidence of factual innocence within fresh evidence applications and habeus corpus processes. Findley argues that the only definition of innocence that should be used by the innocence movementFootnote 18 is an ‘objective’ legal definition: ‘the rules established by our legal system for determining who counts as guilty and who counts as innocent’ (Reference Findley2011, p. 1184). Focusing on the concept of exoneration, which he defines as an innocent person wrongly convicted, Findley would ‘exclude individuals whose convictions were overturned and dismissed based on procedural errors unrelated to innocence … or based on evidence that the defendant played some role in the crime, albeit a lesser role than would support the original conviction’ (Reference Findley2011, p. 1185). Findley suggests that this definition is justified by the fact that so-called procedural errors, although ‘significant and legitimate’, are different types of errors, and that including them within the definition of exonerations would ‘skew the analysis of errors’ that contribute towards convicting the innocent (pp. 1185–1186). Nonetheless, he acknowledges that ‘[t]he demand for clear proof of innocence is a misfit in our system of justice’ (p. 1190) and sees the risk of creating ‘two classes of innocents’: those who can, and those who cannot, bring evidence in support of their claims to positive innocence (p. 1201).

Findley's discussion of innocence is the most thoroughly reasoned that we have encountered within the literature. His efforts to construct a principled response to prosecutors and state officials who persist in declaring their subjective belief in the guilt of exonerated defendants are hugely important within the politicised environment of contemporary criminal justice systems. Even so, like many researchers in this field, Findley seems too ready to accept a shift in the asymmetric principle of the criminal process. Accepting (even reluctantly) the importance of inquiring into whether the convicted defendant can offer positive evidence of innocence risks allowing a subset of cases to define our sphere of concern with the treatment of criminal defendants generally. The basic tenet of criminal justice in an adversarial process is that the state bears the burden of proving guilt beyond a reasonable doubt, in accordance with constitutional and ethical standards of behaviour. A reasonable doubt may remain because of factual uncertainty, inadequate evidence, the application of a legal rule (including a rule of evidence or a defence or constitutional right), the presence of a plausible alternative suspect, the absence of any crime, positive evidence of factual innocence, or some combination of these factors (Raymond, Reference Raymond2001, p. 456). Partitioning factual innocence risks undermining our commitment to this burden of proof and its consequent allocation of the consequences of uncertainty, in relation to all criminal defendants.Footnote 19 The attacks made by innocence scepticsFootnote 20 on the work done by innocence projects and their allies should not knock legal researchers from the main goals of holding the state to its burden of proof and to the highest standards of probity.Footnote 21

Reclaiming the proposition that the phrases ‘miscarriage of justice’ and ‘wrongful conviction’ can apply to those who cannot prove their factual innocence is an important pillar of the wrongful conviction research programme. It gives rise, for example, to the question of what responsibility criminal justice system participants might have to study and prevent wrongful convictions that are not susceptible to factual validation. Susan Bandes (Reference Bandes2008) has argued that part of the challenge faced by the ‘innocence movement’ is conveying the importance of procedural protections:

‘The notion of fair process, as distinguished from the notion of fair results in particular cases, is always a hard sell. The notion of process is abstract, complex, and not very media friendly. The notion that a process needed to protect the innocent will unavoidably protect the guilty on occasion is a sophisticated notion. An even harder sell is the idea that all suspects … should receive fair process, not just as a windfall but because our constitutional protections are not meant to protect only the innocent.’ (Bandes, Reference Bandes2008, p. 16)

Bandes argues that the innocence movement must complicate the simple stories offered by the DNA exoneration paradigm, and find more nuanced ways to discuss state actors' roles in and responsibility for wrongful convictions.

The prevailing focus of the published literature on wrongful convictions is to identify and correct flaws in criminal or police investigatory procedures that allow factually innocent people to be convicted. For example, a Canadian law journal paper summarises the most commonly discussed causes of wrongful convictions (MacFarlane, Reference Macfarlane2006). The former Deputy Attorney-General of Manitoba, Bruce MacFarlane, lists four ‘predisposing circumstances’Footnote 22 and eight ‘immediate causes’ of wrongful convictions, and identifies a number of recommendations culled from various reports and commissions of inquiry into wrongful convictions. The ‘predisposing circumstances’ are described as:

  • • public pressure to convict;

  • • an unpopular defendant;

  • • conversion of the adversarial process into a game; and

  • • ‘noble cause corruption’ when the police believe an accused is guilty. (MacFarlane, Reference Macfarlane2006, pp. 435–444)

The immediate causes of wrongful conviction are listed as:

  • • eyewitness misidentification;

  • • police mishandling of the investigation;

  • • inadequate disclosure by the prosecution;

  • • unreliable scientific evidence;

  • • criminals as witnesses (particularly ‘jailhouse informants’);

  • • inadequate defence work;

  • • false confessions; and

  • • misleading circumstantial evidence. (MacFarlane, Reference Macfarlane2006, pp. 444–483)

This list is similar to several published by US scholars associated with the innocence movement (e.g. Scheck, Neufeld and Dwyer, Reference Scheck, NEUFELD and DWYER2001). Findley has documented US judicial responses to similar revelations about the systemic causes of wrongful conviction, and has highlighted instances in which constitutional rights have been extended or stiffened to guard against identified causes of wrongful conviction.Footnote 23 These responses demonstrate the power that such lists have had within jurisdictions that are otherwise broadly unsympathetic to the rights of accused.

Research is beginning to shed light on some of the ways that women may experience wrongful conviction differently from men. A recent report surveying exonerations in the US in 2013 found that women exonerees were much more likely than men to have been convicted for crimes that never occurred (as opposed to the great majority of cases where a crime was committed but the wrong person was convicted).Footnote 24 One paper has compared the common factors identified in the literature as producing wrongful convictions generally with those that could be discerned from forty-two wrongful convictions of women tried in the US between 1970 and 2007.Footnote 25 Fifteen of these convictions were for murder, and fifteen were for child abuse. Eight women were wrongly convicted for drug crimes – these convictions arose from the same police investigation. Two women were convicted of rape – a proportion which is, as expected, notably different from those identified in relation to the wrongful convictions of men. Ruesink and Free found that the most common causes of women's wrongful conviction were unethical police and prosecutorial behaviour (present in 86 per cent of cases), and eyewitness error. False confessions only constituted a factor in a single case. As we explain below, this may represent an underlying systemic bias in the identification of women's wrongful convictions, rather than a true difference between the rates at which women ‘falsely’ confess.Footnote 26 Other studies have found that women are more likely than men to speak to police when suspected of criminal activity,Footnote 27 and that women are slightly more likely to plead guilty (Daly, Reference Daly1994, p. 19). Ruesink and Free (Reference Ruesink and Free2005, p. 18) also found racial differences in the wrongful convictions of women in the US: White women were more likely to be wrongly convicted of child abuse, and Black women were overrepresented among those wrongly convicted of murder and drug offences. Finally, Ruesink and Free (p. 19) suggested that a new category of cause should be added to those commonly cited: improperly suggestive police techniques when interviewing children.

Richard Leo (Reference Leo2005, p. 213) has criticised the focus on legally defined causes such as those listed above, suggesting that while they may constitute common factors in wrongful convictions, the focus on these factors too often substitutes for a theoretical and empirical exploration of psychological, sociological and institutional ‘root causes’ that dispose the criminal justice system towards making common identified errors.Footnote 28 Leo calls for the generation of more theoretically and methodologically sophisticated approaches to studying wrongful conviction. We do not dispute the importance of attending to predisposing factors and immediate errors, and should not be understood as intending to minimise the injustice caused to the victims of such errors by flawed police and justice system practices. Much more needs to be done to expose and prevent these cases, and resistance to implementing police and justice system reforms is troubling.Footnote 29 However, we suggest that understanding these causes and circumstances cannot be achieved by focusing on demonstrable factual innocence, and we agree with Leo that it is crucial to go beyond common factors to a consideration of root causes. Recall, for example, Ruesink and Free's observation that women are disproportionately likely to be convicted of child abuse and their identification of a ‘new factor’ of improper interview techniques of children. These observations resonate with Gross and Shaffer's (Reference Gross and Shaffer2012) claim that women are disproportionately likely to be convicted of crimes against children ‘that never occurred’.Footnote 30 We will return to Gross and Shaffer's claim below.

Leo's call to focus on root causes also resonates with the realisation, set out in our ‘Introduction’, that at least nineteen women have identifiably been wrongly convicted of child homicide in the UK, Canada, the US and Australia since 1980. In absolute numbers, nineteen does not sound like very many. As a proportion of all cases, the numbers may well be more significant. For example, in Canada the average annual number of child homicides (children less than 12 years old) reported between 1994 and 2004 (the period in which the relevant wrongful convictions arose) was forty-six (Statistics Canada, 1995–2005). Not all of these homicides would result in charges being laid or convictions obtained – for instance, reported homicides include murder-suicides, which is a common pattern for child homicides committed by a parent. Unfortunately, however, Statistics Canada does not separately report the number of charges laid or convictions obtained in child homicide cases. It is now officially recorded that fourteen convictions obtained in this period were wrongful, and charges were laid against twenty-three accused who have since been exonerated in the deaths of a total of twenty children (see also Cunliffe, Reference Cunliffe and Kilty2014, p. 223).

Based purely on the identified wrongful accusations, police and prosecutors erroneously laid charges in 4.3 per cent of initially reported child homicides. Wrongful convictions were obtained in 3 per cent of all child deaths that were classified by police as homicide in this period – including deaths that later proved natural, murder-suicides in which the perpetrator killed both the child and him- or herself, and cases in which charges were never laid. Child homicide is a field in which there is significant attrition between police-reported crime and conviction rates. A fuller discussion of the factors that produce considerable attrition between police-reported child homicides and child homicide convictions can be found in Cunliffe (Reference Cunliffe2011, pp. 39–43).

The wrongful accusations and convictions that arose in Canada between 1994 and 2004 disproportionately affected women, who form two-thirds of those who were wrongly charged (Cunliffe, Reference Cunliffe and Kilty2014). Given the proportion of deaths that are screened out of the criminal justice process, and the uncertainties inherent within ascribing cause of death in infancy, an identified rate of 3 per cent of wrongful convictions within the universe of child deaths originally attributed by investigators to homicide is cause for very significant concern. Other studies that have attempted to estimate the rate of wrongful convictions suggest that the rate is 2.3 per cent of those convicted of capital crimes (Gross and O'Brien, Reference Gross and O'brien2008), or 3.3–5 per cent of capital rape-murders convictions obtained in the 1980s (Risinger, Reference Risinger2007). These rates from American innocence scholars compare identified wrongful convictions against all convictions (and not against the larger universe of police-reported crime). It is not possible to discern from reported Canadian figures what percentage of child homicide convictions were wrongful convictions between 1994 and 2004, but the number would be considerably higher than 3 per cent.

