A New Self-defence Framework for Domestic Abuse Survivors Who Use Violent Resistance in Response

This article criticises the government’s rejection of proposals by the Prison Reform Trust that would have extended self-defence in householder cases to victims/survivors of domestic abuse. The authors argue that the Prison Reform Trust proposals should be enacted, and further supported by novel complementary reform of the option to retreat,and the exclusion of intoxicated mistaken belief in self-defence claims. Specifically, the authors advance a statutory rebuttable presumption regarding the option to retreat in cases involving domestic abuse, namely, an assumption that the victim/survivor was not realistically able to retreat safely, unless it is proven otherwise. The authors also examine the appropriateness of the policy decision to exclude in-toxicated mistaken belief in all self-defence cases and advocate for its removal. It should be replaced with a requirement that all mistaken beliefs must be reasonable regardless of the presence of intoxication. Procedural recommendations are also advanced, including amendment of the Crown Court Compendium to include judicial directions on self-defence which adopt a social entrapment approach in domestic abuse cases,and supported by the admissibility of non-medical expert evidence on the nature and impact of coercive control.


INTRODUCTION
The Domestic Abuse Act 2021 (the 2021 Act) was hailed by commentators as a landmark opportunity to improve responses designed to prevent, protect from, and punish perpetrators of domestic abuse in England and Wales.It introduced an expansive legal definition of domestic abuse, recognising coercive and controlling behaviours and encompassing forms of emotional, financial and psychological injury. 2 It also set out a range of additional measures to, inter alia, increase enforcement powers in the event of breached protective orders,3 provide greater protection from homelessness to victims seeking to exit abusive relationships, 4 and create a clearer framework in relation to the operation of domestic abuse disclosure schemes. 5Despite the opportunity that this legislation afforded to create more substantial transformation in relation to understandings of, and responses to, domestic abuse, the Act was notably silent in respect of recognising the influence that abuse experiences may have on victims'/survivors' offending, and how this should be addressed.
This reflects a missed opportunity to address the widely acknowledged inability of existing criminal law defences to recognise the circumstances of abuse victims/survivors who are driven to offend. 6That abuse drives female offending in heterosexual abusive relationships was highlighted by the Prison Reform Trust (PRT) during the Act's legislative passage. 7Domestic abuse can drive some women to respond with violent resistance against their abuser; and yet, the Centre for Women's Justice (CWJ) Double Standard report confirmed that, despite decades of campaigning on the issue, victims/survivors are rarely acquitted using self-defence. 8Increasing access to such defences would ensure justice is served and be an important step to meeting a key priority in the UK government's 2018 Female Offender Strategy, namely to reduce the number of women in prison. 9ur article focuses on the government's rejection of proposals by the PRT that would have extended self-defence in householder cases to victims/survivors of domestic abuse.In standard self-defence cases the force used must be necessary and proportionate in the circumstances as the defendant believed them to be. 10 In householder cases, a discretionary area of judgment is applied which permits the defence where the force is necessary and not grossly disproportionate in the circumstances as the defendant believed them to be. 11hilst the literature is extensive on the issue of defences for women who kill, 12 we are equally concerned with self-defence claims where the violent resistance has non-fatal consequences.
We suggest that the PRT proposals on extending self-defence as it applies to the householder to victims/survivors of domestic abuse should be enacted.The proposals also provide an opportunity to consider the broader legal principles and procedures impacting upon decisions in this context.In terms of substantive law at trial stage, we recommend that the relevance of any apparent option to retreat in self-defence claims should be statutorily clarified to reflect the challenges that retreating presents for abuse victims/survivors.We suggest that there should be a rebuttable presumption regarding the option to retreat in cases involving domestic abuse, where it is assumed the victim/survivor did not have a realistic opportunity to retreat unless proven otherwise.In addition, the policy decision to exclude intoxicated mistaken belief in all self-defence cases ought to be revisited.For householders, their alcohol consumption is legal and socially acceptable, and they are not anticipating a hostile interaction with an intruder. 13In the context of domestic abuse, a correlation has been found between the abuse and drinking to cope, and the intoxicated mistake exclusion denies defendants in these circumstances access to self-defence. 14We examine the appropriateness of this exclusion and contend that it should be removed, requiring all mistaken beliefs to be reasonable regardless of the presence of intoxication.
We also advance a complementary procedural recommendation for the Crown Court Compendium to include judicial directions on self-defence which support a social entrapment approach in domestic abuse cases. 15This change ought to be supported by the admissibility of non-medical expert evidence on the nature and impact of coercive control.

DEFINING DOMESTIC ABUSE AND UNDERSTANDING VIOLENT RESISTANCE AS A RESPONSE
The 2021 Act moved legal boundaries in terms of defining domestic abuse, but these changes did not extend to the operation of defences in the context of victims/survivors who use violent resistance against their abuser.Underlying the PRT proposals is the view that this wider legal understanding of domestic abuse should inform how defences are applied.The government rejected the proposals, concluding that existing defence doctrine responds appropriately to such cases. 16This is a position we challenge throughout this article.
is a cage": a new partial defence for primary victims who kill' (2015) 66 Northern Ireland Legal Quarterly 149.13 R (Collins) v Secretary of State for Justice [2016] EWHC 33 (Admin) (Collins) at [30].14 Gail Gilchrist and others, 'The association between intimate partner violence, alcohol and depression in family practice' (2010) 11 BMC Family Practice 1. 15 Crown Court Compendium Part 1 (2022) at https://www.judiciary.uk/guidance-andresources/crown-court-compendium/[https://perma.cc/6846-YTW2](Compendium).16 HC PBC Deb (12 th sitting) cols 439-440 17 June 2020 (Victoria Atkins).Domestic abuse is now legally defined in sections 1-3 of the 2021 Act.Section 1 17 defines abusive behaviour as including physical or sexual abuse; violent or threatening behaviour; economic abuse (namely behaviour which has a substantial adverse effect on the victim's ability to acquire,use or maintain money or other property); 18 and psychological, emotional, or other abuse. 19Many of these types of abuse will form a pattern of coercively controlling behaviour. 20Coercive and controlling behaviour sits at the heart of the new statutory definition.Stark explains that coercive control is an ongoing pattern of behaviour consisting of both physical and non-physical tactics. 21The behaviour is specifically designed to make the victim subordinate through intimidation, isolation, taking away means of independence that prevent escape and the micro-regulation of everyday behaviour. 22Failure to comply with the abuser's demands results in punishment which can include physical or non-physical attacks.The abuser knows the form of punishment that will be effective as they are informed by discoveries made during the initial stages of the relationship. 23Victims/survivors become fearful of their safety at all times, and depending upon the nature of threats made, fearful for the safety of others in their family or friends' network. 24hey experience a sustained attack by perpetrators on their self-confidence and are often isolated from family or community support.Consequently, the victim's autonomy and independence are eroded, and they become trapped within a world of the perpetrator's making. 25s a result of this social isolation, the victim's/survivor's perception of their options for escape and support become increasingly limited, and the danger she is in is both continual and heightened. 26An outsider to the relationship may believe that the victim's/survivor's best coping strategy is to leave the relationship.However, leaving can place the victim/survivor in further danger. 27In fact, women use a diverse range of protective strategies against their abuser shaped by the abuser's behaviour, personal values, and financial, institutional, and social obstacles. 28Use of protective strategies is often problematic.For example, compliance furthers the isolation and entrapment.Calling frontline services can lead to retaliation by the abuser, child protection issues and immigration enforcement.Restraining orders can be breached and provide no guarantee of future safety.Isolation, humiliation, and subordination impact upon all facets of a person's life and non-physical threats, such as those directed at losing custody of children 'may in some cases be greater than the threat of physical injury.' 29 Using violent resistance as a strategy is high-risk, as it can lead to a greater risk of more severe future violence from the abuser. 30ichter and others discovered a link between experiencing coercive control and women's use of violence in that context.They found that women who experienced coercive control were more likely to use physical violence towards their abuser than those who had not and concluded that experiencing coercive control increases the victim's use of physical violence as a survival strategy. 31One may determine from these findings that from the perspective of the victim/survivor, violent resistance becomes the only option to escape the entrapment created by the coercive and controlling behaviour of the abuser and must be done with sufficient force to prevent further harm to the victim.The 'chronic, cyclical, and often inescapable' 32 abuse is, therefore, crucial in understanding women's use of violent resistance in abusive relationships and needs to inform assessments of self-defence claims, given that determinations are assessed on the basis of whether reasonable force was used.This can be achieved using a social entrapment approach to coercive control and self-defence claims by victims/survivors.

