I. Introduction
In October 2010, the UK Parliament implemented a package of homicide law reforms for England, Wales, and Northern Ireland, a main aim of which was to tackle serious concerns with the gendered operation of the law. The reforms sought to address a long-standing criticism that the English law of homicide had failed adequately to accommodate the contexts in which women kill an abusive male partner, whilst simultaneously all too readily accommodating the excuses of jealous and controlling men who kill a female intimate partner.Footnote 1 Introduced by the Coroners and Justice Act 2009 (“the 2009 Act”), the reforms saw the abolition of the much criticised partial defence of provocation in its old guise, and the formulation of a new partial defence of “loss of control” that incorporates some features of, but reformulates and goes beyond, the old partial defence. In the wake of the 2009 Act, it is important to evaluate the extent to which the reforms have led to meaningful change in practice. In focus here is the way in which sexual infidelity-related conduct triggering the killing is considered by judges properly to influence convicted murderers' culpability, through the sentence imposed. Such analysis is particularly significant in the light of (1) recent research highlighting the unintended consequences of homicide law reform in comparable jurisdictions, such as Victoria (Australia),Footnote 2 and (2) concerns that the abolition of provocation may merely lead to a transfer of similar gendered discourses and narratives of excuse to the sentencing stage of the justice process.Footnote 3
In examining the impact of the reforms, a key focus is how the operation of the new loss-of-control partial defence has addressed the long-held concerns just mentioned about the treatment of sexual infidelity, when it has led to homicide. The 2009 Act sought dramatically to reduce the relevance of sexual infidelity-related evidence as a basis for excusing murder, following a loss of self-control by the perpetrator. When words or conduct constituting sexual infidelity triggered the defendant's loss of self-control in killing the victim, the jury is now to disregard this evidence in deciding whether murder is to be reduced to manslaughter on the grounds of loss of control.Footnote 4 This article examines the implications of this legal change for sentencing in murder cases. In particular, we focus on post-2009 cases in which a jury rejected the loss-of-control plea and convicted of murder, where the sole or main evidence for the loss of control related to sexual infidelity. We argue that, in sentencing offenders in the post-reform period, the higher courts have failed to carry forward the spirit of the reforms respecting the substantive law, by effecting a corresponding change in sentencing practice. Disappointingly, the English higher courts have treated the change in the substantive law as a purely “technical” one, relevant only to the legal grounds on which murder may or may not be reduced to manslaughter. They have not regarded the change as entailing or demanding a more general shift in moral thinking concerning the relative seriousness of murders committed in response to sexual infidelity-related evidence. In consequence, the courts have continued to regard evidence of sexual infidelity as in principle having the potential to constitute grave provocation, justifying a significantly lower minimum term of imprisonment in murder cases. We believe that this approach to sentencing wrongly ignores the spirit, if not the letter, of the change in the substantive law governing the relevance of evidence of sexual infidelity to the loss-of-control defence in murder cases.
II. Sexual Infidelity and the Partial Defence of Loss of Control
In English law, by virtue of reforms brought about by the 2009 Act, murder will be reduced to manslaughter if the partial defence of “loss of control” applies. To have this effect, s. 54 of the 2009 Act requires amongst other things that the defendant's loss of control at the relevant timeFootnote 5 must have had one of two qualifying triggers.Footnote 6 A qualifying trigger has two elements to it but, for the purposes of this analysis, only one is significant. Under s. 55, the trigger can be a fear of serious violence from the victim, an extension beyond the scope of the old law which dealt only in the currency of provoked anger at something already said or done, and not fear of something anticipated. The inclusion of “fear of serious violence” as a qualifying trigger in the new loss-of-control defence sought to cater primarily for circumstances in which an abused woman kills, by recognising “the close connection between the emotions of anger and fear and thus between provocation and self-defence”.Footnote 7
Alternatively, the trigger can be something “done or said” (or a mixture of actions and words) that constituted “circumstances of an extremely grave character, and … caused D [the defendant] to have a justifiable sense of being seriously wronged”. So far as this second trigger is concerned, the 2009 Act adopts a special position in relation to what it calls “sexual infidelity” as a potential source of something “done or said” that might meet the qualifying trigger condition. Section 55(6)(c) stipulates that, when deciding whether a “qualifying trigger” is present, “the fact that a thing said or done constituted sexual infidelity is to be disregarded”. In justifying the inclusion of this exclusionary section in the new partial defence, and in distancing the new law from the problems associated with its predecessor (the provocation defence), at the time of its introduction, the Ministry of Justice commented: ‘The Government does not accept that sexual infidelity should ever provide the basis for a partial defence to murder. We therefore remain committed to making it clear – on the face of statute – that sexual infidelity should not provide an excuse for killing.’Footnote 8
This provision clearly has important implications for the scope of the loss-of-control defence to murder in law, some of which have been explored by the Court of Appeal in EnglandFootnote 9 as well as by commentators.Footnote 10 It has been held by the Court of Appeal that the provision does not make evidence of sexual infidelity wholly irrelevant to a plea of loss of control. The provision only bites with full force when evidence of sexual infidelity in itself, or as such, provides the trigger for the defendant's plea.Footnote 11 Where, by contrast, such evidence is simply a part of what might be called a broader or more complex “provocation narrative”, the evidence may be admissible as a part of the narrative that constitutes the qualifying trigger for the defendant's loss-of-control plea. So, on the one hand, if the account of the defendant's actions ran no further than saying “I lost control and killed her when she admitted adultery”, the jury would be obliged to disregard the admission as evidence of a qualifying trigger.Footnote 12 On the other hand, if the defendant were to say “It was when she admitted having had an affair with my 14-year-old son that I lost control and killed her”, the position would be different. In the latter kind of example, in the words of the Court of Appeal, evidence of “sexual infidelity is integral to and forms an essential part of the context in which to make a just evaluation whether a qualifying trigger properly falls within the ambit of subsections 55(3) and (4)”.Footnote 13
This brief discussion of the relevant provisions within the new defence provides the legal background that frames the main focus of our analysis: an examination of the implications of s. 55(6)(c) for sentencing in murder cases where a defendant has killed in response to prolonged family violence, or where the lethal violence was preceded by an act (actual or alleged) of sexual infidelity. As this analysis is closely tied to and influenced by sentencing patterns for homicide offences in England and Wales, we will start with a broader examination of the sentencing regime for murder as it affects abused women, rather than jealous and violent male partners.