However, identified wrongful convictions also carry an importance that goes well beyond their numbers. Others have written about the challenges of extrapolating from identified wrongful convictions to the total set of criminal convictions, particularly given the structure of innocence projects (which focus only on the most serious cases and only on claimants who assert a complete lack of involvement with the crime for which they were convicted, and are frequently also limited to those with many years left to serve on a sentence of incarceration) and the moral luck associated with the presence or absence of evidence that may help to exonerate a given claimant (Gould and Leo, Reference Gould and Leo2010, pp. 832–836; Roach, Reference Roach2012, 1470–1476). The bottom line, as stated by Dan Simon (Reference Simon2006), is that ‘the overall rate of error in the criminal justice system is unknown, and unknowable’. As we explain, the difficulties associated with estimating the rate of wrongful conviction apply even more acutely to women's cases. It seems necessary to move beyond observations of patterns within the wrongful convictions of women and consider why women are so vulnerable to certain types of suspicion.

To answer the why question, it is necessary to first revisit some of the assumptions that underpin the focus on factual innocence, and to consider how well these assumptions fit with the lives of women. We believe that an appeal to rationality and self-control lies at the heart of the innocence movement's focus on factual innocence and the associated search for common causes of error. The implicit liberal message of most literature on wrongful convictions is that our criminal justice system is (or is capable, with some tinkering, of being) one that effectively punishes those who have freely chosen, as autonomous individuals, to commit crime while offering confidence to those who choose not to commit crime that they will not be wrongly convicted. For example, Michael Plaxton (Reference Plaxton2002) argues for a rigorous ‘proof beyond a reasonable doubt’ standard and procedural safeguards to prevent wrongful convictions on the basis of the ‘moral independence’ of citizens:

‘the moral independence thesis suggests that the legitimacy of the criminal law flows directly from its ability to predictably foreclose miscarriages of justice … [T]o the extent that rules of criminal procedure … raise the spectre of wrongful conviction, encourage or tolerate arbitrary arrest or detention … [they] destroy individuals' ability to reliably execute meaningful plans … If the public and the State expects only guilty people to suffer criminal sanctions, individuals will be equipped to plan their lives and pursue conceptions of the good that are not harmful but nonetheless unpopular.’ (Plaxton, Reference Plaxton2002, p. 426)

Such characterisations of the ‘wrong’ in wrongful convictions are predicated on assumptions that the logic and fact-determination processes that structure the criminal justice system are inherently fair and capable of error-free implementation, and on beliefs that criminal offending is primarily a matter of choice.

Margaret Davies (Reference Davies2002, p. 14) described the paradigmatic liberal actor as one who ‘sees himself as a free agent relating to a world of discernable analytical structures, and thinks that this is the condition of everyone’. This actor is both constructed and protected by an account of criminal law as a tool that fosters personal autonomy by enforcing responsibility on a case-by-case basis for presumptively choice-predicated actions taken in violation of criminal rules that are known to the actor.Footnote 31 Wendy Brown (Reference Brown1995, p. 157) has suggested that the autonomous liberal actor requires the subjection of women and is ‘quite at odds historically with what men have wanted women to be, with what women were allowed to be, with what families have required of women, and with what women have been socially constructed to be’. The observation that liberal conceptions of autonomy actively exclude women has particular relevance for our concern about the criminal justice system's capacity to judge the actions of women as mothers, and especially the actions of women who have experienced domestic violence (a topic to which we return below). As will become apparent, we believe that exploring the troubled relationship between liberal discourses of autonomy and criminalised women is crucial to understanding the wrongful convictions of women.

Scholars have demonstrated that the liberal reliance on formal equality and individualised discourses of choice and responsibility to justify criminal punishment operates particularly punitively against Aboriginal and other racialised women (Roberts, Reference Roberts1993; Lawrence and Williams, Reference Lawrence and Williams2006; Sugar and Fox, Reference Sugar and Fox1989). A detailed example of the punitive operation of ostensibly neutral rules is supplied by Elizabeth Sheehy (Reference Sheehy2014) in her recent study of ninety-one Canadian women who were tried for killing intimate partners in circumstances where the female accused had been subjected to violence by the man she allegedly killed. Having identified that at least 41 per cent of the women whose cases she studied were Aboriginal, Sheehy concludes on the basis of a close study of transcripts and other court documents that Aboriginal women who kill White male abusers, in particular, ‘face a credibility contest framed by discrimination’ (Sheehy, Reference Sheehy2014, p. 127).Footnote 32 Identifying the difficulties Aboriginal women face in naming discrimination when on trial, Patricia Monture (Reference Monture, Monture and McGuire2009, p. 418) has observed that ‘a criminal court is not interested in hearing about this long trail of individualized and systemic colonialism which leads to conflict with the law’. While Monture is particularly concerned with sentencing and correctional practices, we believe that further study will show that the judicial lack of interest in discrimination and colonisation to which Monture refers also exacerbates Aboriginal women's systemic vulnerability to wrongful conviction in the broader sense in which we use the term. Sheehy also documents this possibility, identifying that the Aboriginal women in her study may have been especially vulnerable to being denied a valid claim to self-defence (Sheehy, Reference Sheehy2014, p. 198).

For other (non-Aboriginal) racialised women, and women in other jurisdictions, it remains important to attend carefully to the potential operation of racism, poverty and discrimination. Elizabeth Comack (Reference Comack and Comack2014, p. 50) provides a helpful reminder of the inappropriateness of speaking generally about the operation of racism and discrimination, emphasising ‘the need for historical and theoretical specificity in our efforts to situate racialized groups in our theorizing’. When considering the wrongful convictions of women, it is therefore important to attend closely to the operation of stereotypes about race and racialised women as well as to the particular context and experience of racialised women who are charged with criminal offences. However, numerous scholars have documented the challenges that arise when one seeks to find traces of race and racism in official records (e.g. Backhouse, Reference Backhouse1999, pp. 12–14). These difficulties are produced in part by the official position that race is irrelevant to the adjudication of guilt – a position which may well operate to deny the social relevance of race in a given case. This effacement of race offers yet another example of the ways in which liberal ideals (recast as empirical truths) operate unevenly in relation to differently situated accused.Footnote 33

So far, our analysis of the concept of wrongful conviction – like most analyses of wrongful convictions – has focused on the paradigmatic serious contested criminal trial. However, we believe that a deeper engagement with wrongful convictions demands attention to the statistical reality that such trials are anomalous in contemporary criminal justice systems. It seems likely that the vast majority of wrongful convictions are produced as part of the routine processing of minor offences, most of which are disposed of by way of guilty pleas (Gross and Shaffer, Reference Gross and Shaffer2012). William J. Stuntz argued that the practices of charge stacking and plea-bargaining – together with the prevailing trial focus on procedural rights rather than substantive adjudication – dilute prosecutors' incentives to avoid wrongful convictions.Footnote 34

Dianne Martin (co-founder of the Osgoode Hall Law School Innocence Project and a key figure in the activist and legal work done to expose wrongful convictions in Canada) has said:

‘Indeed, it is important to be aware that wrongful convictions are not just a product of the high-profile, pressure-to-convict paradigm, so familiar from the notorious cases of wrongful conviction. All miscarriages of justice have their roots in the more common problem of institutional cynicism and neglect – the cases of the same old suspects “obviously guilty” of the same old offences investigated primarily by means of interrogation and informer manipulation and resolved through negotiated guilty pleas. Many of the known cases turn out to be neither aberrations nor difficult to prevent.’ (Martin, Reference Martin2001a, p. 520)

Martin's focus is on factual innocence, noting that identity is only an issue in a minority of cases. Nonetheless, Martin's work is a good place to begin such an inquiry as she highlights systemic failures of policing and enforcement; for example, patrols and police training geared towards identifying ‘who or what does not fit, whether it is a black youth in a white neighbourhood or some other shorthand picture of “what should not be there”’ (Martin, Reference Martin, Westervelt and Humphrey2001b, p. 89). She also insists on exposing some of the cherished myths of the criminal justice system:

‘The ideological functions of criminal prosecutions are enhanced by preserving the myth that guilt or innocence is resolved through fair and public trials where the facts are contested and the law is challenged. That myth in turn rests on others: that the presumption of innocence actually guides criminal justice determinations; that the accused “criminal” has more rights than do crime victims; and that the criminal justice system is soft on crime … The reality that lays bare the myth about the protections of due process safeguarding convictions is something darker and more troubling. Most cases in the criminal justice system involve targeted, racialized, marginalized people who are known to each other and known to the system, and most of these cases are resolved through plea negotiations.’ (Martin, Reference Martin2001a, p. 527)

Martin's work is significant as it extends concerns about wrongful convictions into the mundane world of the thousands of guilty pleas made every day. Her call to unpack the assumption that the rights notionally allotted to criminal suspects are in fact accorded as a matter of course informs our consideration of the wrongful convictions of women.

III. Taking responsibility: women and wrongful convictions

It is now known that, for a variety of reasons, both women and men may plead guilty when they are, in fact, innocent.Footnote 35 The most recent study of known exonerations in the US reports that 17 per cent of exonerees were convicted of offences for which they pleaded guilty (National Registry of Exonerations, 2014, p. 3). The idea of innocent people pleading guilty sits uneasily with the presumption of rationality, which may help to explain the obstacles placed before a convicted defendant who wishes to retract a guilty plea and challenge a conviction. In the Canadian child homicide cases reviewed by Justice Goudge, seven of the accused (out of twenty-four) pleaded guilty to child homicide offences in circumstances where the children probably died of natural causes. Four of these individuals were women (Cunliffe, Reference Cunliffe and Kilty2014). There may be different pressures on some women to plead guilty due to their primary responsibility for the care of children, or other social and economic realities.Footnote 36 Similarly, the experiences that some poor and racialised women and men have had with the police, other state officials and lawyers may create particular pressures to plead guilty.Footnote 37

In her review of Canadian cases involving women charged with killing their male partners where self-defence was at least arguable, Sheehy found that 54 per cent of the accused women entered guilty pleas to manslaughter (Sheehy, Reference Sheehy2014, pp. 9–10). Over half of women who pleaded guilty to manslaughter were Aboriginal (p. 189). Sheehy highlights the case of Jamie Tanis Gladue, an Aboriginal woman in respect of whom the Supreme Court of Canada issued a landmark decisionFootnote 38 on the need to consider the particular circumstances of Aboriginal people in sentencing, and to consider alternatives to incarceration. Gladue pleaded guilty to manslaughter in the death of her abusive common-law partner, but Sheehy's review of the preliminary inquiry transcript and other information available about the case reveals that she likely had a strong self-defence claim which would have entitled her to an acquittal (pp. 190–192).