Social entrapment: contextualising coercive control
Scholars now argue that the concept of coercive control must be understood through a social entrapment lens. 33Coercive control articulates the nature of the abusive behaviour and its impact on the victim, but it is not able to make visible the limitations of institutional safety responses and structural inequities that abused women routinely experience and how that can shape the coercive control used by the abuser.We agree with Douglas and others and Tolmie and others that a social entrapment approach is necessary to support self-defence claims by women, or others, who use violent resistance against an abuser.To achieve this understanding,three dimensions need to be considered by the court to establish the level of the threat from the perspective of the defendant and the reasonableness of the force used in response to it. 34The first dimension examines the tactics of coercive control employed and how the coercive control reduced the victim's space for action. 35Secondly,the frontline service response is assessed in terms of what it was or could realistically have been.This is an essential stage as responses influence the victim's decision whether to seek assistance or not.Finally, intersecting inequalities in the victim's/survivor's life circumstances, such as cultural gender norms, disability or institutionalised racism, are considered as they affect which coercive and controlling behaviours are adopted by the abuser, how victims/survivors respond to the behaviour and how they engage or otherwise with the frontline services which become involved. 36For example, mistrust in the authorities may extend beyond personal experience to common views regarding how victims/survivors may be treated or discriminated against. 37More broadly, the institutional response impacts upon the perpetrator's capacity to continue the coercive control,as they become emboldened by a failed attempt to reach out for support and could also use the response to further control the victim, suggesting that she will not be believed or supported in the future.

THE PRISON REFORM TRUST (PRT) PROPOSALS
The PRT proposals, considered during the passage of the Domestic Abuse Bill, would have enabled a greater appreciation of the domestic abuse victim's/survivor's circumstances by extending the more generous householder defence to such cases.At present, domestic abuse victims/survivors who use disproportionate force to defend themselves against an abuser (inside or outside the home) will be found guilty of the offence, since disproportionate force is automatically unreasonable in standard self-defence.In contrast, disproportionate force may still be regarded reasonable in householder cases when considering all the circumstances including that the defendant has come upon an intruder. 38he PRT proposals suggested the phrase 'or a domestic abuse case' should be inserted into section 76(5A) of the Criminal Justice and Immigration Act 2008 to read: 'in a householder case or a domestic abuse case, the degree of force used by D is not to be regarded as having been reasonable in the circumstances as D believed them to be if it [the force] was grossly disproportionate in those circumstances.'The provision would apply where 'D is, or has been, a victim of domestic abuse' as defined in sections 1 and 2 of the 2021 Act, and where the force is directed against the abuser. 39There would be no requirement to demonstrate the defendant (and victim/survivor of domestic abuse) believed the perpetrator/victim was a trespasser at the time of the incident.
The nuanced approach advocated by the PRT would have brought the reasonableness test in domestic abuse cases in line with the householder defence allowing a much-needed discretionary area of judgment.It is worth stating at this point, however, that those advocating for this reform understood its limitations.Howes and others argue that, notwithstanding suggestions that the householder test has made 'negligible difference to self-defence', extension of this defence to victims/survivors who use violent resistance against an abuser, 'would be of normative significance, as the current householder' defence 'applies in practice mostly to men and is arguably discriminatory.' 40

The Prison Reform Trust proposals: limitations
This section considers some of the potential limitations and the discriminatory approach parliament has adopted with respect to the PRT proposals.In terms of limitations, the PRT proposals did not address the wider context of selfdefence.For example, though there is no duty to retreat in self-defence cases, retreating remains a factor for consideration.Retaining this element of selfdefence provides a double-edged sword for the victim/survivor where previous examples of retreating from the abuser may be used to suggest that retreating was an option notwithstanding escalating abuse.Though the issue was raised by Lord Paddick during parliamentary debate, 41 no proposal to amend current substantive law in respect of retreating was advanced.
Similarly, the proposals did not address the exclusion of an intoxicated mistaken belief in self-defence, 42 notwithstanding the correlation between victims/survivors experiencing domestic abuse and substance misuse to cope. 43t could also prove more difficult to establish that the defendant is an abuse victim/survivor than establishing that the defendant believed the victim was a trespasser, as required in householder cases, 44 given that many victims/survivors do not disclose the abuse suffered to third parties. 45It is the prosecution that must have sufficient evidence to rebut self-defence, but the less evidence there is of domestic abuse, the more likely the prosecution will pursue the offence charged, and self-defence will be rejected. 46Victims/survivors are often reluctant to contact authorities or inform family and friends about their situation.Fear of authorities or previous poor experience of authority involvement may result in non-disclosure by victims/survivors of their experiences, which subsequently affects credibility at trial. 47urther, the victim's/survivor's account of the abuse provides an evidential gateway for adduction of their bad character, which would only be discretionarily excluded by the trial judge where it is requested by the defence, and this would be dependent upon their understanding of the nature of coercive control. 48As Tolmie and others explain, coercive control requires a social entrapment lens which can make visible the realities of all women's experiences of it by emphasising the importance of both the abuse by the perpetrator and the institutional inequities involved. 49We advance recommendations to address these limitations, but before doing so we provide a deeper analysis of the current position, addressing Howes and other's claim that refusal to extend the householder defence to victims of domestic abuse is discriminatory.

Parliamentary debates
There was some exposition of the government's decision to introduce the householder defence prior to the enactment of the Crime and Courts Act 2013, 50 and equally the government proffered some, albeit weak, arguments for rejecting the PRT proposals during the passage of the 2021 Act. 51The government enacted the householder defence based on the argument that householders are in a unique position due to being confronted by an intruder in their own home. 52Consequently, disproportionate force may be reasonable in a householder case when considering the circumstances of coming upon an intruder. 53Thus, this householder defence introduced a greater latitude in respect of an interference to the right to life. 54The effect is a more nuanced approach to a specific category of defendant, recognising that circumstances do have a bearing on a person's criminal responsibility, and thereby permitting greater access to defences.In theory, extending the householder defence to victims/survivors of domestic abuse represents recognition of the unique circumstances of domestic abuse victims/survivors, which according to the PRT should be enacted given the latitude already afforded to householders.Why should householders have greater protection when they act to protect themselves against an intruder than a domestic abuse victim/survivor acting against the perpetrator she lives with? 55or the government, the distinction was necessary as a householder, whose 'home is his castle' would not know who the trespasser was should they take a weapon from a drawer and use it fatally, compared with the 'victim of domestic violence taking a weapon in her hand.' 56This suggests that in the government's view the key difference between householder and domestic abuse cases is that the householder is acting 'on instinct' in response to a stranger whereas in 'domestic abuse cases, the response may not be a sudden instinctual one but may follow years of physical and/or emotional abuse.' 57 The government's basis for providing different thresholds of reasonableness and proportionality to householders compared with women responding to domestic abuse, and other defendants is a symptom of what Naffine describes as the man problem in criminal law. 58The criminal law was created and interpreted by powerful men protecting their own interests and denying legal status to married women who were femme covert and assimilated into the legal status of their husbands.This masculinity of the criminal law was disguised by the semblance of objectivity, with the objects of criminal law described as 'persons' or 'individuals' rather than men.Naffine explains that in doing this the criminal law 'sustains the practice of keeping men under cover and therefore not the subjects of open consideration.' 59 With cultural and social change, women gained legal status, but the inherent masculinity of the law was not openly theorised or addressed by the men of law. 60Instead, as Conaghan suggests, 'the judicial tendency is to approach instances of gender bias' and past inequalities 61 'in law as remnants of a patriarchal legal past which law is gradually casting off.' 62 Effectively women are deemed to be equal to men. 63Legal privileges once denied to women and afforded to the husband, became applicable to all individuals, but that individual 'was permitted to emerge … with barely a nod to the unseemly and illiberal past of criminal law, to its gendered history.' 64 This process of 'deeming' and the abstraction of a 'person' in criminal law relies on minimal characteristics, but includes self-government, typically and historically associated with 'man.' Therefore, 'it is very difficult to pluck gender out of the abstraction of the person or to neutralise its effects.' 65 Indeed,the 'modern shape and form' of the law was not created 'under conditions of gender-neutrality but in the context of a legally sanctioned gender hierarchy.' 66The householder in self-defence is a useful example of this in practice as the 'individual' cum householder is not a neutral individual, but perceived as male entitled to protect 'his' castle.This compares with the domestic abuse victim perceived, not unjustifiably given that they are more likely to be female, as woman. 67The government's explanation for permitting a wider threshold to the householder who is acting on instinct compared with the domestic abuse victim responding to years of abusive behaviours is based on gendered beliefs and expectations about the role of men and women and how each should react and consequently determines what degree of force should be used to protect themselves. 68The PRT proposals are therefore a challenge to the masculinism inherent in the substantive law of self-defence. 69The proposals seek to make gender specificity matter in the context of self-defence, as the force used against an abuser of domestic abuse requires considerations of gendered factors.These include the general disparity between the physical strength of a man compared with a woman and a woman's trained incapacity for self-defence, which explains a woman's instinctive use of weapons to protect herself, and a woman's knowledge of the man's capacity to use violence. 70he discriminatory nature of self-defence towards women defendants was central to the US case of S v Wanrow 71 which involved a five feet four inch woman with a broken leg who used a gun against an intoxicated six feet two inch unarmed intruder who she believed had molested her friend's daughter and was returning for another child.The Washington Supreme Court found that the trial court erred in its direction to the jury by applying an objective approach that limited the acts and circumstances pertinent to the defendant's perception of the threat to her.Utter, J held that the persistent use of the masculine gender in the trial court direction implied that the reasonableness standard to be applied was that of an altercation between two men.The majority considered that aspects of gender were essential factors in determining the extent of the threat perceived by the defendant, including perceptions which were the product of our nation's 'long and unfortunate history of sex discrimination.'72 Accordingly, the court concluded that self-defence directions must 'afford women the right to have their conduct judged in light of the individual physical handicaps which are the product of sex discrimination' and a failure to do so amounted to a lack of equal protection of the law between men and women.73 Whilst this was considered to be a significant decision at the time, Bennett suggests its influence was limited. 74Put simply, if standard self-defence is ill-equipped to address the circumstances of the householder who when 'acting on instinct' is more likely to use a weapon, it follows that standard self-defence is ill-equipped to address the relative size disparity and power imbalance between a perpetrator and abuse victim/survivor.75 As articulated during parliamentary debate, we should not and 'cannot accept, that there can be a householder defence -the Englishman's home is his castle', which some called for -but not an equivalent defence in the extreme cases dealt with by' the PRT proposals.76 The government also indicated its concern that domestic abuse, as a broad term, might be interpreted to enable a victim/survivor defendant to avoid liability where there had been 'any level of abuse' 77 or to afford recognition (that could then be relied upon to ground a defence) to too broad a range of victims/survivors.Those party to a domestic abuse relationship must be personally connected either as former or current intimate partners or relatives.78 This type of 'flood-gate' argument is not uncommon in the terrain of defences where there is an imperative to maintain order, ensure a robust approach to assessing criminal responsibility, and avoid a mandate for casual law-breaking.However, as understandings of the complexity of domestic abuse evolve, with legal recognition of the types of harm it can entail, it is imperative that the appreciation is also afforded to the experiences of the victim/survivor in respect of defence claims.

