The Vulnerable (M)other and the Autonomous Legal Subject: Rethinking Vulnerability in Criminal Law

Abstract

In this article I apply Fineman’s vulnerability thesis to explore the ways in which vulnerability is constructed and mobilised in a criminal law context. Using a ‘failure-to-protect’ offence as a case study reveals contemporary constructs of vulnerability as both a problem to be solved and gendered. Constructing women as pathologically vulnerable allows the state and its institutions to downplay the situational vulnerability of women, evading responsibility for tackling VAWG. Responsibilising women to manage risks to children posed by male violence requires that women undertake ‘safety work’, rendering them vulnerable to both moral and legal sanction if not performed adequately. Replacing the autonomous subject with the relationally vulnerable subject generates new understandings of the ways entwining femininity and vulnerability shores up the (male-coded) autonomous legal subject. Moreover, reconceiving vulnerability as universal reveals the potential of the vulnerable subject for a more inclusive criminal subject who is both embedded and embodied.

1 Introduction

Drawing on Fineman’s theory of vulnerability, I argue that expanding criminal liability of women who fail to protect their children results from the feminisation of vulnerability, and the consequent responsibilisation of women to manage and avoid risk of male violence (towards themselves and their children). Having demonstrated in previous work the potential of vulnerability theory to challenge gendered tropes which deny women agency when on trial, this article takes up this theory in more depth, exploring what may be gained by replacing the autonomous criminal subject with the vulnerable criminal subject (Singh, 2017).

In advocating a more responsive state which builds resilience in citizens and institutions, Fineman’s theory calls on the state to take responsibility for the safety of women and children, rather than holding abused women responsible for the safekeeping of their children. More fundamentally, I argue that reframing vulnerability allows us to confront and challenge criminal law’s production and perpetuation of the gendered nature of care, as well as core criminal law concepts of risk and autonomy. Applying Fineman’s theory to this specific criminal law context and bringing it into conversation with other feminist work, particularly Naffine’s work on the criminal legal subject, generates new understandings of the dynamic between vulnerability and autonomy. Moreover, I show how depicting women as pathologically vulnerable shores up the myth of the autonomous legal subject. Finally, given that Fineman’s vulnerable subject is embedded in relationships both socially and institutionally, teasing out links with Nedelsky’s theory of relationality reveals how prevalent constructs of the autonomous legal subject downplays the role of relationships in increasing our resilience or exacerbating our vulnerability.

The offence of ‘causing or allowing a child or vulnerable adult to die or suffer serious harm’ was initially intended to allow prosecutions where a child had been injured and neither parent would admit to causing the harm, nor incriminate the other. ¹ However, its final form, as enacted in section 5 of the Domestic Violence, Crime and Victims Act 2004 (s.5) (as amended by the Domestic Violence Crime and Victims (Amendment) Act 2012, section 1), ² permits the prosecution of so-called ‘passive abusers’: those who knew, or ought to have known of a significant risk of serious harm to a child or vulnerable adult and failed to stop it. As I have argued elsewhere, this provision disproportionately criminalises abused mothers for allowing harm to come to their children (Singh, 2021). S.5 offers an ideal case study as it demonstrates how stigmatising notions of vulnerability are grounded in neoliberal ideals which attribute responsibility to the vulnerable to take steps to mitigate against the risk of harm from others (Munro and Scoular, 2012). This is significant given that so-called ‘failure-to-protect’ provisions are dominant features of Anglo-American child protection systems (Strega et al., 2013). Moreover, increasingly carceral responses to child abuse means that a failure to intervene is increasingly encoded in criminal law responses, as evidenced by the enactment of similar omission-based offences in other jurisdictions including the US and New Zealand (Panko, 1995; Tolmie et al., 2019). The criminalisation of women for failing to protect their children is important and topical given the global ‘shadow’ pandemic of violence against women and children that resulted from Covid-19 lockdowns (United Nations Women, 2020).

While vulnerability has been theorised extensively, not least in human rights (Heikkilä 2020; Timmer, 2013), family (Gordon-Bouvier, 2021; Mant and Wallbank, 2017; Wallbank and Herring, 2013) and health law contexts (Lindsey, 2019; Travis, 2019), it remains under theorised in a criminal law scholarship. Where such analysis exists, it has predominantly focused on how vulnerability has been used against women to advance paternalistic agendas regarding sexual assault and sex work (Munro 2017; Munro and Scoular, 2012) or addressed responses to vulnerability in the criminal justice process (Brown, 2014; Dehaghani, 2021). Moreover, while feminist philosophers, such as Friedman, have drawn on the vulnerability of abused mothers to consider their moral culpability for failing to protect their children, this article considers how constructs of vulnerability inform the substantive criminal law and understandings of the legal subject (Friedman, 2014).

The article is split into three sections. Section 2 begins by outlining key aspects of Fineman’s theory, acknowledging her thesis as inherently relational. I then explore how traditional constructions of vulnerability have led to problematic legal interventions which frequently exacerbate the position of those labelled ‘vulnerable’ (Mackenzie et al., 2014, p. 9). By positing that vulnerability is universal, Fineman’s thesis provides a lens to explore this issue, offering the potential to expose the link between gendered expectations of dependency and care and the imposition of criminal responsibility. This reveals how ‘failure-to-protect’ provisions epitomise a culture that privileges inevitable dependency (such as that associated with age), but refuses to acknowledge, indeed punishes, derivative dependency caused by situational factors such as caregiving and intimate partner abuse.

Section 3 analyses the offence of ‘causing or allowing a child or vulnerable adult to die or suffer serious harm’, revealing both how problematic notions of vulnerability underpin this offence and highlighting incongruities between how the state frames vulnerability and responds to it. Using this offence as a case study, I apply a relational theory of vulnerability to explore the dissonance between Parliamentary and judicial approaches to the question of who is considered vulnerable. I argue that s.5 is borne out of a belief that mothers should be able to ameliorate the vulnerability of their children, irrespective of circumstances. Exploring case law addressing who may be considered a vulnerable adult for the purposes of this offence exposes the ways in which the vulnerability of children is portrayed as justifying the criminalisation of abused women who might otherwise themselves be considered vulnerable victims.

