Back to blog
Domestic Abuse Victim Turned Cold-Blooded Killer? Reforming Loss of Self-Control in the Domestic Homicide Context
📅 Published on: February 25, 2025

Image taken from One Thousand Words, a bank of free stock images for the meaningful representation of domestic abuse in media. © Laura Dodsworth

Introduction 

When victims of domestic abuse, particularly female victims, kill their abusers, the law often provides scant protection against a murder conviction. With the Law Commission undertaking a consultation in the summer of 2025 to review the defences available in domestic homicides, it is time to look at the law afresh.1  This article will focus on the defence of Loss of Self-Control (LOSC).2 Just like its predecessor, the doctrine of provocation, LOSC is a partial defence which only applies to murder. It does not absolve criminal liability, unlike defences such as self-defence or insanity. Instead, it reduces the offence from murder to voluntary manslaughter. 

LOSC operates in an array of different factual contexts, but we will focus on the unique disadvantages faced by domestically abused women who kill their abusers.3 Such women frequently do not conform to outdated preconceptions regarding the impact of domestic abuse. Contemporary psychiatry understands domestic abuse as cyclical, with each stage (forgiveness, tension accumulation, abuse) able to take months. Research shows that women in abusive relationships, where they are subjected to coercive control, are likely to kill their abuser because of a sense of entrapment where they see no other means to escape.5 It is well-recognised that abused women often plan the killing of their abuser for many months.6 Improving the law is crucial because where LOSC is successfully raised, a lifelong murder sentence is substituted with a voluntary manslaughter sentence.7 This article will align the law with contemporary forensic psychology and propose modernising reforms that aim to remove the gender disparity.

Context

In the year ending March 2023, out of 100 victims of domestic homicide, 70% were women.8 Only 1 was not killed by a man.9 In the rare cases where a woman, specifically a domestically abused woman, kills their abuser the law offers meagre protection. Defences to murder should be limited. Intentional killing, that is not done in self-defence, should result in criminal punishment. However, women who kill their abusers are killing in a context which warrants an adequate defence to shield against, what many would regard as an unjust, murder sentence. A recent independent review found that out of 120 domestic homicides, LOSC was raised 11 times, and was successful a mere 2 times.10 This meagre 18% success rate does not include all the times defence counsel will have (understandably) decided against raising the defence in the first place.

While the Coroners and Justice Act 2009 (CJA 2009) abolished LOSC’s predecessor11 – provocation – the problems for abused women have seemingly been replicated.      

The Current Law

The Law Commission (LC) proposed12 that where D (1) acted in response to (i) gross provocation, (ii) a fear of serious violence, or a combination of both, and (2) a person of D’s age and ordinary temperament might have acted in the same or a similar way to D, then D would have a partial defence to murder. In determining (2), the court should consider D’s age ‘and all the circumstances’ except those which only relate to D’s capacity for self-control. D would be barred from relying on the defence when acting out of a considered desire for revenge or where the provocation was incited to provide an excuse for violence. In recognising provocation’s gender bias,13 the LC recommended removing provocation’s “unnecessary and undesirable” temporal limitation.14 To prevent the defence from being misused, the LC proposed a specific restriction against the defence being applied to revenge killings.15 Notably, the LC did not even call the defence LOSC. The defence would still be ‘provocation’ but would have existed in a modernised form.

However, the Government was not entirely persuaded.16 It was concerned that the LC’s approach would overly broaden the defence’s applicability. The Ministry of Justice (MoJ) stated that it “remain[ed] concerned” of the defence being misused to provide a partial defence for “cold-blooded” killing, organised crime, and “honour” killings. Furthermore, the MoJ considered that the LC’s proposal still included a “fundamental problem” because it provided D with a partial defence when D “has killed while basically in full possession of his or her senses, even if he or she is frightened, other than in a situation which is complete self-defence.” Consequently, the Government decided to retain both the LOSC element and the provision against revenge. 

The current LOSC defence is contained under ss54-55 of the CJA 2009. The defence has three elements. Firstly, the killing must result from D’s loss of self-control.17 Secondly, that the LOSC must be caused by a qualifying trigger18 – either, or both, (a) fear of serious violence19 or (b) things done and/or said – which constitute ‘circumstances of an extremely grave character’ and ‘caused D to have a justifiable sense of being seriously wronged’.20 Thirdly, it must be the case that a person of D’s age and sex, ‘with a normal degree of tolerance and self-restraint and in the circumstances of D, might have reacted in the same or in a similar way to D.21  Our focus will be on the first and third elements because these are the most problematic aspects for domestically abused women seeking to rely on the defence. 