WHAT WE KNOW ABOUT WOMEN'S CLAIMS OF SELF-DEFENCE
A victim's/survivor's criminal responsibility for using violent resistance cannot be meaningfully ascertained until their social circumstances are properly considered, which we argue requires a social entrapment approach (discussed above).We explore this through a focus on female offending in heterosexual relationships, given the power imbalance and use of a weapon which often arises in this context.However, many of the concerns raised apply equally in homosexual/transexual relationships, and where the victim/perpetrator dynamic is reversed in heterosexual contexts. 79n individual is permitted to protect themselves and is justified in doing so according to the law of self-defence. 80Whilst the defence is in theory gender neutral, emerging research indicates that women's access to self-defence claims are very narrow.McPherson found that female defendants in Scotland were rarely successful in raising self-defence.There were no successful claims in homicide cases and the defence was infrequently used in cases of a woman killing her abuser. 81Howes and others found that women are seldom acquitted using self-defence in England and Wales. 82In contrast, 'women are … advised against raising contextual domestic abuse in mitigation or … defence, because it is seen as making an excuse.' 83 Gender stereotypes make the use of violent resistance by abused women challenging.Assumptions may be made that her use of violence indicates that she was mutually violent or controlling, rather than understanding her use of violence as a protective strategy. 84ur knowledge about domestic abuse victims'/survivors' access to self-defence is hindered by the unavailability of data regarding self-defence claims at trial, thereby making the extent of the problem invisible. 85A statutory reporting requirement is needed to enable the gathering of information that will lead to insights into the operation of self-defence for women who use violent resistance against their abuser.Admittedly, there are significant challenges with a reporting requirement as defence disclosure may not always include detail on the defences to be run, and where it does the information is likely incomplete.In some cases, the decision to run a particular defence may be made by the trial judge towards the end of the trial.Where more than one defence is run, in many cases it will not be possible to ascertain the basis for the jury decision.These problems, however, do not detract from the need for more informed data regarding the operation of the defences.Rather they highlight the need to carefully plan such a requirement, and to recognise the limitations on data collected.

