1 Introduction

It is well recognised within existing literature that the law is not simply about ‘rules and policies [but also about] stories, explanation, performances [and] linguistic exchanges’ [8: 2]. Narrative, more specifically story-telling, is a central component of the law and legal discourse, particularly in case law and sentencing remarks [see for example, 11, 13, 33]. Here, it is often the judge in particular who acts as final storyteller, constructing a macro-narrative of the case through their lexical choices and rhetoric which reflects and creates particular discourses around the case. Indeed as Rackley explains, it is through ‘the way in which the judge tells the story [and] the form and language of their opinion’ [33: 46] (emphasis added), that particular narratives emerge in cases.

This article explores some of the ways in which the language and narratives of law discursively constructs lived experiences and dominant identities for women who kill their children. Focusing on this group of homicidal women provides an area that is ripe for analysis, not least because within criminal legal discourse, women continue to be constructed as legal objects, rather than subjects, and thus more easily have dominant identities constructed for them through narrative creation [36]. Drawing upon Jackson’s work on narrative [13], I explore the socially constructed narrative models of the experiences of filicidal women which emerge in statutes, sentencing remarks, and case law. In particular I explore what Jackson terms as the different ‘narrative typifications’ [13: 32–33] that emerge in these cases. Narrative typification ‘refers to evaluative judgments based on their perceived likeness to collective, prototypical images of criminals as well as other narratively organised forms of social knowledge’ [11: 373]. In cases of women who kill, this typification typically involves evaluation of the extent to which these women’s actions can be consolidated or rehabilitated within understandings of appropriate femininity. Narrative typification also allows for a consideration of the micro-narratives related within a trial e.g. through competing evidence given by the defendant and witnesses, and the macro-narrative of the trial itself [13: 33]. Although competing micro-narratives will emerge within the court room, the macro-narrative that emerges from the case is as a result of evaluative judgments made on the completeness of the stories told. Of particular interest in this article are the macro-narratives which emerge in judgments and sentencing remarks in cases of filicidal women and how they discursively create dominant narrative identities for the women concerned. As will be evidenced, these judicial macro-narratives do not necessarily reflect or incorporate the multiple micro-narratives related within the cases, nor the intricacies of the lived experiences of the female defendants involved.

Drawing upon the existing literature which recognises the labelling of filicidal women as ‘bad’, ‘mad’ or ‘sad’ [see for example, 22, 23, 25, 35], I take this further by exploring how these dominant narrative identities emerge in the macro-narratives of sentencing remarks, and case law, as well as critically engaging with the consequences of such narrative typifications and identity constructions. More specifically, I argue that in cases involving filicidal women, judicial and legal rhetoric typically creates a macro-narrative that is focused primarily around agency denial and the discursive construction of narrative identities for the filicidal mother as either ‘bad’, ‘mad’ or ‘sad’. Typically, rather than the true complexity of circumstances in such cases being acknowledged (i.e. multiple micro-narratives), the macro-narrative invoked in judicial rhetoric typically reflects evaluations made of the female defendants’ perceived degree of deviance from gendered stereotypes, particularly those surrounding motherhood.

It is important to note at the outset that in this article I am not arguing that the macro-narratives of the ‘bad’, ‘mad’ and ‘sad’ woman are problematic per se. Indeed, in some cases these narratives may accurately reflect the lived experiences and actions of the filicidal women. Rather, I am arguing that these narratives emerge in such a way that they discursively construct new dominant identities for these women, which is problematic for three main reasons. Firstly, these narrative identities are singular and thus absolute when they emerge, making them pervasive for the women they are attached to, and in some instances preventing the micro-narratives and nuances of the female defendants’ lived experiences from being fully considered. Secondly, they deny the agency of filicidal women, albeit each in subtly different ways, by denying their ability to make any degree of choice in relation to their filicidal actions. Finally, these identities—and the subsequent agency denial which occurs—reinforce pejorative gender stereotypes which are damaging for all women, constructing them as pathological, weak, passive, and pitiful, as well as deviant in failing to adhere to the expectations associated with motherhood.

Before commencing a critical analysis of each of the discursively constructed identities of the ‘bad’, the ‘mad’ and the ‘sad’ filicidal woman, a note must be made of the language and terminology used in this article. The term ‘filicide’ will be used as a general one to refer to the act of a mother killing her child(ren). In contrast, the term ‘infanticide’ is used here only in its legal sense in relation to the specific offence/defence of infanticide found in the criminal law of England and Wales (discussed in more detail later in the article). Finally, it is necessary to set out the mechanics of the term agency, which will be defined here as; the ability of an individual to choose to act or behave in a particular way [35: 338]. When referring to agency denial here, it is more specifically active agency denial that is occurring and thus being referred to. This is because discursively constructing a singular, dominant, narrative identity for these women is a positive, ‘active’, act of doing. This can be contrasted with the passive denials of agency that occur due to the continuing masculine gendering of the legal subject, and construction of women as legal objects who are acted upon within criminal legal discourse [36: 7]. Thus, it is because of these women’s status as objects, whose agency is passively denied, that the ‘mad’, ‘bad’, and ‘sad’ woman narratives can discursively construct new dominant identities for them which actively deny their agency.

2 The Motherhood Mandate

Before exploring the narratives and identities discursively constructed in, and by, the law in cases of women who kill their children, it is first necessary to briefly identify the social context within which these emerge and exist. The gendered meanings, stereotypes and norms of appropriate femininity which are attached to women’s body are extensive. However, within the context of filicidal women the ideology of motherhood for all women is most prominent. Indeed, ‘western society has strong beliefs about motherhood … mother is one of the most fundamental archetypes of woman … motherhood is viewed as central to a woman’s identity and fulfilment’ [2: 317–318].

The ideology of motherhood for all women is also known as the “motherhood mandate” or the “motherhood mystique”; a ‘[c]ulturally proscribed belief that to be complete and successful in the female role, a woman must have children and must spend her time with them’ [24: 223]. The mandate comprises a number of myths including the notions that those women who do not want to be mothers are disturbed or deviant, and those who cannot have children are being fundamentally deprived of a central aspect of what it means to be a woman. Within the mandate, there is a particular focus on voluntarily childless women, who ‘[h]ave been understood in ways that emphasise their selfishness and their deviance, as aberrant, immature, and unfeminine’ [9: 225]. For women who do have children, so-called ‘good’ mothers enjoy caregiving and self-sacrificing to the needs of their children; those that do not are ‘bad’ mothers. Mothers should devote themselves full-time to their children and therefore those who work are often considered to be inferior [2: 318–319].