III. Sentencing in Murder Cases: The Invisible Abused Woman
Following a murder conviction in England and Wales, the trial judge must impose the mandatory life sentence and, within that, a minimum term in prison that the offender must serve before being considered for release.Footnote 14 In setting this term, the trial judge must bear in mind (aside from time already spent in custody) the seriousness of the offence and of any others associated with it, and the guidelines on sentence lengths in murder cases provided by Sch. 21 to the Criminal Justice Act 2003 (“the 2003 Act”).Footnote 15 Schedule 21 to the 2003 Act sets out in considerable detail starting points in sentencing for murder, along with aggravating and mitigating factors to be taken into account so far as these were not already considered when determining the starting point. It is not necessary to set out the entirety of Sch. 21 here, but some key points should be mentioned.
To begin with, the starting points are largely determined by a combination of two factors: the defendant's age at the time of the offence and the presence (or absence) of key aggravating factors. So, for example, at the top end of the scale, if the offender was over 21 years old at the time of the offence, and the judge considers the seriousness of the offence(s) to be “exceptionally high”, the right starting point is a whole-life order.Footnote 16 At the other end of the starting-point scale, if the offender was under 18 at the time of the offence, the appropriate starting point is 12 years' imprisonment. Other than age, it is significant that there is no starting point in Sch. 21 dictated by a mitigating factor. So, for example, that the defendant acted in fear of serious violence (but had not lost their control at the time of the killing, and was thus not eligible to plead the loss-of-control partial defence) will not in itself justify a lower starting point. Schedule 21(11)(e) establishes merely that mitigating factors that “may be relevant” once the starting point has been determined include, alongside evidence of mental disorder or disability (Sch. (11(c))), “the fact that the offender acted to any extent in self-defence or in fear of violence”. This arguably very weak attempt to take into account circumstances that will include those in which abused women may kill their abusive partners hardly matches the effort devoted to carving out a partial defence to murder based, when a loss of control is added to the picture, on this very ground.Footnote 17 What is more, that mitigating factor must be seen in the light of the countervailing provision in Sch. 21(10)(a) indicating that one aggravating factor that may be relevant to the sentence is “a significant degree of planning or premeditation”. It is, of course, possible that the courts may take the view that, where an abused woman has had to plan the killing of her abuser, because she is hardly likely to prevail in a spontaneous confrontation, s. 10(a) will not be relevant. The difficulties are compounded, though, by the addition of a new (higher) starting point for murder by the Criminal Justice Act 2003 (Mandatory Life Sentence: Determination of Minimum Term) Order 2010. For an offender aged 18 or over at the time of the offence, a starting point of 25 years' imprisonment is to be regarded as normal where the defendant “took a knife or other weapon to the scene intending to – (a) commit any offence, or (b) have it available to use as a weapon, and used that knife or other weapon in committing the murder”.
In examining the negative impact that this approach to sentencing for murder in England and Wales is likely to have on women who kill an abuser, despite attempts to reform homicide law to better cater to this unique category of defendant, a consideration of past cases is useful. In the well-known case of Ahluwalia,Footnote 18 for example, the female defendant – a victim of very serious abuse over a long period at the hands of her husband – took a can of petrol that she had stored in a garage and set light to him, killing him. She was initially convicted of murder, although the conviction was quashed and a retrial ordered following the emergence of new evidence that severe depression had affected her actions. On retrial, the prosecution accepted her plea of diminished responsibility. Her original tariff sentence for murder was set at 12 years – surprisingly high given the circumstances of the offence. Her sentence for manslaughter on the grounds of diminished responsibility on retrial was set at three years and four months.
