BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

The Law Commission


You are here: BAILII >> Databases >> The Law Commission >> Homicide: Murder, Manslaughter And Infanticide (Report) [2006] EWLC 304(5) (28 November 2006)
URL: http://www.bailii.org/ew/other/EWLC/2006/304(5).html
Cite as: [2006] EWLC 304(5)

[New search] [Context] [Printable version] [Help]


    PART 5
    PROVOCATION AND DIMINISHED RESPONSIBILITY
    THE EFFECT OF A PLEA OF PROVOCATION

    5.1      We recommend that:

    provocation should be a partial defence, with a successful plea having the effect of reducing first degree murder to second degree murder.

    5.2      We undertook a thorough review of the defence in 2004. We continue to believe that the recommendations we made at that time for reform of the defence are the right ones.[1] However, at that time we had not been asked to consider the role of the defence in a reformed law of murder. For that reason, along with the need to take account of recent developments and consultees responses, some of the arguments in favour of reforming the defence along the lines we recommended will be repeated here.

    5.3      The defence is a confusing mixture of common law rules and statute, but the gist of the defence can be found in section 3 of the Homicide Act 1957:

    Where on a charge of murder there is evidence on which the jury can find that the person charged was provoked (whether by things done or by things said or by both together) to lose his self-control, the question whether the provocation was enough to make a reasonable man do as he did shall be left to be determined by the jury; and in determining that question the jury shall take into account everything both done and said according to the effect which, in their opinion, it would have on a reasonable man.

    5.4      Provocation is currently a partial defence that, if successful according to the terms of section 3, reduces murder to manslaughter. Following a successful provocation plea, the judge has discretion over the sentence and can impose a sentence up to and including life imprisonment. The Sentencing Advisory Panel has given advice to the Sentencing Guidelines Council that sentences for manslaughter in successful provocation cases should be broadly (and in outline) as follows:

    Low degree of provocation: sentencing range of 9 to15 years.
    Substantial degree of provocation: sentencing range of 4 to 9 years.
    High degree of provocation: sentencing range of up to 4 years.[2]

    5.5      Research conducted by Professor Barry Mitchell and Dr Sally Cunningham showed that, either on its own or in combination with another defence, provocation was the second most popular plea in the sample of murder cases examined (22.3% of cases) after denial of intent (39.4%).[3] It is our belief that the adoption of our recommendations for reform of the defence should ensure that, in practice, cases involving a low degree of provocation never reach the jury (or are unsuccessful if they do reach the jury). The details of our recommendations are set out from paragraph 5.11 onwards.

    5.6      During the consultation process, some consultees continued to express the view that there is no place in the law of murder for the provocation defence. We believe that the restrictions on the defence that we recommend will meet many of these concerns. Moreover, the narrowing of the fault element for first degree murder, under our recommendations, means that the provocation defence would come into play in fewer cases than it does at present. The fact that a provocation plea will have to provide some excuse for an intention to kill, or an intention to do serious injury aware of a serious risk of causing death, should also mean that the hurdle that must be surmounted successfully to plead the defence will be higher.

    5.7      In defence of the provocation plea, it is important to bear in mind the great range of offenders and kinds of provocation that may be at issue in murder cases. A life sentence[4] for first degree murder in a case involving a high degree of provocation may be considered especially unjust when the offender is a young person or is of low intelligence.[5] In Camplin,[6] for example, D was a 15-year-old boy who claimed he had been provoked to lose self-control and kill a man who had raped him in spite of his (Camplin's) resistance, and then laughed at him. If a jury believes that a story like Camplin's may be true, it is hard to see what purpose would be served in passing a mandatory sentence for first degree murder of indefinite detention at her Majesty's pleasure.

    5.8      In Part 2 we explained why we are recommending that provocation (along with the other partial defences) should operate as a defence only to first degree murder, reducing the offence to second degree murder. The main reason was the intrinsic nature of the link between the plea's existence and the mandatory life sentence for murder (first degree murder, under our proposals). The fact that there is a mandatory life sentence for murder is the raison d'(tre of the provocation plea in England and Wales, although we recognise that the defence exists in a minority of jurisdictions in which there is no mandatory life sentence for the top tier offence.[7] We do not believe that it would serve the interests of justice to extend the application of this complex defence to any crime where the existence of sentencing discretion already makes it possible to reflect the nature and degree of the provocation in the sentence itself. We gave our reasons for this in Part 2. Under our recommendations, second degree murder, attempted murder, and manslaughter would all have a discretionary life sentence maximum. So, there is no compelling reason to extend the defence to these offences.

    5.9      By way of contrast, there should be a defence of provocation in some form to first degree murder. The defence of provocation has played a part in helping a small minority of offenders to avoid the mandatory penalty for murder for over 400 years. As we indicated in Part 2, whilst the mandatory sentence is retained, abolition of the defence might create a greater set of problems for the law of homicide than are posed by its continued existence. It is worth briefly re-iterating the argument to that effect here.

    5.10      Imagine a case involving what the Sentencing Advisory Panel regards as a high degree of provocation, warranting a sentence for manslaughter of 4 years' imprisonment or less (see paragraph 5.4 above). If there were no defence of provocation to reduce the offence to manslaughter, the offender would have to be sentenced to life imprisonment. However, the initial period the offender would spend in custody before being eligible for release might only be a year or two, given the Sentencing Advisory Panel's guidance. Accordingly, an offender given a life sentence would spend the vast majority of that sentence (perhaps in excess of 40 years) out on licence. The proliferation of such sentences would only increase the public's confusion over the nature and significance of the 'life' sentence.