Women in Control

Women were often disadvantaged by the temporal aspect under the old law of provocation because they often do not respond by killing in the immediate, or near aftermath of the provoking event. In Duffy,22 the defence of provocation failed because D waited until her abusive husband was asleep before killing him. Devlin J’s jury direction that the killing had to be sudden and without time for deliberation for provocation to succeed was upheld by the Court of Appeal. The fact that D killed her abuser this way because, if it were sudden and undeliberated, D would likely have been killed herself was legally redundant. The need for sudden LOSC has been abolished23 but problems with the old law persist under the CJA 2009. 

Research has proven that women subjected to coercive control are likely to kill their abuser because of a sense of entrapment.24 The law does not reflect that research. LOSC is a conceptually inadequate tool to encompass the impact of domestic abuse. In fact, the prevalence of premeditation for women who kill their abusers means that the LOSC formulation structurally inhibits the defence’s utility for domestically abused women. Furthermore, it was held in Clinton that for the defence to be put to the jury, there must be sufficient evidence ‘that a properly directed jury could ‘reasonably conclude that the defence might apply’’.25 While there is no definition of LOSC itself, a preplanned attack while the abuser is defenceless, such as asleep, cannot be reasonably considered to be a LOSC.

To illustrate the problems inherited from provocation in the new defence, it is worthwhile applying the new law to the old law case of Ahluwalia.26 Ahluwalia was in an arranged marriage with an abusive husband. She endured years of violence, including an attempt to kill her. One evening, he threatened to, among other things, throw an iron at her. That evening, while he was asleep, she doused him in petrol and set it alight, resulting in his death. It is unlikely any jury would dispute the existence of a qualifying trigger; there was evidence of a fear of serious violence because of her husband’s threats and that she was in circumstances of an extremely grave character that caused a justifiable sense of being seriously wronged. Did the qualifying trigger(s) result from a LOSC though? Ahluwalia waited until her husband was asleep and planned the method; it was not a ‘loss of control’ but rather premeditated. In these circumstances, the defence would be highly likely to fail, either due to the bar against revenge or insufficient evidence for LOSC. Clinton would mean that the jury could not have the matter put to them anyway. 

This is instinctively unfair. The vile abuse Ahluwalia was subjected to over the years likely led to a sense of entrapment from being so isolated that she had nowhere else to turn. The killing may have been premeditated and intentional, but it was a partially justified killing because of the bleak circumstances Ahluwalia lived in. The inability of LOSC to assist Ahluwalia epitomises the human consequences of law embedded with gender disparity. While the temporal ‘sudden’ limitation was removed, the effectiveness of this change is greatly diminished because whether there was a ‘loss of self-control’ will inexorably involve assessing the time between the qualifying trigger and the killing. 

‘Battered Woman’s Syndrome’27

A core problem with the old law of provocation was the effect of the Privy Council’s judgment in Holley.28 If a domestically abused woman had ‘battered woman’s syndrome’, then the syndrome would not be relevant when assessing whether the reasonable person would have acted in the same or similar way. Its only relevance was regarding assessing the gravity of the sting. The current law has essentially replicated Holley because a jury can assess D’s degree of tolerance and self-restraint in all the circumstances, which is defined as all the circumstances ‘other than those whose only relevance to [D’s] conduct is that they bear on [D’s] general capacity for tolerance or self-restraint’.29 In the conjoined appeals of Rejmanski; Gassman,30 the Court of Appeal held that where post-traumatic stress disorder and emotionally unstable personality disorder only reduced D’s capacity for self-control, those psychiatric conditions were not a relevant consideration for the jury.

In Sargant,31 S killed her mother by stabbing her to death following an argument about finances. S gave evidence of sexual, mental, and physical abuse by her mother. It was an agreed fact between counsel that S’s history of domestic abuse “resulted in” an adjustment disorder with elements of post-traumatic stress disorder, autism spectrum disorder, and a mild intellectual disability.32 All clinical evidence noted how the effects of abuse, and lack of support, meant S had a markedly lower than ordinary emotional threshold.33

However, the disorders were held “not relevant” to objectively assessing S’s self-restraint. The Court of Appeal, endorsing the trial judge, referred to Holley as authority for the irrelevance of S’s psychiatric and cognitive impairments. Quite frankly, these disorders could not be more factually relevant; without the abuse, the disorders would likely not exist, and without the disorders, one can be sure that S would not have killed her mother because of a financial argument. Holley prevails in spite of the CJA 2009. As if to soften the blow, the Court of Appeal stated the disorders “may be relevant to the gravity of the qualifying trigger”.34 This is precisely the same approach the Privy Council took in Holley. However, regardless of the severity of the qualifying trigger, if someone in S’s position would not have acted in the same or similar way, the defence fails. Since Clinton requires that all elements of the defence are present before the matter can be put before the jury, it is hard to see what has changed. 