Finally, section 4 questions what is gained by replacing the autonomous criminal subject with the vulnerable legal subject. Here I explore how relational vulnerability exposes the violence needed to maintain the illusion of the male-coded liberal subject. Focusing instead on the vulnerable subject destabilises neoliberal constructs of autonomy which pervade the criminal-justice system, perpetuating inequality along gender, race and class lines. In exposing autonomy as a gendered myth, the relationally vulnerable subject resists the feminisation of care and individualisation of responsibility which facilitate endemic violence against women and children. In situating the legal subject as embedded, relational vulnerability also begins to unpack foreseeability and risk, key constitutive elements of criminal offences and the attribution of criminal responsibility. Finally, by positioning resilience as the counterpoint of vulnerability this analysis brings structural disadvantage to the fore, revealing patterns in criminality which bolster calls for the state to take a more active role in building the resilience of institutions and citizens and promote more ethical political, social and legal responses to responsibility borne out of dependency.

2 Vulnerability; an inevitable aspect of the human condition or the problem to be solved?

The central premise of Fineman’s thesis is that vulnerability is inherent to the human condition, thus challenging conventional understandings as an individual characteristic that is undesirable and avoidable (Fineman, 2008). In emphasising that vulnerability is universal, Fineman’s mobilisation of the concept resists hierarchies; no one is more or less vulnerable, just differently so (Fineman, 2008). This avoids labelling particular groups as vulnerable due to a shared characteristic which frequently has the ‘paradoxical effect of exacerbating existing vulnerabilities or generating new ones’ (Mackenzie et al., 2014, p. 9).

Group or identity-based notions of vulnerability are also problematic in that they reify the existence of the invulnerable liberal legal subject (Fineman, 2019, p. 355). Fineman’s notion of vulnerability is contrasted to what she terms the ‘myth’ of autonomy (Fineman, 2000; Fineman, 2005). Due to her emphasis on shared dependency her conceptualisation of autonomy is Foucauldian in that it implies responsibilisation of the legal subject, rather than simply as a right to do whatever one pleases (Mackenzie and Rogers, 2013; Morgan and Veitch, 2004). Fineman argues that current responses to vulnerability are problematic because they derive from a notion of individualistic autonomy that is symptomatic of capitalism. Based on an unachievable state of being (i.e., invulnerability), the autonomous legal subject is an ideological construct that, over time, has become naturalised and accepted; a kind of ‘idealised ordinary’ (Fineman, 2016). The crux of her argument is that by perpetuating the myth of the autonomous legal subject (along with emphasis on independence and self-sufficiency) the state fails to engage meaningfully with dependency and is thus exonerated from responding to the vulnerability of its citizens. In contrast to neoliberal constructs of resilience as individualised responsibilisation through self-management, Fineman consequently argues for states to take a greater role in increasing the resilience of their populations (as opposed to ameliorating our vulnerability) (Fineman, 2010). Specifically, she advocates redistributive welfare policies, as opposed to reactionary legislation in response to social challenges (Fineman, 2010).

Providing a much-needed counterpoint to individualistic constructs of autonomy, Fineman’s thesis extends beyond vulnerability in an (inter)corporeal sense; recognising that, as dynamic beings in a dynamic society, we are embedded in and thus individually and collectively vulnerable to social, institutional and legal frameworks (Fineman, 2008). This is of particular significance in the failure-to-protect context since, as Nedelsky highlights, intimate partner violence remains ‘a problem that cannot be solved without a restructuring of relations’ (Nedelsky, 2012, p. 307). The marginalisation of mothers in ‘failure-to-protect’ discourse is fundamental to maternal ideology at large, as well as suggesting that tragedies only befall those on the wrong side of the ‘good mother/bad mother’ dichotomy. Diprose notes how this focus on ‘a politics of blame’ misses the bigger picture. By blaming individuals, it loses focus on building the collective resilience required to challenge dominant socio-legal constructs such as that of the ‘good’ mother or, even more fundamentally, the ideal woman (Diprose, 2013, p. 190). A vulnerability framework shifts the focus from the characteristics and behaviour of the individual, thus moving away from the temptation to punish individual women for failing to mitigate their vulnerability to violence. Instead, it questions why the abuse of women and children remains a normalised part of our culture; both socially and legally.

Fineman’s thesis is not, however, beyond critique. Her emphasis on vulnerability as universal has led to claims that she does not sufficiently acknowledge the role the state plays in actively creating and perpetuating structural vulnerabilities in specific groups, and particularly how race affects the inequal distribution of vulnerability (Davis and Aldieri, 2021, p. 322). While this is a valid criticism, emphasising the inevitability of vulnerability unlocks what Murphy terms its ‘ambivalent potential’ and, in doing so, encourages policies which build resilience through the provision of resources as opposed to often punitive interventions which seek to eradicate vulnerability based on specific characteristics (Murphy, 2012, p. 86). Fineman’s call for a more responsive state is also criticised for being utopian, given that resources are inherently limited, and naïve in that it idealises humanity, expecting unrealistic levels of altruism (Cloud, 2013). However, importantly, her argument for redistribution of resources to increase resilience challenges the privatisation of dependency that occurs under neoliberalism. In doing so, this thesis reminds us that, (just like constructs of motherhood and femininity), capitalism, and the injustices it necessitates, are not to be passively accepted: ‘Because neither inequalities nor the systems that produce them are inevitable, they can also be objects of reform’ (Fineman, 2008, p. 5). Furthermore, it is claimed that in ‘devaluing’ autonomy she fails to recognise the ways in which our independence shores up our feelings of dignity, safety and security (Kohn, 2014, p. 15). This criticism, however, fails to consider that we perhaps only value autonomy so much because vulnerability (and, by association, dependency) are portrayed as both negative and to be avoided. As Butler comments, ‘If we oppose vulnerability in the name of agency, does that imply that we prefer to see ourselves as those who are only acting, but not acted upon?’ (Butler, 2016, p. 14).

Notwithstanding these criticisms, Fineman’s theory is particularly suited to challenging the criminalisation of abused mothers for failing to adequately manage risk of violence to both themselves and their children on account of three key aspects. First, her framework is designed to address a legal context. Secondly, although she has not applied vulnerability theory to a criminal law context, she has written extensively on the gendered privatisation of care (Fineman, 1995; Fineman 2005). This theme is relevant to the issues which I seek to address given that the expectation of women to manage risk relates not only to themselves, but those who may be dependent upon them. More recently, Fineman has begun to link her vulnerability thesis back to her earlier work on motherhood and dependency. In doing so, she argues that formal equality gives an illusion of shared responsibility for dependency, which does not exist in reality (Fineman, 2017).

Describing her conception of the vulnerable subject, she writes:

‘This vulnerability approach both expands upon and complements earlier work I have done in theorising dependency … Reconsidered, the concept of vulnerability can act as a heuristic device, pulling us back to examine hidden assumptions and biases that shaped its original social and cultural meanings. Conceiving of vulnerability in this way renders it valuable in constructing critical perspectives on political and societal institutions, including law.’ (Fineman, 2008, p. 9).