    THE SUBSTANCE OF THE DEFENCE

    5.11      In our review of the defence of provocation in 2004, we concluded that the circumstances in which it should in future be available ought to be changed, in the ways indicated below.[8] Our conclusions were reached after widespread and detailed consultation. We see no compelling reason to depart from them in substance, although we will indicate below where our conclusions remain controversial and, therefore, where there is an issue that could profitably be taken further in the next stage of the review. We are recommending that the defence be reformed as follows:

    (1) Unlawful homicide that would otherwise be first degree murder should instead be second degree murder if:
    (a) the defendant acted in response to:
    (i) gross provocation (meaning words or conduct or a combination of words and conduct) which caused the defendant to have a justifiable sense of being seriously wronged; or
    (ii) fear of serious violence towards the defendant or another; or
    (iii) a combination of both (i) and (ii); and
    (b) a person of the defendant's age and of ordinary temperament, i.e., ordinary tolerance and self-restraint, in the circumstances of the defendant might have reacted in the same or in a similar way.
    (2) In deciding whether a person of the defendant's age and of ordinary temperament, i.e. ordinary tolerance and self-restraint, in the circumstances of the defendant, might have reacted in the same or in a similar way, the court should take into account the defendant's age and all the circumstances of the defendant other than matters whose only relevance to the defendant's conduct is that they bear simply on his or her general capacity for self-control.
    (3) The partial defence should not apply where:
    (a) the provocation was incited by the defendant for the purpose of providing an excuse to use violence; or
    (b) the defendant acted in considered desire for revenge.
    (4) A person should not be treated as having acted in considered desire for revenge if he or she acted in fear of serious violence, merely because he or she was also angry towards the deceased for the conduct which engendered that fear.
    (5) A judge should not be required to leave the defence to the jury unless there is evidence on which a reasonable jury, properly directed, could conclude that it might apply.

    5.12      We asked consultees whether they still supported these recommendations and a clear majority of them did, either as a whole or in part. There was broad support for them from (amongst others) a number of individual judges in the higher courts; the Criminal Cases Review Commission; the Crown Prosecution Service; the Law Society; the London Criminal Courts Solicitors' Association; the Association of Chief Police Officers; the Police Federation; the teachers of criminal law at the London School of Economics and Political Science; Justice; Liberty; Rights of Women; Refuge; Southall Black Sisters; and the Royal College of Psychiatrists.

    5.13      Those who supported our proposals only in part rarely disagreed with the reforms we proposed for the basic elements and structure of the defence. One area of disagreement centred on the scope of the defence. Some consultees thought it ought to apply only in cases of so-called 'excessive defence' (1(a)(ii) above, in paragraph 5.11), and not in the wider set of circumstances (permitted by 1(a)(i) above in paragraph 5.11). 14 consultees responded to the specific question whether the defence should be restricted to 'excessive' defence cases. Seven responded saying it should be so restricted and seven responded saying it should not be so restricted. Those who disagreed with our own view that it should not be so restricted included the Crown Prosecution Service, Justice for Women, and Victims' Voice. We take up the issue in paragraph 5.61 onwards below.

    5.14      The range of legal problems to which the current defence gives rise have led us to adopt these recommendations for change.[9] We will concentrate here on the most important of these problems and on how we recommend dealing with them.

    An absence of judicial control over when the plea is considered by the jury

    5.15      Under the existing law, the judge must put the defence to the jury, whenever there is evidence that the defendant ("D") was provoked to lose self-control, however unlikely the defence is to succeed. This judicial obligation was probably not intentionally created by Parliament when the doctrine of provocation was reformed in 1957. The current position does not serve the interests of justice because the need to put the defence to the jury in these circumstances increases the likelihood that an unmeritorious claim may succeed. The current position may not even serve the interests of every D. Even if there is evidence of a loss of self-control, D may not want the jury to be side-tracked by a partial defence if his or her main claim is for a complete acquittal.

    5.16      We address this issue in (5) in paragraph 5.11 above. (5) seeks to restore the common law position prior to 1957, under which the trial judge had the task of filtering out purely speculative and wholly unmeritorious claims. Clearly, a judicial ruling that a provocation plea should not be put to the jury could have a major influence on the way in which the defence case is run. Accordingly, we suggest that consideration is given to the creation of an interlocutory appeal against a judge's ruling that the defence should not be put to the jury. This appeal should be permitted only before the trial (the basis of the defence should be considered at a pre-trial hearing under modern case management procedures), so that the trial itself is not substantially delayed by the use of the interlocutory procedure. The Higher Court Judges' Homicide Working Party could see no special problems arising from the provision of such an appeal.

    The unnecessary and undesirable loss of self-control requirement

    5.17      For 250 years or more, the law took the uncomplicated view that the defence of provocation could be pleaded whenever D was provoked into a towering rage or temper and killed before the rage or temper subsided. In the 19th century, this subjective requirement was turned into a requirement that D 'lost self-control' at the time of the killing.[10] Judges have since struggled to interpret and apply this notion as a description of the necessary state of mind.[11] It remains unclear to what extent a delay between the provocation and the loss of self-control will undermine a provocation plea.[12]

    5.18      In addition, the requirement of a loss of self-control has been widely criticised as privileging men's typical reactions to provocation over women's typical reactions. Women's reactions to provocation are less likely to involve a 'loss of self-control', as such, and more likely to be comprised of a combination of anger, fear, frustration and a sense of desperation. This can make it difficult or impossible for women to satisfy the loss of self-control requirement, even where they otherwise deserve at least a partial defence.