Women are also expected to protect their children from harm, even if that is at the expense of their own personal safety and autonomy. This expectation is reflected within legal discourse, for example the offence of ‘causing or allowing the death of a child or vulnerable adult’ [3]. The expectation of protection from harm has also been emphasised within case law in R v Sally Lorraine Emery [32], where the mother, a victim of domestic abuse, failed to protect her child from violence by the father. In her appeal, Lord Justice Taylor CJ stated:

It cannot be too clearly emphasised … that a parent’s paramount duty is to protect his or her child. Failure to do so … cannot be excused by other pressures that may have been upon the mother, unless they were such as to render her incapable of action. Still less can such failure be excused by a mother putting her relationship with her partner, or even her own protection, before the life and health of her child [32: 396].

Therefore, when women have children it is not enough for them to be “just” mothers, they must be “good mothers” and conform to the stereotypes and expectations associated with, and symbiotic to, the role. Perhaps most importantly, they must be self-effacing, self-sacrificing and put the well-being of their child(ren) before everything, anything, and anyone else.

The prevalence and pervasiveness of the socially constructed norms within the motherhood mandate means that when women kill their children ‘they present a profound challenge to accepted notions of motherhood and the protection offered by mothers to their children’ [5: 585]. I now turn to the legal response to such gendered deviance, in particular critically engaging with the macro-narratives that emerge in such cases, which discursively construct dominant identities for filicidal women that deny their agency.

3 The ‘Bad’ Woman?

The ‘bad’ filicidal woman is perhaps most frequently conjured up as the image of women who kill their children. However, when the ‘bad’ woman narrative emerges in cases involving women killers, the discourse around the term ‘bad’ differs from simply referring to criminal or deviant behaviour. Instead, when the term ‘bad’ is used in relation to filicidal women it is referring to their extreme deviance from appropriate feminine behaviour, and more specifically their deviance from the motherhood mandate. This deviance is perceived as being unexplainable or inexcusable as a result of mental illness or life circumstances. Consequently, ‘what is being alleged is these women are perceived as being wicked, an ‘extra element’ of bad that goes beyond their actual crime’ [35: 346], which cannot be explained or justified in any way, for example through the use of defences such as diminished responsibility or infanticide.

Unsurprisingly then, ‘bad’ women who kill their children are viewed as doubly deviant; ‘these women not only break the law, but by breaking the law they transgress their own female nature and their primary social identity as a mother or potential mother’ [34: 107]. These homicidal ‘bad’ mothers demolish the construction of motherhood for all women. Therefore, the macro-narrative of such cases focuses on their failures as mothers and thus also as women, as well as the form of ‘monstrous’ motherhood that they engaged in when killing their child(ren). Indeed, in the context of mothers who kill their children, the ‘bad’ woman narrative vilifies the filicidal woman as an abomination. More specifically, in such cases the women have their status demoted firstly from ‘mothers’ to ‘women’ and then to non-women, non-human monsters. Indeed, as noted by Huckerby; ‘the act of killing one’s child is considered so antithetical to the behavioural norms of motherhood to justify the “demotion” of status from “mother” to the pre-maternal state of “woman”’ [10: 151].

For women whose identity is discursively constructed as that of the ‘bad’ woman and thus ‘bad’ mother, their status is further demoted from woman to non-woman, to that of a monster lacking humanity. As Morrissey argues, women ‘murderers considered to have performed especially dastardly acts are transformed into monsters … and are thus disconnected from their societies and from the human race in general’ [23: 16]. Thus within the ‘bad’ woman narrative, these women demonstrate a monstrous, non-human “form of mothering” with their depravity, selfishness, callousness, and violence, which is dichotomous to that expected within the motherhood mandate. The filicidal ‘bad’ woman is so non-feminine, that she is non-woman, non-human, and thus discursively constructed as a monster. This reflects what Jackson calls society’s ‘stock of substantive narratives … which provides the framework for understanding particular stories presented in discourse’ [13: 20]. In the case of the ‘bad’ woman the macro-narrative that emerges constructs a dominant identity for the filicidal woman as a ‘medusa-style’, mythical, non-human character.

3.1 R v Amanda Hutton and R v Rebecca Shuttleworth

This construction of filicidal women as ‘bad’ is illustrated in the sentencing remarks of two recent cases; R v Amanda Hutton and Tariq Khan [27] and R v Rebecca Shuttleworth and Luke Southerton [30]. In R v Amanda Hutton [27], Hutton was convicted of manslaughter for the unlawful killing of her four-year-old son, Hamzah, whom she neglected and starved to death. She was also convicted of preventing his proper burial for 2 years, having left his body to mummify in his cot, as well as being found guilty of several other counts relating to child cruelty of her other young children. In R v Rebecca Shuttleworth [30], Shuttleworth was convicted of murdering her 2 year-old son, Keanu, after seriously assaulting him and then subsequently failing to arrange appropriate medical attention to treat his life-threatening injuries. She was also convicted of cruelty towards another child.

Neither woman successfully made use of any defences which could have mitigated or ‘excused’ their actions. Moreover, the heinousness of their actions was perceived as being so deferent to that expected within the motherhood mandate that the overriding narrative in both of these cases focused on the ‘inexcusable’ actions of these women, their failures as mothers and thus as women and as humans, and the consequences that this had for their children.

Both Hutton and Shuttleworth have specific mention made to their failures as mothers, their inadequacies in caring for and protecting their children, as well as their selfishness [27: 1, 3–4, 25: 2–3]. In particular, attention is drawn to breaches of the duty of care associated with, and assumed to arise from, motherhood. In Hutton the judge noted that starving her son, Hamzah, to death demonstrated her ‘longstanding and wretched breaches of [the] most simple and fundamental requirements’ of being a mother [27: 3] and that her actions had to ‘[b]e regarded as a bad a case of unlawful killing of a child by a parent as it is possible to imagine’ [27: 5]. Similarly, in Shuttleworth her inadequacies as a mother were persistently affirmed through the use of phrases such as; ‘you failed to attend appropriately to his needs … you persistently neglected … you laughed it off …’ [30: 2–3]. It was explained that her actions were termed to be ‘the grossest abuse of a position of trust’, with the judge noting it was Shuttleworth’s most basic and natural duty as a mother to protect her son [30: 6]. This focus on the maternal failings of these women constructs both of them as no longer being mothers, but instead as failed pre-maternal women.

Further demotion of these ‘bad’ women to non-human monsters occurs as a result of a narrative that focuses on their deviant and depraved actions. Indeed, in both Hutton and Shuttleworth the language used by the judge is emotive, vivid, and pejorative in the extreme, invoking imagery and thus constructing a macro-narrative of these women as wicked and evil; as monsters. The semantic fields used in both cases are evocative and mythological, for example; ‘mummified’; ‘squalor’; ‘dreadful’; ‘callous indifference’. Interestingly, the monstrous creature construction is invoked in subtly different ways in both of these cases. Within the sentencing remarks in Hutton there was a more sustained focus on her actions, what she did, as well as her omissions in providing for her child(ren). In contrast, in Shuttleworth, the monstrosity narrative emerges in relation to a focus on her son’s injuries, and thus more clearly on the outcome(s) of her actions.