Modernisation 

The gender disparity that hinders abused women from raising LOSC persists. The elements of provocation that restricted the availability of the defence have seemingly been reiterated in the CJA 2009. Clear statutory reform is needed to ensure that women who are subjected to horrendous domestic abuse are not destined for a murder sentence because of outdated, arbitrary, and gendered laws. The modernising zeal seen in the Domestic Abuse Act 2021 must be embodied in the LC’s 2025 consultation. By enacting the offence of intentional strangulation35 and controlling and coercive behaviour,36 the criminal law has proven its capacity to encompass conduct that is uniquely despicable in the domestic context. For criminal law to have a comprehensive approach to domestic abuse, partial defences to homicide must be modernised.

The LC noted two potential theoretical bases for LOSC:37 an excusatory or justificatory view. The former is adopted by Mousourakis.38 He argues that defendants who are provoked are less morally culpable because they are acting in the heat of the moment, as opposed to in cold blood. The issue here is that such a view inadvertently embodies the gendered disparity. Women often do not act in a ‘heat of the moment’ manner. The latter justificatory approach is advocated for by Ferzan.39 She submits there is a partial justification for the conduct as the deceased contributed to their death by their wrongdoing towards the defendant. Thus, the deceased must share some of the responsibility. This is not necessarily problematic. However, it is hard to see how the law could incorporate a theory of contributory moral culpability into the law of murder. Something either is criminal or is not criminal. The actus reus and mens rea elements of an offence are either fulfilled or they are not. For the former, a defendant is guilty and for the latter, the defendant is not guilty. Clearly, a contributory analysis is incongruous with the binary determination of culpability in the criminal trial, at least insofar as determining if the offence itself was committed.

Luckily, there is a much more appealing middle way. It is one the LC left out of their consultation, but one that is extremely helpful. Norrie characterises the theoretical approach of the LC, and of the CJA 2009, as one that seeks to encompass ‘conduct [that] is imperfectly rightful, and therefore both condemned and partially vindicated’.40 One benefit of this conceptualisation of a reasonable excuse is that it need not entail any LOSC element, and it can simultaneously recognise the imperfect conduct of the defendant who takes a human life, while also noting the partial rightfulness of the homicide in light of the abuse suffered by the defendant at the hands of the deceased.      

By retaining the LOSC requirement, both as an element and the defence’s name, the current law’s theoretical foundation seems more premised on the excusatory view than the middle ground of a reasonable excuse. The LC’s approach was seemingly founded upon Norrie’s conceptualisation, but the substantive law has ended up with an approach more aligned with Mousourakis’ theory. This, as illustrated above, has not produced a satisfactory outcome, but it is fixable by removing the LOSC requirement and renaming the defence to something like gross provocation (like the LC seemed to).41

Moving from theory into substantive law, the temporal restriction is a relatively simple fix and is solved the same way: remove the LOSC requirement. This was the LC’s initial recommendation and rightly so. Removing it means women in circumstances like in Ahluwalia are not prevented from raising LOSC purely because they waited until a physically stronger man was unable to fight back. These women are terrified and have been controlled by despicable men.  Their conduct occurred in response to severe psychological entrapment arising from the cumulative impact of domestic abuse. Getting rid of the LOSC component permits much greater flexibility to recognise this. It allows the focus to shift to the qualifying trigger’s gravity rather than arbitrary temporal analysis.

Now, we turn to allowing the psychological impact of domestic abuse, in particular its effect on one’s capacity for self-restraint, to be a consideration for the objective assessment. Here, we must ensure that women are not disadvantaged, but also avoid allowing any individual’s angry temperament to be considered. A way to avoid such difficulties is for the substantive law here to replicate the modernising ambition behind the proposals of Wade KC in her Domestic Homicide Sentencing Review. She recommended, among other things, that the defendant having been subjected to controlling and coercive behaviour should be a mitigating factor at sentencing and denote lower culpability under sentencing guidelines.42 Applying this approach to the substantive defence, a provision could be inserted to the effect that a jury may consider the cumulative psychological effect of domestic abuse on a defendant when undertaking their objective assessment. 

The pre-existing statutory definition of domestic abuse under the Domestic Abuse Act 2021 provides a useful foundation.43 Some changes are needed though. The 2021 Act’s inclusion of a single incident44 would be insufficient. The psychological sense of entrapment which has featured so prominently in this article does not apply when just one incident has arisen. A series of persistent abuses must have occurred. The abuse must also be ‘significant’. These requirements should act as safeguards against misuse. 