    5.19      This is why in our previous report and in the CP, we took the view that a positive requirement of loss of self-control was unnecessary and undesirable. Our current recommendations have not sought to resurrect it.[13] Professor Mitchell and Dr Cunningham's research[14] shows that, in cases where provocation has been pleaded, whilst there may be uncorroborated evidence that the killing took place in anger, evidence of the 'loss of self-control' that the law requires was much harder to find:

    whilst sometimes there was evidence of a loss of self-control as the law requires, it was impossible to know whether the court felt that the defendant had lost his self-control at the critical moment, or whether more reliance was being placed on the reasonableness of the defendant's reaction. In at least nine cases where provocation was pleaded it was unclear whether the defendant had lost his self-control at any stage. However, it appeared that there was usually very little time lapse between the provocation and the defendant's reaction to it… .[15]

    5.20      That being so, as (3) above in paragraph 5.11 makes clear, we have sought to express the so-called subjective condition negatively, avoiding reliance on a positive requirement of loss of self-control. D's reaction must not have been 'engineered' by him or her through inciting the very provocation that led to it. Further, it should not reflect a considered desire for revenge. This second element of the restriction will be given further consideration here.

    5.21      It may be argued that our approach to the problems caused by the loss of self-control requirement goes too far the other way, by not replacing the requirement with a more broadly based subjective condition. An example of such a condition would be the term used in the American Model Penal Code, 'extreme mental or emotional disturbance'.

    5.22      In our report on provocation in 2004, we followed the views of a majority of judges and academics in taking the view that the phrase 'extreme mental or emotional disturbance' was unduly vague and indiscriminate.[16] Further, the introduction of a new phrase of this kind at the heart of the defence would lead to intense legal scrutiny and, no doubt, to a number of cases in the appeal courts to determine its meaning and scope. Whether the law would be improved as a result of this is open to question. However, we recognise that the phrase has formed the basis for a provocation defence in at least some American state jurisdictions,[17] and cannot therefore be dismissed as unworkable (see paragraph 5.32 below).

    5.23      We do accept that excluding cases in which there was a 'considered desire for revenge' ((3) in paragraph 5.11 above) may by itself be insufficient to exclude from the trial every unmeritorious case. However, we believe that juries can simply be trusted to reject such claims when they slip through the net and into the trial itself. It is worth exploring this issue in more detail because to do so will show how difficult it is to provide a clearly superior alternative.

    5.24      The exclusion in (3) in paragraph 5.11 above was phrased so as to allow some cases in which there had been an element of premeditation to fall in principle within the scope of the defence. Examples are cases in which a battered woman has killed her violent partner.[18] Such cases clearly call for the jury's consideration of a partial excuse, despite the element of premeditation. The premeditation will typically reflect no more than D's reasonable fear that an immediate and direct confrontation with the abusive partner will lead to violence being inflicted on her.[19] It would not be appropriate to describe D's premeditation in such cases as constituting a 'considered desire for revenge'.

    5.25      Setting such cases on one side, it has been put to us that there are other cases that, whilst not involving cold-blooded revenge, should nonetheless be ruled out as potential provocation pleas. Examples are so-called 'honour' killings, in which D may say that he (and it will normally be a 'he') planned the killing of the victim to uphold the honour of the family rather than to take revenge on the victim for something said or done. We believe that there is likely to be a strong motive of revenge in such cases. The offender is seeking to make an example of the victim because she (and it normally will be a 'she') has defied tradition, custom or parental wishes in her choice of boyfriend, spouse or life-style.[20]

    5.26      It may be that such cases – or others like them - involve mixed motives (other than those contemplated in (4) in paragraph 5.11 above). An example might be Baillie,[21] in which D's motive could be said to have been both the taking of revenge on a drug dealer and the prevention of further crime by that drug dealer. Such cases may or may not be ruled out under (3) in paragraph 5.11 above.

    5.27      In drawing up our recommendations, we were not engaged in statutory drafting and so our use of phrases such as 'considered desire for revenge' was only suggestive of an approach. It seems likely that in many mixed motive cases the judge might take the view that, even if there is no 'considered desire for revenge', it is nonetheless a case where no reasonable jury would find that the defence applies. It would thus be ruled out under (5) in paragraph 5.11 above. We regard it as significant that of the provocation cases studied by Professor Mitchell and Dr Cunningham, in the two involving honour killing both the accused were convicted of murder.[22] We are confident that the result would be no different under our recommendations.

    5.28      Nonetheless, the question arises whether such cases should, or should not, be ruled out through a restriction of the kind set out in (3) in paragraph 5.11 above and, if so, how.

    5.29      It was clear to us that when a battered woman uses excessive force against her abusive partner only because she fears for her safety in any direct confrontation, it would be wrong to rule out her plea simply because there was no evidence of a loss of self-control. Such cases are discussed further from paragraph 5.47 onwards.