Finally, one of Fineman’s key arguments which I take up in this article is that replacing the autonomous subject with the vulnerable subject helps us to envision a legal subject more aligned to achieving both criminal and social justice.

2.1 Vulnerability: the problem criminal law needs to solve?

In a criminal law context, vulnerability has been perceived as both undesirable and preventable, which is unsurprising given the centrality of the carefully constructed invulnerable, bounded man of legal discourse (Naffine, 2003; Smart, 1990). Criminal law does not exist in a vacuum; it absorbs and reflects cultural meanings. The negative connotations of vulnerability in part stem from a refusal to engage with or accept corporeal vulnerability as an inevitable aspect of the human condition (Diprose, 2013). As Shildrick acknowledges, ‘in western modernity at least, vulnerability is figured as a shortcoming, an impending failure’ (Shildrick, 2022, p. 71). Moreover, the negative connotations of what it means to be ‘vulnerable’ have been exacerbated by the co-option of the term by certain powerful (often male dominated) groups, either in an attempt to secure their own privileged position in society or to try and regulate the undesirable behaviour of others. For instance, Butler cites the attempts of those who oppose racial integration and minority and/or women’s rights on the basis that they render the dominant population ‘vulnerable’ (Butler, 2015, p. 13). These problematic associations are compounded by the frequent co-option of vulnerability by governments to justify a particular moral agenda and/or to regulate the behaviour of a group who are in some way deemed undesirable (Cowan, 2012; Munro, 2017). For example, legal reforms regarding sex work are often justified by emphasising the need to protect ‘vulnerable’ sex workers, while pushing a moral agenda that perversely exacerbates the vulnerability of those they purport to protect (Fitzgerald and Munro, 2012).

Using criminal law in this way maintains the illusion that vulnerability is something which can largely be avoided if risk is managed adequately. In criminalising omissions, ‘failure-to-protect’ provisions go beyond criminalising those who exacerbate vulnerability by causing harm, punishing those who fail to adequately manage this risk by intervening. Over time, these attempts to use criminal law to ameliorate or eradicate vulnerability have inevitably become linked with responsibilisation.

2.2 The feminisation of vulnerability

In contrast to the bounded male body, legal responses to female embodiment have derived from and perpetuated essentialist assumptions of female bodies as volatile, permeable and thus inherently vulnerable (Armstrong, 1999; Sheldon, 1999; Shildrick, 1997). The construction of female vulnerability as pathological pervades criminal law; the vulnerable victim of crime is pathologized and feminised as exemplified by societal and legal responses to gendered crimes such as rape and domestic abuse, which disproportionately focus on prevention, attempting to ameliorate the vulnerability of the victim rather than addressing the violence of the perpetrator (Campbell, 2005; Carline et al., 2017). This is also reflected in the gendering of criminal defences; it is well documented that women who kill often rely on bio-psychiatric defences such as diminished responsibility. Requiring the diagnosis of a recognised medical condition (commonly ‘battered woman syndrome’ as a subcategory of Post-Traumatic Stress Disorder (Carline, 2005; Nicolson, 1995)) this defence mitigates culpability for women’s violence on account of being pathologically vulnerable to the extent that they could not be expected to act rationally. This remains in stark contrast to men who kill, as they can more readily rely on defences such as self-defence and loss of control. Such defences are grounded in a recognition of situational vulnerability, acknowledging that his environment or circumstances rendered him vulnerable thus justifying, or at very least excusing, his violent response (Centre for Women’s Justice, 2021; MacPherson, 2022; Wu, 2019).

Stanko details how this enshrined misogyny derives from and perpetuates what she calls the criminology of the self and of the ‘other’. The ‘criminology of the self’ dictates that, to be recognised as respectable and responsible, women (i.e., the ideal victim) must avoid male violence (Stanko, 1997, p. 490). As women, an awareness of our vulnerability to violence and thus our relational insecurity is critical to the performance of traditional constructs of femininity which, in turn, ensures our ontological security (Hollander, 2011). Lennox notes that ‘through performances of self that embrace defencelessness, docility and reliance on men, women secure men’s recognition of their virtue (thus) maintaining their fragile hold on the material and symbolic resources that accrue to women whose behaviour conform to standards of femininity’ (Lennox, 2022, p. 652). Moreover, the ‘criminology of the other’ maintains that the ‘otherness’ of abusers cannot be recognised (or addressed) and thus women continue to be condemned when our assessment of risk is proved wrong (Stanko, 1997, p. 490). This provides the bedrock of the legal culture of victim blaming which pervades western democracies; women have failed to manage risk appropriately and this failure means that we are deserving of the abuse suffered (Douglas, 1992). As this is subsumed and naturalised into our gender role, this means that as women we are socialised into seeing everyday engagement in public life as inherently risky. As a result, we are socialised to perform safety work including engaging in safety talk to mitigate our own vulnerability, as well as that of those who depend on us (Stanko, 1997, p. 492).

This responsibilisation of women to ameliorate our own vulnerability is exacerbated by gendered constructs of good mothering which situate mothers as responsible for mitigating any risk posed to her and her child (Ward, 2004, p. 369). The ‘deficit model’ of motherhood, whereby if anything undesirable happens to a child their mother is blamed, means that, as enshrined in ‘failure-to-protect’ provisions, women are held accountable for the actions of men (Lapierre, 2008). The focus is on her failure to stop the violence, rather than on his infliction of harm. Neoliberal expectations of so-called ‘intensive mothering’ demand that mothers are always physically and emotionally available to their children (Hayes, 1996). As a result, anything short of omnipresence is portrayed as ‘risky’, depicting the mother as feckless and therefore deserving of moral and criminal sanctions (de Benedictis, 2012; Grewal, 2006). The assumption of maternal omnipresence allows mothers to be held responsible not only for the wellbeing of their children, but also for the behaviour of their partners, again reifying and perpetuating women as risk managers for both their own vulnerability and that of their dependents.