An interesting question arises concerning how the sentencing issues in a murder case mirroring the circumstances in Ahluwalia (prior to the discovery of the evidence of severe depression) would be addressed in the wake of the 2009 Act and the application of Sch. 21. At the time of the offence Ahluwalia was over 21 years old, and took a weapon to the scene of the crime with a view to using it to commit not just any offence, but murder.Footnote 19 Further, there was no evidence of a loss of control. In theory then, according to Sch. 21, the starting point in sentencing should be 25 years' imprisonment. There was also a further aggravating feature, namely the element of premeditation (Sch. 10(a)) demonstrated by her conduct in storing the petrol in the garage in the first place, although this would be offset by the mitigating factor in Sch. 11(d) provided by “the fact that offender was provoked (for example by prolonged stress)”. It seems contrary to the spirit of the 2009 reforms that the application of Sch. 21 should mean that sentencing in a case of this kind could proceed in such a manner. It raises the very real possibility that the minimum term would be set at, perhaps, 20–22 years' imprisonment, not far short of double the tariff sentence that Ahluwalia originally received in 1989. No one has explained why, in cases of this kind, such a dramatic increase in the starting point for the minimum term is warranted.
This example highlights the problematic reality that the sentencing starting points and accompanying guidance set out in Sch. 21 are shaped almost exclusively by thinking about offenders who will in all probability be male and have committed the worst kinds of murder. No attention whatsoever was paid in the development of the starting points to the typical circumstances in which women are most likely to kill an abusive male partner (some 5% of male homicide victims are killed by their partner or ex-partner).Footnote 20 The vague and exiguous provisions relating to mitigation in Sch. 21 do almost nothing to make up for this glaring omission, and now sit very uneasily alongside Parliament's aims in crafting the loss-of-control partial defence.
IV. Sexual Infidelity and Sentencing under Schedule 21
In evaluating how the reforms have affected judicial consideration of sexual infidelity-related evidence at the sentencing stage for murder, the following section first traces the consideration of such evidence in English courts from the nineteenth century up to the time immediately prior to the 2003 Act. This analysis is used to chart the historical view taken on the extent to which mitigation should be afforded to men who kill in response to an (alleged) act of sexual infidelity, but also to contextualise the subsequent examination of sentencing practices and judicial discourse post 2009. The latter analysis considers the extent to which, when the 2009 reforms are examined alongside the sentencing provisions of the 2003 Act, meaningful change has been achieved in the law's response to intimate homicides motivated by sexual infidelity.
A. Sentencing Practices and Judicial Attitudes Prior to the 2003 Act
You could be forgiven for thinking that, the further back in time one goes, the more lenient one is likely to find the treatment by judges of men provoked to kill by, in some form, the sexual infidelity of (ex) partners. This view finds superficial support in the well-known rule that for a man to catch his wife in the very act of adultery was provocation “exceeding great”,Footnote 21 jealousy being “the rage of the man” and adultery “the highest invasion of property”.Footnote 22 In fact, the later historical picture is more complex. Whilst the old view long persisted – and may still persist – in folk memory, Martin Weiner has argued that judges in the latter part of the nineteenth century often defied popular opinion by adopting a hard line with such offenders.Footnote 23 The Prisoners' Counsel Act of 1836 had, for the first time, allowed defence counsel not only to tackle matters of law, together with examining and cross-examining witnesses (both practices developed in the eighteenth century), but also to address the jury in felony cases.Footnote 24 That led judges to seek to counter-balance this important pro-defendant influence by themselves becoming pro-prosecution and hence “pro-authority”, not least through the development of the power to sum up cases.Footnote 25 While this was an across-the-board development, it had particular implications for domestic homicide cases where a man had killed his wife or partner. This was because, in their capacity as moral as well as legal authorities, late-nineteenth-century judges saw themselves as entitled, and indeed bound, to play a part in the use of denunciation and deterrence to root out what was widely taken to be the “lower-class” understanding of marriage: “… not [a relationship] of mutual dependence and intercourse of protection and comfort, but of absolute control on the one hand, and abject submission on the other.”Footnote 26 Consequently, the rule that words alone could not be sufficient provocation was used at the time to prevent a partial defence of provocation being pleaded by men who had killed wives for being verbally abusive or insulting, or merely for being drunk.Footnote 27 This included, for example, a case in which the words in question expressed the victim's jealousy of the defendant's interest in former girlfriends.Footnote 28 The trial judge, Baron Parke, stated to the jury that the “law was clear” that, even where accompanied by minor blows, abusive words would not be sufficient provocation.Footnote 29 The jury found the defendant guilty of murder, following this direction, but recommended mercy. Nonetheless, the man was hanged.Footnote 30
This judicial approach remained relatively constant during the early part of the twentieth century, with between a third and a half of men executed for murder annually between 1900 and 1950 having been found guilty of killing their wives or partners,Footnote 31 and was still evident in the famous mid-twentieth-century House of Lords case of Holmes v D.P.P.Footnote 32 In Holmes, the defendant killed his wife by strangulation, having subdued her with a blow from a hammer. He had arranged to meet a lover shortly afterwards. According to his account (there was no corroboration), his wife had admitted being unfaithful to him at the time of the fatal quarrel. In cross-examination, he admitted intending to kill his wife, following a “loss of temper”. The judge refused to put the issue of provocation to the jury, and was later held to have been correct in law to refuse to do so. In a well-known passage, Viscount Simon held that:
[A] sudden confession of adultery without more can never constitute provocation of a sort which might reduce murder to manslaughter . . . . [W]e have left behind us the age when the wife's subjection to her husband was regarded by the law as the basis of the marital relation . . . . [A]s society advances, it ought to call for a higher measure of self-control in all cases.Footnote 33
From the mid-twentieth century onwards, the understanding that sexual infidelity-related provocation could be grave provocation began to have more influence. This was not only because, from the implementation of s. 3 of the Homicide Act 1957 onwards, words – however trivial a provocation they constituted – had to be put to the jury as provocation that might reduce murder to manslaughter.Footnote 34 We may also surmise that judges ceased to regard it as part of their role to act as moral educators as well as legal authorities, and hence no longer attempted to reduce, in summing up, the growing influence of sexual infidelity-related provocation on jury verdicts. To intervene to such an end would now be inconsistent with judges' increasingly significant role as “impartial moderators” in criminal trials.Footnote 35 Further, following conviction, late nineteenth-century judges saw themselves as relatively free to ignore a jury's manslaughter verdict or recommendation for mercy when deciding on their approach to sentence: the “jury contempt” approach. By contrast, in the second half of the twentieth century, judges felt more obliged to respect jury verdicts in their sentence: the “jury deference” approach. That meant to some extent rationalising, rather than condemning, the defendant's actions when sentencing, when those actions had been treated leniently in law by juries bringing in a manslaughter verdict. Hence, we find an increasing emphasis on (in Viscount Simon's words) “the effect of provocation on human frailty”.Footnote 36 In this way, the typically male view of the gravity of provocation constituted by sexual infidelity-related evidence – disapproved of by judges in the late nineteenth centuryFootnote 37 – became a basis for significant mitigation of sentence.
A good example of this is R. v Melentin. Footnote 38 In this case, the defendant's wife (and eventual victim) had an affair with another man or men whilst the defendant was in prison for a dishonesty offence. The defendant sought a reconciliation. According to the defendant's unchallenged account, they went upstairs to have sexual intercourse, but the victim taunted him about his sexual prowess. She extended a piece of sash cord into a taught, upright position to represent her lover's penis, then shortened it and allowed it to droop to represent the defendant's penis. In response to this conduct, Melentin lost self-control and strangled his wife with the sash cord. He was acquitted of murder on the grounds of provocation. The trial judge sentenced him to five years' imprisonment for manslaughter – a sentence that was appealed by the defence. The Appeal Court remarked of the victim's alleged provocation “to taunt a man about his lack of sexual inclination or prowess does involve striking at his character and personality at its most vulnerable”.Footnote 39 Taking this into account, along with the victim's alleged boasting about previous affairs, the original sentence imposed was reduced to a four-year term.
In cases such as Melentin, a late-nineteenth-century tradition of judicial condemnation for engaging in such morally degraded and violent conduct has been replaced by a practice in which the same behaviour is cast, for sentencing purposes, in a highly favourable excusatory light. That light is cast by the re-emergence of the much older tradition of viewing sexual jealousy as a man's rage, in modern psychiatric garb:
If a man's wife sleeps with someone else … [h]e will be compared, he will be judged in that one place where he was secure, most vulnerable because most himself … . The fantasy is that [sex] may give possession of the person … but why should sexual relations be thought to be the key to such extraordinary power? It is because it is thought to be … an assurance of unconditional, unjudgmental attentive acceptance.Footnote 40
More recently, this sympathetic (and arguably highly problematic) approach was in effect endorsed in the important decision, following the Attorney General's reference to the Court of Appeal, in Suratan, Humes and Wilkinson.Footnote 41 In these cases, the Attorney General sought a finding of unduly lenient sentences in three cases, two of which involved jury findings of manslaughter by reason of provocation constituted by sexual infidelity-related evidence. Relying in part on the observations of Viscount Simon in Holmes v D.P.P.,Footnote 42 the prosecution case was that the normal starting point for sentencing in such cases that had developed in recent years – sentences of between five and seven years' imprisonment – was too lenient and that jealousy and possessiveness were no longer acceptable reasons for losing one's self-control and committing lethal violence.Footnote 43 The Court of Appeal, however, disagreed, ruling that each case was an example of “uncharacteristic violence” and should be sentenced as such.Footnote 44
Further, in the two relevant cases, Humes and Wilkinson, the Court of Appeal largely airbrushed the sexual infidelity basis for the provocation pleas out of the picture, in discussing the applicable sentencing principles.Footnote 45 The Court of Appeal chose instead to place its main emphasis on characteristics that militated in favour of sentencing leniency.Footnote 46 Adopting the “jury deference” approach outlined above, the Court of Appeal stressed that, in sentencing for manslaughter under provocation, the judge must take into account, for example, not only the fact that the defendant was found to have lost control, but that the jury must also have found that:
the defendant's loss of control was reasonable in all the circumstances, even bearing in mind that people are expected to exercise reasonable control over their emotions, and that as society advances it ought to call for a higher measure of self-control … [and] that the circumstances were such as to make the loss of self-control sufficiently excusable to reduce the gravity of the defendant's offence from murder to manslaughter.Footnote 47
This approach sits uneasily alongside the Court of Appeal's expression of the view, only a few paragraphs earlier, that matters relevant to the availability of the provocation defence are one thing, and matters relevant to sentence are another: “… we cannot see how this [the development of a less forgiving attitude towards jealous rage] provides an argument that there should be heavier sentences once a verdict of manslaughter by reason of provocation has been entered”.Footnote 48
Moreover, the Court of Appeal's approach entirely abdicates what ought to be its role in reflecting, through its sentencing practice, deep-seated (i.e. not merely populist or transient) shifts in opinion on relevant moral matters, even when such shifts appear to contradict an individual jury's opinion in an individual case.Footnote 49 Indeed, that most judges value this role has been subject to empirical testing and verification.Footnote 50
So far as we are concerned with the individual cases that were the subject matter of the Attorney General's reference, they provide (as we shall see) an instructive contrast, at least in some respects, to the approach more recently taken following the changes to sentencing principles in murder cases post 2003. Due to space constraints, only one of these cases can be considered in detail here. In the case of Humes, the defendant, a solicitor whose marriage was in difficulty because he was a workaholic, discovered that his wife (the eventual victim) was on intimate terms with her karate instructor. When Humes called at the family home (he had been staying at a hotel after the victim had asked him to leave), the victim told him that their relationship was finished and that she had switched her affections to another man. The defendant alleged that his wife said to him, “By the time you get back, in a week, I'll have slept with him”. In response, Humes lost control and stabbed his wife 11 times with a bread knife, continuing his attack as she sought to escape from the kitchen to the dining room. Two of the stab wounds penetrated the victim's whole body, including the fatal wound which involved a double thrust through the heart and then the lung. Part of the attack was witnessed by the couple's eldest daughter (aged 14), who became covered in blood attempting to revive her mother, and by the other three children, aged between 12 and two and a half. For reasons considered below, the prosecution accepted the defendant's offer to plead guilty to manslaughter by reason of provocation.
The defendant's sentence of seven years' imprisonment was upheld by the Court of Appeal, which described the sentence as not even lenient, let alone unduly lenient. On the one hand, it was conceded that the judge had been entitled to regard as an aggravating factor the fact that the children had to witness such a brutal attack on their mother (one child removed the bread knife from the defendant's stomach when he stabbed himself following the attack), together with the impact on the victim's family, and on her twin sister in particular. That justified a sentence at the upper end of the normal five-to-seven-year range. On the other hand, the Court of Appeal pointed to the stress that the defendant had been under at the time of the offence, that he had a good character and no previous history of violence, adding: “In accepting the offender's plea, the prosecutor did not dispute that the offender's loss of control was reasonable in all the circumstances and was sufficiently excusable to reduce the gravity of the offence. We find it difficult to understand how consistently with that the degree of provocation can be said to be slight.”Footnote 51
With respect, this argument is technical and obfuscatory. It is true that the prosecution can accept a plea of guilty of manslaughter in such cases, if there is “insufficient evidence” to press on with a murder charge.Footnote 52 However, in this case, the prosecution was no doubt moved to accept a plea of guilty because the defendant had agreed in exchange not to press a plea of diminished responsibility, and in part because to do so would avoid any prospect of the defendant's children having to appear as witnesses. As we will see, not only would such a case necessarily now end in a conviction for murder, but it would attract a minimum prison term greatly in excess of that which Humes was then expected to serve. Most disappointing of all is the Court of Appeal's steadfast refusal in the case to comment in any significant way on the sexual infidelity basis for the defendant's plea, and its determination instead, as a basis for mitigation, to hide behind the jury's (or the prosecution's) acceptance that all the elements of the provocation defence were or might have been present. As Mandy Burton argues, the appeal gave the Court “a clear opportunity to state that jealousy should afford no mitigation; however, their ruling implicitly approved the mitigation afforded to jealous men who kill”.Footnote 53
B. Sexual Infidelity Evidence and Sentencing Post 2003
The new partial defence of loss of control, and specifically s. 55(6)(c) of the 2009 Act, is, of course, designed to arrest the twentieth- and early-twenty-first-century development of leniency in provocation cases involving sexual infidelity-related evidence, so far as the substantive law is concerned. The 2009 Act was passed in response to a growing body of scholarship recognising the inherently gender-biased nature of the law of provocation and the injustice that stems from its operation.Footnote 54 Post 2009, cases with facts such as those in Humes will involve the sentencing principles applicable to murder, not those applicable to manslaughter. So, how, if at all, will the reforms, and the ensuing re-categorisation of these cases as ones of murder rather of manslaughter, change the position with regard to sentencing?