    5.30      A different approach could be adopted for cases falling solely within 1(a)(i) in paragraph 5.11 above, those that do not involve the excessive use of force motivated by a fear of serious violence (1(a)(ii) in paragraph 5.11 above). It would be possible to insist on satisfaction of an extra requirement of 'immediacy' (as in Scottish law[23]), or of, say, 'a spontaneous display of anger'[24] in cases falling exclusively within this limb (1(a)(i) in paragraph 5.11 above) without relying on the notion of 'loss of self-control'. Satisfaction of such an extra requirement would not be necessary under 1(a)(ii) or 1(a)(iii) in paragraph 5.11 above (where only (3) and (5) in paragraph 5.11 would have to be satisfied).

    5.31      It must be said that such a 'positive' requirement, added to 1(a)(i) in paragraph 5.11 above, would complicate the law, not least in cases where the question whether the case falls into 1(a)(i) alone is disputed. It might be simpler to rely on the judge's power to withdraw unmeritorious cases, under (5) in paragraph 5.11 above, to weed out cases in which someone has committed a premeditated honour killing or has taken the law into his or her own hands by acting as self-appointed judge, jury and executioner, as in Baillie.[25]

    5.32      This is a difficult issue, and there are no perfect solutions. If the answers provided by the present law are inadequate, then the following alternatives may be something to be considered further in the next stage of the review. We recommended alternative (1) below in 2004, for the reasons given above, but alternatives (2) and (3) below will also have supporters:

    (1) A negative requirement, that D must not have acted out of considered desire for revenge, whether in response to gross provocation or in response to a fear of serious violence, or both together (our conclusion in 2004: see para 5.11 above).
    (2) In addition to (1), a positive requirement, that D must have acted in a state of extreme emotional disturbance whether in response to gross provocation or in response to a fear of serious violence, or both together (rejected by us in 2004: see para 5.22 above).
    (3) In addition to (1), a positive requirement that D must have acted immediately following gross provocation, this additional requirement being inapplicable both in the case of a response to a fear of serious violence alone and in the case of a response to both gross provocation and a fear of serious violence together.
    Uncertainty over the 'reasonable person' requirement

    5.33      At present a provocation plea cannot succeed unless the jury decide that a reasonable person might have responded to the provocation in question by doing as D did, namely losing self-control and killing. We set out above, in paragraph 5.11, 1(b) and 2, our recommendation for the way in which this 'reasonable person requirement' should be understood. We believe the law needs clarification on this point for the following reasons.

    5.34      In a series of cases in recent years, the Court of Appeal, the House of Lords, and the Privy Council, have disagreed over how broadly or narrowly to construe the reasonable person requirement. The result has been uncertainty over its scope and nature, although in no case has the justification for having the restriction in some form been doubted. In the Privy Council case of Attorney-General for Jersey v Holley,[26] departing from an approach to the reasonable person requirement adopted by the House of Lords only four years previously,[27] Lord Nicholls said:

    In expressing their conclusion…their Lordships are not to be taken as accepting that the present state of the law is satisfactory. It is not. The widely held view is that the law relating to provocation is flawed to an extent beyond reform by the courts…Their Lordships share this view. But the law on provocation cannot be reformulated in isolation from a review of the law of homicide as a whole.[28]

    5.35      Disagreement in the courts has focused on the extent to which D's own characteristics, or other factors, can, or must, be taken into account in judging how the reasonable person might have responded to the provocation.

    5.36      One key question in making that judgement is 'how gravely provocative really was the provocation'? It is obvious that D's own characteristics must be relevant to this question. To give a simple example, D's own height would be relevant in assessing the gravity of the provocation constituted by an accusation that he or she was 'a midget'.[29] The courts have not encountered significant difficulties in recent years in deciding how such characteristics or factors affecting the gravity of provocation should be dealt with in law.[30] The jury is obliged to take such characteristics into account.[31]

    5.37      More controversial has been the question whether the jury should be required, or permitted, to take into account individual characteristics of D (or other factors) liable to affect the level of self-control that he or she can be expected to show in the face of any provocation. It may be, for example, that a drunken D, an immature D or a mentally deficient D, is unable to exercise the same level of self-control, in the face of provocation generally, as a sober adult with normal mental capacities. The courts have disagreed over whether the jury should be required or permitted to take such factors into account.

    5.38      We will not retrace the history of the courts' attempts to introduce clarity to the law on this question.[32] In our view, the function of the reasonable person requirement is to test D's own reaction against the standards of someone of his or her age possessed of an ordinary temperament: someone who is neither intolerant nor lacking in a reasonable measure of self-restraint when facing provocation. Unless the jury concludes that D's reaction might have been that of such a person, the defence ought to fail, even if D only killed as a result of a provoked and momentary loss of temper.

    5.39      We are reluctant to speculate on how the courts would interpret the provisions in 1(b) and (2) in paragraph 5.11 above. Still less would we wish to insist that they interpret them in a given way. None the less, the following examples may provide some guidance on the kinds of distinctions we think that it would be helpful to draw.

    5.40      Our provisions impose a duty on the judge to instruct the jury to ignore factors that affect D's general capacity to exercise self-control. Alcoholism, for example, or another mental deficiency or disorder that is liable to affect temper and tolerance are obvious examples.[33] A person who has killed because his or her capacity for self-control was reduced by such a characteristic must look to the defence of diminished responsibility for a partial defence, because such characteristics constitute an abnormality of mental functioning, unlike, for example, D's age.