APPLYING A DOMESTIC ABUSE LENS TO THE CURRENT HOUSEHOLDER DEFENCE
As the parliamentary debates during the Domestic Abuse Bill illustrated, the law does not provide an equivalent householder defence to victims of abuse who act to defend themselves and the government argued that the existing defence framework is sufficient.Interestingly, in an extremely limited number of abuse perpetrator-cum-trespasser cases, domestic abuse victims/survivors may claim the householder defence, thus clouding the issue as to what gives rise to the additional latitude for force and potentially undermining the government's refusal to extend the provision further.In Ray 86 for example, the householder defence was raised in the abuser-cum-trespasser context when Ray stabbed his partner's (Allen's) abusive ex (Hemmings) during a fight after Hemmings returned to Allen's home as a trespasser.
Allen had ended a relationship with her abusive partner, Hemmings, and had begun a relationship with Ray.Hemmings allegedly hated the relationship leading him on one occasion to threaten to 'smash both their faces in.' 87 After sending abusive texts to Allen in the evening, he went to Allen's house, awaking her by banging loudly on the door.When she opened the door, Hemmings burst in shouting and swearing.Ray attempted to defuse the situation, and asked Hemmings to leave, which led to a fight.Fearing for his safety, and believing Hemmings to be armed, Ray picked a knife up from the drainer and fatally stabbed Hemmings.Ray was convicted of murder and sentenced to life imprisonment. 88n appeal Ray argued that the trial judge had wrongly relied on the interpretation of the householder defence in Collins 89 which had erroneously placed householders in the same position as non-householders acting in self-defence.The Court of Appeal in Collins had endorsed Blackstone's succinct summary: 'The new provision merely affects the interpretation of "(un)reasonable in the circumstances" so that force is not by law automatically unreasonable in householder cases simply because it is disproportionate, provided it is not grossly disproportionate.' 90 By contrast, according to Ray's counsel, the householder defence should be interpreted such that where the jury are satisfied that force was not grossly disproportionate, it would automatically be reasonable. 91The Court of Appeal rejected Ray's assertion, reaffirming Collins, and confirming that the householder defence had merely, 'slightly refined the common law in that a degree of force used that is disproportionate may nevertheless be reasonable.' 92hough the defendant in this case was the victim's/survivor's new partner, the jurisprudence highlights the possibility that the lower threshold test would apply to the victim/survivor defending themselves against an abuser under current law in extremely limited cases, but only if she knew or honestly believed the abuser to be a trespasser.The question is not whether the victim 86 Ray n 38 above.87 ibid at [3]. 88 ibid at [2]-[6] and [8]- [10].89 ibid at [24].90 ibid at [19] citing Collins n 13 above at [34].91 ibid at [24].92 ibid at [27].
was a trespasser or became a trespasser in law, but rather whether the defendant honestly believed they were a trespasser. 93iven the dearth of data on the issue, it is unclear whether any cases involving domestic abuse victims/survivors who defend themselves against an abuser-cum-trespasser have had the householder defence considered over standard self-defence.Two separate and experienced defence counsel teams admittedly overlooked the potential that the householder defence might apply in R v Magson 94 despite a neighbour's account that they overheard Magson saying that she did not want her ex-partner to enter the house because he had been imbibing cocaine. 95We simply do not know how often such cases have arisen in practice and whether the potential that they could be considered householder cases is being overlooked.Yet, the difference is important given that under standard self-defence disproportionate force is automatically unreasonable whereas in householder cases disproportionate force may be reasonable in the circumstances.
Significantly, in Ray, Lord Thomas of Cwmgiedd, CJ outlined the 'context that differentiates the householder case' from standard self-defence, namely, the particular 'dilemma that would confront any householder when an intruder enters his or her house.' 96Whilst we recognise that it is generally unhelpful to draw parallels across different forms of victimisation, we have already illustrated the government's artificial and discriminatory distinction across householders and domestic abuse survivors.Accordingly, we suggest that Lord Thomas' ruling provides an opportunity to consider afresh the 'validity' of the distinction between householders and all other defendants acting in self-defence in greater depth, and to further assess the importance of applying a social entrapment lens to domestic abuse cases.
Below we outline Lord Thomas' guidance on directing jurors in relation to the reasonableness assessment in householder self-defence claims and advance a comparison with each element as it might apply to the victim/survivor of domestic abuse who acts in self-defence, adopting a social entrapment approach.Our analysis demonstrates that the distinction across householder and domestic abuse cases is unsustainable, and rests on a lack of understanding of all the circumstances.We aim to show the 'context that differentiates the domestic abuse case', namely, the particular 'daily dilemma that would confront any [domestic abuse victim] when [the abuser] enters his or her house.' 97ord Thomas, in Ray, provided a list of factors designed to assist jurors in assessing whether a householder's response, which may be disproportionate, but not grossly so, was reasonable in the circumstances.According to Lord Thomas, jurors may be directed to consider, 'the shock of coming upon an intruder,' and 'the time of day.' 98 We submit that for domestic abuse victims/survivors the threat of attack is unabating and persists, compared with a person shocked for a moment by an unknown intruder in their home.Lord Thomas noted 'the presence of other help' 99 may be a relevant factor in assessing reasonableness of force used by a householder.We have similarly highlighted that accessing support from authorities is difficult for many victims/survivors of domestic abuse.Adopting a social entrapment lens, institutional responses to the abuse and the perceived responses by the defendant given individual and community experiences of negative interactions with authorities must be considered.In addition, how these responses, perceived or real, exacerbate the coercive control employed by the perpetrator must be understood to make the challenges of gaining assistance visible.
In householder cases, Lord Thomas highlighted the 'desire to protect the home and its occupants,' including 'the vulnerability of the occupants, particularly children' as relevant to the reasonableness assessment.Both factors are equally pertinent to domestic abuse cases, and there is an argument that an abused parent could be supported by the criminal law to react with force against the perpetrator-parent, as doing nothing runs the risk of prosecution under section 5 of the Domestic Violence, Crime and Victims Act 2004.
In addressing use of a weapon in householder cases, Lord Thomas reiterated the government position that 'picking up of an object (such as a knife or stick that would lawfully be to hand in the home)' distinguishes householder and standard self-defence cases. 100However, a key problem for the victim/survivor seeking to persuade the jury that their use of violence against the abuser was 'reasonable' and 'proportionate' arises where a weapon is used, and she is more likely to use a weapon because she knows through experience that the abuser is physically stronger. 101Here the reasonableness limb of self-defence both in its standard and householder form reflects Naffine's argument that the criminal law is constructed around a male 'person' and blind to the female individual's position.As an equal to a man, she should be able to protect herself from bodily violation. 102As Edwards argues, however, the use of body force against attacks in inter-male cases has been more readily deemed reasonable in practice, and excused when used against a female partner, compared with a female's use of a weapon as a response to an abusive partner's attack. 103Those who can afford to defend themselves instinctively and with their bare hands are most often men who have the physical power to do so. 104As Howes and others conclude, 'unless there is a good understanding of the dynamics of violence against women, the use of a weapon is likely to be interpreted as disproportionate by a jury, even 99 ibid.100 ibid at [37].101 Edwards, n 6 above.102 Naffine, n 58 above.103 Edwards, n 12 above, 267.104 Lord Paddick explained: 'I have seen misogyny described as the hatred of women who fail to accept the subordinate role ascribed to them by a patriarchal society, who fail to conform to the misogynist's belief that women should be no more than compliant and decorative, whose role is to serve the needs of men.Out of such a false and outdated narrative comes the idea that physically stronger men should stand and fight while physically weaker women should run away.I am very sad to say that this appeared to be the Government's position [in relation to the "self-defence"', HL Deb vol 809 col 1750 10 March 2021.
in circumstances where a woman has just been subject to an attack where she feared for her life.' 105n householder cases, Lord Thomas said jurors might be directed to consider 'the conduct of the intruder at the time (or on any relevant previous occasion if known to the defendant)' 106 a point that is equally relevant to victims/survivors who know the abuser and have bitter experience of his conduct.Significantly, the court in Ray explained that '[e]ach of these [factors] might lead to the view that what was done, such as using a knife, which otherwise in a different context might be unreasonable, in the circumstances of a householder coming on an intruder might, in all the circumstances of such a case, be reasonable.' 107These factorisations highlight the extent to which the latitude granted to householders is at best weak and at worst discriminatory in comparison with victims/survivors of domestic abuse.It is essential to consider gender, 'age, race, religion and nationality to address intersectional discrimination.' 108The weak distinction across householder and domestic abuse cases is unsustainable.Adoption of the PRT proposals would have assisted in 'mollify[ing] the criminal law's privileging of male experience and addressed the exclusion and invisibility of coerced and abused women.' 109 As substantive law's application is reliant on judges and jurors, these beliefs as to male and female roles and reactions will not ensure a woman's violent resistance to an abuser would be deemed reasonable, proportionate or less than grossly disproportionate. 110Therefore, substantive legal reform should be accompanied by procedural change that embeds a social entrapment approach to domestic abuse self-defence cases.
Before exploring potential procedural changes to the Crown Court Compendium and rules governing the use of non-medical experts at trial, there are two further matters of substantive law in relation to self-defence that require analysis -the duty, or otherwise, to retreat and intoxicated mistaken beliefs.

THE RELEVANCE OF RETREATING AND THE CHALLENGE IT POSES TO VICTIMS/SURVIVORS OF DOMESTIC ABUSE
There may be options other than violence available to the defendant in response to an imminent attack, although the context may mean these options are unrealistic and potentially unsafe.The law recognises this situation stating,'A possibility that D could have retreated is to be considered (so far as relevant) as a factor to be taken into account, rather than as giving rise to a duty to retreat.' 111 Therefore, there is no specific duty to retreat, and the court identified that the option to retreat is unlikely to arise in many householder cases.As such, force which might in other cases appear disproportionate may be deemed reasonable in the circumstances. 112Retreating may be unlikely in abuse cases, 113 where the cumulative impact of abuse and the futility of former (if any) retreats signify to the victim/survivor that retreating is not a viable option, or alternatively that it exacerbates the abuse. 114Leaving is not easy when considering broader concerns regarding support, protecting children, and finding an alternative place to live, but more specifically women are most at risk when they attempt to leave, at which point the abuse may escalate. 115These problems are compounded by 'issues relating to disclosure, women providing false or inconsistent accounts of what has happened, violence on "both sides", the use of weapons, and the existence of harmful myths and stereotypes.Collectively these common features serve to undermine women's accounts of abuse [and any ostensible capacity to retreat], creating a barrier to self-defence claims.' 116e suggest that there should be a statutory rebuttable presumption regarding the option to retreat in cases involving domestic abuse, namely, an assumption that the victim/survivor was unable to realistically retreat safely.In practical terms, this rebuttable presumption could apply to the householder in the sense that in many cases retreating will be unlikely to be a viable option. 117Further, if the social entrapment concept were embedded within broader Compendium self-defence directions (discussed below), explicit reference could be made to the challenges associated with retreating.For example, a victim/survivor may be most at risk when they attempt to leave or retreat from the dangerous situation in which they find themselves. 118