Taking first the case of Hutton, her actions were described as ‘wicked’ [27: 3] and her attitude towards her son as ‘cruel’ [27: 4]. Within the sentencing remarks, there was a clear and detailed focus on the way in which she neglected her children, particularly Hamzah, with the judge centring on how she purposefully starved him, whilst also noting that his siblings ‘in fact received less than adequate food themselves’ [27: 4]. Attention was also drawn to the living conditions of the children, with the judge noting that Hutton was bringing all of the children up in ‘breathtakingly awful’ squalor [27: 3]. It was made clear that Hutton’s actions were purposeful and no explanation or excuse could be provided for her lack of care for Hamzah or her other children; ‘your deviousness and entirely purposeful conduct in keep various Agencies away from you and your children, so that you were doing was not discovered, must be a feature of this case …’ [27: 3].

Perhaps the most powerful passage within the sentencing remarks, which reinforced and reflected the discursive construction of Hutton as monstrous and non-human, focused on the starvation of her son:

The most telling and awful fact in this case that speaks volumes about how you starved Hamzah is that when his mummified remains were found he was comfortably clothed in a ‘babygro’ which was designed for a 6 to 9 month old child. Moreover, he was found in a cot wearing, at the age of 4.5 years, a nappy [27: 4].

This passage, although noting the outcomes of her behaviour, clearly highlighted her actions in failing to feed her child and the tremendous neglect and suffering inflicted on her son, in particular the way in which after he died his remains were left to mummify. This combined with the horrific imagery painted by such a detailed focus on her actions (and omissions) not only explicitly reinforced the reference made to Hutton as being a ‘bad’ mother, but also constructed a macro-narrative based around a dominant identity of her as a depraved monster.

This construction of Hutton as non-human and non-woman was solidified by explicit recognition that Hutton would never be allowed to care for a child again; ‘you Amanda Hutton must be regarded as a real danger to any child with whom you may live, or in any way have care of in the future, the possibility of you ever in fact being in the situation of caring for a child is so remote …’ [27: 2]. No opportunity was given either during the sentencing, or indeed in the future, for Hutton to be rehabilitated to the status of full womanhood. Her status as a non-human, monstrous creature was absolute and unyielding.

In contrast in Shuttleworth, it was the vivid description of the injuries suffered by her ‘defenceless’ [30: 1] two-year-old son that demoted her from explicitly being recognised as a ‘bad’ mother, to discursively constructing her as a monster. Although there was also some consideration of her actions, particularly in relation to the facts of the case, by focusing in such detail upon the distress and injury suffered by her son, Shuttleworth’s dominant identity as being non-human was cemented.

The judge focused not only upon the fatal injuries sustained by Keanu, but also previous serious injuries he had received from Shuttleworth. For example; ‘There was a nasty burn to the sole of his foot … consistent with his foot being held by force against a hot radiator or other hot surface … you left Keanu to limp in pain all day until he had to go to hospital that evening when even more serious injuries had been caused’ [30: 4]. In particular, the descriptions of Keanu’s fatal injuries were detailed and graphic;

No one could look at the photographs of Keanu taken after his death, or even the body mapping images of the multiple injuries he suffered, without being appalled at the brutality they represent. There must have been many separate blows … There were injuries inflicted to his abdomen, his chest, his face, his head, and his back … The depth and extent of bruising to Keanu’s back strongly suggests he must have been thrown or slammed against the floor or a wall or another firm surface for at least part of the attack upon him. One can scarcely imagine the pain and distress Keanu must have suffered from this outburst of violence … [30: 5].

The detailed descriptions and imagery of the violence and extent of the injuries Keanu sustained, created a narrative within which her behaviours were constructed as being so deviant and antithetical not only to motherhood and femininity, but also to humanity, that she could only have been acting as a monster. For it is only a monster who would inflict such injuries on any child, let alone their own child, in such a way.

This non-woman, non-human, monstrous construction was further reinforced by contrasting her murderous actions with the innocence of her child. Indeed, the judge made specific reference to Keanu’s character at length, as well as his relationships with others;

Keanu was a delightful little boy, described as a real character, a little entertainer, who remained cheerful despite everything. Joyful was the word that more than one witness used to describe him. He was a much loved brother, grandson, nephew and cousin. He was a favourite at nursery as well as in the family. His death has left a void in many lives which will never be filled [30: 1–2].

Alongside this it was noted that ‘Keanu relished the attention [Shuttleworth] gave him when [she] could be bothered to do so and there is no doubt that there was a strong bond between [the two of them] for the good’ [30: 3]. The dichotomy between Shuttleworth’s violent, neglectful, and homicidal actions and the descriptions of Keanu further enhanced Shuttleworth’s depravity and the extent to which she deviated not only from expectations of motherhood, but those of humanity more generally.

3.2 Agency Denial

The ‘bad’ woman construction is absolute; it is a singular narrative identity. Within this ‘bad’ woman macro-narrative there is no space for the micro-narratives of the case to emerge and be fully considered. Thus, the dominant identity of the female defendant becomes that of the ‘bad’ woman/mother/monster. This is clearly demonstrated in Hutton, where it was acknowledged in passing that she was an alcoholic, and thus there was potential to include some divergence within the ‘bad’ woman narrative. Instead the judge used this to further enhance the macro-narrative which constructed her as a ‘bad’ woman, mother, and monster. He commented: ‘your use of alcohol over the years gives a real insight into you as a parent … you placed your own selfish addiction to drink well before your responsibilities to your many children’ [27: 4].

Similarly, in Shuttleworth approximately three of the twenty-nine paragraphs of the sentencing remarks [30: 2–3] are dedicated to acknowledging the difficult childhood and abusive relationships she experienced. Such an approach highlighted the micro-narratives relayed and related within the case. Indeed, the judge explicitly noted his acknowledgment of her ‘troubled upbringing’, the domestic violence she experienced, and ‘the assistance provided to the police recently in connection with another distressing criminal investigation’ [30: 7]. However, these micro-narratives within the case were ultimately used to reinforce the macro-narrative of the case by further highlighting her inadequacies as a mother. Indeed, reference was made to the fact that it was her father, and not her, who took over the parental role for the twins born out of her abusive relationship [30: 2]. The judge noted that the mitigating factors justified a ‘substantially smaller’ reduction in the sentence, than the aggravating factors when taken with the facts of the case as a whole [30: 7]. Thus it is apparent that the macro-narrative of the ‘bad’ woman, which constructs a dominant identity for the female defendant of a monster, occurs as a result of an evaluation by the judge of the competing micro-narratives within the case. When invoked, it is the dominant ‘bad’ woman/mother/monster narrative that overrides all others to become the macro-narrative, with the other competing micro narratives failing to be truly considered when constructing an identity for the filicidal woman.