    5.41      Abnormal states of mind, such as intoxication or irritability, should also be left out, as should other factors that affect a general capacity to exercise adequate self-control, like a claim that D is 'more jealous or obsessive than most'.[34] This approach to the general capacity to exercise adequate self-control will produce some hard cases. Examples might be ones in which, at the time of the provoked killing, D's general capacity for self-control was temporarily impaired by the effect of taking prescribed medicine, by having suffered a stroke, by involuntary intoxication, by an allergic reaction of some kind or by a bang on the head.

    5.42      In such cases, the individual might well be accurately described in general terms as someone with adequate powers of self-restraint. It is just that their reaction could (albeit for good reason) not be expected to be the reaction of such a person in the particular circumstances. We believe that if their reaction was not one that might have been the reaction of a person exercising ordinary powers of self-restraint in the particular circumstances, that very fact means the provocation defence must fail. The fact that the individual in question could, putting on one side their reaction in the circumstances, be described as someone with adequate powers of self-restraint is quite irrelevant to the provocation plea. They must instead look to diminished responsibility for a defence.[35] As Professor Gardner expresses the point:

    [T]he question, for excusatory purposes, is obviously not whether the person claiming the excuse lived up to expectations in the predictive sense of being true to form … The question is whether that person lived up to expectations in a normative sense … In the face of … taunts, did this person exhibit as much self-restraint as we have a right to demand of someone in her situation? The character standards which are relevant to these and other excuses are not the standards of our own characters, nor even the standards of most people's characters, but rather the standards to which our characters should, minimally, conform … those standards cannot be capped according to the capacities (be they past, present or even future) of the person to whom the excuse is supposed to apply. For, such incapacity, far from militating against unfitness [for playing the role of someone with adequate powers of self-control], is a mode of unfitness in its own right.[36]

    5.43      By way of contrast, a low IQ could be taken into account as part of the circumstances of D (see 1(b) in paragraph 5.11 above) if it meant, for example, that D misinterpreted a provocation, thinking it to be more grave than a person of higher intelligence might have done.[37] To give a different example, the fact that D was dumb and thus unable to respond verbally, is a factor that might legitimately be taken into account when considering D's reaction to a particular provocation given on a particular occasion. In each example, the characteristic is not being used as evidence that the D lacked a general capacity to exercise adequate self-control.

    5.44      By way of contrast, some of the evidence given by a psychiatrist in Roberts[38] would not be relevant to the provocation plea, under our recommendations. This was evidence that 'irrational violence was to be expected from some immature prelingually deaf persons when emotionally disturbed'. This is evidence relevant to a plea of diminished responsibility, rather than to a plea of provocation, because it is evidence of an impaired general capacity for self-control.

    5.45      In many instances, the circumstances liable quite properly to influence the jury in D's favour will bear on how 'gross' the provocation was, or on how justifiable it was for D to feel seriously wronged (see 1(a)(i) in paragraph 5.11 above). An example is the cumulative effect of repeated provocations given, quite possibly over many years, in circumstances where it may also have been impossible for D to escape the provocation's effects. There is usually no theoretical difficulty about taking such background factors into account because they do not necessarily suggest that D is someone with a reduced general capacity to exercise self-control. A classic example would be the intimidated spouse who has been subject to abuse, the cumulative effect of which has become intolerable over the years.[39]

    5.46      This area of law will always remain difficult. As we indicated in the CP, however, a trial judge is under a duty to explain to the jury the full context in which a provoked killing has taken place, and the form of his or her direction ought to be discussed in advance with prosecution and defence advocates.[40] These safeguards should go some way towards minimising the chance of misdirections, and hence appeals.

    A NEW BASIS FOR THE DEFENCE: FEAR OF SERIOUS VIOLENCE

    5.47      The problems dealt with in the preceding section are in some respects problems generated by the terminology of the existing law ('loss of self-control'; 'reasonable man'). The provision in 1(a)(ii) in paragraph 5.11 above, however, extends the defence of provocation in a novel way. We explain how in this section.

    5.48      In 1(a)(ii) we recommend that:

    the partial defence of provocation should be expanded to encompass cases in which the defendant over-reacted to a fear of serious violence.

    5.49      Historically, the common law treated as provocation (sufficient to reduce murder to manslaughter) an overreaction to illegal conduct, such as violence towards or (threatened) false imprisonment of the accused.[41] This line of cases lost its authority when the 'loss of self-control' and 'reasonable person' requirements became established. As we have said, the provocation defence is currently available only when there is evidence that D was provoked to lose his or her self-control. The defence is concerned with angry, spur-of-the-moment reactions to provocation. It is not concerned with reactions prompted by fear, unaccompanied by a loss of self-control, even if the fear in question was that the victim would have inflicted serious violence on D if the victim had not been killed.

    5.50      There is a contrast here with Scottish law, which permits murder to be reduced to culpable homicide in a case where someone's fatal overreaction to physical violence was not disproportionate to the nature and degree of violence confronted.[42] It should be made possible in English law to plead provocation when the killing was in response to a fear of serious violence, if someone of an ordinary temperament but otherwise acting in circumstances facing the accused, might have reacted in the same or in a similar way.[43] A rigid insistence on confining the provocation plea to angry reactions, as opposed to actions taken out of fear, has prevented this development occurring at common law (through a resurrection of the older cases). Here are two examples where the law is currently deficient:

    Example 1: D and V live together, but their relationship is a violent one. V frequently hits D when he (V) comes home drunk. One night, when V comes home drunk and threatens to beat D yet again, she goes to the kitchen, fetches a knife, and stabs V in the chest while he is off his guard. V dies.
    Example 2: D is an armed police officer called to a house where a neighbour has said there is a man (V) with what looked like a gun. When the officer enters the house, V appears to have something in his hand. D demands that V show him what is in his hand but V does not respond. D shoots V and V dies. It turns out that V had a small metal bar in his hand. V may not have heard what D said because he (V) suffered from deafness.