A neoliberal agenda of responsibilisation and individualism means that the parameters of the ideal responsible mother continue to expand (Garrett et al., 2016; Rich, 1976). Imposing safeguarding obligations on women, first for their own safety and later as mothers, exposes a tension in neoliberal patterns of criminalisation. Women are coded as pathologically vulnerable yet are simultaneously responsibilised. This implies agency if agency is understood as the self-management of risk by the autonomous individual (Kelly, 2001). Upon closer analysis however, responsibilisation is revealed as being more concerned with enabling the state to evade responsibility as opposed to conferring autonomy. ³ Enacting ‘failure-to-protect’ legislation was part of what Brown terms New Labour’s ‘re-moralising’ agenda which involved using the criminal law to ‘taper’ the contours of vulnerability to fit a political strategy (Brown, 2012, p. 45). New Labour sought to eradicate the vulnerability of children by ascribing this responsibility back to the family (more specifically mothers) rather than it falling on the state; The offence of ‘causing or allowing a child or vulnerable adult to die or suffer serious harm’, as articulated by the then Home Secretary, aimed to ‘change the culture of sheer selfishness’ by ‘[building] responsibility in the family … rather than allowing responsibility to lie solely with the Government’ (UK Hansard, HC Deb., 14 June 2004, col. 537–8). This responsibilisation of motherhood allows collective social responsibility to be diminished, as individual responsibility (both culturally and criminally) is emphasised (Diquinzio, 1999; Lacey, 1998). Ultimately, responsibilising the individual mother ir-responsibilises the state. Making child safety the sole responsibility of mothers not only exonerates fathers from caregiving, but also allows the state to justify the chronic underfunding of the welfare state, specifically children’s services, social work and domestic violence support (Ridge, 2013, p. 406). This in turn exacerbates the vulnerability of caregivers, whose derivative dependency renders them reliant on (often scarce) resources for resilience (Kittay, 1999).

3 s.5 as a case study into criminal law’s understanding of vulnerability

Due to vulnerability being commonly understood as ‘the problem that we need to solve’, legal interventions usually seek to eradicate the vulnerability of a certain group based on a certain characteristic (Drichel, 2013; Fineman, 2010). Crude constructs often stigmatise those labelled as vulnerable, attracting responses that are paternalistic to the extent of provoking harm to those they purport to help (Diprose 2013, p. 185) As outlined above, s.5 was initially intended to allow for prosecution where a child had been injured and neither parent would admit to causing the harm, nor incriminate the other. However, in criminalising ‘causing or allowing’ harm, it encompasses acts and omissions, allowing for the prosecution of so-called ‘passive abusers’ those who knew, or ought to have known, of a significant risk of serious harm to the victim and failed to stop it. In applying this theory to the ‘failure-to-protect’ context, I argue that when the notion of vulnerability is used in a reactionary way to respond to the plight of a particular group, (in this context children), it is difficult to resist problematic and reactionary interventions that derive from patriarchal constructs of maternity and responsibility. As I have demonstrated elsewhere, this offence is often used to coerce abused mothers to testify against their partners to secure a murder conviction regarding that partner (Singh, 2021).

Focusing on how vulnerability is mobilised and responded to in this context reveals the incongruity of how the state frames vulnerability both as the problem to be solved and its proposed ‘solution’ (Carline, 2012, p. 208). While UK legislature considered this offence necessary to meet international obligations ⁴ to ameliorate the vulnerability of children, it justified the criminalisation of women on the premise that this would ‘support’ women to leave abusive relationships without explaining how they envisioned this being the case (UK Hansard, HC Deb., 14 June 2004, col. 543). Thus, the attempt of the state to frame itself as concerned with the vulnerable (m)other (in this context abused women) is what Carline refers to as the ‘con’ which allows them to justify increased criminalisation. (Carline, 2012, p. 210).

Applying Fineman’s thesis to s.5 and the Parliamentary debates preceding its enactment, it becomes clear that criminal law sought to address the ‘inevitable’ vulnerability of children (Singh, 2017, p. 523). It does this, however, by punishing the ‘derivative’ dependency of their abused mothers, blaming them for failing to manage the risk posed to themselves and their children (Fineman 1995, p. 2200). ‘Failure-to-protect’ provisions are based on the fundamental assumption that the innate vulnerability of children on account of their dependency outweighs the situational vulnerability of their abused mother. To this extent, the vulnerability of children, ‘provides the normative justification for coercing and punishing people for their dangerousness’ (Ramsay, 2012, p. 1).

Used in this reactionary way, vulnerability silences those who are labelled as ‘vulnerable’, depriving them of agency. Ignoring the ways in which intersectional factors including race, class and gender affect the distribution of vulnerability (and conversely, the unequal distribution of privilege which is key to resilience) renders certain groups disproportionately vulnerable to harmful interventions (Butler, 2004, p. 33). For example, in Western societies, black and ethnic-minority children are often framed as particularly vulnerable, meaning that their mothers are disproportionately vulnerable to persecution by both the child-welfare and criminal justice systems (Landertinger, 2015; Roberts, 2022). Moreover, while we are all vulnerable to political decision making, this vulnerability is exacerbated the more an individual must rely on the state for basic necessities such as shelter and access to legal representation.

While enacted to address the vulnerability of children, in its final form s.5 also criminalises those who cause or allow vulnerable adults to die or suffer significant harm. For the purposes of s.5, a vulnerable adult is defined as an adult ‘whose ability to protect himself from violence, abuse or neglect is significantly impaired through physical or mental disability or illness through old age or otherwise’. ⁵ This definition recognises the inability to defend oneself from violence renders a person vulnerable. However, there is an underlying assumption that this is attributable to old age, or alternatively, to physical or mental disability or illness. Constructing vulnerability in this way perpetuates the idea that a person’s resilience and vulnerability are both only dependant on innate characteristics, rather than situational factors. Furthermore, this depicts vulnerability as an absolute state of being – that a person either is, or is not vulnerable, maintaining the idea that vulnerability is a perpetual status only experienced by certain groups.

The inclusion of ‘or otherwise’ in the statute reflects opinions expressed in the House of Lords who felt this ‘future-proofed’ the provision by catering for forms of domestic abuse not yet recognised, allowing judges to use their ‘common sense’ to protect ‘the most vulnerable in society’ (UK Hansard, HL GC, 21 January 2004, col. 334). The circumstances which may otherwise render an adult vulnerable has been subject to considerable attention by the courts, notably in R v. Khan (2009) ⁶ and R v. Uddin (2017) ⁷ . In Khan, the victim was a 19-year-old woman who spoke no English and had travelled to England to marry her cousin, S. Upon their marriage they lived with S’s family including his mother, his two sisters and his brother-in-law. After a significant period of abuse, S fatally attacked V and was consequently convicted of her murder. While it was accepted that none of Khan’s co-defendants saw the victim in the last twelve weeks of her life, it was successfully argued that earlier injuries inflicted, including broken ribs, should have been sufficient for the defendants to foresee a risk of harm to the victim, leading to their conviction under s.5. In what remains a leading judgment in this area, the Court of Appeal considered that the victim’s language barrier and consequent isolation rendered her a ‘vulnerable adult’ for the purposes of the act.