Discursively constructing this ‘bad’ woman singular narrative identity is an attempt within legal discourse to nullify the challenge that these homicidal women pose to appropriate femininity as dictated by gender discourse, and more specifically, to understandings of motherhood. However, in creating a dominant identity focused upon non-human, monstrous mythology, not only are these gender norms being reinforced, but socio-legal discourse is also failing to explicitly acknowledge the ability of all women and mothers to choose to be violent towards, and ultimately kill, their child(ren). Such an approach denies the agency of not only these filicidal ‘bad’ women, but also of all women.

Within such cases it appears, prima facie, that there is space for some agency recognition to occur, particularly where the judge focuses upon the woman’s actions without invoking explicit excuses or justifications for her behaviours, for example as in Hutton, discussed above. In both Hutton and Shuttleworth, the judge described their actions in an active voice, suggesting agentical wrong-doing. For example, Shuttleworth ‘failed to attend’ [30: 3], ‘persistently neglected’ [30: 2], ‘laughed it off’ [30: 3], and Hutton is condemned for her ‘entirely purposeful conduct’ [27: 3], with her alcoholism being constructed as an intentional choice; ‘you placed your own selfish addition to drink well before your responsibility to your many children’ [27: 4]. Thus, prima facie, it seems that in cases involving ‘bad’ women, the female defendants are actually viewed as hyper-agents, who have made condemnable choices and are thus are fully responsible for their actions. They are unable to sufficiently benefit from any micro-narratives in their cases which suggest any mitigation may be appropriate.

However, when this hyper-agency is combined with the macro-narrative of these ‘bad’ women as non-women, non-human, monsters, a deeper analysis shows that agency denial is still occurring. Although agency is being recognised, more specifically here hyper-agency, both the agent and the type of agency is unrecognisably feminine or human. The ‘doer behind the deed’ in cases of ‘bad’ filicidal women, is not the woman herself, but the non-human monster which is constructed as her dominant identity within the legal narrative. The agent here is the non-human monster constructed by the macro-narrative of the case, rather than the filicidal woman herself. This explains the hyper-agency evident in the judicial language: women are not capable of being hyper-agents due to the masculine gendering of the legal subject and thus their status as ‘objects’ within legal discourse [36]. The hyper-agency is thus not being attributed to the female defendants themselves, but rather to the non-human, mythical creatures they are transformed into. Therefore, in cases of ‘bad’ women who kill their children, it is not these women as mothers or indeed even as women or humans who have chosen to act. Rather it is within the context of their singular narrative identities as non-women, inhuman monsters that they have acted. They have not acted as human women but rather as monstrous creatures whose humanity has been rejected. Thus ‘bad’ filicidal women have their agency, as women and as humans, denied.

This is reinforced by Morrissey [23: 25] who explains; ‘the agency denial which takes place in this technique is specifically that of human agency. The murderess is considered to have acted, but not as a human woman.’ Therefore, a filicidal woman discursively constructed as ‘bad’ is ‘[n]ot just monsterised but [is] transformed into the living embodiment of mythic evil through her relation to figures traditionally interpreted in this way’ [23: 25]. In Hutton and Shuttleworth the vivid dichotomies of the expected behaviour of a good mother, and thus a human woman, were contrasted with continued references to the abhorrent acts perpetrated against, and injuries suffered by, their children, thus evidencing their lack of adherence to any measure of appropriate femininity or humanity. These dichotomous constructions combined to deny the agency of these ‘bad’ filicidal women, specifically as women and as humans [35: 352].

Filicidal women whose human agency is denied but non-human monstrous agency is recognised, are severely punished. These bad women are viewed as being doubly deviant [16] not only have they broken the law and thus offended against society, they have also violated appropriate feminine behaviour and thus have offended against gender norms. They are punished more formally and severely than women whose behaviour can be more readily constructed in relation to appropriate femininity, and whose humanity is recognised and thus agency denied in gendered ways (discussed later in the article in relation to the ‘mad’ and ‘sad’ woman). Both Hutton and Shuttleworth received lengthy prison sentences: Hutton was sentenced to 15 years imprisonment and Shuttleworth was sentenced to life imprisonment with a minimum term of 18 years. Although these sentences may arguably be viewed as appropriate, the consequence of their humanity, and thus human agency, being denied is little chance of rehabilitation to the status of a human, let alone a woman, for these female defendants. The requirement of punishment for these ‘bad’ women whose human agency is denied reflects the non-human, monstrous singular narrative identity created for them.

The agency denial in cases of ‘bad’ women is also problematic in relation to the impact it has on all women. Indeed, the identity transformation of the ‘bad’ filicidal woman into a non-human monster whose human agency is denied, also arguably denies the agency of all women. Transforming these women into non-human creatures when no justifiable excuse or explanation can be offered for their filicidal actions, suggests that every woman is incapable of making the choice to kill their own child. That is to say that every woman is incapable of exercising her agency as a woman. Such an approach only serves to uphold and reinforce the motherhood mandate and gender stereotypes around all women, requiring them to be caring, compassionate, and home-makers.

4 The ‘Mad’ Woman?

The discursive construction of the filicidal woman as ‘mad’ is dichotomous to that of the ‘bad’ woman. The ‘mad’ woman construction is typically invoked in cases where the female defendant successfully utilises the plea of infanticide. The ‘mad’ woman construction as a response to filicidal women is unremarkable, as throughout history a common response to women’s violence has been to invoke pathological and medical discourses to explain such deviance. In 1895, Lombroso and Ferrero [17] were amongst the first proponents of pathologising women offenders’ behaviour. They theorised that women were less highly developed and less intellectually capable than men and therefore less likely to commit crime. Consequently, they labelled offending women as ‘abnormal’ and ‘pathological’ [17].

Despite advances in scholarship and universal criticism of Lombroso and Ferrero’s work, both contemporary society and the law continues to locate women’s criminality within the “psy” discourses in particular, as it offers a tolerable explanation for their deviant behaviours. Although it may indeed be the case that some female defendants are suffering from a mental disorder, in many cases, where pathological macro-narratives are invoked this occurs at the expense of a fuller exploration of the circumstances of the case and thus consideration of the multiple micro-narratives involved. Indeed, as Lambie has argued, a macro-narrative focused around mental disturbance often disguises ‘[t]he relevance of social and economic factors, including dysfunctional parenting and inadequate knowledge about contraception’ [14: 76]. This results in the construction of a dominant identity of the ‘mad’ woman for these women, which is pervasive and overwhelming.