    5.51      In both examples, assuming he or she cannot plead diminished responsibility, D has only two effective choices under the existing law if he or she admits having acted with the fault element for murder. D can plead self-defence, a justificatory plea that, if successful, will end in complete acquittal. Alternatively, D can seek a manslaughter verdict by pleading provocation.

    5.52      To succeed in a plea of self-defence, D must make a case that what he or she did was within the bounds of reasonableness, as a means of averting a threat posed by V.[44] In both examples it is possible, given the circumstances, that a jury will accept this. It is also quite possible, however, that the prosecution will persuade the jury in each case that acting with intent to kill or do serious harm was not within the bounds of reasonableness as a response to the threat posed by V.

    5.53      Under the present law, if D's killing of V is regarded as an overreaction in self-defence, he or she must be convicted of murder unless he or she can succeed in a plea of provocation (with the result that the offence is reduced to manslaughter).[45] In example 1, D would be required to frame the partial defence in terms of the provocation constituted by V's threat, in the context of V's history of violence. In example 2, the provocation would have to be framed in terms of V's failure to respond to D's request, coupled with the fact that V is holding what D believes to be a gun. In both examples, that is an artificial way to analyse the basis for a partial defence. Further, D will in both examples have to show that the provocation caused him or her to lose self-control at the time of the killing. It is not enough that D was frightened, but still in control of himself or herself.

    5.54      In some circumstances, cases on facts such as those in examples 1 and 2 should end in a first degree murder verdict. In our view, however, a rational approach to reaching the right verdict is currently hampered by arbitrariness and unfairness in the way that the provocation defence is structured. In particular, D should not be prejudiced because he or she over-reacted in fear or panic, instead of over-reacting due to an angry loss of self-control.

    5.55      Consequently, we are recommending that the provocation defence should be available where D killed in response to a fear of serious violence.[46] D will be allowed to say that the effect of the fear of the threat, or of the fear of the threat coupled with the impact of the gross provocation received, was such that, in the circumstances, someone of D's age and of an ordinary temperament might have reacted in the same or in a similar way. The frequently close relationship between anger and fear in someone's reaction makes us confident that it is right to link these elements together in a single partial defence of provocation.

    5.56      This reform would have the additional benefit of giving Ds (such as those in examples 1 and 2) more flexibility in how they choose to run their defence. If they are prepared to accept nothing less than total vindication of what they did, then they can plead nothing other than self-defence. In such a case, the outcome sought is complete acquittal, to which the alternative is conviction for first degree murder.[47] If D is not so confident that the jury will find his or her actions to have been fully justified, he or she can plead provocation – in the form of a fear of serious violence – instead of or alongside a plea of self-defence.

    5.57      Having the latter option reduces the chance that D will be harshly adjudged to have committed first degree murder because the jury finds that he or she overreacted. The jury can opt for the middle course: guilty of second degree murder on the grounds of provocation,[48] leaving the judge with discretion over sentence. As Lord Bingham has indicated:

    The interests of justice are not served if a defendant who has committed a lesser offence is either convicted of a greater offence, exposing him to greater punishment than his crime deserves, or acquitted altogether, enabling him to escape the measure of punishment which his crime deserves. The objective must be that defendants are neither over-convicted nor under-convicted, nor acquitted when they have committed a lesser offence of the type charged.[49]

    5.58      In making our recommendation, we have considered whether the law's understanding of "serious violence" should be extended, as it has in the state of Victoria.[50] In Victoria, "violence", in relation to a child, can include causing or allowing the child to see or hear physical, sexual or psychological abuse. Therefore, in Victoria, D can say that he or she was reacting to a fear of serious "violence", if he or she killed to protect a child from having to see or hear physical, sexual or psychological abuse. We have concluded that "violence" should not be given such an extended meaning.

    5.59      It only seems necessary to give violence such an extended meaning if (as in Victoria) a partial defence plea cannot be framed in terms of a retaliatory response to gross provocation generally, and must be framed in terms of a defensive response specifically to violent conduct. Under our recommendations, 'gross provocation' is a basis for reducing first degree to second degree murder. Causing or allowing a child to see or hear physical, sexual or psychological abuse might well amount to gross provocation to another family member. That would be the basis for reducing the offence to second degree murder. Therefore, there would be no need to portray such conduct as a form of "violent" conduct from which the defendant was defending the child.[51]

    5.60      Almost all consultees welcomed the extension of provocation to cover cases in which the killing was motivated by a fear of serious violence.

    HOW NARROW SHOULD THE PROVOCATION DEFENCE BE?

    5.61      As we indicated in paragraph 5.13 above, some consultees would prefer the defence of 'provocation' to be confined to cases of excessive self-defence. They would like to see 1(a)(i) in paragraph 5.11 above (gross provocation such as led D to have a justifiable sense of having been seriously wronged) omitted from legislative reform. This would leave only 1(a)(ii): ('fear of serious violence'). Consultees who held this view did so for one main reason.

