Policy Papers

The ‘rough sex’ defence: lessons from history

Adrian Williamson |

  • RSS Feed Icon

Executive Summary

  • The recent tragic case of Grace Milane has drawn attention to cases where the defendant to a murder charge relies in his defence upon the allegation that the death occurred by accident during consensual ‘rough sex’.
  • This defence has echoes of two other areas of the criminal law, which each took decades to resolve. Learning from their histories should lead to a much more rapid resolution in this case.
  • In rape cases, the focus in the past was often upon the previous sexual behaviour of the victim, rather than the acts of the defendant.
  • In the law of provocation, defendants relied upon the alleged conduct of the deceased, who could not answer for herself, to reduce a charge of murder to a conviction for the lesser offence of manslaughter.
  • In both situations, Parliament ultimately has had to step in, to pass legislation which severely limited the scope of these defences, in order to protect female victims and enable effective prosecution of very serious offences.
  • Legislators should now consider taking a similarly prescriptive approach in relation to cases like Milane’s.


The recent case of Grace Milane has drawn attention to the so-called ‘rough sex’ defence. In November 2019, a man who had strangled the British backpacker and hidden her body inside a suitcase was found guilty of murder following a trial in New Zealand. His defence was that she had died accidentally during ‘rough sex’. In February 2020, the man was jailed for life for the murder.

The nature of this defence provoked outrage in some circles and calls for legal reform. Joan Smith complained in The Guardian that the case had ‘exposed something profoundly disturbing about the way women are treated in modern criminal justice systems. Many people in the UK and in New Zealand have followed the trial with visceral horror, disturbed by the ‘intense focus on the victim.’ Harriet Harman MP, and other politicians, have called for a change in the law, Harman observing that, ‘The new version of men being able to blame the woman they killed for her own death is by saying “she wanted the violence". And the difficulty is that there were two people there and only one of them gets to speak because he has killed her, and therefore her version of events is not heard.’

As these comments make clear, this defence combines two potentially objectionable components. The first is the concentration on the behaviour of the victim rather than the perpetrator. The second is that the relevant evidence comes entirely from the defendant. The deceased cannot contest what is being said about her. In fact, similar issues have arisen in the past in the law pertaining to rape and to provocation. Reviewing this legal history may provide assistance in dealing with cases such as Milane’s. What follows discusses only male perpetrators and female victims although, of course, both rape and provocation can occur in other contexts.

In rape cases there was, prior to legislative intervention in the 1970s, likewise ‘intense focus on the victim’, particularly where the complainant was alleged to be sexually promiscuous. Also, in the law of provocation many cases, until very recently, were dominated by evidence of alleged nagging or taunting by the deceased, evidence where ‘only one of them gets to speak because he has killed her’.

Before turning to these two areas of the law it may be helpful to summarise the broad legal context in which they each arise (it should be noted that in each case this is no more than a brief survey of the law as it stood in the late twentieth century, and not intended to be an authoritative guide to a complex and ever-changing area of the law).

Rape and homicide in law

Rape has been defined as ‘a person intentionally penetrat[ing] another's vagina, anus or mouth with a penis, without the other person's consent’. In some cases, the issue is identity, i.e. whether the defendant had sex with the complainant or with someone else. The defendant can, of course, maintain a complete denial that sex took place at all. But in most cases, the central issue is consent. The defendant accepts that he had sex with the complainant, but says that she agreed to what took place.

Murder is essentially the intentional and unlawful killing of another. There are numerous defences to the charge. Some are complete defences, such as accident, self-defence, and complete denial (‘I wasn’t there, I didn’t do it’). Others are partial defences, such as diminished responsibility and provocation. Where a partial defence is put forward, the defendant, if successful in his defence, will see the charge of murder reduced to one of manslaughter. This distinction matters very much, because murder continues to attract an automatic sentence of life imprisonment, whether mercy killing or gangland execution. Manslaughter, on the other hand, receives a discretionary sentence, and, in practice, sentences vary widely. In this, and other, respects the historic shadow of the gallows continues to hang over the law of murder. Until 1957 all – and until 1964 many – convictions for murder were automatically visited with a death sentence (albeit many of these sentences were commuted in practice). The murder/manslaughter distinction was, therefore, a matter of life and death. This distinction has been jealously guarded in the post-1964 world, with politicians anxious to avoid charges of softness on crime by insisting that ‘life must mean life’ for murder.