Uddin is factually similar to Khan in that the victim was the defendant’s 19-year-old cousin who suffered sustained and ultimately fatal injuries, predominantly at the hands of her sister-in-law, with numerous extended family members complicit in either actively perpetrating or failing to intervene in the abuse. Uddin sought to appeal his conviction for causing or allowing her death on the ground that the victim was not a ‘vulnerable adult’ on account of her suffering no illness or disability. ⁸ Furthermore, he argued that Khan suggested an adult may only be considered vulnerable if rendered ‘utterly dependent on others’. ⁹ Rejecting his appeal, the Court of Appeal clarified that all that is required is ‘a cause significantly impairing his ability to protect himself from violence, abuse or neglect’. ¹⁰ While this recognition of how women may be rendered situationally vulnerable by domestic abuse is welcome, it is noteworthy that the court does not accept that women are rendered vulnerable solely on account of extrinsic factors, such as domestic abuse. Rather victims must be reduced to a state of vulnerability as a result of intrinsic psychological/emotional damage which impairs their ability to secure protection. ¹¹

Furthermore, the limited recognition of abused women as vulnerable adults in Khan and Uddin highlights the dissonance underlying this offence; abused women may be deemed a victim under s.5, but, due to the aforementioned patriarchal expectations of maternal responsibility, abused mothers are constructed as culpable perpetrators (Singh, 2021). Imposing a duty on abused mothers in legislation which was touted as supporting female victims of domestic abuse creates a vulnerable woman/culpable mother dyad. This is highlighted by contrasting R v. Khan and R v. Uddin with R v. Walker and Henry (unreported, 2018). ¹² In June 2018 Alistair Walker, 27, was convicted of manslaughter of his 3-month-old son Ah’Kiell. Ah’Kiell died of a non-survivable head injury, thought to have been caused by shaking or submersion in water. A pathologist also found evidence of a broken shoulder and four older rib fractures. In addition, the jury heard evidence of Walker’s physical, financial and emotional abuse towards Ah’Kiell’s mother, Hannah Henry, 22. This included evidence that Walker had broken Henry’s jaw and left her lying injured in the middle of a road. Rather than recognising this as indicative of a context of fear and abuse, the prosecution portrayed her as selfish and feckless. Ultimately, this narrative resonated with the jury since Henry was convicted of ‘causing or allowing’ Ah’Kiell’s death and sentenced to three years in prison. Following Khan and Uddin, in different circumstances, had Walker killed or seriously harmed Henry she could herself have been considered a vulnerable adult if it was accepted that the abuse suffered had resulted in psychological or emotional harm rendering her unable to leave him. Yet, because her abuser hurt their child, Henry is no longer considered vulnerable, but a feckless mother who failed to manage both her child’s vulnerability and her partner’s violence and is thus deserving of punishment. In the context of this offence (as well as more broadly) vulnerability translates into both a moral and legal judgement of deserving or underserving (Brown, 2017, p. 498).

The caveat that under 16s are only guilty of this offence if they are the child’s parents reflects how this provision further enshrines and perpetuates the idea that vulnerability is hierarchical; the unique vulnerability of young children is seen as warranting the prosecution of defendants who are themselves children. A child (under 16) is deemed implicitly vulnerable on account of their age alone unless they are a parent. Founded on the belief that vulnerability is transient and attributable to particular characteristics (in this case age), this provision reveals Parliament’s refusal to appreciate the myriad of ways the care expected of parents (particularly mothers) creates and exacerbates the vulnerability of the caregiver (Kittay, 1999). Furthermore, these expectations are even more difficult to negotiate if a parent is under 16 due to maternal ideology largely excluding or disparaging teenage mothers, thus further exacerbating their ontological vulnerability (Hogeveen and Minnaker, 2015; Selman, 2003). Cultural and legal expectations of mothers mean that if harm befalls their child, women find themselves deemed undeserving and subject to moral and legal condemnation. At present, criminal law renders children’s lives exceptionally ‘grievable’; while mothers like Hannah Henry, positioned on the wrong side of the vulnerable victim/culpable perpetrator dyad, remain the ungrievable other (Butler, 2009).

‘Vulnerable adult’ cases are also revealing in exposing the nexus between relationality and vulnerability; s.5 is symptomatic of a general trend of expanding omission liability in English criminal law. Cases such as R v. Stone and Dobinson (1977) ¹³ and R v. Evans (2009) ¹⁴ show that the vertical familial ties which used to denote responsibility (i.e., parent to child but not sibling to sibling), are increasingly blurred by the mobilisation of more diffuse concepts such as voluntary assumptions of responsibility for another and/or contributing to the ‘danger’ to which the victim was exposed (Crosby, 2018). As enacted, s.5 imposes a duty upon anyone in the household who has frequent contact with the victim, thus imposing a broad notion of responsibility seemingly deriving from physical proximity rather than the relationship between parties. Moreover, in Khan and Uddin courts considered the victims’ social isolation as fundamental to why both women were rendered vulnerable. However, as I will outline in Section 4, due to the dominance of the autonomous legal subject, when on trial for this offence there is little recognition of how a defendant’s relationships (and/or social isolation) may have diminished defendants’ resilience and thus their ability to intervene.

4 Replacing the liberal legal subject with the vulnerable criminal subject

In this final section I explore the ramifications of a shift from autonomy to vulnerability in terms of reconceiving the gendered criminal subject. Moving from the abstract legal subject I illustrate the concrete difference this would make in a ‘failure-to-protect’ context. In this section, I first consider how vulnerability promotes a thicker conceptualisation of the legal subject. I then argue that situating the legal subject as vulnerable is critical in destabilising constructs of autonomy which derive from and perpetuate the gendered criminal subject. I outline how a more nuanced mobilisation of vulnerability exposes the extent to which women are coded as pathologically vulnerable victims. This in turn allows for the abrogation of state responsibility towards the safety of women and children, perpetuating a culture of physical, ontological and carceral precarity. I then consider how this theoretical framework is suited to envisioning a more inclusive criminal legal subject, one which is compatible with notions of social justice, rather than deriving from and perpetuating inequality. Destabilising this neoliberal construct of autonomy and focusing instead on relational vulnerability thereby affects our understanding of key cornerstones of omission-based offences. By being both embodied and embedded the relationally vulnerable subject lends itself to a more nuanced framing of understanding of responsibility (in both a moral and legal sense), foreseeability and risk. Moreover, by re-responsibilising the state to build resilience in citizens and institutions, this theory outlines how reform of the criminal subject is critical to any move towards substantive equality.