    5.62      There are currently no restrictions on the kinds of provocation that may, in principle, be considered by the jury. So, provocation could in principle include, for example, a confession of adultery, the crying of a baby, or a simple refusal by one spouse to obey the other unconditionally. This state of the law is thought to be especially 'user-friendly' to men seeking to plead provocation, because they may be more likely than women to lose their temper and respond violently over such matters. When women kill, it tends to be in response to an extreme situation involving a fear of violence to themselves or their children.[52] As Dr Quick and Professor Wells have stated:

    Male killing is about power and control. Women killing abusers is about avoiding power and control…Women do not often kill from anger, while anger is what fuels many male killings.[53]

    5.63      The argument from this perspective is that the existing law goes wrong twice. First, it makes no provision for fear of serious violence to reduce murder to manslaughter. It then compounds this error by permitting reduction in cases where the provoked murder may have been little more than a reflection of the continuing cultural acceptability of men's use of violence in anger. For example, in their response to our CP on partial defences to murder, Justice for Women cited cases in which provocation had been successfully used by a male D who had killed his partner because, 'she was having an affair', 'she was going to leave me', and 'the baby wouldn't stop crying'.[54]

    5.64      Our recommendation that the defence of provocation be extended to cases in which there was a fear of serious violence (1(a)(ii) in paragraph 5.11 above) meets the first of these criticisms of the existing law. We also believe that the way in which we have unambiguously restricted the scope to the first limb of the provocation defence (1(a)(i) in paragraph 5.11 above) does enough to meet the second criticism.

    5.65      The belief that the criticism has been met is reinforced by adding to the picture the new power the trial judge will have to withdraw a provocation claim from the jury if he or she thinks no reasonable jury would accept it ((5) in paragraph 5.11 above). The requirement that provocation must be 'gross' in that it must be such as to give rise to a justifiable sense of having been seriously wronged (1(a)(i) in paragraph 5.11 above) is meant to ensure that the kinds of provocation pleas to whose success Justice for Women rightly objected (in paragraph 5.63 above) are not successful.[55] Such pleas should fail either because they are ruled out by the judge right from the start, using the new power in (5) (paragraph 5.11 above), or because the jury accepts that they cannot satisfy the test in 1(a)(i) in paragraph 5.11 above.

    5.66      We recognise, however, that this is a controversial recommendation, even though it received widespread support. It may be a matter to be taken further in the next stage of the review of the law.

    5.67      We believe that restricting 'provocation' cases to instances in which someone overreacted to a fear of serious violence would probably not introduce greater certainty or justice. Trying to exclude some kinds of provocation from the jury's consideration, whilst permitting the jury to consider other kinds, is unlikely to prove any more successful a strategy now than it was 300 years ago, for the following reasons.

    5.68      First, a recommendation to introduce this severe restriction on the provocation defence plea would make the courts' interpretation of the term 'violence' especially significant. Inevitably, even if a statutory definition – be it partial or complete – is provided as it is in New Zealand and Victoria, the courts would come under pressure to extend it, especially in cases on the borderline between 'violent' conduct and 'abusive' conduct more generally.[56]

    5.69      Secondly, there might also be uncertainty over when and whether other kinds of provocation become relevant to a plea, once the "fear of serious violence" threshold has been crossed. This raises the possibility of Ds making a claim of "fear of serious violence", simply in order to ensure that evidence of other kinds of gross provocation is then placed before the jury. Once before the jury, such evidence – if it involves gross provocation – is likely to persuade a jury to reduce the crime to second degree murder, whether or not they are also satisfied that there was a genuine fear of serious violence. So, the 'serious violence' restriction could come to be evaded. It may be, of course, that such evasion could be prevented by implementation of the judge's power to withdraw the issue from the jury, when no reasonable jury could find that there was a fear of serious violence (see (5) in paragraph 5.11 above).

    5.70      Thirdly, it should be kept in mind that the restriction of the formal provocation defence to cases where there is evidence of a fear of serious violence will not put an end to the relevance and admissibility of evidence of other kinds of provocation in murder cases. The restriction could, in fact, very easily be avoided in many cases.

    5.71      This could happen if the defence, as part of their case, were to use evidence of provoking conduct (other than threats of violence) to support a denial of the fault element for first degree murder. The defence will say: "The deceased's outrageously provocative conduct caused D to completely lose his temper, with the result that he cannot be proved to have intended to kill, or to do serious injury aware of a serious risk of causing death". There could be no restrictions on the use of evidence of provocation to this perfectly legitimate end, as a means of securing a verdict of second degree murder.[57]

    5.72      If evidence of provocation were used in this way (and it would have to be, to be relevant at all), none of the safeguards built in to the formal provocation defence – such as the role of the judge in withdrawing the defence, and the "ordinary person" restriction – would apply. Juries could, of course, be trusted to reject specious claims of this nature. Nonetheless, we believe that it is important to highlight what will be the continuing importance of evidence of all kinds of provocation to murder cases, whether or not the formal defence of provocation is itself more severely restricted than we are already suggesting that it should be.