The 2003 Act has ensured that a radical transformation has been effected in the way such cases are to be approached at the sentencing stage. To begin with, Sch. 21 to the 2003 Act indicates that the starting point – were the Humes case to occur now – would have to be a minimum term of 15 years' imprisonment.Footnote 55 Mitigating features in the case would certainly include the spontaneous and unplanned nature of the attack,Footnote 56 although the absence of any history of violence in the defendant's past should not, as such, be a “mitigating” factor.Footnote 57 Schedule 21(11)(d) also refers to the need to consider whether the defendant was provoked “for example, by prolonged stress”.Footnote 58 However, these factors would be counter-balanced by the fact that Humes's attack was a “sustained”Footnote 59 – even “savage”,Footnote 60 or “ferocious”Footnote 61 – attack. In R. v Genestin (also an adultery case),Footnote 62 such an attack warranted the passing of a minimum term of 20 years' imprisonment even though the attack was also unpremeditated.Footnote 63 Even when the attack was not ferocious, an element of premeditation may also warrant a minimum term higher than 15 years' imprisonment. For example, in R v Taylor,Footnote 64 where the defendant strangled his former girlfriend after she refused to resume their relationship, a minimum term of 18 years' imprisonment was upheld in light of (amongst other things) the defendant's premeditated use of a belt to kill his estranged partner.Footnote 65
Furthermore, a murder committed in front of children of the family now seemingly counts as a more substantial aggravating factor than it was treated as being in Humes. In Attorney General's Reference (No 23 of 2011),Footnote 66 the defendant had been in a relationship for four years with the victim, and they had a three-year-old child. During the period of separation, the victim had a relationship with another man. The defendant crept into his estranged partner's home in the early hours, following which neighbours heard sounds of screaming and violence (and a child crying) over a period of about 40 minutes. The victim was later found to have 27 injuries, and died from blunt impact injuries to the head. The trial judge set a minimum term of 15 years' imprisonment. On appeal against that sentence as unduly lenient, the Court of Appeal raised the minimum term to 20 years. Lord Judge L.C.J. said:
There are a number of aggravating features present … a history of violence … some evidence of sexual possessiveness by the offender; an invasion of the deceased's home at night; a prolonged, determined and persistent beating; the presence of the young child at the beating … . We cannot guess the long-term damage that will have been caused … . There is nothing in this case which can be said to amount in any way to provocation by the deceased or which would in any way serve to mitigate the offence.Footnote 67
We take no stand on whether, in such cases, the minimum term is currently too harsh, or about right (or even, for some people perhaps, still too lenient). That issue aside, these remarks appear at first sight to constitute important evidence that infidelity-related evidence is now no longer the ground for substantial mitigation that it once was.Footnote 68 For example, in Attorney General's Reference (No 73 of 2009),Footnote 69 where, to punish his wife, the defendant stabbed his 15-year-old stepson to death, and his stepdaughter non-fatally, a 16-year minimum term was increased on appeal to a 25-year minimum. Lord Judge L.C.J. said:
Our attention today has been drawn to the fact that there is no evidence that the offender had ever used violence in the house before. However, it is clear, whether he had done so or not, that he did not need to use violence. The house was filled with fear of a dominating man … This was a remorseless killing of a defenceless boy … the offence was aggravated in the extreme by the circumstances of the associated offence.Footnote 70
Such changes in the way that domestic abusers' killings are evaluated, for sentencing purposes, are welcome in that they recognise a wider range of non-physical behaviour as abusive. Moreover, they do something to align the approach to sentencing for murder under the 2003 Act with the spirit and intent of the 2009 Act in relation to its formulation of the new partial defence of loss of control. However, they must be set against what we regard as a problematic approach to sentencing in post-2009 cases of this kind, where the question is whether s. 55(6)(c) has in itself any implications for sentencing offenders whose lethal loss of control was triggered by something said or done constituting sexual infidelity.
In R. v Haywood,Footnote 71 the defendant, 69, began a relationship with W, after his wife died. He bought them a house together. W took a lover, and then a second lover, and informed the defendant that she was going to set up home with her second lover. The defendant armed himself with an iron bar and sought to disguise himself. He then fatally struck his partner's new lover several times on the head and neck as the victim was leaving work. On appeal, in justifying a reduction of the minimum term imposed from 11 years' imprisonment to nine, Aitkens L.J. said:
There is no escape from the fact that, albeit under immense emotional strain, the appellant deliberately went to the hotel that evening, armed with an iron bar and dressed in a manner that he hoped would disguise him … . However, the judge described carefully and properly, in our view, the emotional turmoil and the mental state of the appellant which … was none of his doing. That was not provocation such as to amount to a defence to murder because the jury rejected that defence. But, in our view, there was the greatest possible provocation in the non technical sense.Footnote 72
In our view, the strong element of premeditation in this case ought to have been regarded as largely nullifying the effect of any provocation, but our focus is the view taken by Aitken L.J. of the proper relationship between what is regarded as “technical” provocation, namely provocation as it is relevant to the substantive law – the defence to murder – and “non-technical” provocation as it bears on the sentencing process.Footnote 73 Can it be right to say, in particular, that, although as a matter of law words or conduct constituting sexual infidelity are to be wholly disregarded by the jury, in so far as they were the trigger for the loss of control, such words or conduct can be regarded, when sentencing for murder, as “the greatest possible provocation” (to use Aitkens L.J.'s phrase)?