4.1 Infanticide

The legal offence/defence of infanticide was introduced in 1922, reflecting the social and legal dynamics of recognising filicidal women as being partially exculpated as a result of mental incapacity [18: 216]. In its original form the defence only applied to women who killed their new-born child(ren), when their ‘“balance of mind” was “disturbed” by the effect of childbirth’ [18: 216]. The Infanticide Act 1922 was repealed and re-enacted, with modifications, by The Infanticide Act 1938, which is still in in force today. ‘By contrast to the 1922 Act’s reference to “newly born”, the 1938 Act set a 12-month age limit for the child and introduced an additional clause providing that a defendant’s mind could be disturbed by “the effects of lactation” (breastfeeding) as well as “by the effect of giving birth”’ [18: 217].

The current law on infanticide can be found in section one of The Infanticide Act 1938:

Where a woman by any wilful act or omission causes the death of her child being a child under the age of twelve months, but at the time of the act or omission the balance of her mind was disturbed by reason of her not having fully recovered from the effect of giving birth to the child or by reason of the effect of lactation consequent upon the birth of the child, then, notwithstanding that the circumstances were such that but for this Act the offence would have amounted to murder, she shall be guilty of felony, to wit of infanticide, and may for such offence be dealt with and punished as if she had been guilty of the offence of manslaughter of the child.

Before a more detailed analysis of the ‘mad’ woman narrative that emerges in cases of infanticide, several particulars relating to the statute must be noted. Firstly, infanticide is a gendered offence; it can only be committed by female defendants who kill their own child(ren), thus encoding ‘what was … the long-standing special status accorded to women who kill their infants’ [18: 217]. Secondly, infanticide is a partial defence to murder, as well as being recognised as a distinct homicide defence. Thus the terms ‘offence’ and ‘defence’ will be used interchangeably, alongside ‘the law/doctrine of infanticide’ throughout this article. Thirdly, infanticide is ‘the only offence in English law for which a pre-condition is the possession of an abnormal mental state’ [20: 664].

Finally, there is no requirement that the female defendant’s disturbance of mind caused her to kill her child, rather a temporal connection is required [19: 704]. ‘So, any murder of a child under the age of 12 months by its biological mother, whatever the reason, is capable of amounting to infanticide as long as at the time of the murder the balance of her mind was disturbed’ [15: 166]. This simplified relationship between the defendant’s mental incapacity and the act of killing her child, as Loughman argues, ‘forecloses the question of the defendant’s responsibility for her offence … the infanticidal woman is in effect decreed to have attenuated responsibility for her actions’ [19: 704].

As can be seen in both the definition and initial observations made of infanticide, it is an offence that invokes discourses of medicalisation, pathology and abnormality in relation to the postnatal female body and biology. This is reflected in the terminology used within the statute, through phraseology such as ‘the balance of her mind was disturbed’ and ‘not having fully recovered’ [12]. These discourses are reinforced in the legal basis for the plea of infanticide; a diagnosis of puerperal psychosis. This is a ‘severe mental disorder which affects one or two out of every 1000 women within the first few weeks of childbirth. The symptoms span a number of categories of psychosis … from mania to delusions to acute depression’ [22: 206–207].

Although a diagnosis of puerperal psychosis is in theory required to convict a woman of infanticide, research has highlighted that it ‘[i]s very rarely the cause of a mother killing her child. Estimates are that this occurs in around five cases a year’ [22: 207]. Therefore, the courts have traditionally interpreted this strict criterion more liberally to include any sort of psychiatric disorder, or indeed in some cases to include women who are not suffering from any sort of mental disorder. This is reflected in research cited by Morris and Wilczynski which suggests that ‘[a]bout half of the women who … are convicted of infanticide are not suffering from any identifiable mental disorder at all’ [21: 35]. The use of infanticide in such contexts allows for an explanation to be secured for the actions of these filicidal women. More specifically, this ‘mad’ woman macro-narrative reflects gender discourse around appropriate femininity and the idea that all women are inherently irrational, and emotionally unstable, particularly as a result of their biological functions.

4.1.1 R v Sainsbury and R v Sultan

The use of psychiatric and pathological language and discourse, perhaps unsurprisingly, extends outside of the infanticide statute itself into the macro-narratives that emerge from the case law. In R v Sainsbury [31], the appellant, Joan Sainsbury, gave birth ‘to a baby boy in squalid circumstances sitting on a lavatory in the bathroom of her boyfriend’s flat’ [31: 534]. She received no medical assistance or pain relief. Following the birth, Sainsbury and her boyfriend wrapped the baby in a blanket and disposed of the child ‘across the country in an isolated part of the River Frome’ [31: 534]. At trial, she was convicted of infanticide.

During the trial and subsequent appeal against her sentence of 12 months detention in a young offender’s institution, the narrative in the case focused almost exclusively upon her fragile mental condition, thus discursively constructing a dominant identity for her as being that of a ‘mad’ woman. Of the 72 lines of Lord Justice Russell’s judgment in the Court of Appeal, 22 of them referred explicitly to Joan’s mental state, with the majority of the remaining lines of the judgment focusing on the facts of the case and the grounds of appeal relating to her sentence. Joan was referred to as having the ‘mentality of a 12 year old’, as being ‘a woman in body but a child in mind’ and ‘quite unable to cope with pregnancy’ [31: 534]. The birth of her son was described as leaving ‘the balance of her mind disturbed so as to prevent rational judgment and decisions’ [31: 534].

A similar linguistic approach was taken in the more recent case of Natasha Sultan [4], where again the macro-narrative of the case focused upon her mental deficiencies, thus constructing a dominant identity of her as a ‘mad’ woman. Sultan fatally fractured her six-week-old daughter’s skull following a ‘sudden explosion of violence’ [4: 4]. She then put the baby into the cot and went to sleep, with her partner finding the child unconscious when he returned home the following morning from working his night shift. During sentencing, Sultan was referred to as an ‘utterly broken woman’, who had acted as a result of ‘an uncontrolled explosion born from a disturbance of the mind associated with childbirth’ [4]. She was explicitly acknowledged as being in, what was termed as, a “maelstrom of fatigue and mental disorder”, and as suffering from postnatal depression at the time of killing her daughter [4]. In both Sainsbury and Sultan pathological and medical narratives dominated and were pervasive within the judicial rhetoric, with reference being made, and analysis conducted throughout the judgements, in relation to their mental disturbances. Both women were subsequently constructed as mentally disturbed and imbalanced; as ‘mad’ women.