Mistaken belief in self-defence
The approach to intoxicated mistaken belief in standard self-defence and householder cases, enshrined in section 76(5) Criminal Justice and Immigration Act 2008, has attracted criticism, and requires statutory reform. 119An individual is entitled to rely on a genuine mistaken belief about the need to use force in relation to both defences. 120The mistake need not be reasonable,but reasonableness is considered in assessing whether the belief was genuinely held. 121Mistaken beliefs attributable to self-induced intoxication are excluded. 122No amendment recommendation was made by the PRT on the matter, however, it may have significant implications for victims/survivors claiming self-defence, particularly those who self-medicate as a coping strategy.Whilst we focus on the circumstances of domestic abuse victims and householders in the analysis that follows, we are of the view that our proposed amendments to the law on intoxicated mistaken belief ought to be considered equally relevant to all self-defence claims.
A genuine mistaken belief permits 'any stupid or objectionable ground for believing oneself to be under attack (racism is one such example, and not far-fetched!)… [as] legitimate.' 123 In contrast, public policy militates against mistaken belief in self-defence where the defendant is voluntarily intoxicated and/or where they are no longer intoxicated but the mistake is 'immediately and proximately consequent upon earlier drink or drug-taking.' 124rombag and others have categorised criticisms of 'prior fault' as 'definitional' based upon the understanding of the defendant's 'fault', and 'functional' in assessing whether it is appropriate for prior fault to preclude a defence. 125On basic intent offences (those requiring recklessness as the mental element), 126 by becoming intoxicated, the criminal law considers that the defendant 'voluntarily made himself dangerous in disregard to public safety, [and] that is morally equivalent to having the fault element of recklessness as to others' safety.Consequently, the defendant is to be regarded as having acted with a sufficient fault element to warrant conviction for the offence.' 127 The 'functional' exclusion of intoxicated mistaken beliefs is an extension of the policy principle not to enable defendants to escape liability based upon their intoxicated state.Foresight of the risk,however,that an individual might commit a criminal act while intoxicated is different from 'foresight of the likelihood of acting in putative self-defence due to intoxication', which would be an unlikely state of mind. 128As Crombag and others suggest, it would appear that sufficient fault to prevent mistaken belief claims in self-defence amounts to becoming intoxicated enough to make the mistake. 129This delimiting approach differs considerably from other defences, such as duress, where the defendant must have foreseen that they may be forced to commit a criminal offence (voluntary association). 130his inconsistency across imputation of prior fault is exacerbated when considering the often inextricable link between intoxicant use as a coping mechanism, and circumstances of intoxicant induced mental disorder.There is some correlation between being an abuse victim/survivor and self-medication. 131verup and others found 'drinking to cope is an important predictor of drinking problems, as well as an outcome of experiencing violence in the relationship.' 132Their results identified a greater risk that women would suffer alcoholrelated problems under the 'self-medication model', potentially 'because they are drinking as a means of coping with the negative effects associated with their victimization, including depression, anxiety, and social problems.' 133It is not uncommon for abuse victims/survivors to suffer 'multiple disadvantage', a term referring to any combination of homelessness, violence and abuse, substance misuse, poor mental health, poverty and contact with the criminal justice system.' 134 She is placed at a further disadvantage, where mistaken belief in selfdefence is excluded based upon a victim's/survivor's self-medication.
The extension of the intoxicated mistaken belief exclusion to the short term effects of voluntary intoxication, such as paranoia, is potentially problematic for the victim/survivor of domestic abuse. 135Victims/survivors may experience abuse-related paranoia, and it remains unclear how abuse-related paranoia versus intoxication-related paranoia might be legally delineated, given the possible causal connections across abuse/intoxication/paranoia.
A distinction is made between 'psychosis caused by acute intoxication' which precludes self-defence and 'psychosis resulting from primary mental illness separate from and co-morbid with drug use' which may be relevant to the question of whether force was necessary, but not the reasonableness assessment which is likely to mean the force used is regarded disproportionate. 136istinguishing mistaken belief across these different categorisations of intoxication/intoxicated effects/mental disorder is arguably a delineation by mud rather than crystal. 137Moreover, any attempt to rely on mistaken belief attributable to mental illness induced by intoxication or otherwise is counterintuitive to the move away from pathologising the abuse victim/survivor, who may, in any event, be reluctant to engage with medical professionals. 138 social entrapment lens is required which fully considers the circumstances defence, intentionally inciting a qualifying trigger for the purposes of providing an excuse to use violence is statutorily barred; Coroners and Justice Act 2009, s 55(6).See Qurat-ul-ain Jahangir, John J. Child and Hans S. Crombag of the victim/survivor, and the authority response to any earlier attempts at intervention in addition to any intersecting inequalities.139 In a qualitative study, Douglas interviewed 65 women who engaged with the legal system following domestic abuse.140 The study revealed that whilst some consulted medical professionals for support, others were reluctant to do so because of fear of negative court outcomes,particularly on parenting arrangements.141 Some refused to take prescribed medication.142 A key concern was the potential for medical records to be used to question parental capability.143 Fear of psychiatric reports led others to self-medicate.
These concerns are compounded when considering the stark reality of the difficulties associated with claiming self-defence under the current model, and the impact that self-medication might have in relation to a self-defence claim.Crombag and others argue that the only way to sensibly delineate prior fault in voluntary intoxication from mental conditions precipitated by or separate from drug use is to narrowly define intoxication 'in terms of drug on board state', but ultimately, they reject 'intoxication, even drug-on-board intoxication, as an accurate proxy for prior fault.' 144e are in measured agreement with this rejection.Narrowing the confines of voluntary intoxication alone does not address the inconsistency in allowing self-defence based on honest albeit repugnant mistaken beliefs and overcriminalising defendants by precluding mistaken belief in self-defence based upon legal and socially accepted alcohol consumption.'Reasonshould recoil' 145 that an 'incorrigible racist' 146 should leave the 'Court without a stain on his character' 147 where he injures another consequent on his racist mistake.The current law, however, only applies that rationale to the drunk individual who injures a third party consequent on their drunken mistake, capturing under the same umbrella, inter alia, addicts, binge drinkers, householders who imbibe in a bottle of wine in their home on a Friday evening, and abuse victims/survivors who consume alcohol to cope.
On the householder defence, Leveson, in Collins explained 'it is at least arguable [that the approach to self-induced intoxication] is unduly restrictive for householders.' 148According to Leveson, those who go 'outside their own homes … must take responsibility for their level of intoxication … Why that should be so in the defendant's own home in circumstances where he is not anticipating any interaction with a trespasser is, perhaps, a more open question but that remains part of the test even in a householder case.' 149 That a householder's self-defence claim should not be precluded where his violent resistance was predicated on an intoxicated mistaken belief following lawful consumption of alcohol in his own home is relatively uncontroversial.There is, in contrast, an 'intuitive appeal' to the view that those who go 'outside their own homes … must take responsibility for their level of intoxication' 150 yet, as Crombag and others identify, it is incredibly unclear what should amount to 'sufficient' fault to preclude self-defence. 151eveson's criticism assumes a clear distinction between intoxicated mistaken belief inside and outside the home which, as Rogers explains, fails to recognise that the policy approach 'is not restricted to drunken brawlers: it could … apply to a woman who walks home by herself, having had a few drinks, and who may react with mace spray against a man who appears to be following her, but who turns out to have been walking home in the same direction.' 152Fault should not be attributed to the woman who has been taught from a young age that walking home alone is dangerous, to avoid drinking too much, to wear modest clothes including trainers to run faster,'be alert but not alarmed',to flag a bus, 153 and to 'consider in terms of the legal process, to learn just a bit about the legal process.' 154 The implication is consistently on limitations women 'should' place upon their own lives to make themselves safer from men. 155 A woman who, having consumed alcohol (against the background of warnings regarding how unsafe society is), makes a mistake regarding the need to act in self-defence or the level of force used should not be prevented from accessing self-defence.
As Rogers explains, if the issue concerns 'wholly unreasonable' mistakes, the law should be amended to require a reasonable mistake, aligning the position with other areas of criminal law, such as sexual offending, and ensuring Article 2 ECHR compatibility. 156The reasonable belief as a sober belief concept in sexual offending would need to be re-evaluated in self-defence contexts so a more nuanced approach can be applied in householder and victim/survivor cases.Should the mistaken belief requirement in self-defence be amended to require a reasonable belief, there is no reason that this change should be limited to householder and domestic abuse cases.

PROPOSED AMENDMENTS TO THE CROWN COURT COMPENDIUM 2022
Having considered proposed substantive law changes to both standard selfdefence and householder defence claims that would increase access to the defendant who uses violent resistance against an abuser, and those who make a mistake in using force to defend themselves even when intoxicated, we now turn to consider procedural changes.The most important procedural change, as we see it, would be to include an example of a defendant's use of violent resistance towards an abuser in the Crown Court Compendium (2022) which provides a route to verdict based on the social entrapment approach.The Compendium is designed to provide guidance on directing juries in Crown Court trials, and currently provides several example routes to verdict applicable to self-defence. 157The current guidelines make no reference to domestic abuse or coercive control.We have adapted the facts in example five of the Compendium to convey a situation which involves a victim/survivor of abuse using force against her ex-partner who at the material time does not live in her home.
We apply the questions provided by the Compendium, to illustrate the current state of self-defence as it applies to abuse victims/survivors and the limited cases in which a householder self-defence claim might succeed.With this analysis we support our arguments that the defence proposed by the PRT should be enacted and that the concept of social entrapment should be embedded within the Compendium guidelines to provide greater access to these defences by victims of domestic abuse who use violent resistance as a protective strategy.