There is relatively little qualitative or quantitative research on gender or race differentials in the prevalence of guilty pleas. However, a study by Kathleen Daly of felony court proceedings in Connecticut in the 1980s found that women pleaded guilty at higher rates than men (Daly, Reference Daly1994). In Canada, the Aboriginal Justice Inquiry of Manitoba analysed Provincial Court data in that province in the 1980s and found that Aboriginal people pleaded guilty in 60 per cent of the cases, compared to 50 per cent for non-Aboriginal persons (Hamilton and Sinclair, Reference Hamilton and Sinclair1991, chapter 13). In addition, Aboriginal and other racialised accused persons are denied bail at a higher than average rate (Hamilton and Sinclair, Reference Hamilton and Sinclair1991, chapter 4; Tam, Cole, Gittens, Williams, Ratushny, and Sri-Skanda-Rajah, Reference Tam, COLE, GITTENS, WILLIAMS, RATUSHNY and SRI-SKANDA-RAJAH1995, p. 119), and other research has shown that denial of bail is a powerful factor in decisions to plead guilty (Cheng, Reference Cheng2013, p. 270). Research also suggests that women, children and Aboriginal persons are more likely than other suspects to waive their right to silence during police investigations (New South Wales Law Reform Commission, 2000, paragraph 2.118).

Of course, not all people who plead guilty under pressure are ‘innocent’, however that term may be defined. What is troubling is the relative lack of discussion of the relevance of differential pressures on variously situated suspects to identifying, correcting and preventing wrongful convictions. This lack is particularly evident in the relatively limited notion of ‘false confessions’ generally employed within this literature. False confessions are most often understood to mean confessions that are coerced by illegal police interrogation tactics and may produce a wrong-person error. This definition excludes confessions that may be produced by social or ideological pressures to take responsibility for harms for which one may not be legally responsible, but for which one may feel morally responsible. Richard Ofshe and Richard Leo's work suggests the possibility that a sense of moral responsibility may be leveraged in the service of obtaining false confessions (Ofshe and Leo, Reference Ofshe and Leo1997). These researchers found that manipulative tactics were ubiquitous within a large study of false confessions. They have traced the manner in which police structure the options available to a suspect to lead that suspect to perceive that confession is the optimal choice whether or not she committed the relevant offence.Footnote 39

3.1 Women and homicide

Most studies of women who commit homicide identify that women are far less likely to commit homicide than men, but that when women kill, they are overwhelmingly likely to kill family members – either intimate partners or children.Footnote 40 The identification of a large number of miscarriages of justice in child homicide cases prosecuted in Canada and the UK demonstrates that women may well be vulnerable to conviction when they are factually innocent. However, it is possible that an even larger number of women are denied access to legally relevant defences, including self-defence, or to lesser-included offences, such as infanticide, that recognise the diminished responsibility of some women. We would include these women among the wrongly convicted, on the basis that the legal system has failed to ensure that they have access to a defence to which they are legally entitled. By the eligibility criteria used by many Innocence Projects and the discussion offered in much of the associated innocence literature, these women would be excluded from the ranks of the wrongfully convicted. We explain the basis for our belief that women may systematically be denied access to relevant defences in this subsection.

A notable discussion of the pressure imposed on women to plead guilty is supplied within Judge Lynn Ratushny's Self-Defence Review: Final Report (SDR) (Ratushny, Reference Ratushny1997). The SDR was an en bloc judicial examination of cases of women serving life sentences for murder when they may have killed in self-defence. The SDR reviewed ninety-eight applications and made recommendations in relation to the conviction or sentence of seven of those women, although no woman was ultimately released from prison.Footnote 41 However, of note for our inquiry into women and guilty pleas, Judge Ratushny drew a straight line between the ‘irresistible forces’ on all accused to plead guilty in homicide cases and the mandatory minimum sentences for murder, while also noting some particular pressures on women. She said in her Report:

‘there may be additional factors that exert even more pressure on a woman to plead guilty, including the fact that she may have a young family to care for; she may have been a victim of abuse and is reluctant to testify publicly about that abuse; she may be genuinely remorseful and even though she feels she had to act to defend herself she has difficulty justifying taking another person's life even to herself. For a woman in this situation, the forces impelling her to plead guilty are considerable. This situation causes me serious concern. It means that these guilty pleas are influenced in whole or in part by forces extraneous to the merits of the cases. It also means that women (and men) may be pleading guilty to manslaughter when they are legally innocent because they acted in self-defence.' (p. 10; see also pp. 83–84; emphasis added)

Ratushny suggested in the body of the Report that defence counsel would not advise a woman whom the counsel believes innocent to plead guilty, but that they are duty bound to explain to women the risks and benefits associated with choosing to plead guilty versus proceeding to trial. She concluded that there is a real risk that women plead guilty because of systemic ‘forces extraneous to the merits of their cases’ (Ratushny, Reference Ratushny1997, p. 84). It is worth noting that the SDR was one of the few times that the legal system has opened itself up to a relatively broad-based inquiry into whether a whole group of people were ‘legally innocent’ of serious crimes.

When viewed in that context, it is perhaps unsurprising that the SDR's recommendations were modest, most were not implemented, and no substantial reforms have emerged from it. At virtually all stages of the process, Judge Ratushny encountered substantial opposition to the SDR, both from provincial prosecutors who were opposed to this ‘second guessing’ of ‘their’ convictions, as well as from those who rejected the very idea of a review of women's cases only, rather than a review of all homicide convictions (Ratushny, Reference Ratushny1997, p. 12). This opposition reveals the strong pull of gender neutrality in law (Conaghan, Reference Conaghan2013, chapter 1; Daly and Chesney-Lind, Reference Daly and Chesney-Lind1988, pp. 524–526; Boyle, Reference Boyle1985, pp. 101–106), as well as the strength of resistance to a potentially broader notion of innocence that was implicit in the SDR, since factual innocence was not claimed by the women seeking review of their convictions.Footnote 42

Sheehy has described one post-Lavallee case of a woman, Kim Kondejewski, in Brandon, Manitoba who was ultimately acquitted of murder on the basis of self-defence in the light of substantial evidence of years of horrific abuse at the hands of the deceased. Sheehy notes how this woman very nearly pleaded guilty to manslaughter and accepted a twelve-year sentence (Sheehy, Reference Sheehy2001, p. 548; see further Sheehy, Reference Sheehy2014, pp. 88–125). In a cross-jurisdictional study, Sheehy, Stubbs and Tolmie (Reference Sheehy, Stubbs and Tolmie2012) reported that of thirty-six cases identified between 2000 and 2010 in which battered women were prosecuted for homicide in Canada, twenty resulted in guilty pleas. Nineteen of these women pleaded guilty to manslaughter, and one pleaded guilty to murder. Twelve women were acquitted, or their cases did not proceed to trial (p. 486). In Australia, relatively fewer women were acquitted and a greater proportion of women were convicted of manslaughter – in many cases, on acceptance of a guilty plea (Sheehy et al., Reference Sheehy, Stubbs and Tolmie2012). The authors conclude that ‘the question of plea bargaining … requires greater scrutiny. There are substantial pressures on battered defendants who are facing murder charges to plead guilty to manslaughter rather than proceed to trial in order to run self-defence’ (p. 489). Sheehy et al. also identify a trend towards charging women who kill abusive partners with murder despite the fact that murder convictions are infrequently obtained (p. 490).

Judge Ratushny's suggestion that defence counsel may find it difficult to advise women who may have sound claim to a defence against homicide charges is supported by an interview-based study of Aboriginal women who had served time in federal prisons. Jennifer Dyck interviewed ten Aboriginal women, most of whom had served time in federal Canadian prisons. Two of these women had killed men who were sexually and physically violent towards them and, in one case, to the woman's young daughter. One pleaded guilty to manslaughter despite the fact that there was photographic evidence of injuries caused by the assault committed against her; another pleaded guilty to a murder that was committed while she was so drunk that she was initially arrested by police for public intoxication (Dyck, Reference Dyck2013).

Turning to women who are accused of killing children, similar dynamics emerge. Emma Cunliffe found in 2009 that Canadian women who killed children in circumstances that seemingly fit the statutory definition of infanticide were most likely to be charged with murder in prosecutions conducted since 2000.Footnote 43 This represented a marked shift in charging practice compared with cases conducted between 1954 and 2000 (Cunliffe, Reference Cunliffe2009, pp. 107–108). Prior to 1990, no accused woman in these circumstances had been charged with murder since the introduction of the infanticide offence. Of the seven women in Cunliffe's study who were charged with murder between 1990 and 2008, only one was convicted of that crime. Most were convicted of a lesser-included offence such as manslaughter or infanticide.

Reviewing the transcripts of police interviews with women who are suspected of killing children, we have found that there are strong similarities between the interviews of women who are considered ‘rightly’ convicted, and those who are now considered to be factually innocent. In particular, we have observed a pervasive police strategy that utilises a mother's presumed sense of responsibility for the welfare of her children as a lever for inducing a suspect mother to confess to performing the actus reus, or admit to the highest form of intent. While invariably designed to elicit evidence that can later be used to demonstrate a mother's criminal guilt, this tactic works slightly differently from case to case. The differences appear to be produced by the police goal of compensating for other weaknesses in the evidence in a given case – for example, when the medical investigation into cause of death reaches an inconclusive diagnosis, the importance of obtaining a confession to having performed the actus reus is correspondingly increased. Conversely, a mother's decision to confess that she feels responsible for a child's death appears to result in a less thorough medical investigation of cause of death, and may result in alternative explanations being missed.Footnote 44 A universal feature is the interviewing officers' attempts to secure admissions that would increase the gravity of the alleged behaviour, thereby justifying a more serious charge.

Our review of more than a dozen police interviews with suspect mothers reveals that police interviews pass through several stages.Footnote 45 In the initial stage, a mother is encouraged to accept inchoate responsibility for a child's death, in the sense of admitting that she feels guilty or blames herself for the death. When such an admission is offered, the police officer will press for a more detailed confession or direct admission that the mother performed an action or omission such as shaking, or ignoring signs of distress, with the underlying presumption that this behaviour (could have) caused the child to die. If a suspect mother admits to performing such an act or omission, she is repeatedly encouraged to admit that she acted deliberately and with an intention that is sufficient to constitute murder. Throughout such interviews, a mother's sense of responsibility towards her dead child and any surviving children, and towards her partner and family, are leveraged to encourage honesty, candour, and an acceptance of criminal consequences. However, the candour sought at this stage is singularly directed towards certain topics. Information that may be relevant to a defence, or that may decrease the seriousness of the applicable charge (such as information that a suspect mother was subjected to intimate abuse or that she was depressed or suicidal) is discouraged either tacitly by not being pursued when volunteered by the suspect, or actively by being derogated by the interviewing officer as failing to relieve her of responsibility for her purported actions. On the other hand, information that tends to increase a mother's culpability or relieve the prosecution of the duty to prove cause of death receives positive reinforcement and encouragement from the interviewing officer, and is actively elicited.