Creating a narrative based around the discourse of the pathological ‘mad’ woman who kills her child(ren) reflects and reinforces existing socio-legal and gender discourse which constructs women as irrational, emotionally abnormal, and seen to be ‘in need of psychiatric help more than men’ [6: 20]. The use of language within both the statute and case law, which focuses on mental abnormality, imbalance, and disturbance, unsurprisingly constructs a macro-narrative of the infanticidal woman as being ‘mad’. Within this macro-narrative, there is no space for competing micro-narratives to emerge, with the semantics of particular lexical choices in infanticide cases focusing almost exclusively on the mental illness and imbalance of the female defendant. Thus, the primary concern with the ‘mad’ woman construction is not that these women are discursively constructed as ‘mad’ per se. Rather it is that they are constructed as ‘mad’ within a macro-narrative that does not allow for the emergence of micro-narratives that more extensively examine their circumstances, instead automatically pathologising their actions and thus denying their agency.

Indeed, in both Sainsbury and Sultan, there was no significant consideration given to the wider circumstances within which their filicides took place. In particular, in Sainsbury there was potential for more significant attention to be given to the context within which the case arose, for example the young age of the defendant (she was only 15 years old) and the apparent lack of supervision and neglect by her parents [31: 534]. Instead, during the appeal these issues received very brief consideration, a total of approximately 8 of the 72 lines of the judgment, and largely within the context of the facts of the case. For example, the judge briefly noted ‘that having been removed from the custody, care and control of her parents this girl was thriving with the assistance of caring foster parents’ [31: 534]. Despite this micro-narrative focusing on neglect from her parents potentially being an important factor in the case, Sainsbury’s actions were instead pathologised with the majority focus, as previously noted, being on her mental disturbance.

4.2 Agency Denial

The agency denial in cases of ‘mad’ filicidal women differs to that of ‘bad’ women. Unlike the ‘bad’ woman, the ‘mad’ woman’s humanity is not removed or replaced with that of a non-human monster. ‘Mad’ women maintain their humanity, and indeed their femininity, throughout, and thus do not have their human agency denied, rather it is their agency as women that is being denied.

In both Sainsbury and Sultan the defendants were presented as not truly intending to kill their children and as not knowing what they were doing. In particular, in Sultan, the judge made explicit reference to the fact that she was a ‘“loving mother who deeply wanted a child” and “adored” her daughter’ [4] and that her actions ‘were not of a calculating cruel woman, rather, an uncontrolled explosion born from a disturbance of mind associated with childbirth’ [4]. Such judicial analysis and rhetoric denies the agency of Sultan herself as a woman, instead associating any agency with the ‘madness’ being experienced. Indeed, rather than acting as an agent, she is discursively constructed as acting without any real intention or understanding of what she was doing. The agent, if any, in this scenario becomes the so-called ‘explosion of madness’.

This non-agentic discourse which emerges in relation to the ‘mad’ woman narrative is perhaps unsurprising, as within both law and society women are not typically and readily recognised as agents [36: 1], instead having their decision-making placed within the context of the biological and hormonal. Thus the agency of women, and particularly that of those women who kill their children, is denied within criminal legal discourse. Indeed, if agency is understood as referring to the ability of individuals to choose to act or behave in a particular way [35: 338], as discussed earlier in the article, in the case of infanticidal women, this is denied by a macro-narrative that focuses exclusively on the influence of these women’s biological functions on their choices. Rather than being understood as agents, the actions of these women are narrated and thus constructed as ‘[a]bherrant “tragedies” for which they are not responsible’ [37: 424]. The ‘mad’ woman macro-narrative within judicial rhetoric constructs infanticidal women as inherently abnormal and thus as lacking agency; rather than exercising choice over their actions. The sentiment is that they ‘must have been “mad” to kill [their] own child[ren]’ [22: 36].

The erasure of the agency of these filicidal women is arguably not problematic per se. Indeed, one consequence of the macro-narrative of the ‘mad’ woman and subsequent denial of agency is more lenient treatment for these women within the criminal justice system, reflecting the ‘chivalry hypothesis’. Within this hypothesis it is argued that female offenders are treated preferentially in a male dominated criminal justice system and are therefore protected from punishment. This is particularly the case where gendered excuses and explanations can be offered for their actions, for example, suffering from mental illness [26]. This is reflected in the overwhelming use of non-custodial sentences for infanticidal ‘mad’ women which was noted in Sainsbury; of the ‘59 cases of infanticide recorded in the … 10 years between 1979 and 1988, not one … resulted in a custodial sentence. There [were] 52 orders either of probation or supervision [and] six hospital orders, one of which was restricted’ [31: 534]. Looking more recently at the period between 1989 and 2000, ‘[o]f the 49 women convicted of infanticide … only two were jailed; the rest were given probation, supervision or hospital orders’ [7]. Non-custodial sentences were also given in both Sainsbury and Sultan, with the former receiving a three-year probation order, and the latter a three-year supervision order.

For women convicted of infanticide the construction of a dominant ‘mad’ woman identity and subsequent erasure of agency may therefore be beneficial in the mitigation of punishment. The underlying pathological discourses reflect the notion that rather than punitive treatment, these ‘mad’ women need ‘“help to come to terms with” what they have done’ [37: 424]. Indeed, in Sainsbury the appeal judges explicitly documented their disagreement with the trial judge’s ‘reasoning to the effect that the welfare of society in this case demanded a custodial sentence’ [31: 535]. Highlighting the need to support and treat Sainsbury, rather than protect society from a dangerous agentic criminal, they noted how ‘without any hesitation whatever’ [31: 535], a probation order was viewed as being the appropriate response in the case, ‘to best serve the interests not only of this appellant but of society as well’ [31: 535]. Thus it is support and treatment for non-agentic, mad, filicidal women that is viewed as being most appropriate to all concerned.

Despite the ‘mad’, non-agentic woman stereotype that is constructed, it is likely that many of these female defendants will not care that they are being stereotyped in this way, as long as the result is a more lenient sentence. Although narratives constructed around gender stereotyping and agency denial have instrumental value in the short term for individual infanticidal women, they also ensure that gender stereotypes surrounding all women’s mental health remains firmly entrenched, i.e. that all women are inherently emotional, irrational, and mentally unstable. As Allen explains, ‘[c]enturies of experience should have taught us that the overall outcome … is invariably the reinforcement of inequality, inferiority and disadvantage’ [1: 431], which takes the form of erasure of agency and the reinforcement of pejorative feminine gender stereotypes. Thus, whilst the ‘mad’ woman may be less punitively treated, she attracts other stigma around her mental health, as well as reinforcing damaging stereotypes surrounding femininity and the denials of women’s agency.