In criminal law, vulnerability analysis contributes to well documented feminist critiques of the gendered criminal subject. Fineman’s universally vulnerable legal subject necessitates a shift away from traditional (masculinised) constructs of autonomy towards recognition of law itself as a relational mechanism, rather than sovereign and bounded (Naffine, 2019, p. 121). Theorising the potential of Nedelsky’s theory of relationality for reconfiguring the criminal legal subject, Naffine notes two ‘models’ of the criminal legal subject co-exist. ‘Model 1’, the dominant model, describes a legal subject who is ‘snipped out of place and time’ (Naffine, 2021, p. 55). This thin construct of the autonomous invulnerable legal subject allows the context of criminality to be downplayed, since offenders and victims are depicted as individuals deprived of socio-political or historical context. Building upon her previous work on the man of criminal law, model 1 resonates with the specter of the bounded criminal subject, presented as without gender, while, in reality, being male-coded (Naffine, 1997). Naffine is one of several feminist legal theorists, who have long argued that portraying the male experience and perspective as objective renders women the gendered legal subject (Smart, 1990; Wells, 2004). According to Duncan, (drawing on Irigaray) ‘[the woman of legal discourse] has no reflection. She is the substance of which the mirror is made, not the subjectivity it reflects’ (Duncan, 1996, p. 173). This metaphor not only encapsulates the denial of female subjectivity but is pertinent in illustrating how defining women as ‘other’ to man plays into the portrayal of women as fragile. Thus we are coded as pathologically vulnerable as opposed to men who are bounded and inherently resilient. From a feminist perspective the problems with this dominant construct are both deeply entrenched and far reaching. Socio-legal constructions of femininity have never resonated with this concept of the bounded self as evidenced by the numerous ways female bodily autonomy is frequently undermined by criminal law’s regulation of reproductive rights (Goodwin, 2020).

Naffine terms the second, less common, construct of the legal subject the ‘demographic’ or ‘social’ model. This more relational approach to subjectivity rarely comes into view as most crimes are presented as gender neutral acts committed by deviant individuals (as per model 1). Rape however is a notable exception, hence this being the focus of Naffine’s analysis (Naffine, 2019). Unusually, as encoded in English and Welsh criminal law, rape acknowledges gendered patterns in offending (what she terms criminal law’s ‘man problem’) by encoding who the perpetrator of the crime is. Relationality is therefore a counterpoint to traditional notions of autonomy which present the criminal subject as an isolated ‘bad apple’, instead engaging with the legal subject as a social being and thus acknowledging demographic patterns of criminal behaviour (Lacey, 2016, p. 205). A relational analysis reveals that men are only positioned as men i.e. one of two sexes rather than universal, when their crime has been in their capacity as a husband, partner, or father (Naffine, 2019, p. 42). This recognition is important since acknowledging men as men reveals how masculinity is constructed and how femininity shifts alongside it to maintain the illusion of the bounded male subject. However, in ignoring the socio-historical context of crime and the criminal legal subject, the dominance of model 1 allows for the erasure of historical legacies which expose the true extent to which women continue to lack independent identity in criminal law. While the doctrine of coverture and marital rape exemption have been consigned to the past, the relentless policing of pregnancy exposes how women’s identity and legal subjectivity have always been, and continue to be, relational (Goodwin, 2020).

While Naffine has explored the potential of relationality for reconfiguring the criminal legal subject, vulnerability has been under theorised in this context. Although relationality is helpful in critiquing dynamics of power and responsibility, Lacey highlights that it is ‘naïve’ to assume contextualisation is necessarily favourable to women, given that the legal subject is situated in and is shaped by cultural discourses which often allow women to be situated less favourably (Lacey, 1998, p. 201). The use of sexual history evidence in rape trials is just one instance of context being used to responsibilise and ultimately blame women for the violence they have suffered (McGlynn, 2017). Given these limitations of relationality, bringing it into conversation with Fineman’s conceptualisation of vulnerability mitigates some of the gendered problems of contextualisation. Providing a counterpoint to traditional constructs of autonomy, Fineman’s thesis reveals the critical role constructs of vulnerability play in maintaining the traditional criminal subject. Moreover, this highlights how resilience is crucial in re-envisioning a more inclusive legal subject, one who is both embodied and embedded.

In focusing on building collective resilience as opposed to individualising blame, a relationally vulnerable subject starts to unpick the complex dynamic between autonomy and responsibility. As outlined above, the liberal legal subject rests on notions of responsibility and vulnerability which bolster ‘fault models’, allowing the state to evade responsibility for its citizens and institutions. More specifically, s.5 exemplifies how the state designs and interprets laws which responsibilise women for male violence. When drafting this offence the Government rejected amendments designed to exonerate those suffering from abuse, as politicians felt it vital to guard against domestic violence being relied on as ‘an excuse’ (UK Hansard, HC Deb., 14 June 2004, col. 543).

Fineman’s notion of vulnerability as both unavoidable and universal exposes the extent to which safety work keeps vulnerability and femininity entwined, again shoring up the liberal bounded legal subject, leaving little room for female agency. Resonating with Foucault’s theory of governmentality, the criminal legal subject is a ‘technology of gender’, bringing the gendered self into being and shaping subjectivity by shifting explanations or concerns from external agents onto the self (Pyysiainen et al., 2017, p. 216; Smart, 1995, p. 125). In maintaining the fallacy of the autonomous legal subject, the feminisation of vulnerability (and by extension safety work) diminishes women’s resilience in several tangible ways.

First, envisioning a relationally vulnerable subject reveals the extent to which the autonomous subject of criminal law is so imbued in misogyny it derives from and perpetuates a culture of violence against women. At a structural level, Naffine notes that by entrenching men’s dominance ‘criminal doctrines have removed key duties critical to legal personhood, most ostensibly the duty to respect the autonomy of others which [has led to] reduced responsibility to the female ‘other’’ (Naffine, 2019, p. 90). Moreover, citing Nedelsky, Naffine suggests that men find the level of dissociation demanded by the bounded legal subject unattainable, causing distress and ultimately fostering ‘a controlling and possessive nature’ thus creating and perpetuating conditions for abuse (ibid., p. 41).

Second, in denying gendered patterns of offending, the autonomous legal subject allows for historical revisionism regarding both the extent of violence against women and girls (VAWG) and the circumstances in which this occurs. This in turn impacts women’s agency and perceived vulnerability in the public sphere. The refusal to acknowledge endemic violence against women sustains the illusion that women undertake ‘safety talk’ due to the risk of interpersonal injury from strangers in public spaces, obscuring the reality that we are most at risk from partners in our own homes. Moreover, research indicates that women perceive our physical vulnerability to male violence as so unavoidable that we do not engage in risk management behaviours solely as a way of avoiding physical harm. Rather, as noted in Section 2, women recognise adequate performance of risk management as a vital way of ensuring our ontological security, securing access to public space and resources which are only available to those who adequately perform gender (Lennox, 2022, p. 644). A failure to adequately perform risk management renders the undeserving, irresponsible subject vulnerable to further curtailment of their autonomy as they become ‘subjugated to various re-responsibilisation techniques and procedures which include, among other things, sanctions accompanied by schemes for “naming, shaming and blaming”’ (Rose, 2000, p. 323).