Given that it is often difficult to determine with medical certainty how a very young child has died (Cunliffe, Reference Cunliffe2011, chapter 4), the police strategy of using a mother's sense of responsibility for the welfare of her children to elicit confessions to criminal acts presents a very real risk of error. When a mother admits that she treated a child harshly – for example, by placing an infant roughly into a bassinet – this information seems to substitute for a more comprehensive medical investigation of cause of death even though reputable medical research suggests a realistic possibility of natural causes of death (Cunliffe, Reference Cunliffe2011; Findley, Barnes, Moran and Squier, Reference Findley, Barnes, MORAN and SQUIRE2011). Such an admission does not constitute a false confession in the sense in which this term is usually understood – instead, it risks misdirecting resources away from a thorough investigation.

In 2008, the Canadian Association of Elizabeth Fry Societies and the Native Women's Association of Canada (NWAC) published a discussion paper in which they suggested that:

‘Canadian law is often built around expectations that individuals take responsibility for their actions and nowhere is this truer than in criminal law. What we have noticed over the years is a number of situations where women (particularly when they are racialized, have a disability or a mental illness, are poor or a sexual minority) are expected by the legal system to take more responsibility than others. This is the situation we are referring to as hyper-responsibility.’ (CAEFS and NWAC, 2008, p. 94)

It seems to us that the operation of hyper-responsibility is very apparent when a woman is investigated for killing an intimate partner or a child. The strategies that are used by police in seeking to encourage women to admit to criminal wrongdoing against family members are identifiably different from those used in other contexts, as are the strategies used by the prosecution to persuade a trier of fact of a charged woman's guilt. These strategies seem likely to produce distinctive dangers of wrongful convictions in the broad sense in which we have argued this term should be understood. Casting a critical eye on the prevalence of guilty pleas by women to homicide charges is key,Footnote 46 but we suggest that the problem of wrongful conviction by guilty plea goes well beyond homicide cases.

3.2 Domestic violence

The problem of innocent women and men pleading guilty goes well beyond the high-profile murder cases which are most often identified as wrongful convictions. The literature pays far less attention to the more common, everyday nature of wrongful convictions based on guilty pleas. This omission is particularly important given that the overwhelming majority of cases are resolved by a guilty plea.Footnote 47 Debra Parkes began thinking about this issue after speaking to a friend who is a Crown Attorney in Toronto, and who was working in the Domestic Violence Court at the time of their conversation. This lawyer said it is common knowledge among Crown Attorneys that women charged with assault offences (often in ‘dual charge’ situations or, increasingly, in situations where the woman is the only one charged) will generally plead guilty more quickly and more often than men who are so charged.

However, the lawyer did not consider these convictions to be ‘wrongful’ as, in her view, the reasons for women pleading guilty were outside the scope of criminal justice system actors (i.e. police and prosecutors) in the case at hand. By contrast, it is precisely this idea – the lack of state responsibility or accountability, for example, to provide safety to women and children who are regularly abused in their homes and not safe when they try to leave – that we are interested in interrogating. What might be gained by broadening our concept of wrongful conviction to include women convicted in such situations?

A study by the Woman Abuse Council of Toronto, consisting of qualitative interviews with women arrested for domestic violence-related offences, found that women report pleading guilty in a variety of dubious situations (Pollack, Green and Allspach, Reference Pollack, GREEN and ALLSPACH2005, pp. 11–14). According to the report:

‘[one] woman reported that since she was unfamiliar with the criminal justice system, her partner convinced her to plead guilty:

I pled guilty. (Her partner) told me to do that. He said, “this is easier, you've got no record, you'll just get probation or something. Otherwise, a trial could take 2 years, you'll be going back to court … they'll just keep putting it over and putting it over …” he knows the system so well, so he advised me.' (Pollack et al., Reference Pollack, GREEN and ALLSPACH2005, p. 11; italics in original)

The study goes on to note that the women interviewed ‘identified a number of pressures influencing their guilty plea, such as a desire to return home to their children, lack of financial resources, and pressure/manipulation from their partners. Nine women accepted responsibility and pled guilty early on as they lacked the financial resources to go to trial’ (Pollack et al., Reference Pollack, GREEN and ALLSPACH2005, p. 14). In this regard, constraints on the availability of legal aid may play a particularly pernicious role. Here is just one example of the women's comments:

‘I spoke to a lawyer, it was too much money, he wanted $4500, I said “No, even if I had wanted, I cannot afford.” I pay a high rent, my kids … I can't I went to Legal Aid, I was not approved, because they said “you don't go to jail for this problem, that's why you can't get a lawyer” … Because I just want to finish the problem, to finish the court, that's why I said “Ok, I'm going to plead guilty, just to finish this”.' (Pollack et al., Reference Pollack, GREEN and ALLSPACH2005, p. 14)

This is not to suggest that all of these women were wrongfully convicted. However, the prevalence of guilty pleas and the dubious circumstances in which many of them are entered leads to the likely conclusion that at least some of the women are innocent, whether factually or legally. Studies conducted in the US suggest that the provision of legal aid and other services to victims of domestic violence reduces the incidence of that violence to a significant extent.Footnote 48 Conversely, the denial of legal aid may reasonably be expected both to increase victimisation and to make it more difficult for women to access legal and procedural rights within the criminal justice system. The gendered impact of these factors is compounded by the statistically pervasive inequitable allocation of income and material assets within and after the breakdown of heterosexual relationships.

The tensions between women's paradigmatic role as victims of violent crime and the increasing frequency with which women are charged with violent crime are most evident when women are charged with crimes associated with domestic violence. It has proven difficult for courts and commentators to avoid understanding these cases through the ‘either/or’ lens of victimhood and aggressor, particularly when the accused is an Aboriginal woman (Kaiser-Derrick, Reference Kaiser-Derrick2012). Patricia Monture-Angus argues that a broader conception of victimhood, which incorporates attention to the harms of colonisation as well as the pervasiveness of racism within Aboriginal women's lives, is necessary to understanding Aboriginal women's experiences of violence (Monture-Angus, Reference Monture-Angus1995, p. 171). Elspeth Kaiser-Derrick demonstrates in her comprehensive study of reported Canadian sentencing judgments that this broader conception of victimhood is not reflected in judicial approaches to Aboriginal women who plead to or are found guilty of violent crime (Kaiser-Derrick, Reference Kaiser-Derrick2012).

Lisa Muftic, Jeffrey A. Bouffard and Leana Allen Bouffard (Reference Muftic, Bouffard and Bouffard2007) found in a study of seventy women charged with intimate partner violence that women who were charged alongside their male partners were more likely to be cohabiting with their partner, more likely to have been injured in the incident and more likely to be identified in police reports as having acted in self-defence. By contrast, women who were charged as sole assailants were more likely to be dating their partner and less likely to be injured or identified as acting in self-defence. These authors concluded that women who are arrested in dual-charge situations have more in common with victims of crime than with other accused. Other literature emphasises that the relationship between women's victimisation and their criminal behaviour is contextual rather than deterministic.Footnote 49 Similarly, Toni Williams (Reference Williams, Grabham, Cooper, Krishnadas and Herman2009) warns that judges who seek to understand the victimisation of criminalised Aboriginal women tend to depict Aboriginal families and communities as ‘incubators of risk’, and prisons as the source of healing intervention. Like Williams, Monture-Angus and Kaiser-Derrick, we consider that the common thread of victimisation of Aboriginal women speaks to the need for greater resources to be directed to securing the social and economic equality of Aboriginal women and families. The constraints imposed by this lack of resources and by the effects of victimisation on women's capacity to offer a vigorous legal defence may well produce wrongful convictions when women are charged with intimate personal violence offences.

Cases involving women charged in domestic violence situations raise some different issues from those that tend to pervade the wrongful conviction literature. They implicate social and economic policy, ‘gender-neutral’ justice system responses to gendered crimes, and the implications of inadequate or unavailable legal aid for accused charged with lower-level offences. There are other categories of convictions (such as administration of justice offences)Footnote 50 that also seem highly likely to produce undetected wrongful convictions. Looking for wrongful convictions outside the homicide context is challenging – both at a practical level and in terms of the extent that the justice system is implicated in producing those wrongful convictions – but we suggest it is crucial to developing an inclusive innocence movement.

IV. Conclusion

In this paper, we have argued that the prevailing conception of the term ‘wrongful conviction’ as requiring demonstrable factual innocence is unduly narrow. The prioritisation of demonstrable factual innocence serves an important rhetorical purpose for the innocence movement, as it enlists a moral consensus about who should (not) be subjected to criminal sanctions.Footnote 51 As Abbe Smith has written: ‘who does not love a story about innocence?’ (Smith, Reference Smith2010, p. 316). Demonstrable factual innocence has the benefit of focusing public attention on the risk that a ‘real criminal’ remains undetected, often for a serious and violent crime. This appeal to the importance of catching the real offender establishes a model for wrongful convictions in which there is no doubt that a crime has occurred. However, the focus on factual innocence also excludes certain types of cases – and certain types of claims – from the domain of legitimate claims for review.

When one examines cases in which women have been unjustly convicted, the risks associated with an insistence upon demonstrable factual innocence become more apparent. These convictions arise most often for crimes allegedly committed against children and intimate partners. Samuel Gross and Michael Shaffer (Reference Gross and Shaffer2012, p. 30) have found that women are disproportionately included among the ranks of those who have been exonerated for crimes that ‘never occurred’.Footnote 52 This claim that no crime occurred illustrates the distortions that are introduced by the innocence movement's cleavage to the concept of demonstrable factual innocence. In one sense, a crime has taken place on completion of the actus reus with the requisite mens rea in the absence of an applicable defence. However, many actions that qualify by this definition as a crime are never charged, and so some would argue that a crime is an act (etc.) which is proven to the requisite criminal standard or which elicits a guilty plea. Gross and Schaffer's claim signals a belief that no actus reus was performed. Proving the truth of this belief, however, is complicated by the inherent uncertainty of cause-of-death determinations when a young infant dies, and by the difficulties of proof in child abuse. While it is plainly just to acquit a defendant when the state's proof is inadequate, the claim that no crime occurred fits uneasily with the evidentiary challenges of such cases. This claim risks shifting the focus away from the state's responsibility to demonstrate criminal responsibility and towards the question of whether an accused can show what really happened.