This issue was addressed directly by the Lord Chief Justice in Attorney General's Reference (No 23 of 2011),Footnote 74 the facts of which were given above.Footnote 75 Lord Judge C.J. said:
In short, it [55(6)(c)] is concerned with the substantive criminal offence of murder, not with the determination of the minimum term where murder is admitted or proved. Paragraph 11 of Schedule 21 remains in force. Even if not amounting to a defence of provocation, provocation may provide relevant mitigation to murder. That accords not only with common sense, it reflects the sentencing principle which allows for mitigation when the same material could not constitute a defence as, for example, provocation in the context of attempted murder or provocation in the context of causing grievous bodily harm. The circumstances in which provocation may serve to provide mitigation for an offence of murder are not closed as a result of section 55 of the 2009 Act.Footnote 76
Whilst, in a broad sense, this understanding of the law is correct, in our view, a reading more fully informed by a gendered perspective would have led to this understanding being modified in significant ways.
To begin with, it must be kept in mind that s. 11(d) of Sch. 21, whilst referring to “provocation”, appears to focus in point of mitigation neither on the gravity of the provocation offered nor on the role of the victim in being provocative. Rather, the focus is placed on the role of the defendant's mental state in making him or her perhaps peculiarly susceptible to provocation. The example given is provocation producing “prolonged stress”. We believe that this is a legislative steer towards an approach to the provocation issue in sentencing for murder that rightly places emphasis on the cumulative deleterious impact on the mind of repeated stressful events. In other words, it invites the sentencing judge to treat “provocation” as a background issue. The real mitigation in the foreground – the effect of prolonged stress – is analogous to (albeit not the same as) evidence of “any mental disorder or mental disability … [that] lowered [the defendant's] degree of culpability”, that forms the substance of the adjoining s. 11(c) of Sch. 21.Footnote 77 Of course, it will be said that the example of “prolonged stress” given in s. 11(d) is just that: an example, and an example that is only there to guide in any event; but, to repeat, our aim is to construct a gendered analysis of the applicable law, and so such objections will not detain us.
If this analysis is capable of gaining legal traction, then it opens up a new way of looking in principle at provocation-as-mitigation in murder cases post 2009. Central cases calling for mitigation will be those in which – and the drafters of s. 11(d) of Sch. 21 may well have had this in mind – women (or men) have suffered abuse at the hands of a partner or former partner over a long period, even though there was no loss of control at the time of the offence.Footnote 78 Whether or not such people can be described as suffering from a recognised medical condition, the enormously damaging mental effects caused by prolonged abuse are well known.Footnote 79 By way of contrast, relegated to the mitigation periphery will be cases in which, whatever the supposed “gravity” of the provocation, the defendant's response is aptly described in terms of a more or less spontaneous outburst of anger or rage – a reaction not attributable to the long-term build-up of stress. Perhaps some defendants who kill their (former) partners when their reaction is triggered by sexual infidelity-related evidence will claim to fall within the former category, when the infidelity-related conduct or words have been repeated over a long period and have had deleterious effects on their mental state in general. We suspect such cases will be very much in the minority, but this issue takes us back to the principal question: the relationship between s. 55(6)(c) of the 2009 Act and sexual infidelity-related evidence as mitigating evidence in sentencing for murder.
Quite simply, if Lord Judge's approach in Attorney General's Reference (No 23 of 2011) Footnote 80 permits a sentencing judge, as in R v Haywood,Footnote 81 to regard sexual infidelity-related evidence as “the greatest possible provocation”, then something has gone badly wrong in the operation of the law as a whole. Sentencing principles appear scarcely to have moved on, morally, from the seventeenth century, when the Court in R. v Manning Footnote 82 directed that Manning – convicted of manslaughter having lost self-control when he caught his wife in the act of adultery – should be punished only by light burning of the hand, as there “could not be greater provocation than this”.Footnote 83 Most problematically, this position is likely to remain unchanged so long as judges continue to point to the relevance of s. 55(6)(c) of the 2009 Act to the substantive law, merely then to contrast that with the relevance of Sch. 21 to sentencing. What is now required is an integrated, holistic approach to the issue. As David Thomas remarked long ago, “A reconstruction of the law of homicide [ought to] begin with a decision on the nature of the sentencing structure which is to be attached to the offences concerned”.Footnote 84
One basis for such an approach can be found in s. 55(6) itself. Alongside sexual infidelity-related evidence, also to be disregarded as a possible qualifying trigger is a fear of serious violence, if the violence itself was incited by the defendant (s. 55(6)(a)). Further, a sense of being seriously wronged by something done or said is not to be regarded as justifiable – and hence a qualifying trigger – if the defendant him or herself incited the thing done or said (s. 55(6)(b)). The latter rule, in particular, changed the common law, which had previously permitted evidence of “self-induced” provocation at trial.Footnote 85 Little, if any, credit in point of mitigation at the sentencing stage is likely ever to be given to a defendant who him or herself engineered an opportunity to take offence, worked themselves into a rage, and then killed in response to that offence.Footnote 86 Accordingly, judges should adopt the holistic view that the placement of sexual infidelity-related evidence directly alongside self-induced losses of self-control in s. 55(6) has implications not only for the directions given to juries, but also for sentencing in all such cases. The grouping together of these kinds of so-called “provocation”, as a matter of substantive law, should not be regarded as an accident. It can and should come to be regarded as reducing the seriousness of the provocation constituted by sexual infidelity-related evidence, as such, to insignificance: that is, to the level of “self-induced” and/or incited losses of self-control.