Finally, in terms of ‘blaming’, a failure to adequately perform safety work is so pervasive as to leave the female ‘other’ exposed to criminal sanction. As Cobb notes, imposing criminal liability for omissions is unusual in part because of masculine anxieties over infringing the autonomy of the bounded legal subject (Cobb, 2008). However, in practice, ‘failure-to-protect’ provisions further shore up the gendered nature of dependency, allowing the fallacy of the autonomous male self to persist. The feminisation of care and associated risk avoidance expected of the relational unbounded woman means that, in practice, far more women are tried and convicted for these offences (Anon, 2021). The existence of such offences therefore reiterates that women are expected to mitigate their own vulnerability and their children’s vulnerability and that, if they fail to manage this adequately, they will be vulnerable to criminalisation. This highlights the role vulnerability to criminal sanction plays in maintaining the precarious criminal legal subject (Ramsay, 2013). As Lacey acknowledges, the dominant autonomous criminal subject does not result in a blanket decontextualization so much as an insidious oscillation between contextualisation and abstraction which exacerbates the vulnerability of those who come into contact with the criminal justice system (Lacey, 1998, p. 200). Characteristics and context, particularly those which distance perpetrators from the universal male-coded subject are emphasised and elided at different parts of the criminal process. These characteristics remain largely absent from legislation but play a key role in who is charged with which offence. They then slide in and out of focus in the courtroom. While the actus reus of offences is usually gender neutral (for example, s.5 appears to impose liability on anyone in sufficient physical proximity who fails to take reasonable steps to protect a child or vulnerable adult) the mens rea of offences is often highly gendered (Rollinson, 2000). The status of female defendants as women and as mothers increases the level of both criminal and moral responsibility attributed to them, thus impacting the level of foresight and mitigation of risk expected (Anon, 2021). Moreover, as the criminal legal subject reflects male experiences and motivations, the female criminal subject is denied agency in the courtroom. When on trial, abstract tropes of female criminality based on performance of femininity (including safety work) determine whether women are convicted or acquitted (Singh, 2017). Even where women are victims (or witnesses), conformity to those roles determines whether they are deemed credible and/or deserving of retribution or feckless to the extent that they are ultimately blamed for their own victimisation. If convicted, the defendant’s context then re-emerges to some extent at the point of sentencing.

5.1 Moving beyond critique; envisioning the vulnerable subject of criminal law

In focusing on how states and institutions, including law, can increase future resilience, Fineman’s vulnerability thesis helps to move beyond critique, towards envisioning the potential a relational theory of vulnerability may have for a more inclusive legal subject. Replacing the autonomous criminal subject with the relationally vulnerable subject also has implications for shaping the attribution of criminal responsibility more generally. As Stychin acknowledges, (in the context of negligence law).

‘Vulnerability provides a discursive means by which to articulate demands for the legal responsibility of those in positions of privilege who otherwise can claim that they have not assumed responsibility and that they are not responsible. It shifts the focus to responsibilities that arise from ‘encounters with others’ and potentially widens and enriches our understanding of the ethical and legal duties we owe’ (Stychin, 2012, p. 350).

This also demonstrates that proximity gives rise to duties, implying a broader category of persons to be responsible. In some ways s.5 highlights a tension here, but potentially one which can be managed by a more responsive (and responsible) state. As outlined above, ‘failure-to-protect’ provisions derive from and perpetuate, gendered expectations of care and risk management, thus underpinning paternal and state ambivalence towards addressing violence against women and children. While s.5 imposes a duty to rescue on any member of the household who has frequent contact with the victim, Parliamentary debates around the drafting of this provision reveal that this was not borne out of recognition of the ways proximity promotes an ethical responsibility towards the other. Rather the Government viewed this as a way of accounting for the ‘more fluid and flexible … [family arrangements than] in the past’. (UK Hansard, HL GC, 21 January 2004, col 348) This phrasing was intended to capture the transient abusive partners of (implicitly licentious) women who ‘move from one circumstance of tragedy to another’ (UK Hansard, HC Deb., 14 June 2004, col. 546).

By requiring states to take an active role in building the resilience of citizens rather than responsibilising women for their own vulnerability and that of their children this framework shows that re-envisioning the criminal legal subject is vital to attaining social and criminal justice. Primarily, unlike the autonomous criminal subject, the vulnerable subject is ‘embedded’ in social relationships and institutions. This, in turn, exposes both structural inequalities and personal vulnerabilities. Fineman’s call for a more responsive state requires Governments to acknowledge backdrops of inequality, but more importantly, to take active steps to address them, by building resilience in both its citizens and institutions.

On a macro level, this recognition of the criminal subject as embedded shows how institutional forces influence subjecthood. A lack of resilience in institutions, notably the family, welfare and criminal justice system necessitates and perpetuates the thin, gendered legal subject. Failing to acknowledge a backdrop of inequality is key to criminal law’s illusion of legitimacy (Norrie, 1993); an illusion which is increasingly precarious. In contemporary Anglo-American jurisdictions the criminal justice system and associated institutions, including the welfare and healthcare systems are simultaneously enormously powerful and chronically under resourced. Neoliberal cuts to the criminal justice system mean that legal professionals are increasingly responsibilised for the cost effectiveness and outcomes of their work (McEwan, 2011). This impacts the criminal legal subject on an individual level, for example, counsel not being afforded adequate time with clients before trial. More fundamentally, however, this lack of resources influences how fundamental aspects of the criminal legal subject are shaped. As Lacey recognises, a lack of resilience within the criminal justice system leads to a resurgence in character and risk-based approaches to ascribing criminal responsibility, as aspects like capacity require more resources to assess (Lacey, 2016, p. 264). By ascribing responsibility based on the virtues of the offender, character and-risk based notions of responsibility are mired in appropriate performance of gender. Moreover, focusing on the dangerousness of the particular offender (rather than their acts) further maintains the impression that crime is the result of individual depravity rather than structural inequality.