Given that women seem most likely to be wrongly convicted of crimes in which scientific evidence is sparse and uncertain, the unjust convictions of women are excluded at the outset from many aspects of the dominant narrative of the harms of wrongful conviction. For example, mothers convicted of killing children who are now believed to have died naturally, or for whom no cause of death can properly be ascribed, fit uneasily with the pervasive claim that a key ‘wrong’ in wrongful conviction is the fact that the real criminal has not been identified because in such cases the moral imperative to catch a real criminal is non-existent. Likewise, a woman who kills her intimate partner in self-defence has committed the actus reus of culpable homicide, frequently with the requisite mens rea for murder. If denied access to the defence, she is excluded from Bedau and Radelet's definition of a miscarriage of justice (Bedau and Radelet, Reference Bedau and Radelet1987, p. 40), and will in many instances be unable to seek the assistance of innocence projects. Failing to see these cases as wrongful convictions risks further undermining the systemic commitment to protecting accused people by imposing a burden on the state to prove that a crime was committed in circumstances that should result in punishment.Footnote 53 In particular, the focus on demonstrable factual innocence undermines the criminal burden and standard of proof and the expectation that the state should act with the highest standards of probity in criminal investigation and prosecution.

The common law does not differentiate between reasons for acquitting an accused at trial. Correspondingly, we argue that the term ‘wrongful conviction’ applies to any case in which an accused should not have been convicted, whether the source of that error is factual, legal or procedural. This argument has obvious implications beyond the wrongful convictions of women – it applies to any accused whose case is infected with error – but the importance of insisting upon an expansive definition of wrongful conviction is especially clearly demonstrated by the distinctive patterns of error arising in cases where the accused is a woman.

There are at least two obvious implications of taking our inquiry about wrongful convictions beyond factual innocence. The first implication is that such wrongful convictions are not easily remedied or avoided. They will not be solved by prescriptive procedural measures such as videotaping police interrogations or rendering in-custody informant testimony or dubious scientific evidence inadmissible, just a few examples of the many laudable recommendations that have been advanced (and, in most cases, not implemented) in the literature on avoiding wrongful convictions of the factually innocent. Interrogation of this kind by feminists and other critical scholars shows the difficulty of containing wrongful convictions within neat categories. Focusing on women's experiences within the criminal justice system presents a challenge to accepted taxonomies of wrongful convictions. Implicit in that challenge is a desire to expose the danger that innocence narratives may legitimate the liberal legal order by domesticating the threat of error as both anomalous and readily categorised.

The second implication is related to the first. When we open the concept of wrongful conviction to critical interrogation, the state is much more broadly implicated in the production of injustice. The immediate outcome of our review of women's wrongful convictions is that women accused seem to experience a double barrier to justice: they may disproportionately be denied access to appropriate legal advice and representation, and access to defences, at the initial stage; and the criteria for review of criminal convictions disproportionately denies women access to the institutional mechanisms that now exist for remedying unjust convictions. However, if we take seriously the experiences of women, Aboriginal people and other racialised groups, for example, with the very common, everyday operation of guilty pleas, and with the failure of the police and other arms of government to protect women and children from violence, the scope of state accountability and responsibility is expanded in ways that do not fit well with liberal notions of individual freedom and choice that are frequently invoked within the criminal justice system. Sheehy (Reference Sheehy2005) has described the resistance of justice system actors (including lawyers, judges and the police) to feminist claims about state accountability and responsibility in relation to sexual assault and other violence against women made in the Canadian Jane Doe Footnote 54 and Bonnie Mooney Footnote 55 cases. While Jane Doe was significant for its finding of a legal cause of action for discrimination against women in the policing of rape investigations,Footnote 56 Bonnie Mooney foundered in legal terms for lack of causation between the eventual violence committed by Mooney's estranged partner and the police's failure to investigate her reports of his threats against her and her children.

This very preliminary look at some of the ways that the dominant conceptions of wrongful conviction may not address the experiences of women is meant as a starting point for further dialogue and investigation, particularly into the hegemony of factual innocence in the popular and legal understanding of wrongful conviction. Since beginning this project, we have seen increasing attention paid to the circumstances of women who are wrongfully convicted.Footnote 57 However, there is more work to be done to interrogate the intersection of race, gender, class and other marginalisation in the allocation of state and individual responsibility for harms that are susceptible to criminal punishment. This work requires a commitment to the fundamental principles of criminal justice. In particular, academic and activist work on wrongful convictions should take care not to undermine the asymmetric burden of proof that forms the ‘golden thread’ of criminal law.Footnote 58

Footnotes

This research was funded by a SSHRC Standard Research Grant and by the Social Justice and Human Rights Research Project, University of Manitoba. Research assistance was supplied by Zoë Prebble, Jennifer Dyck, Katrine Dilay and Brenda Harvey. Thanks also to Benjamin Berger, Amanda Glasbeek, Imogen Goold, Sonia Lawrence, Kim Pate and Richard Weisman for offering feedback on earlier versions of this paper.

1 This pattern is apparent from the narratives of wrongful conviction supplied in articles and reports, such as Gross and Shaffer (Reference Gross and Shaffer2012). It probably arises in part because of the large number of DNA exonerations that are sexual assault / murder cases.

2 See e.g. the novel The Lovely Bones (Sebold, Reference Sebold2002) and the 2009 movie of the same name (Jackson, Reference Jackson2009).

3 For example, Conviction, a 2010 movie about a wrongful conviction (Goldwyn, Reference Goldwyn2010).

4 Chamberlain v. R (No.2) (1984) 153 C.L.R. 521.

5 Reference under S.433A of the Criminal Code by the Attorney-General for the Northern Territory of Australia of Convictions of Alice Lynne Chamberlain and Michael Leigh Chamberlain No. CA2 of 1988 [1988] NTSC 64 (15 September 1988).

6 R v. Clark [2000] EWCA 54; [2003] EWCA Crim. 1020; R v. Cannings [2004] EWCA Crim. 1; R v. Anthony [2005] EWCA Crim. 952.

7 For example R v. Marquardt 2011 ONCA 281; R v. Sherret-Robinson 2009 ONCA 886; R v. Trotta, [2007] 3 S.C.R. 453, 2007 SCC 49. In addition, Brenda Waudby was wrongly convicted of assaulting a child after initially being charged with murder (Hasham, Reference Hasham2012).

8 Nineteen per cent of adult criminal cases in Canada involve a female accused (Boyce, Reference Boyce2013).

9 For example, women account for eight of fourteen guilty verdicts wrongly obtained in Canadian child homicide cases. Women comprise approximately 11 per cent of those convicted of homicide in Canada (Boyce and Cotter, Reference Boyce and Cotter2013). Gross and Shaffer (Reference Gross and Shaffer2012) found that 7 per cent of exonerees were women in a database that primarily lists those convicted of serious violent crimes, including murder (48 per cent) and rape (35 per cent).

10 See Canadian Association of Elizabeth Fry Societies, online: <http://www.caefs.ca/> (last accessed 6 July 2015).

11 The Criminal Conviction Review Group (CCRG) is the division of the federal Department of Justice that handles applications to the Minister for post-conviction review pursuant to ss. 696.1–696.6 in Part XXI.1 of the Criminal Code. See Department of Justice, Canada, Criminal Conviction Review, online: <http://www.justice.gc.ca/eng/cj-jp/ccr-rc/> (last accessed 6 July 2015). For a discussion of some of the limitations and inadequacies of the Canadian system of criminal conviction review, see Roach (Reference Roach2012, pp. 1494–1501. For an account of the strengths of the Criminal Cases Review Commission adopted in England and Wales, see Hamer (Reference Hamer2014).

12 Kerry Scullion, (Senior Counsel and Manager of the CCRG) 4 July 2007 and 26 June 2013, personal correspondence.

13 ‘False confessions’, usually as a result of a prolonged police interrogation or trickery, are one of the commonly cited causes of wrongful conviction. However, the notion of ‘taking responsibility’ for a crime that we see quite often in cases involving women goes beyond that phenomenon and will be discussed below.

14 Four of eight women wrongly convicted in the Canadian child homicide cases pleaded guilty to the charges against them.

15 For an account of the concept of criminalisation and its importance to contemporary criminology, see Lacey (Reference Lacey2009, Reference Lacey2013). For an intersectional account of criminalisation, see Comack (Reference Comack2006), which discusses the Othering function of the criminalisation of women in particular: ‘crime categories are premised on a dualism between the criminal and the law-abiding, which reinforces the view of women involved in crime as Other and thereby misses the similarities that exist between women. In this way, women who come in contact with the law are in very many ways no different from us. They are mothers, daughters, sisters, girlfriends, and they share many of the experiences of women collectively in society’ (p. 42). Our invocation of the concept of criminalisation also includes attention to the ways that race, class, ability, sexual orientation and other axes of marginalisation operate in the criminalisation of particular women or groups of women.

16 See Hughes (Reference Hughes2011) for a paper that makes a similar point in relation to the dangers of categorising types of innocence.

17 Elsewhere, Lamer describes Parsons as ‘a good person to whom bad things happened’ (Reference Lamer2006, p. 172).

18 A useful working definition of the innocence movement is supplied in Hughes (Reference Hughes2011, p. 1084): ‘the group action of various people throughout the United States, including legal scholars, attorneys and advocates working in various innocence projects, whose primary goals are to raise awareness of wrongfully convicted individuals and to advocate for their release.’ We adopt this definition, but would expand it to include similar work in other common-law jurisdictions such as Canada, Australia and the UK.

19 Raymond (Reference Raymond2001, p. 452) also observes that there is an essential randomness in whether an accused person can muster the evidence that constitutes proof of evidence. We agree, and would go further: beyond the randomness of physical evidence, there is systematic variance in accused persons’ capacity to adduce favourable evidence depending on social characteristics including factors such as wealth, employment status and community; see Cooney (Reference Cooney1994).

20 See Findley (Reference Findley2011, pp. 1164–1166): ‘innocence sceptics’ refers to those commentators who challenge the work done by innocence projects, on the basis that the criminal justice system nearly flawlessly screens out innocent suspects. Such commentators frequently pair this claim with a criticism of innocence projects for allegedly relying on subjective measures of guilt and innocence in their work.