Our example
The hypothetical scenario involves a (former) intimate partner dispute where D claims W entered the home in which she was residing with their son.According to D, W was angry because he found out D had contacted the property owner and arranged for the locks to be changed after W decided to leave her for another woman the week before.D claims when she asked him to leave, W became verbally abusive, threatened to call the police because she 'had no right to have the locks changed,' and then advanced towards her.According to D, that is when she picked up a knife and cut W's arm.In contrast, W claims their son let him into the house, that he intended to collect some belongings, and D stabbed him out of jealousy and anger in response to the affair.There is no dispute that D cut W's arm with a knife.

Necessity of Force
To illustrate situations where jurors may disagree as to whether the victim in a case is a trespasser for the purposes of the 'householder defence' the Compendium provides five questions that the court should apply. 158The first 157 Compendium, n 15 above, 18-1-18-15.158 Compendium, n 15 above.The example is number 5, ibid, 18-9.
question relates to the subjective necessity test: 'Are you sure that D was the aggressor and that she did not believe it was necessary to use force against W? If your answer is Yes, your verdict will be "Guilty."' 159The first (subjective) question in both householder and standard self-defence cases is: 'Did D believe, or may D have believed that it was necessary to use force to defend themselves from an attack or imminent attack on them or others or to protect property or prevent crime?' 160 Responses to physical attacks or threats suggesting an imminent attack will be sufficient to satisfy this limb, however, this question cannot be properly addressed without an appreciation of coercive control through the lens of social entrapment.As self-defence is only available as a response to imminent physical force it may be of limited use to the victim who is experiencing coercive and controlling behaviour characterised by psychological and non-physical tactics.In cases involving coercive control, victims may understand that an attack by an abuser is imminent where the victim has inadequately complied with the abuser's demands.Abusers may no longer need to verbalise the threat, as the victim/survivor has internalised the rules, learning from experience the consequences of non-compliance. 161Gestures representing a threatening symbol indicating an imminent attack to the victim/survivor may appear to be innocent to an outside observer. 162o appreciate imminence in abuse contexts, it is necessary to view the perpetrator's behaviour as coercive control rather than through the bifurcatory lens of 'a bad relationship with incidents of violence' since this assumes 'effective safety options' or other alternatives between violent incidents. 163Similarly, if the victim/survivor's response is viewed through 'battered spousal syndrome' rather than coercive control and social entrapment, her conduct is pathologised rendering it unlikely to be viewed as reasonable. 164To appreciate the level of force used in the second stage of self-defence, it is essential to understand the necessity to use force from the defendant's perspective using a social entrapment lens.
To do this jurors should, in all self-defence cases involving domestic abuse, be directed to consider: the coercive and controlling behaviour the abuser used and how this reduced the victim's/survivor's space for action; 165 the responses of frontline services and what they could realistically be; and, the impact of any intersecting inequalities in the victim's/survivor's life on how they responded to the coercive and controlling behaviour and frontline services and how this further exacerbated the coercive control. 166As Howes and others explain, judges and juries need 'to take account of the wider context in which [these incidents] take place -namely what is in large part arguably a failure of the criminal justice system to protect women or address men's violence towards them.' 167 The threat level experienced by an abuse victim/survivor may fluctuate, but the overarching threat is ever-present, and imminence/necessity and reasonableness/proportionality in abuse contexts may appear very different to standardised perceptions of self-defence. 168his is not to say any level of coercive control would give rise to self-defence, but how a physical threat is communicated based upon the coercion and the experience of it should not be ignored.There remains a broader need to appreciate the circumstances of that threat as a pattern of behaviour where the victim knows they are at risk but equally they are ensnared within a web of coercive control and varying perpetrator tactics designed to prevent them from leaving the relationship.Monckton-Smith developed a risk assessment tool following a study involving 378 intimate partner homicides.In every case, Monckton-Smith identified a typical chronology of events which demonstrate that rather than acting in the 'heat of passion' (and killing the victim spontaneously) abuse perpetrators follow a pattern of coercive control which can frequently, and quickly, escalate into life threatening and fatal violence. 169Monckton-Smith has described this pattern, articulated in an eight-stage process, as the homicide timeline. 170Stage one involves a history of control or stalking.At stage two perpetrators seek early commitment in the relationship.Stage three sees the relationship established where tactics are adopted to entrap the victim/survivor and prevent them from leaving.At stage four, a trigger challenges the perpetrator's control, for example, an attempt to leave or threat to do so.It is at this stage that the risk to the victim/survivor increases.An escalation of controlling behaviour occurs at stage five which may involve physical or non-physical forms, threats of suicide/homicide, begging and crying.Ostensibly small threats which are acted upon demonstrate an escalation of risk and threat.During stage six, a change in thinking occurs and the victim/survivor notices a subtle demeanour change.The change could be increased anger but reports also indicate that a period of calm might ensue.
The perpetrator may leave and begin the pattern with a new partner, revert to using controlling tactics to entrap the victim until another challenge to control arises, plan revenge or resort to homicide.Stage seven challenges the traditional 'heat of passion' killings in light of evidence of planning and often more than a month between the trigger and the killing.The final stage is homicide which may include killing others, such as children and/or be followed by suicide, an immediate confession, attempts to conceal the crime 166 Tolmie and others, n 33 above.167  etc.This risk assessment tool demonstrates that these systemic patterns of control are indicative of an ongoing and heightened risk of danger to the victim/survivor, and their entrapment within the relationship.More broadly, Monckton-Smith's study challenges traditional notions of how an imminent threat is manifest since the consistency of each stage and escalating risk across almost 400 cases provides the clearest indication that imminence and escalation in the pattern of coercive control should not be viewed as mutually exclusive.Instead, escalation should be viewed as falling within the imminence concept in appropriate circumstances given the dangerous situation created by the typical pattern in these cases and how rapidly the situation develops.
Notwithstanding these observations, we acknowledge that the concept of self-defence concerns a physical attack or threat thereof and whilst a full defence should be available where the victim/survivor responds to such an imminent attack, in whatever way it is communicated, we cannot see it possible that the defence extends to victims using violent resistance to escape where there is no such threat, though we appreciate the dire circumstances.In homicide cases, where there is not an imminent threat, a partial defence may be an appropriate alternative. 171There is also clearly a requirement to consider how victim/survivor non-fatal offending in response to abuse is addressed within the defence framework outwith self-defence.