It is important to note two general matters about a criminal trial. Firstly, the burden on most issues remains on the prosecution throughout, and, in respect of matters where they bear the burden, the prosecution must satisfy the jury so that they are sure of the relevant matter. Thus, if consent is raised in a rape case, or provocation in a murder trial, the prosecution must satisfy the jury to this standard (of being sure) that the complainant did not consent or that the defendant was not provoked. The second point, really a corollary of the first, is that there is no burden of proof on the defendant, who need prove nothing. But he can raise matters intended to cast doubt on the prosecution’s case and is, generally, permitted to do so, however fanciful those matters may be. The trial judge is not, ordinarily, allowed to exclude such defence material on the grounds that s/he finds it unconvincing.

Finally, it is worth saying something about the vexed issue of consent. The law has always been that people can consent to the infliction of serious violence in certain circumstances, for example a surgical operation or ‘manly sports’, although you cannot, of course, consent to being killed. However, the law has narrowly confined the limits of this principle.  This issue arose in the ‘Operation Spanner’ case of Brown in 1994. This concerned a group of gay men who engaged in violent, but consensual, BDSM practices. They were charged with various offences of causing actual or grievous bodily harm. The House of Lords held, that, in the absence of a good reason, the victim's consent was no defence and that the satisfying of sado-masochistic desires did not amount to a good reason. This case was, of course, decided in a very different climate from today, and the attitude of the courts to gay sex might now be different. Even then, two of the five law lords dissented, and all of them agreed that a prosecutor had to prove lack of consent to secure a conviction for mere assault.

Trying rape cases in the 1970s

In 1987, the feminist historian Anna Clark wrote a book called Women's silence, men's violence. This examined rape cases tried at the Old Bailey in the period from1770 to1845. She concluded that ‘judges and juries refused to take rape victims seriously’. However, Clark could just as easily have been discussing the treatment of sexual assault charges in the 1970s as the 1770s. Another study at about the same time by Zsuzsanna Adler, Rape on trial, presented a similarly depressing picture, albeit that she was reviewing trials at the Old Bailey in the 1970s. The victims, not the alleged rapists, were put on trial. As Adler observed, to stand a chance of conviction:

“You need to be either very young or very old, preferably a virgin, or at least a married woman of unassailable respectability'…'Your attacker should be a stranger and you should not willingly have found yourself in his company. You should have fought back fiercely and suffered physical injuries, preferably grave ones. And then you should have reported the incident immediately.”

There were many reasons why the treatment of cases of sexual assault was so unsatisfactory, but two misogynistic features were apparent. The first was a widespread social assumption that women who ‘cried rape’ were lying. As an appellate Judge explained in 1968:

“the judge has to…convey to the jury that in cases of alleged sexual offences it is really dangerous to convict on the evidence of the woman or girl alone. This is dangerous because human experience has shown that in these courts girls and women do sometimes tell an entirely false story which is very easy to fabricate, but extremely difficult to refute.” 

The second feature was that women who were victims of sexual violence were often thought to have been ‘asking for it’.  In a 1977 case, Lord Justice Lawton in the Court of Appeal, in upholding modest sentences upon men who had abused a 14-year-old girl, observed that ‘the girl was a wanton…she had become debauched…before she met these three…[they were] treating her as the village whore’. In a later case, he described a young rape victim as ‘a menace to young men’ and ‘very promiscuous’. In a 1982 case, another judge directed the jury that ‘it is not just a question of saying “no”…if she does not want it, she has only to keep her legs shut and he would not get it without force, and there would be marks of force being used’.