21 A growing body of US scholarship argues that the US Supreme Court's focus on the procedural protections of the Constitution have come at the expense of substantive fairness, including fairness to innocent defendants; see e.g. Stuntz (Reference Stuntz2011, Thomas (Reference Thomas2008). This posited linkage between the rise of proceduralism and a decline in concern for substantive outcomes has also been challenged, for example by Bandes (Reference Bandes2009) and Weisberg (Reference Weisberg2012). We share the view that procedural safeguards and substantive outcomes are not a zero-sum relationship, for reasons set out by Bandes and Weisberg.

22 MacFarlane borrows this idea from Martin (Reference Martin, Westervelt and Humphrey2001b, p. 83).

23 Findley (Reference Findley2011, pp. 1174–1177), describing the influence of ‘innocence consciousness’ in the context of a paper that argues for a broader definition of wrongful convictions.

24 According to the National Registry of Exonerations (2014, p. 16): ‘64% of the female exonerees (62/97) were convicted of crimes that never occurred – mostly child sex abuse – but only 19% of the men were convicted in no-crime cases (224/1184).’

25 Ruesink and Free (Reference Ruesink and Free2005, pp. 2–3) defined wrongful convictions in the traditional way, i.e. as requiring demonstrable factual innocence.

26 See below, text accompanying notes 36–37.

27 See the discussion by the New South Wales Law Reform Commission (2000, para. 2.118), drawing primarily on research conducted in England.

28 Italics in original.

29 See, for example, Roach (Reference Roach2012, p. 1466) and Johnson (Reference Johnson2013), citing a National Institute of Justice report finding that four out of five US police agencies have no written policies concerning eyewitness identification.

30 Fifty-four per cent of the women in Gross and Shaffer's study (and 12 per cent of the men) were convicted of crimes that ‘never occurred’ (Reference Gross and Shaffer2012, p. 30). See also Flowers (Reference Flowers2014), reporting that more than half of female exonorees were convicted of a crime against family or a loved one.

31 See further Cunliffe (Reference Cunliffe and Drakopoulou2013).

32 Perhaps equally suggestive of the operation of racist stereotypes in determinations of credibility, Sheehy (Reference Sheehy2014, pp. 195–198) concluded that Aboriginal women who kill Aboriginal men appear to have somewhat less difficulty persuading the trier of fact of the severity of the deceased man's violence towards her.

33 A particularly striking example of this dynamic is supplied by Sherene Razack in her analysis of the Pamela George case (Reference Razack and Razack2002, p. 122).

34 Stuntz (Reference Stuntz2001, pp. 569–572) has factually innocent defendants in mind, though he also offers an interesting analysis of the decreasing judicial focus on defining offence elements and the limits of defences.

35 See Martin's discussion of the numerous innocent men pleading guilty in Los Angeles, revealed in the wake of investigation into police corruption in the Ramparts Division of the LAPD (Reference Martin2001a, p. 526). See also Sherrin (Reference Sherrin2005), Trotter (Reference Trotter2004) and Ericson (Reference Ericson1994). Ericson discusses ‘discount justice’ and the ways that police, prosecutors and defence counsel operate under a presumption of guilt and actively promote guilty pleas. Gross and Shaffer (Reference Gross and Shaffer2012) report that 8 per cent of those exonerated in the US between 1989 and 2012 pleaded guilty.

36 See e.g. Ratushny (Reference Ratushny1997, pp. 13–14), Sheehy (Reference Sheehy2014, pp. 127–128), Jones (Reference Jones2011). Some women may resist these pressures, as described in Cunliffe (Reference Cunliffe2003, pp. 54–61, 156–157).

37 See Sheehy (Reference Sheehy2014, pp. 192–198), describing a variety of pressures – including linguistic and cultural differences, criminal records, stereotypes, and lack of trust in police and lawyers – experienced by Aboriginal women that contribute to them pleading guilty at rates higher than non-Aboriginal women, in her study of battered women charged with homicide offences in relation to male partners.

38 R v. Gladue, [1999] 1 S.C.R. 688.

39 Ofshe and Leo (Reference Ofshe and Leo1997, p. 11). Within Ofshe and Leo's taxonomy, the ‘decision’ to confess given one's sense of moral responsibility for a harm would seem to constitute a ‘Stress-Compliant’ confession. Where a sense of responsibility beyond the charged offence is leveraged (e.g. the need to care for one's children and a corresponding desire to minimise the adverse consequences of a charged offence), a confession may be ‘Coerced-Compliant’. Interestingly, a large proportion of the examples Ofshe and Leo choose to illustrate their taxonomy involve false confessions made by women, but all of their fully worked case-studies in this classic work involve male suspects.

40 For example, Morgan (Reference Morgan2002, pp.16–29) and Schwartz (Reference Schwartz, DeLisi and Conis2012, pp. 190–191). Compare Wilson and Daly (Reference Wilson and Daly1992, pp. 190–911).

41 This landmark review was the product of years of feminist lobbying led by the Canadian Association of Elizabeth Fry Societies (CAEFS). See Sheehy (Reference Sheehy2000), who notes that while some of CAEFS’ goals were achieved by the Self-Defence Review (SDR), the fact that recommendations were only made in seven cases and no woman was released from jail was deeply disappointing. Sheehy argues that ‘the SDR was constrained by its political location and by the women's need for confidentiality such that many feminist aspirations were ultimately compromised’ (Sheehy, Reference Sheehy2000, p. 197).

42 Sheehy (Reference Sheehy2000) is also critical of the Judge Ratushny's ‘presumption of regularity’ in criminal proceedings. She notes that the modest number and scope of recommendations made in respect of specific women has been interpreted as reinforcing the notion that the criminal justice system generally operates fairly. See headline and story from the Ottawa Citizen, cited in Sheehy (Reference Sheehy2000, p. 232).

43 The Canadian offence of infanticide applies to a mother who kills a child under the age of one year while the balance of her mind is disturbed as a result of giving birth or lactation. It carries a maximum penalty of five years’ imprisonment, compared with a mandatory life sentence for murder. See s. 233 of Criminal Code, RSC 1985, c C-46.

44 The Canadian Association of Elizabeth Fry Societies and Native Women's Association of Canada relate a similar phenomenon in relation to another crime to which an Aboriginal woman wrongly confessed to killing her adult nephew because she felt responsible for his welfare (CAEFS and NWAC, 2008, p. 96).

45 These interview transcripts form part of a SSHRC-funded project presently being conducted by one of the authors. The project examines the court records of twenty-two child homicide cases prosecuted in Canada between 1990 and 2010. A list of the court records in which police interviews are included is available on request.

46 As Sheehy does (Reference Sheehy2014, pp. 192–198).

47 Canadian data from 2011–2012 shows that 64 per cent of cases result in a finding of guilt (including guilty of an included offence and cases in which a discharge is entered); 32 per cent are stayed or withdrawn; 3 per cent are acquittals; and 1 per cent result in other verdicts (Boyce, Reference Boyce2013). It is surprisingly difficult to find Canadian statistics that differentiate between findings of guilt and guilty pleas. We suspect that the rate of guilty pleas is similar in Canada to that in the US, where 95 per cent of criminal cases that result in a finding of guilt are resolved by plea (Auerhahn, Reference Auerhahn2012, p. 95). The Ontario Court of Justice reports that 95.5 per cent of all criminal cases on its docket are resolved before trial or on the day of trial, but without a trial being run; See Ontario Court of Justice (2014).

48 See the studies cited in Abel and Vignola (Reference Abel and Vignola2010, pp. 147–148).

49 See Kaiser-Derrick, (Reference Kaiser-Derrick2012, pp. 55–68) and the sources cited therein.

50 See e.g. Natapoff (Reference Natapoff2012), examining the informal and deregulated process of convicting people – including many who are innocent – of misdemeanour offences in the US, a problem that Natapoff sees as a key ingredient of the racialisation of crime by sweeping up poor defendants of colour.

51 A leading example is Thomas (Reference Thomas2008).

52 Fifty-four per cent of the women in Gross and Shaffer's study (and 12 per cent of the men) were convicted of crimes that ‘never occurred’ (Reference Gross and Shaffer2012, p. 30). Eighty-four per cent of women convicted of crimes that never occurred were convicted of violent crimes against children. See also The National Registry of Exonerations (2014, p. 3).

53 Of course, the criminal law technically captures a far wider range of behaviours than can ever be enforced. In relation to the idea that judgment is exercised by police and prosecutors about when, where and for which behaviours they will make arrests and seek criminal sanctions, see Stuntz, (Reference Stuntz2001, esp. pp. 569–578).

54 Doe v. Metropolitan Toronto (Municipality) Commissioners of Police, 1998 CanLII 14826 (ON SC); 39 OR (3d) 487; 160 DLR (4th) 697; 126 CCC (3d) 12; OJ No 2681 (holding the Toronto police liable for sex discrimination, contrary to the Charter, for its negligent conduct of a rape investigation and, in particular, for failing to warn Jane Doe and other women about a man who had raped at least four other women in the area where she lived).

55 B.M. v. British Columbia (Attorney General), 2004 BCCA 402; [2004] 10 WWR 286; 31 BCLR (4th) 61 (dismissing Bonnie Mooney's claim in negligence based on the RCMP's failure to investigate her reports of threats of violence from her estranged partner, Roland Kruska, in violation of the province's policy on violence against women. The BCCA held that Mooney had failed to prove causation – i.e. that the (in)action of the police contributed materially to the harms suffered, namely Kruska's killing of Mooney's friend, attempting to kill and grievously wounded one of Mooney's daughters, traumatising the other daughter, setting fire to Mooney's home, and killing himself.

56 Sheehy notes that the ‘evidence of systemic sexism in the police processing of rape documented by the Jane Doe litigation is breathtaking, yet, it is completely ordinary in terms of everything we know about this aspect of policing in Canada’ (Reference Sheehy2005, p. 94).

57 See e.g. the recently launched Women's Project at the Bluhm Legal Clinic Centre on Wrongful Convictions, Northwestern University Law School, online: <http://www.law.northwestern.edu/legalclinic/wrongfulconvictions/womensproject/> (last accessed 6 July 2015).

58 Woolmington v. DPP [1935] UKHL 1 (House of Lords). Ironically perhaps (given our concern with the state's role in protecting women from violence) in Woolmington the burden of proving malice gave rise to an acquittal after Woolmington shot and killed his estranged wife.