Trespasser
The second question in our example, based on the Compendium, is relevant only to the householder defence: 172 'Are you sure that D knew her son had invited W into the house?'Weconsider this question for two key purposes.First, to illustrate the extremely limited circumstances in which victims/survivors of domestic abuse will be able to claim householder self-defence where she uses violent resistance against an abuser.Second, to highlight the need to appreciate the broader concept of social entrapment experienced by the victim/survivor in such contexts.
W does not have to be a trespasser in law.The question is based on D's genuine belief in the situation, but there must be an evidential basis for the belief. 173If D knew or believed W was invited into the property, then she did not honestly believe W was trespassing at that point.Similarly, a question arises regarding whether D honestly believed W was trespassing when W refused to leave at D's request or whether such a belief could be inferred.It can be argued that in notifying the property owner that W had left and changing the locks, D honestly believed W no longer had a legal right to enter the property, and his refusal to leave at her request might support an evidential basis for that belief.Alternatively, she may still believe that it is his home and that he had a right to be there which would negate a claim.
The limited case law on abuse-cum-trespasser cases indicates that a liberal interpretation in domestic abuse cases may not be forthcoming.appealed her conviction for murder of her former partner arguing that the trial judge had erred in not leaving the householder defence to the jury, following the sentencing judge's remarks that he was sure she decided not to let him into her home, as a result 'of which he started to kick the front door, whilst [Magson was] heard shouting by one of the neighbours that [she was] not going to let him inside.' 174 The application was viewed as a vexatious claim, in part because any putative belief that Magson's former partner was a trespasser was not raised at either trial, nor was Magson asked whether he was trespassing which the Court of Appeal considered 'unsurprising' because he 'had lived with her for months.' 175 As Collins articulates, cohabitation is not a bar to the householder defence but based thereon are likely to be problematic for victims of abuse who may struggle with the notion that the perpetrator is trespassing thereby precluding the defence. 176n practical terms, 'the clearer it is that someone was a trespasser the more readily a jury will not be troubled by the issue whether the defendant did or did not hold the belief.' 177 For example,it may more readily be inferred that D knew or honestly believed W was a trespasser if she had contacted the authorities to ask about their respective legal rights to the property, or had called the police and for a senior police officer to have issued a Domestic Abuse Protection Notice (DAPN). 178This would require, inter alia, the senior police officer to have reasonable grounds for believing W has been abusive towards D, 179 and that it was necessary to give the notice to protect D from domestic abuse, or the risk of domestic abuse, carried out by W. 180 The DAPN would need to stipulate that W may not contact D for whose protection the notice is given; and may not come within a specified distance of any premises in England or Wales in which D lives. 181It is worth noting in relation to this scenario that DAPNs are for 48 hours during which time D may apply for a Domestic Abuse Protection Order (DAPO) prohibiting return to the premises.Either of these would arguably provide strong evidence that D honestly believed W was trespassing. 182t is unsurprising that academics have called for more guidance on when an honest belief may be inferred from the circumstances in such contexts. 183It will 174 n 94 above at [17].175 ibid at [23].Note that the court said that it would not be necessary to ask that question in every case as D's belief might be inferred from the circumstances, ibid at [23] be more difficult to evidence an honest belief in abuser-sum-trespasser cases because of assumptions made within the criminal justice system about the nature of the relationship/living arrangements.Further, an appreciation of the social entrapment concept is required to understand why a victim/survivor may have refused entry to her home but still not view the perpetrator as a trespasser or why more formal evidence is unavailable due to suspicions regarding authority involvement or ineffective official responses.
More broadly, it is irrational that an abuse victim/survivor could have the lower threshold test apply in such limited circumstances, but not in others, adding an unwelcome layer of complexity to the law.Better protection will only be afforded to the domestic abuse victim when she leaves to reside elsewhere and/or he leaves and returns to the premises as a trespasser.A masculinist approach is adopted, placing greater emphasis on the need for the perpetrator to be an intruder rather than on recognition of the petrifying circumstances she lives in every day.
In most cases, D is likely to believe W has (and he may do) a right to be there because of the nature of their relationship, proprietary interests or because the authorities are unaware of the abuse, or their response was ineffective.In cases where she believed he had a right to be there, and it later transpires he did not, there can be no ex post facto justification in which the lower threshold test would apply, and she would remain without self-defence where disproportionate force is used. 184Given the complex nature of relationship cases it is disappointing that the court in Magson did not take the opportunity to outline the types of circumstance from which a belief that W is a trespasser might be inferred. 185In many cases, the standard reasonableness test will apply as per question three of the Compendium.

Reasonable and Proportionate
Should the jury accept D honestly believed W was a trespasser at the time of the incident, they are obliged to apply the lower reasonableness threshold of the householder claim as depicted by questions four and five in the route to verdict for example five.
Question four of the householder self-defence route to verdict asks, 'Has the prosecution made you sure that the force used by D against W was grossly disproportionate in the sense of being completely "over the top"?If your answer is Yes, your verdict will be "Guilty."If your answer is No, go on to consider question 5.' 186 Question five in the route to verdict asks 'Are you sure that, in the circumstances as D believed them to be, and having particular regard to the fact that D was confronted by someone he/she believed to be an intruder in the home, the force used by D was unreasonable?If your answer is Yes, your verdict will be "Guilty."If your answer is No, your verdict will be "Not guilty."'Where D does not believe W is a trespasser standard self-defence applies and the court refers to question three.This asks 'If you are sure D knew W was invited into the house then has the prosecution made you sure that the amount of force used was unreasonable on the facts as D believed them to be?If your answer is Yes, your verdict will be "Guilty."If your answer is No, your verdict will be "Not guilty."' In relation to our posited scenario, the foregoing analysis and the use of a knife highlight that D would be in a better position if the jury accepted that she honestly believed W was a trespasser because the lower threshold test in questions four and five would apply.Even where the lower threshold test applies, the extent to which use of a knife would be considered reasonable is unclear.Adding questions to the route to verdict example that incorporates the social entrapment concept allows several factors to be examined by the jury when determining whether the level of force used was reasonable (standard self-defence) or at least not grossly disproportionate (householder self-defence) in the circumstances as understood by the defendant.These include discussion of the defendant's experience of coercive and controlling behaviours, an examination of the local authorities' response to any prior help-seeking by the defendant, or the defendant's perception of institutional responses towards members of their community (who may share characteristics such as race, gender etc).Consideration would then be given to what the victim believed would be a realistic response to the abuse.
The most progressive way to enable women's access to claims of self-defence in domestic abuse cases would be to accept the PRT proposals and provide a Compendium example with directions and a novel route to verdict that provides a social entrapment enquiry into the defendant's use of violent resistance.Together, the substantive law change and procedural guidance might go some way towards placing the issue of women's limited physical strength that leads to them resorting to weapons on the same footing as male use of body force against a person of similar physical strength. 188owever, given that substantive legal change is likely to be slow to achieve, adoption of a route to verdict modelled on the social entrapment approach could be introduced in the interim, enabling greater access to standard selfdefence claims by women who resort to violent resistance as a protective strategy.These broader directions should challenge abuse stereotypes and myths. 189he amendments we advance would direct jurors to consider essential factors to make an informed assessment regarding whether force was necessary and reasonable in the circumstances as she believed them to be.We view this as a novel way to assist in increasing awareness within the judiciary and juries regarding the dynamics of abuse, given reluctance to allow non-medical experts to provide evidence in court on these issues.190

THE ROLE OF EXPERTS
The potential role of both medical and non-medical experts in these cases is significant.Expert opinion evidence is admissible where it will assist the court in reaching a conclusion because it is outside the knowledge and experience of the judge and jurors.Neither the court nor jurors are bound by the opinion but may take it into account when determining the facts in issue. 191Experts must have sufficient knowledge through qualification and/or experience in the relevant field for their opinion to be considered. 192Our foregoing analysis has indicated that there is a general lack of understanding of domestic abuse and coercive control across parliament and myths and misconceptions regarding why victims/survivors did not leave their abusers remain commonplace.
This lack of understanding has been (partly) addressed by the use of medical expert testimony on Battered Woman Syndrome (BWS).As early as 1990, the Canadian Court in Lavallee 193 considered the relevance (and arguably the importance) of expert medical evidence on battered woman syndrome in the context of a self-defence plea by a defendant who killed her abuser.Citing State v Kelly 194 the Canadian Court of Appeal explained that expert testimony on BWS must be relevant since it addresses an area subject to a great deal of mythology, such as 'why didn't she leave?' and expert evidence is required to help jurors to disregard these myths and potentially wholly incorrect conclusions based on 'purported common knowledge … which may be very much mistaken.' 195Expert testimony on her ability to perceive danger from the perpetrator and why she did not flee may also be relevant in assessing the nature and extent of the abuse and assessing the reasonableness of her perception that killing the perpetrator was the only way to save her own life. 196Although the term BWS has been criticised as pathologising women's violent resistance to domestic abuse,it did serve an important function in opening the door to expert evidence in this context. 197here is, however, an educative function that could be served through the admissibility of general expert evidence (for example, those working with victims/survivors of domestic abuse/coercive control, counsellors, academics, etc) which alerts jurors to the implications of coercive control. 198This evidence could explain the commonality of problematic features of victim/survivor claims, for example, that she believed she could not leave despite having the physical means to do so, and why this should not necessarily lead to the view that alternative options were available. 199These general explanations would not be aligned to BWS and, as such, would avoid the implication that the defendant was operating under some form of diminished responsibility rather than reacting reasonably in the circumstances as she perceived them.The effect would be to provide a 'neutral summary of the relevant research' and leave jurors to determine the case. 200Understanding social entrapment is outside current court expertise, and there should be greater recognition of the need to engage both medical and non-medical experts in self-defence cases. 201his is the position in Victoria (Australia) where, following abolition of partial defences, 202 self-defence was amended to better accommodate the circumstances of victims of domestic abuse who respond with violent resistance.Part of those reforms included expansion of the admissibility of family violence evidence (social framework evidence) to all self-defence claims.Such evidence includes, inter alia, the history of the relationship, cumulative impact of family violence, and social economic and cultural factors which may impact on a family member. 203We suggest that the approach adopted in Victoria provides a model for development and introduction of specific legislation on the admissibility of expert evidence based on the social entrapment concept.