In introducing the legislation, whilst the Government of the day placed some emphasis, in explaining s. 55(6)(c), on the narrow effect it was to have on the substantive law,Footnote 87 Claire Ward M.P. went on to say:
The provision does not reflect a lack of trust in the jury; what it does reflect is the Government's determination to ensure that the law in this matter keeps pace with the times. In this day and age, it should not be possible for any person, regardless of gender or sexuality, to stand up in court and blame their partner – let us not forget that it is the partner that they themselves have killed – for having brought on their own death by having an affair.Footnote 88
This passage is not, of course, a piece of legislation in itself. Even so, we believe it is right to give a broad meaning to Ward's words when she speaks of the need to ensure that the law keeps pace “with the times”. That can be and should be taken to indicate that judges must adjust their sentencing philosophy to match that which now shapes the substantive law. Further, when Ward says that it should not be possible for anyone to “stand up in court and blame” a partner for having brought about their own death by engaging in sexual infidelity, this should be understood as a general moral claim about homicide trials, not just an indication of the Government's reasoning in relation to a particular substantive law provision. Ward's words can perfectly justifiably be understood as rightly applicable to the sentencing stage of the criminal process as much as to the process of reaching a verdict. In this respect, her words can also be construed as a source of implicit guidance to the effect that sexual infidelity-related evidence should have no bearing on mitigation in murder cases by virtue of the application of s. 55(6)(c), except in so far as it is part and parcel of a – necessarily rare – claim of “prolonged stress” bordering on mental disorder. For what is the alternative?
It is true, of course, that those who kill with the fault element for murder, in response to no more than sexual infidelity-related evidence, automatically post 2009 receive the highly stigmatic label of murder (other things being equal). However, in itself, that additional element of punitiveness in the substantive law does nothing to support the view that sexual infidelity-related evidence should continue to be regarded, at the sentencing stage, as evidence as capable of amounting to “the greatest possible provocation”. On the contrary, in our view, such an approach is likely to come to be regarded as running directly contrary to the Government's overall philosophy in this long-controversial area of the law of homicide.
In saying this, of course, we are implicitly endorsing the view that public policy considerations (such as the comparative treatment of men and women by criminal law and in sentencing) should play a highly significant role in influencing decisions on sentence in murder cases, notwithstanding the importance of personal mitigating factors in an individual case. In that, we simply follow the 6th Report of the House of Commons Justice Committee (2008–09) when it expressed the view that:
Parliament sets the framework for sentencing in legislation. Sentencing guidelines are a key element to how this legislation works in practice. It is vital that Parliament, representing the public voice, contributes to sentencing guidelines as they are produced and in doing so identifies the crucial issues of public confidence and the effectiveness of sentencing. We are convinced this is compatible with safeguarding the crucial discretion of sentencers to impose a sentence tailored to the individual case.Footnote 89
V. Conclusion
This article has examined how the English courts have historically sentenced men “provoked” to kill by sexual infidelity-related conduct on the part of their current or estranged partner, and to what extent sentencing practices have changed in the period following the implementation of the 2009 English homicide law reforms. Although we have subjected it to critique, the English courts' approach in sentencing for murder, post 2009, is in one way understandable, when considered in light of the policy underlying the 2003 sentencing legislation and guidance governing minimum starting points for murder in England and Wales. The sentencing regime for murder cases introduced by s. 269 (Sch. 21) of the Criminal Justice Act 2003 is draconian. As is shown throughout this analysis, particularly in relation to the sentencing of persons who kill in response to prolonged family violence, significant steps need to be taken to soften its impact in many cases, so that justice can be better achieved. However, in our view, the treatment of evidence of sexual infidelity-related evidence (almost always on the part of a female partner) as in principle capable of amounting to grave provocation is not a legitimate way to achieve this necessary softening effect. That approach simply threatens the integrity of the moral message that the change in the law in 2009 was in broad terms meant to bring about.
The importance of this analysis is that it highlights the difficulty of achieving meaningful reform to the law of homicide, without also considering the likely impact of sentencing legislation on the success of those reforms in practice. For this reason, and beyond the English context, we emphasise the importance of an approach to reform which considers not only the substantive law of homicide, but also sentencing legislation and guidance. This is essential when attempting to overcome pervasive gender biases in the law's operation, such as those that have come to be associated in many jurisdictions with the controversial partial defence of “provocation” or loss of control. Without such a holistic approach, it appears likely that any attempt to achieve meaningful change in practice at one stage may be undermined by a lack of consideration of the need for change at the other stage.