References

Abel, Laura K. and Vignola, Susan (2010) ‘Economic and Other Benefits Associated with the Provision of Civil Legal Aid’, Seattle Journal for Social Justice 9(1): 139–167.Google Scholar
Auerhahn, Kathleen (2012) ‘“Social Control of the Self” and Pleading Guilty in Criminal Court’, International Review of Sociology 22(1): 95–122.Google Scholar
Backhouse, Constance (1999) Colour-Coded: A Legal History of Racism in Canada, 1900–1950. Toronto: University of Toronto Press for Osgoode Society for Canadian Legal History.Google Scholar
Balfour, Gillian and Comack, Elizabeth (eds) (2006) Criminalizing Women: Gender and (In)Justice in Neo-Liberal Times. Halifax, Nova Scotia: Fernwood Publishing.Google Scholar
Bandes, Susan (2008) ‘Framing Wrongful Convictions’, Utah Law Review 2008(1): 5–24.Google Scholar
Bandes, Susan (2009) ‘Protecting the Innocent as the Primary Value of the Criminal Justice System’, book review of The Supreme Court on Trial: How the American Justice System Sacrifices Innocent Defendants by George C. Thomas III, Ohio State Journal of Criminal Law 7: 413–438.Google Scholar
Bedau, Hugo A. and Radelet, Michael L. (1987) ‘Miscarriages of Justice in Potentially Capital Cases’, Stanford Law Review 40(1): 21–179.Google Scholar
Boyce, Jillian (2013) Adult Criminal Court Statistics in Canada, 2011/2012. Report prepared for the Canadian Centre for Justice Statistics, Juristat. No. 85-002-X. Ottawa: Minister of Industry, Statistics Canada. Online: <http://www.statcan.gc.ca/pub/85-002-x/2013001/article/11804-eng.htm?fpv=2693> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Boyce, Jillian and Cotter, Adam (2013) Homicide in Canada, 2012. Report prepared for the Canadian Centre for Justice Statistics, Juristat. No. 85-002-X. Ottawa: Minister of Industry, Statistics Canada. Online: <http://www.statcan.gc.ca/pub/85-002-x/2013001/article/11882-eng.htm?fpv=2693> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Boyle, Christine (1985) ‘Sexual Assault and the Feminist Judge’, Canadian Journal of Women and the Law 1(1): 93–107.Google Scholar
Brown, Wendy (1995) States of Injury: Power and Freedom in Late Modernity. Princeton: Princeton University Press.Google Scholar
Caddell, Paul G. (1999) ‘The Guilty and the “Innocent”: An Examination of Alleged Cases of Wrongful Conviction from False Confessions’, Harvard Journal of Law and Public Policy 22(2): 523–603.Google Scholar
Caefs (CANADIAN ASSOCIATION OF ELIZABETH FRY SOCIETIES) and Nwac (NATIVE WOMEN'S ASSOCIATION OF CANADA) (2008) ‘Women and the Canadian Legal System: Examining Situations of Hyper-Responsibility’, Canadian Woman Studies 26(3/4): 94–104.Google Scholar
Cheng, Kevin Kwok-yin (2013) ‘Pressures to Plead Guilty: Factors Affecting Plea Decisions in Hong Kong's Magistrates’ Courts’, British Journal of Criminology 53(2): 257–275.Google Scholar
Comack, Elizabeth (2006) ‘The Feminist Engagement with Criminology’, in G. Balfour and E. Comack (eds.), Criminalizing Women: Gender and (In)Justice in Neo-liberal Times. Halifax, Nova Scotia: Fernwood Publishing, pp. 22–55.Google Scholar
Comack, Elizabeth (2014) ‘Theoretical Excursions’, in Comack, E. (ed.), Locating Law: Race/Class/Gender/Sexuality Connections, 3rd edn. Halifax, Nova Scotia: Fernwood Publishing, 9–62.Google Scholar
Conaghan, Joanne (2013) Law and Gender. Oxford: Oxford University Press.Google Scholar
Cooney, Mark (1994) ‘Evidence as Partisanship’, Law and Society Review 28(4): 833–858.Google Scholar
Cunliffe, Emma (2003) ‘Weeping on Cue: The Socio-Legal Construction of Motherhood in the Chamberlain Case.’ LLM, Law, UBC, online: <https://circle.ubc.ca/handle/2429/14550> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Cunliffe, Emma (2009) ‘Infanticide: Legislative History and Current Questions’, Criminal Law Quarterly 55(1/2): 94–119.Google Scholar
Cunliffe, Emma (2011) Murder, Medicine and Motherhood. Oxford: Hart Publishing Ltd.Google Scholar
Cunliffe, Emma (2013) ‘Ambiguities: Law, Morality and Subjectivity in H. L. A. Hart's The Concept of Law’, in Drakopoulou, M. (ed.), Feminist Encounters with Legal Philosophy. Abingdon: Routledge, 185–204.Google Scholar
Cunliffe, Emma (2014) ‘Women and Child Homicide: Exploring the Role of Stereotypes about Gender, Race and Poverty in Contemporary Canadian Cases’, in Kilty, J. (ed.), Within the Confines: Women and the Law in Canada. Toronto: CSPI, 223–244.Google Scholar
Daly, Kathleen (1994) Gender, Crime, and Punishment. New Haven, CT: Yale University Press.Google Scholar
Daly, Kathleen and Chesney-Lind, Meda (1988) ‘Feminism and Criminology’, Justice Quarterly 5(4): 497–538.Google Scholar
Davies, Margaret (2002) Asking the Law Question: The Dissolution of Legal Theory. Pyrmont, NSW: Lawbook Company.Google Scholar
Duncan, Ian (2013) ‘After Murder Exoneration, Freed Man Wonders How Case Went Wrong’, Baltimore Sun, 9 November. Online: <http://data.baltimoresun.com/wrongful-conviction/> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Dyck, Jennifer (2013) ‘Stories from the Front: Realities of the Over-Incarceration of Aboriginal Women in Canada.’ LLM, Law, UBC, online: <https://circle.ubc.ca/handle/2429/45347> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Ericson, Richard (1994) ‘The Decline of Innocence’ [audio-recording]. Lecture delivered to the Vancouver Institute, 12 March, 28 UBCLR 367, online: UBC cIRcle <http://circle.ubc.ca/handle/2429/21814?show=full> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Ewick, Patricia and Silbey, Susan S. (1995) ‘Subversive Stories and Hegemonic Tales: Towards a Sociology of Narrative’, Law and Society Review 29(2): 197–226.Google Scholar
Findley, Keith A. (2011) ‘Defining Innocence’, Albany Law Review 74(3): 1157–1208.Google Scholar
Findley, Keith A., Barnes, Patrick D., MORAN, David A. and SQUIRE, Waney (2011) ‘Shaken Baby Syndrome, Abusive Head Trauma, and Actual Innocence: Getting it Right’, Houston Journal of Health Law and Policy 12(2): 209–312.Google Scholar
Flowers, Alison (2014) ‘Commentary: Is the Exoneree Movement Leaving Women Behind?’, WBEZ91.5, 20 May, online: <http://www.wbez.org/series/exoneree-diaries/commentary-exoneree-movement-leaving-women-behind-110206> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Goldwyn, Tony (2010) Conviction. Fox Searchlight Pictures. Online: <http://www.foxsearchlight.com/conviction/> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Goudge, Stephen T. (2008) Inquiry into Pediatric Forensic Pathology in Ontario: Report. Toronto: Queen's Printer for Ontario, Ontario Ministry of the Attorney General. Online: <http://www.attorneygeneral.jus.gov.on.ca/inquiries/goudge/report/index.html> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Gould, Jon B. and Leo, Richard A. (2010) ‘One Hundred Years Later: Wrongful Convictions After a Century of Research’, Journal of Criminal Law and Criminology 100(3): 825–868.Google Scholar
Gross, Samuel R. and O'brien, Barbara (2008) ‘Frequency and Predictors of False Conviction: Why We Know So Little, and New Data on Capital Cases’, Journal of Empirical Legal Studies 5(4): 927–962.Google Scholar
Gross, Samuel R. and Shaffer, Michael (2012) ‘Exonerations in the United States, 1989–2012: Report by the National Registry of Exonerations’, Michigan Law School and Northwestern University School of Law. Online: <www.law.umich.edu/special/exoneration/Documents/exonerations_us_1989_2012_full_report.pdf> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Hall, Michael (2012) ‘Another Tale of Wrongful Conviction’, Texas Monthly, November, online: <http://www.texasmonthly.com/story/another-tale-wrongful-conviction> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Hamer, David (2014) ‘Wrongful Convictions, Appeals, and the Finality Principle: The Need for a Criminal Cases Review Commission’, UNSW Law Journal 37(1): 270–311.Google Scholar
Hamilton, Alvin and Sinclair, Murray (Commissioners) (1991) Report of the Aboriginal Justice Inquiry of Manitoba. Winnipeg: Province of Manitoba, the Aboriginal Justice Implementation Commission. Online: <http://www.ajic.mb.ca/volumel/chapter13.html> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Hasham, Alyshah (2012) ‘Brenda Waudby Officially Innocent of Child Abuse’, Toronto Star, 27 June, online: <http://www.thestar.com/news/gta/2012/06/27/brenda_waudby_officially_innocent_of_child_abuse.html> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Hughes, Emily (2011) ‘Innocence Unmodified’, North Carolina Law Review 89(4): 1083–1124.Google Scholar
Jackson, Peter (2009) The Lovely Bones. DreamWorks Pictures. Online: <http://www.lovelybones.com/#home> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Johnson, Kevin (2013) ‘Eyewitness Rules Ignored, Wrongful Convictions Result’, USA TODAY, 11 June, online: <http://www.usatoday.com/story/news/nation/2013/06/11/eyewitness-wrongful-convictions-exonerate-dna/2411717/> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Jones, Stephen (2011) ‘Under Pressure: Women Who Plead Guilty to Crimes They Have Not Committed’, Criminology and Criminal Justice 11(1): 77–90.Google Scholar
Kaiser-Derrick, Elspeth (2012) ‘Listening to What the Criminal Justice System Hears and the Stories It Tells: Judicial Sentencing Discourses about the Criminalization and Victimization of Aboriginal Women.’ LLM, Law, UBC, online: <https://circle.ubc.ca/handle/2429/43725> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Kennedy, Helena (2004) Sudden Unexpected Death in Infancy: A Multi-Agency Protocol for Care and Investigation. London: Royal College of Pathologists and Royal College of Paediatrics and Child Health.Google Scholar
Lacey, Nicola (2009) ‘Historicising Criminalisation: Conceptual and Empirical Issues’, Modern Law Review 72(6): 936–960.Google Scholar