These attitudes meant that a complainant was likely to face intrusive cross-examination concerning her sexual history. Until 1976 evidence might be adduced that the complainant was ‘of notoriously bad character for want of chastity or common decency’. Although not required in law, the courts often demanded corroboration of rape allegations. It was not surprising that a woman involved in a high profile 1976 case reported that ‘if it happened again, I wouldn't even tell the police…it was another sort of nightmare’.

In the 1970s, the climate was not always favourable for those who sought to campaign on issues relating to sexual violence, not least because most of the decision makers in the judiciary and the political world were men, and many of those men shared these attitudes. However, this was not true of all male politicians. Some took up the issue. A catalyst for change was the case of Morgan, a very important case both legally and politically.  Morgan invited the other three defendants to his house and suggested that they should have sex with his wife, telling them that she was 'kinky' and any apparent resistance on her part would be pretence. They had intercourse with her despite her protests. They were charged with rape. All four defendants were convicted. The House of Lords, although upholding the convictions, held that when a defendant had sex with a woman without her consent, genuinely believing that she did consent, he was not guilty of rape, even though he had no reasonable grounds for so believing. 

The case produced a massive public uproar, and was condemned as a ‘charter for rapists’. Politicians and public alike were concerned that a man could not be convicted of rape provided he thought the woman was consenting, even if she was not and that belief was unreasonable. Jack Ashley MP, therefore, introduced a Private Member’s Bill to reverse Morgan and also to restrict disclosure of the woman’s sexual history. To pre-empt this, the Home Secretary (Roy Jenkins) announced the appointment of a small advisory group chaired by a female judge (Heilbron J), to consider the law. Heilbron reported in December 1975. She limited her recommendations to the issues raised by Ashley, including the definition of the offence, and evidence.

Following Heilbron’s recommendations, in due course, section 2 of the Sexual Offences (Amendment) Act 1976 provided that:

“(1) If at a trial any person is for the time being charged with a rape offence to which he pleads not guilty, then, except with the leave of the judge, no evidence and no question in cross-examination shall be adduced or asked at the trial, by or on behalf of any defendant at the trial, about any sexual experience of a complainant with a person other than that defendant.

(2)The judge shall not give leave in pursuance of the preceding subsection for any evidence or question except on an application made to him in the absence of the jury by or on behalf of a defendant; and on such an application the judge shall give leave if and only if he is satisfied that it would be unfair to that defendant to refuse to allow the evidence to be adduced or the question to be asked.”

Subject to the remaining judicial discretion, the effect of the 1976 Act was greatly to reduce the questioning of complainants as to their previous sexual behaviour. The scope for such investigation was further reduced by section 41 of the Youth Justice and Criminal Evidence Act 1999. This states that, save in the very limited and carefully delineated circumstances set out in the Act, no question may be asked in cross-examination, on behalf of any accused at the trial, about any sexual behaviour of the complainant.

The 1976 regime, as further developed since 1999, has become an important part of the ‘toolkit’ for trying sexual offences. Those advising defendants charged with a sexual offence will be aware that questioning about sexual history will only be allowed exceptionally, on application, and by reference to the statutory criteria. This is not to say that such provisions are perfect or free from controversy. In the recent case of the footballer Ched Evans, the Court of Appeal quashed a conviction for rape and allowed a retrial, at which Evans was acquitted, on the basis that sexual history evidence should have been admitted. However, Lady Justice Hallett made clear that this was an exceptional case, and that the courts had to walk at tightrope in such cases between the need to provide a fair trial for the defendant and the obligation to protect the complainant.

Certainly, few would recommend a return to the pre-1976 world, in which complainants were routinely subjected to what Hallett described as ‘intrusive and unnecessary questioning about their sexual history’. This experience suggests that political action, in the form of legislation, and very heavily confined judicial discretion, can provide the means to limit offensive and unnecessary investigations during criminal trials.