Conclusion

The gender disparities under the old law of provocation have been replicated in the CJA 2009. Recognising the disproportionate effect of violence against women requires modernising the law as it applies to women who kill their abusers. The reforms suggested here seek to remove outdated legal conceptions that are out of step with contemporary forensic psychology. Modern understanding recognises that psychological entrapment, rather than LOSC, is more likely to prompt an abused woman to kill their abuser. This proposal offers a way for the law to recognise the undoubted psychological impact of controlling and coercive behaviour when considering partial defences to murder. The proposed reforms recognise that women who kill their abusers are imperfectly rightful in the bleak circumstances in which they act, and, thus, do not deserve the mandatory life sentence for murder.      

Endnotes

  1. Domestic Abuse Act 2021, s 1(3)
  2. Law Commission, ‘Defences for Victims of Domestic Abuse Who Kill Their Abusers’, (November 2023) < https://lawcom.gov.uk/law-commission-to-review-defences-in-cases-of-domestic-homicide/ > accessed 14 July 2024
  3. Coroners and Justice Act 2009, s 54-56
  4. Alan Norrie, ‘The Coroners and Justice Act 2009 – Partial Defences to Murder (1) Loss of Control’, [2010] Criminal Law Review 275 at 277-9
  5. Jane Monckton Smith, ‘Intimate Partner Femicide: Using Foucauldian Analysis to Track an Eight Stage Progression to Homicide’ (2020), Violence Against Women, pp 1267-1285
  6. Evan Stark, Coercive Control: How Men Entrap Women in Personal Life: The Entrapment of Women in Personal Life, (1st edition, Oxford University Press, 2007) p 267
  7. British Medical Association, ‘Domestic Abuse’, (BMA Board of Science, June 2007, updated September 2014) page 22
  8. Coroners and Justice Act 2009, s 54(7)
  9. Office of National Statistics, ‘Domestic Abuse Victim Characteristics, England and Wales: Year Ending March 2023’, (November 2023) < https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabusevictimcharacteristicsenglandandwales/yearendingmarch2023 > accessed 15 July 2024
  10. Ibid 
  11. Clare Wade KC, Domestic Homicide Sentencing Review (CP 814 March 2023) para 9.5.2
  12. Coroners and Justice Act 2009, s 56(1)
  13. Law Commission, Murder, Manslaughter and Infanticide (Law Com No. 304, 2006) para 5.11
  14. Ibid at 5.18
  15. Ibid at 5.19
  16. Ibid at 5.20
  17. Ministry of Justice Consultation Paper 19/08, Murder, Manslaughter and Infanticide: Proposals for Reform of the Law, (2008) at para 36
  18. Coroners and Justice Act 2009, s 54(1)(a)
  19. Coroners and Justice Act 2009, s 54(1)(b)
  20. Coroners and Justice Act 2009, s 55(3)
  21. Coroners and Justice Act 2009, s 55(4)
  22. R v Duffy [1949] 1 All ER 923
  23. Coroners and Justice Act 2009, s 54(1)(c)
  24. Coroners and Justice Act 2009, s 54(2)
  25. Evan Stark, Coercive Control: How Men Entrap Women in Personal Life: The Entrapment of Women in Personal Life, (1st edition, Oxford University Press, 2007) p 267
  26. R v Clinton [2012] QB 1
  27. R v Ahluwalia [1992] EWCA Crim 1
  28. Lenore. E. Walker, ‘Battered Woman Syndrome’, (2009) Psychiatric Times Vol 26 No 7
  29. A-G for Jersey v Holley [2005] EWCA Crim 1880
  30. Coroners and Justice Act 2009, s 54(1)(c), (3)
  31. R v Rejmanski; Gassman [2017] EWCA Crim 2061
  32. R v Sargeant [2019] EWCA Crim 1088, [45]
  33. Ibid, [8]
  34. Ibid, [16]-[20]
  35. Ibid, [45]
  36. Serious Crime Act 2015, s 75A(1)(a)
  37. Serious Crime Act 2015, s 76
  38. Law Commission, Partial Defences to Murder, (Law Com No. 290, August 2004) para 3.22
  39. George Mousourakis, Criminal Responsibility and Partial Excuses, (1st edition, Routledge, 1998) Ch 3-5
  40. Kimberly Kessler Ferzan, ‘Provocateurs’, All Faculty Scholarship (2013) 7 Crim. L. & Phil. 597
  41. Alan Norrie, ‘The Coroners and Justice Act 2009 – Partial Defences to Murder (1) Loss of Control’, [2010] Criminal Law Review 275 at 277-9
  42. Law Commission, Partial Defences to Murder, (Law Com No. 290, August 2004) para 3.15
  43. Clare Wade KC, Domestic Homicide Sentencing Review (CP 814 March 2023) page 105
  44. Domestic Abuse Act 2021, s 1

This site has 0 comments...

0 Comments