5 The ‘Sad’ Woman?

The narrative typifications of ‘mad’ and ‘bad’ filicidal women ‘are not immutable’ [10: 159]. An alternative macro-narrative is that of the ‘sad’ woman. The ‘sad’ woman narrative has emerged in cases of filicide where the mother has not presented with any psychiatric disorder, and it is clear that she was not acting in a particularly heinous or cruel and calculated manner in killing her child. Thus, the ‘sad’ woman narrative typically emerges in those cases where more lenient treatment and empathy is perceived as being appropriate, but where this subjectively cannot take the form of the ‘mad’ woman narrative.

The ‘sad’ woman macro-narrative emerges as a result of a consideration of the multiple micro-narratives that consider the wider circumstances of the case, allowing an explanation for the woman’s filicidal actions to be offered. Thus in such cases it is often the particularly tragic circumstances surrounding the child’s death, and the mother’s role in it, that take prevalence in the narrative that emerges. In some ways, this makes the discursive construction of the ‘sad’ woman identity less problematic as it may be an accurate reflection of the multiple micro-narratives in, and wider context of, the case. However, these micro-narratives are often considered through a gendered lense which results in the macro-narrative reflecting, reproducing, and reinforcing the gendered norms associated with appropriate femininity. Indeed, the ‘sad’ woman narrative in the context of filicidal women typically discursively constructs these women as pitiful, as ‘social casualties’ [37: 424], as being unable to ‘cope with social pressures’ [22: 199] and thus reacting, often irrationally, in response to their circumstances and inadequacies as women and as mothers [37: 424]. Much like the ‘mad’ and ‘bad’ woman constructions, the ‘sad’ woman narrative is singular and absolute, thus discursively constructing this as the female defendants’ dominant identity.

5.1 R v Nkosiyapha Kunene and Virginia Kunene

The emergence and invocation of the ‘sad’ woman narrative is perhaps most evident in the recent case of R v Nkosiyapha Kunene and Virginia Kunene [29]. The defendants were the married parents of Ndingeko Kunene who died less than six months after birth from florid rickets, resulting from a severe vitamin D deficiency. He had medical problems from birth, with fluctuating health over the short period of his life. However, because of their religious beliefs as strict Seventh Day Adventists, his parents did not seek medical assistance when he became unwell at home. Therefore, on the day he died, although both parents realised that he was very unwell and at risk of death, they continued not to seek medical assistance. They both pled guilty to gross negligence manslaughter. In particular, Mrs Kunene pled guilty ‘only on the basis of the omission to call for medical attention’ [29: 4].

The judicial rhetoric in this case focused upon the micro-narratives surrounding the continuing ill health of the child and the difficulties faced by the parents in reconciling their religious beliefs with caring for their son. Indeed, it was context that was particularly important and thus focused upon in relation to Mrs Kunene’s role in her son’s death, and her subsequent construction as a ‘sad’ woman. For example, it was noted that she was not given advice about taking supplements to support her breastfeeding despite the fact that she was vegan and was exclusively feeding him that way, that her son was not initially diagnosed with a vitamin D deficiency, and that she submitted to her husband, whose religious views in particular were ‘very extreme’ [29: 2–3]. It was also explicitly noted that Mrs Kunene did wish to seek medical attention for her son but was advised by her husband ‘that it would be a sin in the eyes of God to call for medical assistance as he had made a vow that he would await guidance from God before doing so’ [29: 4]. It was acknowledged that her ‘religious convictions were a significant factor in deference to her husband’s view … and her omission to call for medical attention’ [29:4].

The case was explicitly referred to in the opening lines of the judge’s sentencing remarks as being ‘on any view a tragic [one]’ [29: 1] and reference was also made to the fact that Mrs Kunene subsequently had a miscarriage following the death of her son [29: 7]. By creating a macro-narrative reflecting the multiple micro-narratives within the case and thus the context within which Mrs Kunene failed to act in providing medical assistance to her son, the judge discursively constructed her as a woman who had succumbed to social, cultural and religious pressures, particularly from her husband. She was constructed as a caring mother who gave birth to a baby who had numerous medical problems, and as a woman who deferred to her husband’s judgment regarding medical treatment due to the pressures and expectations of her religion. The tragedy for her as a ‘sad’ woman was multiple; oversights in support that should have been offered regarding breast-feeding, the delayed diagnosis of vitamin D deficiency in her son, responding inappropriately in relation to his ill health because of her submission to religious beliefs and familial pressures, her son’s subsequent death, and finally, suffering a miscarriage after Ndingeko died.

Her ‘sad’ woman dominant identity was further reinforced in sentencing, with the judge passing different durations of imprisonment for the two defendants. The judge explicitly accepted that ‘Mr Kunene’s culpability was higher than that of Mrs Kunene, who wanted to call for medical assistance that day but deferred to his view’ [29: 8]. Subsequently, Mr Kunene was sentenced to 3 years imprisonment, whereas Mrs Kunene received a sentence of 2 years and 3 months imprisonment [29: 9].

5.2 The Relationship Between the ‘sad’ and ‘mad’ Woman: R v Kai-Whitewind

The ‘sad’ woman narrative has also emerged in cases of infanticide where women may not have the requisite mental illness diagnosis to successfully plead the defence, but the facts of the case and the context within which they killed their child(ren) are such that it is felt that they should be treated more leniently. Similarly, in cases where the female defendant’s deviance from appropriate femininity and/or the facts of the case are not perceived as being heinous or reprehensible enough to construct a macro-narrative around the ‘bad’ woman, the ‘sad’ woman narrative construction offers a palatable alternative. The relationship between the ‘mad’ woman and ‘sad’ woman narratives in the context of infanticide is perhaps unsurprising, because, as was noted in the initial discussions on the offence, there is no requirement of a causal link between the defendant’s mental disorder and their filicidal actions.

Despite a requirement within infanticide that women should be suffering from a disturbance of the mind, as has already been noted, the judiciary are traditionally liberal in their interpretation of the requirement of puerperal psychosis, to include any sort of recognised mental disorder or indeed include those women who are not suffering from any sort of psychiatric illness at all. In such cases it is possible that the ‘sad’ woman narrative is being invoked, which suggests that the use of this macro-narrative in cases of filicide by mothers is far more subjective than those of the ‘mad’ or ‘bad’ woman. This reflects the consideration of multiple micro-narratives in constructing the ‘sad’ woman macro-narrative, and this as the dominant identity of the filicidal woman. This can be contrasted with the dominant identities of the ‘bad’ and ‘sad’ woman where, as noted earlier, the macro-narrative of the case may not reflect the multiple micro-narratives within it. Therefore, it is arguable that in relation to the ‘sad’ woman identity the judiciary is subjectively incorporating within the parameters of the infanticide defence those women who are not suffering from a recognised psychiatric condition by considering the multiple micro-narratives which emerge. As such I would suggest that this discursive construction of filicidal women as ‘sad’ is almost a non-pathologised extension of the ‘mad’ woman narrative acknowledged earlier.