We now turn to a historical review of provocation. Since the early modern period, English law has recognised that certain behaviour on the part of the deceased might be sufficient to reduce a charge of murder to one of manslaughter. As the law developed in the eighteenth and nineteenth centuries, the courts set firm limits upon the provocation defence: otherwise, it was feared, those who had killed might get away with murder and avoid execution. The necessary degree of provocation was that which would inflame the reasonable man, ‘something which might naturally cause an ordinary and reasonable-minded man to lose his self-control and commit such an act’. Moreover, the ‘act’ had to be proportional to the ‘something’. And, ‘no words or gestures, however opprobrious or provoking, will be considered in law to be provocation’.

These restrictions meant that, at least in principle, the partial defence of provocation was more likely to be available to female than male killers. The former might be provoked by brutish (male) behaviour, but this was less of a prospect for male defendants. They might claim that they had been driven to the end of their tethers by ‘shrewish’ or ‘slut-like’ conduct, but this would, on analysis, generally consist of verbal affront rather than violence. Indeed, the appellate courts sought expressly to remove gender biases from this area of the law. In the 1946 decision of the House of Lords in Holmes, where the deceased wife had (allegedly) admitted infidelity, provocation defence was rejected in the following resounding terms:

“a sudden confession of adultery without more can never constitute provocation of a sort which might reduce murder to manslaughter…the rule, whatever it is, must apply to either spouse alike, for we 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… in no case could words alone, save in circumstances of a most extreme and exceptional character, so reduce the crime…as society advances, it ought to call for a higher measure of self-control in all cases.”

This was how the law stood when, in January 1949, the Attlee government set up a Royal Commission on capital punishment. The Commission reported in 1953, and recommended various changes, such as raising the minimum age for execution from 18 to 21. As to provocation, the Commission proposed that the old rule that words alone were incapable of constituting provocation should go. The evidence, which they heard and accepted, was couched in highly gendered terms. Lord Goddard and Lord Justice Denning, both senior judges, ‘cited the case of a husband whose patience is at length exhausted by a persistently nagging wife and who suddenly strikes and kills her’. The latter, in addition, thought that manslaughter was ‘the appropriate verdict in cases where a returning solder killed his wife on her confessing adultery’.

This report eventually gave rise to the Homicide Act 1957, which significantly extended the scope of the provocation defence to ‘mere words’. Whether intended or not, the consequence of this change was to open a new way to successful defences by men who had killed their wives and lovers. Thus, in an early case decided under the Act, the defendant relied on provocation by his wife who had nagged him, prevented him from sleeping for two nights and finally offered him the choice of divorce or paying her £4 a week to live separately. He was convicted of manslaughter and sentenced to 4 years in prison. This was one of many cases where judges were ‘merciful’ towards men claiming they had been provoked beyond endurance by their wives and in various subsequent cases in the 1960s, male defendants received short sentences for manslaughter in such circumstances. In an extreme example, a man who had strangled his ‘unfaithful’ wife was absolutely discharged: he was ‘a thoroughly decent man with a strong sense of the sanctity of marriage and the unity of family life’.

Thus, it became apparent that the doctrine of provocation was being applied in a highly unfair and gendered way to women. In her 1989 study Misogynies, Joan Smith perceived an assumption ‘that men live their lives on a hair trigger and can be provoked to violence by the most insignificant stimulus’. She cited a 1985 case, in which a court accepted that, where a ‘hard-working’ man had been provoked to kill his nag of a wife, he should be sentenced on the basis that ‘a man of reasonable self-control might…have done what you did’. What he had done included ‘cut[ting] her up with a saw and boil[ing] her skin and bones’. As Smith observed, ‘the picture of family life conjured up by this judgment is mind-boggling’. In the 1991 case of McGrail, a man who ‘had been through ten years of torment’ before killing his wife received a suspended sentence for manslaughter and a period of supervision to help him ‘overcome his feelings of remorse’. The judge observed that ‘this lady would have tried the patience of a saint.’ In the same year, in Singh, a man who had strangled his wife in front of their children ‘to stop her incessant nagging’ likewise escaped with a suspended sentence: the judge did not ‘see that sending you to prison is going to do you any good’.