Lacey, Nicola (2013) ‘The Rule of Law and the Political Economy of Criminalisation: An Agenda for Research’, Punishment and Society 15(4): 349–366.Google Scholar
Lamer, Antonio (2006) The Lamer Commission of Inquiry Pertaining to the Cases of: Ronald Dalton, Gregory Parsons, and Randy Druken. St John's, NL: Office of the Queen's Printer.Google Scholar
Lawrence, Sonia N. and Williams, Toni (2006) ‘Swallowed Up: Drug Couriers at the Borders of Canadian Sentencing’, University of Toronto Law Journal 56(4): 285–332.Google Scholar
Leo, Richard A. (2005) ‘Rethinking the Study of Miscarriages of Justice: Developing a Criminology of Wrongful Conviction’, Journal of Contemporary Criminal Justice 21(3): 201–223.Google Scholar
Macfarlane, Bruce (2006) ‘Convicting the Innocent’, Manitoba Law Journal 31(3): 403–487.Google Scholar
Martin, Dianne L. (2001a) ‘Distorting the Prosecution Process: Informers, Mandatory Minimum Sentences, and Wrongful Convictions’, Osgoode Hall Law Journal 39(2/3): 513–527.Google Scholar
Martin, Dianne L. (2001b) ‘The Police Role in Wrongful Convictions: An International Comparative Study’, in Westervelt, S. and Humphrey, J. A. (eds), Wrongly Convicted: Perspectives on Failed Justice. New Brunswick, NJ: Rutgers University Press, 77–95.Google Scholar
Medwed, Daniel S. (2008) ‘Innocentrism’, University of Illinois Law Review 2008(5): 1549–1572.Google Scholar
Milgaard, Joyce and Edwards, Peter (1999) A Mother's Story: The Fight to Free My Son David. Toronto: Doubleday Canada.Google Scholar
Monture, Patricia A. (2009) ‘Women and Risk: Aboriginal Women, Colonialism and Correctional Practice’, in Monture, P. A. and McGuire, P. D. (eds), First Voices: An Aboriginal Women's Reader. Toronto: Inanna Publications and Education Inc, 414–422.Google Scholar
Monture-Angus, Patricia (1995) Thunder in My Soul: A Mohawk Woman Speaks. Halifax, Nova Scotia: Fernwood Publishing.Google Scholar
Morgan, Jenny (2002) Who Kills Whom and Why: Looking Beyond Legal Categories (Victorian Law Reform Commission Occasional Paper). Melbourne: Victorian Law Reform Commission.Google Scholar
Morling, Justice T. R. (1987) Report of the Commissioner the Hon. Mr. Justice T. R. Morling / Royal Commission of Inquiry into Chamberlain Convictions. Canberra: Government Printer, the Parliament of the Commonwealth of Australia.Google Scholar
Muftic, Lisa R., Bouffard, Jeffrey A. and Bouffard, Leanna Allen (2007) ‘An Exploratory Study of Women Arrested for Intimate Partner Violence: Violent Women or Violent Resistance?’, Journal of Interpersonal Violence 22(6): 753–774.Google Scholar
Natapoff, Alexandra (2012) ‘Misdemeanors’, Southern California Law Review 85: 1313–1376.Google Scholar
National Registry Of Exonerations (2014) ‘Exonerations in 2013’, Michigan Law School and Northwestern University School of Law. Online: <https://www.law.umich.edu/special/exoneration/Documents/Exonerations_in_2013_Report.pdf> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
New South Wales Law Reform Commission (2000) Report 95: The Right to Silence. Sydney: NSW Government. Online: <http://www.lawreform.justice.nsw.gov.au/Documents/report_95.pdf> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Ofshe, Richard J. and Leo, Richard A. (1997) ‘The Social Psychology of Police Interrogation: The Theory and Classification of True and False Confessions’, Studies in Law, Politics and Society 16: 1–99.Google Scholar
Ontario Court Of Justice (2014) Criminal Court Statistics, ‘Clearance Rate Statistics: Ontario Court of Justice, All Criminal Cases, Ontario Overview April 2012–March 2014’. Online: <http://www.ontariocourts.ca/ocj/files/stats/crim/2014/2014-Q1-Clearance-Rate-Criminal.pdf> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Plaxton, Michael (2002) ‘Are Wrongful Convictions Wrong? The Reasonable Doubt Standard and the Role of Innocence in Criminal Procedure’, Criminal Law Quarterly 46(3/4): 407–446.Google Scholar
Pollack, Shoshana (2000) ‘Outsiders Inside: The Social Context of Women's Lawbreaking and Imprisonment.’ PhD, Social Work, University of Toronto, online: <http://www.collectionscanada.gc.ca/obj/s4/f2/dsk2/ftp03/NQ50006.pdf> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Pollack, Shoshana, GREEN, Vivien and ALLSPACH, Anke (2005) Women Charged with Domestic Violence in Toronto: The Unintended Consequences of Mandatory Charge Policies. Toronto: Woman Abuse Council of Toronto.Google Scholar
Poveda, Tony G. (2001) ‘Estimating Wrongful Convictions’, Justice Quarterly 18(3): 689–708.Google Scholar
Rattner, Arye (1988) ‘Convicted but Innocent: Wrongful Conviction and the Criminal Justice System’, Law and Human Behavior 12(3): 283–293.Google Scholar
Ratushny, Judge Lynn (1997) Self-Defence Review: Final Report (Submitted to the Minister of Justice of Canada and the Solicitor General of Canada). Ottawa: Ministry of Justice.Google Scholar
Raymond, Margaret (2001) ‘The Problem with Innocence’, Cleveland State Law Review 49(3): 449–463.Google Scholar
Razack, Sherene H. (2002) ‘Gendered Racial Violence and Spatialized Justice: The Murder of Pamela George’, in Razack, S. H. (ed.), Race, Space, and the Law: Unmapping a White Settler Society. Toronto: Between the Lines, 121–156.Google Scholar
Risinger, D. Michael (2007) ‘Innocents Convicted: An Empirically Justified Factual Wrongful Conviction Rate’, Journal of Criminal Law and Criminology 97(3): 761–806.Google Scholar
Roach, Kent (2012) ‘Wrongful Convictions in Canada’, University of Cincinnati Law Review 80: 1465–1526.Google Scholar
Roberts, Dorothy (1993) ‘Motherhood and Crime’, Iowa Law Review 79(1): 95–141.Google Scholar
Ruesink, Mitch and Free, Marvin D. Jr. (2005) ‘Wrongful Convictions Among Women: An Exploratory Study of a Neglected Topic’, Women and Criminal Justice 16(4): 1–23.Google Scholar
Scheck, Barry, NEUFELD, Peter and DWYER, Jim (2001) Actual Innocence: When Justice Goes Wrong and How to Make it Right. New York: Signet.Google Scholar
Schwartz, Jennifer (2012) ‘Comparing Women and Men Who Kill’, in DeLisi, M. and Conis, P. J. (eds), Violent Offenders: Theory, Research, Policy, and Practice. Burlington, MA: Jones and Bartlett Learning, 105–206.Google Scholar
Sebold, Alice (2002) The Lovely Bones. New York: Back Bay Books.Google Scholar
Sheehy, Elizabeth A. (2000) ‘Review of the Self-Defence Review’, Canadian Journal of Women and the Law 12(1): 197–234.Google Scholar
Sheehy, Elizabeth A. (2001) ‘Battered Women and Mandatory Minimum Sentences’, Osgoode Hall Law Journal 39(2/3): 529–554.Google Scholar
Sheehy, Elizabeth A. (2005) ‘Causation, Common Sense, and the Common Law: Replacing Unexamined Assumptions with What We Know about Violence against Women or from Jane Doe to Bonnie Mooney’, Canadian Journal of Women and the Law 17(1): 87–116.Google Scholar
Sheehy, Elizabeth A. (2014) Defending Battered Women on Trial: Lessons from the Transcripts. Vancouver: UBC Press.Google Scholar
Sheehy, Elizabeth, Stubbs, Julie and Tolmie, Julia (2012) ‘Defences to Homicide for Battered Women: A Comparative Analysis of Laws in Australia, Canada and New Zealand’, Sydney Law Review 34(3): 467–492.Google Scholar
Sherrin, Christopher (2005) ‘False Confession and Admissions in Canadian Law’, Queen's Law Journal 30(2): 601–659.Google Scholar
Simon, Dan (2006) ‘Are Wrongful Convictions Episodic or Epidemic?’ Paper presented at the Annual Meeting of the Law & Society Association, 7–9 July 2006, quoted in Gould and Leo (2010), p. 832.Google Scholar
Smith, Abbe (2010) ‘In Praise of the Guilty Project: A Criminal Defense Lawyer's Growing Anxiety about Innocence Projects’, University of Pennsylvania Journal of Law and Social Change 13(3): 315–329.Google Scholar
Statistics Canada (1995–2005) Homicide in Canada. Ottawa: Minister of Industry, Annual Reports. Online: <www.statcan.gc.ca.pub/85-002-index-eng.htm>..>Google Scholar
Stuntz, William J. (2001) ‘The Pathological Politics of Criminal Law’, Michigan Law Review 100(3): 505–600.Google Scholar
Stuntz, William J. (2011) The Collapse of American Criminal Justice. Cambridge, MA: The Belknap Press of Harvard University Press.Google Scholar
Sugar, Fran and Fox, Lena (1989) ‘Nistum Peyako Séht'wawin Iskwewak: Breaking Chains’, Canadian Journal of Women and the Law 3(2): 465–482.Google Scholar
Tam, Moy, COLE, David P., GITTENS, Margaret, WILLIAMS, Toni, RATUSHNY, Ed and SRI-SKANDA-RAJAH, Sri-Guggan (Commissioners) (1995) Report of the Commission on Systemic Racism in the Ontario Criminal Justice System. Toronto: Queen's Printer for Ontario. Online: <https://archive.org/details/reportracismont00comm> (last accessed 6 July 2015).+(last+accessed+6+July+2015).>Google Scholar
Thomas, George C. (2008) The Supreme Court on Trial: How the American Justice System Sacrifices Innocent Defendants. Ann Arbor, MI: University of Michigan Press.Google Scholar
Trotter, Gary T. (2004) ‘False Confessions and Wrongful Convictions’, Ottawa Law Review 35(2): 179–210.Google Scholar
Weisberg, Robert (2012) ‘Crime and Law: An American Tragedy: Surveying the Problems Plaguing the Criminal Justice System’, book review of The Collapse of American Criminal Justice by William Stuntz, Harvard Law Review 125(6): 1425–1452.Google Scholar
Williams, Toni (2009) ‘Intersectionality Analysis in the Sentencing of Aboriginal Women in Canada: What Difference Does it Make?’, in Grabham, E., Cooper, D., Krishnadas, J. and Herman, D. (eds), Intersectionality and Beyond: Law, Power and the Politics of Location. New York: Routledge-Cavendish, 79–104.Google Scholar
Wilson, Margo I. and Daly, Martin (1992) ‘Who Kills Whom in Spouse Killings? On the Exceptional Sex Ratio of Spousal Homicides in the United States’, Criminology 30(2): 189–216.Google Scholar