On a micro level, using ‘failure-to-protect’ as a lens, it is clear that the mens rea requirement of foreseeability is tied to the calculation of risk. Failing to take reasonable steps to mitigate foreseeable harm essentially equates to mothers failing to adequately manage risk. However, acknowledging the legal subject as embedded requires an awareness that calculations of risk are dependent on time, space and resources, all of which may be compromised, particularly in a context of abuse. When on trial for ‘causing or allowing a child or vulnerable adult to die or suffer serious harm’ assessment of a defendant’s foresight of risk is a qualified objective test. Juries are asked to appreciate the defendant’s actual calculation of risk, but then to decide if that aligns with a reasonable person’s calculation of risk in that context. Although the objective element means that the reasonable person is not similarly situated to the ideal mother; how can a jury begin to appreciate a defendant’s calculation of risk, and decide that they acted irresponsibly in taking this risk, without knowing the context of the defendant’s actions and their environment? The autonomous criminal subject is so grounded in misogyny that it allows the impact of domestic abuse to be simultaneously ignored and emphasised when assessing a defendant’s foresight of harm. Traditional constructs of autonomy assume self-trust, self-respect and self-esteem, paying no heed to how all of these self-affective attitudes may be diminished by abuse, thus affecting a defendant’s calculation of risk (Anderson, 2014). Moreover, a history of abuse is deployed by prosecutors to convince juries that this harm was foreseeable; the more a mother is abused by her partner the more likely it is that the prosecution contend that she ought to have foreseen the possibility that he would also harm her child (Singh, 2021; Herring, 2007).

Risk management also presumes resources and choice: that the defendant had (and recognised that she had) a choice as to how she managed any perceived risk. The parental pay-gap continues to widen and 80% of this deficit is attributable to the ‘motherhood penalty’, the systematic disadvantages faced by mothers in the workplace (Fineman et al., 2017; Whiting, 2022). Unsurprisingly, economic precarity and a lack of basic resources is a common theme throughout s.5 case law, given that it is estimated that financial abuse is present in up to 99 percent cases of domestic abuse (Hughes 2021; Anon, 2017). ¹⁵ Relentless ‘making do’ results in a lack of resilience which, in turn, can lead to a lack of capacity to address present needs and to plan for the future, resulting in crises which may have been otherwise avoided (Emmel and Hughes, 2010, p. 171). The destitution apparent in s.5 cases (including descriptions of having no money for electricity or food ¹⁶ and pawning jewellery and electronic goods to provide necessities ¹⁷ ) underlines this argument and bolsters Fineman’s call for a more responsive state which would allow children’s dependency to be distributed across numerous well-resourced institutions, as opposed to being contained solely within the sexual family (Fineman, 1995).

In addition to acknowledging subjects as embedded, the vulnerable subject is also embodied. While Fineman’s concept of embodiment is focused on corporeality (citing vulnerability to wounding, drought, hunger etc. (Fineman, 2008, p. 9)), envisaging the relationally vulnerable subject necessitates recognising the embodied experience of feelings and emotions. These are notably absent from both ‘failure-to-protect’ discourse and criminal law more generally as the legal paradigm dictates that, as legal subjects, we are ‘abstracted from our own bodies’ (and the thoughts and feelings which live in and govern said bodies) (Lacey, 1998, p. 181). The dominance of the ‘why didn’t she just leave?’ narrative (as enshrined in these provisions) not only derives from and reifies the responsibilisation of women to manage and/or avoid male violence, but also demonstrates law’s inability to appreciate fear and risk as an embodied experience. As Grear argues ‘[o]ur embodiment means that we are always necessarily corporeally positioned, and that our perception is shaped (or deformed) by embodied limitation’ (Grear, 2013, p. 58). The vulnerable subject necessitates a respect for the embodied self and thus recognises the impact of corporeal threats, such as fear of pain, injury and ultimately, death. At present, the downplaying of maternal fear in ‘failure-to-protect’ discourse ignores the nexus between embodiment, choice and harm (Fletcher et al., 2008). Masochistic expectations of abused mothers mean that no credence is given to how fear of harm to her corporeal self (and her children) affects her decision making.

6 Conclusion

Replacing the autonomous criminal subject with the vulnerable criminal subject may not hold all the answers, but it can usefully shift the debate. Theory alone cannot prevent children or vulnerable adults from dying, however it does have the potential to change the social and legal discourses when such events occur. Attempts to use law to ameliorate the vulnerability of citizens the state identifies as vulnerable maintains the illusion that invulnerability is both desirable and achievable, a fallacy further sustained by dominant constructs of the autonomous criminal subject. Legal interventions or policies should therefore be aimed at building resilience, rather than using punitive criminal provisions to attempt to eradicate the vulnerability of groups based on particular characteristics.

I have argued that elements of Fineman’s vulnerability thesis can be deployed to critique contemporary ‘failure-to-protect’ provisions and discourses around them. The insight gained from applying vulnerability theory in this particular, ‘concrete’ context has then led to a more nuanced understanding of vulnerability and allowed me to grapple with the gendered nature of autonomy that underpins the criminal law more broadly. This critique is relevant far beyond the context of English and Welsh criminal law, as many androcentric western legal jurisdictions are based on the foundational ‘myth’ of the gendered autonomous legal subject (Fineman, 2000; Naffine, 1990).

The illusion of the bounded autonomous criminal subject is only maintained by the feminisation of vulnerability and, flowing from this, the responsibilisation of women as risk managers. This is not to claim that all people socialised as women or men experience the social, corporeal or cerebral dimensions of life in the same ways, but simply to highlight the role hegemonic gender (and as part of this, vulnerability) plays in shoring up neoliberal institutions. Replacing the autonomous legal subject with the relationally vulnerable subject reveals the extent to which female identity constantly shifts to sustain the neo-masculine autonomous subject, a critical feature of an increasingly precarious criminal justice system. Moreover, the fallacy of the bounded autonomous subject is so grounded in misogyny that this diminishes female agency and renders women vulnerable in many real ways; ontologically, physically and criminally. By engaging with the socio-historical context of crime the relationally vulnerable subject necessitates acknowledgment of patterns in both criminal behaviour but also more broadly in terms of structural inequalities. Unpicking the complex nexus between autonomy and vulnerability has implications extending far beyond criminal justice responses to women and children. Parliamentary debates preceding s.5 reveal that, in the House of Commons, vulnerability was portrayed as attributable to old age or physical or mental illness (UK Hansard, HC Deb., 14 June 2004, col. 559). This exposes the ways in which dominant constructs of vulnerability continue to undermine the agency of all those positioned as ‘other’ to the illusory autonomous subject, in this context the elderly or those with disabilities. Engaging with embedded and embodied realities (rather than precarious patriarchal fantasies) makes steps towards a much-needed re-envisioning of the criminal legal subject in a way which is more compatible with aspirations of social and criminal justice.