Moreover, such cases often turned into a trial of the deceased woman, who could no longer defend herself, and the courts did not see this as much of a problem in cases where men had killed their wives. In the 1985 case of Mellentin, for example, the court noted of alleged taunting, ‘this of course all comes from the appellant and in the nature of things cannot be corroborated, but we have no reason to suspect that what he ultimately told the police was other than an accurate account of what transpired.’

Once more, the unsatisfactory state of the law was recognised by Parliament, which intervened to restrict judicial (and jury) room for manoeuvre. The Coroners and Justice Act 2009 has, in effect, abolished the common law of provocation. It has  provided that a defendant is not to be convicted of murder if his acts resulted from loss of self-control, which ‘had a qualifying trigger’ and where ‘a person of the defendant’s sex and age, with a normal degree of tolerance and self-restraint and in his circumstances, might have reacted in the same or in a similar way’. The expression ‘qualifying trigger’ is defined in detail in the Act in a way highly relevant to the issues addressed here: for example, it is specified that ‘the fact that a thing done or said constituted sexual infidelity is to be disregarded’. The Minister made clear in advocating the Bill in the Commons that the legislative intention was to tie the hands of the courts on such issues:

“In modernising the law in this matter, we have purposely set a very high threshold for the circumstances in which killing in anger could ever be treated as manslaughter rather than murder. …in relation to sexual infidelity, it is important to set out the position precisely and uncompromisingly - namely that sexual infidelity is not… ever sufficient on its own to found a successful plea of loss of control so as to reduce the verdict from murder to manslaughter.”


Some of the press coverage of the Milane and other cases has suggested that they have arisen from a world in which Fifty Shades of Grey has allowed BDSM practices to enter the mainstream. However, and as ever, one needs to take a longer view. Fifty Shades may or may not be innovative, but misogyny within society and the legal system most certainly is not. After all, the legal system has, until very recently, been a system by and for men. In this system, men had rights of possession and property over women. It is social and political change which has transformed this system into one which offers a greater level of equality between the sexes.

In relation to the ‘rough sex’ defence, it is instructive to see how similar problems have been approached in law in the past. This experience suggests that tightly drawn legislation, which exhaustively defines what evidence may be adduced and what matters are capable of being relied upon, may point the way forward. In particular cases, there will be difficult judgements to be made as to the balance to be struck in order to secure a fair trial, but that is a task with which the courts are familiar. No arrangement will ever be perfect, especially where a key witness is necessarily silent, but such a legislative approach should produce a substantial improvement on the current situation, where female victims are traduced in their absence and on the basis of spurious claims of consent to dangerous sexual practices.

Further Reading

Zsuzsanna Adler, Rape on trial (London: Routledge, 1987).

Susan Atkins and Brenda Hoggett, Women and the law (Oxford: Blackwell, 1984).

Julie Bindel et al, ‘Trials and Tribulations – Justice for Women: A Campaign for the 1990s’, in Gabriele Griffin, Feminist activism in the 1990s (London: Taylor & Francis, 1995) pp.65-76.

Joanna Bourke, Rape: a history from 1860 to the present day (London: Virago ed., 2008)

Anna Clark, Women's silence, men's violence: sexual assault in England 1770-1845 (London: Pandora, 1987).

Susan Edwards, Sex and Gender in the Legal Process (London: Blackstone, 1996).

Jeremy Horder, Provocation and responsibility (Oxford: Clarendon Press, 1992).

Helena Kennedy, Eve was shamed: How British Justice is Failing Women (London: Chatto, 2018).

Joan Smith, Misogynies (London: Faber, 1989).


Papers By Author

Papers by Theme

Digital Download

Download and read with you anywhere!


Sign up to receive announcements on events, the latest research and more!

To complete the subscription process, please click the link in the email we just sent you.

We will never send spam and you can unsubscribe any time.

About Us

H&P is based at the Institute of Historical Research, Senate House, University of London.

We are the only project in the UK providing access to an international network of more than 500 historians with a broad range of expertise. H&P offers a range of resources for historians, policy makers and journalists.

Read More