CONCLUSION
The 2021 Act, which statutorily defines domestic abuse provided a clear opportunity for parliament to harmonise offences, such as, coercive and controlling behaviour, with defences where an abuse victim/survivor offends in response.Our article has focused on the government's refusal to extend the householder defence to domestic abuse victims/survivors on grounds that the householder is acting 'on instinct' whereas in 'domestic abuse cases, the response may not be a sudden instinctual one but may follow years of physical and/or emotional abuse.' 204 The distinction is discriminatory, unhelpful and rests on a misunderstanding of the social entrapment of the victim/survivor and its impact on their protective strategies. 205The distinction is artificial since the lower threshold householder defence may apply in very limited circumstances involving abuse, for example, where the abuser leaves home and returns as a trespasser.
In terms of future developments, we argue that the PRT proposals to extend the householder defence to domestic abuse claims ought to be adopted.Unlike the householder defence reforms, however, a broader contextual analysis of the positioning of these defences within the existing self-defence framework is mandated.We advocate novel recommendations on the option to retreat in self-defence and the public policy blanket exclusion of intoxicated mistaken belief in self-defence.In respect of the former, we advance a statutory rebuttable presumption that victims/survivors of domestic abuse/coercive control are unlikely to be able realistically to retreat safely -a clause that could equally apply in the householder context.In terms of the latter, the exclusion of intoxicated mistaken belief in self-defence is inconsistent with the decision to allow any genuine albeit repugnant mistaken belief to qualify for the defence.It is also insufficiently nuanced to distinguish mistakes made by drunken brawlers from householders who legally consume alcohol in their own home without any thought to being confronted by an intruder, and victims/survivors of domestic abuse who drink to cope with that victimisation.If the concern is about reasonable beliefs, then a reasonable belief, which may, in appropriate cases, include an intoxicated belief, ought to be required.There is no reason why this approach should be limited to domestic abuse/householder cases.
These substantive legal changes ought to be supported by the admissibility of non-medical expert evidence on the nature and impact of coercive control and supplemented by judicial directions on self-defence in abuse cases.We recommend a novel route to verdict in the Crown Court Compendium which addresses the social entrapment concept ie the impact of coercively controlling behaviours on capacity for action; the response of authorities and what they could realistically be; and, the impact of any intersecting inequalities in the victim's/survivor's life on how they responded to the coercive and controlling behaviour and frontline services and how this further exacerbated the coercive control.Non-medical expert evidence (for example, domestic abuse advisors, academics, etc) can play a crucial educative function here in ensuring the wider circumstances of domestic abuse are understood.
Finally, better understanding of self-defence in response to abuse would be achieved through a statutory data reporting requirement, and amendments pre-and-post trial to address diversion and sentencing in a manner which better recognises the circumstances of victims'/survivors' self-defence claims. 206nless we really engage in understanding the differences between traditional self-defence claims and claims involving a history of coercive and controlling behaviour, self-defence will remain out of reach for women who use violent resistance to protect themselves from an abuser.206 It is not substantive law at trial stage alone that can increase the abuse victim's/survivor's access to defences, and several additional process and procedural changes are required at pre-trial, and post-trial stages.For example, Sentencing Council Guideline, 'Overarching principles: domestic abuse' (2018) are relevant where the offender has perpetrated violence against the victim, but not where the victim/survivor defendant has been abused.
199 ibid.200 ibid.201 See Donna K. Cocker and Lindsay C. Harrison, 'The Story of Wanrow: The Reasonable Woman and the Law of Self-Defense' in Donna Coker and Robert Weisberg (eds), Criminal Law Stories (New York, NY: Foundation Press, 2013).Tolmie and others, n 33 above; Douglas, Tarrant and Tolmie, n 33 above.202 Crimes (Abolition of Defensive Homicide) Act 2014.Note defensive homicide was categorised as an offence rather than a partial defence.203 Crimes Act 1958, s 322J as amended by the Crimes Amendment (Abolition of Defensive Homicide) Act 2014.For further discussion see Wake, n 12 above.204 HL Deb vol 810 col 1754 10 March 2021.See also Edwards, n 70 above, 357.205 Douglas, Tarrant and Tolmie, n 33 above.
Naffine describes men of law as the judges, textbook writers, doctrinal and conceptual scholars and legal scholars as philosophers.61 Joanne Conaghan, Gender and the Law (Oxford: OUP, 2013) ch 3.For example, the spousal immunity to rape prior to R v R [1991] UKHL 12;[1991]4 All ER 481.Accordingly, Conaghan states that 'gender features as a historical aberration, a mistake to be corrected within the context of a conception of law as an essentially benign and progressive institution, albeit, as the product of human design, prone to error.' 62 ibid.63 Naffine, n 58 above, 165.64 ibid, 148.
79 Catherine Donovan and Rebecca Barnes, 'Help-seeking among lesbian, gay, bisexual and/or transgender victims/survivors of domestic violence and abuse: The impacts of cisgendered heteronormativity and invisibility' (2020) 56 Journal of Sociology 554.80 Criminal Law Act 1967, as restated in Criminal Justice and Immigration Act 2008, s 76.
85 In Scotland, in theory, there should exist a formal record of self-defence claims considering the requirement to lodge 'defence statements' in advance of trial, McPherson, n 8 above.Pretrial defence disclosure, at Crown Court level, in England and Wales, is governed by Criminal Procedure and Investigations Act 1996, s 6C, as amended by Criminal Justice Act 2003.No defence statement is required at Magistrates' Court level.

112
Ray n 38 above at[38].113 'No mention has been given in this new clause to a defendant's option to retreat from the abuse, and I make that point with due care.I acknowledge, and am well aware, that an abused woman or man may not have that option', n 1 above, col 1754.114 '[I]n my experience, having been physically threatened by an intruder and having been physically assaulted by my then partner, the intense stress is far worse and sustained when the person you rely on for love and affection snaps and attacks you or subjects you to abuse over a prolonged time.My own experience of domestic violence is that retreat just encourages further violence.
139 ibid.140 Heather Douglas, 'Domestic and Family Violence, Mental Health and Well-Being, and Legal Engagement' (2018) 25 Psychiatry, Psychology and the Law 341.141 ibid.142 ibid, 'I didn't take it because I went oh, I don't want people to think I've got bipolar … He's accused me of being the one with a mental illness … and I didn't want him to subpoena my medical charts or Medicare records.' 143 ibid, 'I actually do have a mental health plan that I can get subsidised care … But I was very careful and I guess read up about what you should put on there.I guess it sort of fitted in that the best diagnosis was to say it was post-traumatic stress.' 144 Crombag and others, n 125 above.145 R v O'Grady [1987] QB 995 (O'Grady), 1000 per Lord Lane CJ, cited in R v Hatton [2005] Magson 171 Coroners and Justice Act 2009, ss 52-56.172 For the purposes of the Criminal Justice and Immigration Act 2008, s 76(8).173 Ministry of Justice Circular No 2013/02, 16.Cheeseman n 93 above at [21].
. 176 Jennifer Collins, 'Householder Defence: R v Magson' [2023] Criminal Law Review 81.In Magson the victim 'had a key to the house.It was his home', ibid at [23].177 Cheeseman n 93 above at [21].178 Domestic Abuse Act 2021, ss 22-26.179 The victim must be aged over 16 or personally connected to the perpetrator, Domestic Abuse Until then the existing Domestic Violence Protections Notices and Orders under the Crime and Security Act 2010, ss 24-30 apply.182 Domestic Abuse Act 2021, ss 27-28.For discussion of the relevant provisions see Vanessa Bettinson, 'Adding to the Domestic Abuse Criminal Law Framework: The Domestic Abuse Act 2021' [2022] Criminal Law Review 88. 183 Collins, n 176 above.
187 184 R v Dadson (1850) 4 Cox CC 358.For discussion see, John Cyril Smith, Justification and Excuse in the Criminal Law (Hamlyn Lectures, 40 th Series, 1989).See also, Russell L. Christopher, 'Unknowing Justification and the Logical Necessity of the Dadson Principle in Self-Defence' (1995) 15 Oxford Journal of Legal Studies 229.
191 R v Turner (1974) 60 Cr App R 80. 192 R v Clarke and Morabir [2013] EWCA Crim 162.193 n 164 above.194 478 A.2d 364 (1984), 379.195 Lavallee n 164 above, citing State v Kelly 478 ibid, 378.196 'Dr.Shane prepared a psychiatric assessment of the appellant.The substance of Dr. Shane's opinion was that the appellant had been terrorized by Rust to the point of feeling trapped, vulnerable, worthless and unable to escape the relationship despite the violence.At the same time, the continuing pattern of abuse put her life in danger.In Dr. Shane's opinion the appellant's shooting of the deceased was a final desperate act by a woman who sincerely believed that she would be killed that night', ibid at [1].197 Thank you to Professor Susan Edwards for making this comment.198 Louise Ellison and Vanessa Munro, 'Assessing the Impact of (Mock) Juror Education in Rape Trials' (2009) 49 The British Journal of Criminology 1.