Reflecting this relationship between the narrative constructions of the ‘mad’ and the ‘sad’ woman, there have been suggestions by the judiciary that the definition of infanticide should be expanded to include cases involving the discursively constructed ‘sad’ woman who is pitied because she cannot cope with societal and mothering pressures. In R v Kai-Whitewind [28], the Court of Appeal in obiter commented that ‘the law relating to infanticide is unsatisfactory and outdated. The appeal in this sad case demonstrates the need for a thorough re-examination’ [28: 458].

In this case, Chaha’oh Niyol Kai-Whitewind, the mother of three children, was convicted of murdering her youngest son, twelve-week old Bidzil, who she claimed had been conceived in the course of a rape. At trial, the Court heard how she claimed to be suffering from depression and had discussed this with her health visitor. Despite this, and Bidzil’s age being within that required for infanticide, no evidence was put forward to sustain the defences of either diminished responsibility or infanticide. Instead the defence case rested on the argument that her son had died from ‘natural, even if unexplained causes’ [28: 457]. Passing judgment and dismissing the appeal against her conviction, the Court of Appeal included some observations on reforming the law of infanticide to include the discursively constructed ‘sad’ woman within the offence’s remit. They questioned whether ‘[i]nfanticide should extend to circumstances subsequent to the birth, but connected with it, such as the stresses imposed on a mother by the absence of natural bonding with her baby …’ [28: 484]. They also suggested that infanticide may be an appropriate defence in cases where ‘the mother … has in fact killed her infant [but] is unable to admit it’, suggesting that her denials may in and of themselves, be a symptom of a mental disorder and her filicidal actions [28: 484].

The issues raised in Kai-Whitewind were considered by the Law Commission in their review of the law of infanticide, in particular the dilemma of how the law should deal ‘with cases in which a mother suffering from a postpartum psychiatric disorder kills her child, but denies the killing’ [15: 167]. Their recommendation was that where the issue of infanticide is not raised during trial, the trial judge should be given ‘the power to order a medical examination of the defendant with a view to establishing whether or not there is evidence that at the time of the killing the requisite elements of a charge of infanticide were present’ [15: 167]. It should be noted however that this recommendation was never implemented into the statute. The implication of such a reform could have been to more easily allow those discursively constructed ‘sad’ women, unable to cope with social and mothering pressures, to instead be constructed as ‘mad’ women with a recognised mental disorder. That is to say, that there would be potential for the ‘sad’ woman narrative to become that of the ‘mad’ woman in such cases.

5.3 Agency Denial

Similarly to the ‘mad’ woman, the ‘sad’ woman has her agency denied whilst maintaining her humanity and her femininity. In such cases, although there is no doubt that the female defendant acted (or in the case of Kunene omitted to do so), her ability to make choices with regard to her behaviours is ultimately inhibited because of a consideration of the tragic circumstances surrounding these cases. This ‘sad’ woman construction sits well with gender discourse which constructs women as being ‘subject to control at the hands of their partners and a patriarchal society’ [25: 113], and thus as victims of their circumstances. Thus, their agency is denied due to their construction as victims of circumstance, acting as a result of being unable to cope with motherhood and social pressures, rather than being able to make their own choices with regards to their mothering behaviour. Consequently, they are not seen to have made the choice to act, rather they, and their actions, are merely the products of their limited life experiences, the pressures associated with motherhood and, more widely, societal, cultural, and religious pressures, which they are unable to manage or to resist.

As in the case of the ‘mad’ woman, the discursive construction of the ‘sad’ woman whose agency is denied may be beneficial in securing more lenient treatment. For example, the sentence of just over 2 years received by Mrs Kunene. However, once again this kind of agency denial is harmful for women, and particularly mothers, by reinforcing the stereotypes associated with the motherhood mandate and more widely those of appropriate femininity. The ‘sad’ woman narrative discursively constructs these women as meek, passive, and victims of tragedy and difficult life circumstances, and as such maintains that they should be pitied and not vilified. The ‘sad’ woman narrative reinforces the need for adherence to the motherhood mandate and feminine gender stereotypes by all women by highlighting the tragedy of women who are unable to conform. For example, women who are unable to ‘naturally’ and ‘properly’ bond with their child(ren) [28], women who are not properly advised with regards to the treatment of their child(ren) (Kunene) [29], or where they lack practical skill or knowledge in relation to the care of their child(ren). Whilst agency denial may allow women to win their individual battles within the criminal justice system by securing more lenient sentences, it does little to allow women to win the war against having to conform to feminine gender stereotypes or having their individual agency recognised.

6 Concluding Thoughts

In this article I have critically engaged with narratives that emerge in cases of women who kill their children. In particular, I have argued that three macro-narratives typically emerge which discursively construct dominant identities for filicidal women as ‘bad’, ‘mad’ or ‘sad’ women. Although invoking differing discourses, all three of the dominant identities constructed by the narratives reinforce gender norms associated with appropriate femininity. The ‘mad’ and ‘sad’ representations not only reinforce, but also reproduce, gendered norms associated with appropriate femininity, such as irrationality, passivity, and weakness. The ‘bad’ woman construct is invoked slightly differently. Rather than also reproducing gendered norms, it constructs these women as non-women, non-human, monsters in order to mitigate and nullify the challenge they pose to the motherhood mandate and appropriate femininity. As such the discursively constructed ‘bad’ woman reinforces the gendered norms associated with good motherhood and appropriate femininity, such as care-giving, altruism, and non-violence.

All three of the discursive constructions of the ‘bad’, ‘mad’, and ‘sad’ woman restrict the portrayal of women who kill their children to that of non-agentic beings. Indeed, each narrative clearly denies the agency of the female defendant, either by constructing them as non-human, mythic monsters whose human agency is denied whilst their non-human monstrous agency is asserted, or as women whose agency is denied because they lack understanding in relation to their actions, or are victims of their circumstances. Whilst for the ‘mad’ and ‘sad’ woman denials of their agency may be instrumentally valuable in justifying lenient sentences, such an approach is pejorative for all women, reinforcing damaging feminine gender stereotypes, and expectations around the motherhood mandate. For ‘bad’ women whose human agency is denied, and whose discursively constructed monstrous agency is acknowledged, there is little expectation of rehabilitation to the status of a human, let alone a woman, a position which is uniformly damaging. The ‘bad’ woman narrative identity also reinforces gender stereotypes and agency denials for all women by suggesting that women are incapable of choosing to kill their child(ren), unless this choice is directed by mental illness or tragic circumstances (in which case agency is also denied). What has become abundantly clear is that in the macro-narratives that emerge ‘nowhere, it seems, is the presentation of a woman who is both violent and agentic, responsible and human’ [23: 167].