    5.73      Finally, we believe that it is just to permit the formal defence of provocation to apply beyond cases in which D was responding to a threat of serious violence. It is important to remember the reason for having the first limb of the defence (1(a)(i) in paragraph 5.11 above). It is there to leave the door ajar for cases where real injustice would be done if a reduction from first degree murder to second degree murder cannot be achieved under the second limb, because D did not fear serious violence. In paragraph 5.7 we mentioned, as an illustrative example, the case of Camplin,[58] where a 15-year-old boy lost his self-control and killed after having been, as he claimed, raped and then laughed at in consequence by the victim.

    5.74      This is perhaps the most important argument because it is supported by the survey of public opinion that we commissioned by Professor Barry Mitchell for our previous Report on partial defences to murder.[59] Here is an example from that survey:

    Example 3: An Asian woman returned home to find two white men attempting to rape her 15-year-old daughter. She got a knife from the kitchen. The men shouted racist abuse at her and started to run away. She chased after them and stabbed one of them several times in the back, killing him.

    5.75      Example 3 was the subject of analysis in our survey of public opinion in 2004.[60] Our survey was conducted amongst people across a broad spectrum, with a variety of ethnic, religious and cultural backgrounds. A number of those taking part were the next-of-kin of victims of homicide.

    5.76      70% of those taking part in the survey thought that D in example 3 should receive, at worst, a prison sentence of less than 5 years. Over 40% thought that she should receive a non-custodial sentence or should not be prosecuted at all.[61] If the partial defence of provocation is restricted to cases in which the killing reflects a fear of serious violence, D in example 3 must be convicted of first degree murder if, when she is chasing after the men, she no longer fears at that moment that they will do serious violence to her or to her daughter. On conviction for first degree murder, she will receive a life sentence, a result at odds with public opinion. No doubt other analogous examples could easily be imagined or drawn from the existing case law.

    5.77      In conclusion we believe, along with a majority of our consultees, that the partial defence of provocation should extend beyond cases in which D reacted to a fear of serious violence. It should extend in principle to any case in which there was gross provocation such as caused D to have a justifiable sense of having been seriously wronged. It must be kept in mind that even if provocation satisfies this test, the defence can still fail. It should fail if the jury takes the view that a person of ordinary tolerance and temperament would not have acted in the way D did (the "ordinary person" restriction).

    BARS TO THE DEFENCE

    5.78      We are recommending that:

    It should not be possible to plead provocation either if (a) the provocation was itself incited by the defendant for the purpose of providing an excuse to use violence; or if (b) the defendant acted in considered desire for revenge.

    5.79      We have already considered (b) in paragraphs 5.17 onwards above. We regard (a) as unproblematic, although it requires a change in the law. At common law, "self-induced" provocation was not regarded as sufficient to found a provocation plea. However, the 1957 Act requires the judge to put the defence to the jury whenever there is evidence of a provoked loss of self-control. This requirement – criticised in paragraphs 5.15 to 5.16 above – may have removed the common law restriction relating to "self-induced" provocation. The law is uncertain. The common law position ought to be clearly affirmed by statute.

    5.80      The recommended new basis for pleading provocation – where someone has overreacted to a fear of serious violence (1(a)(ii) in paragraph 5.11 above) - raises the question whether there should be a further bar to the plea. This is in circumstances where someone's fear of serious violence stems from their own conduct, in exposing themselves to threats through engagement in criminal activity. A violent conflict between rival criminal gangs may, for example, give rise to a well-justified fear amongst members of the gangs that they will be targeted with serious violence. If someone, fearing that he or she faces a threat to his or her safety, kills a member of the rival gang as a pre-emptive strike, it is arguable that such a case should be specifically excluded from the scope of 1(a)(ii) in paragraph 5.11 above.

    5.81      It should be noted that the law of justified self-defence does not raise an absolute bar to the use of the plea in cases of this kind.[62] If there was an imminent threat, but the person using force to negate it was themselves to blame for the fact that they faced the threat, the law relies on the tribunal of fact to find that the use of force was not reasonable in all the circumstances. We expect a similar approach to be taken to such cases, in so far as they raise the partial defence in 1(a)(ii), in paragraph 5.11 above. Either the judge will rule that no reasonable jury could find that the defence succeeded, and will not permit the plea to be made; or the jury will themselves reject it on the grounds that a person of ordinary temperament and tolerance would not have responded in that way.

    5.82      There are, of course, cases in which people may have more or less understandably exposed themselves to the risk of criminal violence, before the need to kill due to fear of serious violence arises. Examples might be members of a severely intimidated family who remain with the gang-boss patriarch, or the persecuted individual who, in desperation, accepts help from a criminal gang to flee the regime persecuting him.[63] In such cases, the judge and jury are more likely to find that a person of ordinary temperament and tolerance might have acted as the accused did.

    THE EFFECT OF A DIMINISHED RESPONSIBILITY PLEA

    5.83      We recommend that:

    diminished responsibility should be a partial defence, with a successful plea having the effect of reducing first degree murder to second degree murder.

    5.84      The defence does not play a central role in murder cases, being successful in fewer than 20 cases annually.[64] Nonetheless, a survey of public opinion in 2003 revealed broad public support for treating in a tolerant way those who kill because of serious mental abnormality, so long as there is adequate protection against dangerous offenders.[65] Under the existing law, upon conviction for manslaughter following a successful diminished responsibility plea, the judge has discretion over the sentence to be imposed on the offender.[66] Roughly half of those who plead diminished responsibility successfully are made the subject of hospital orders under the Mental Health Act 1983.[67]

    5.85      Almost all the hospital orders made are coupled with restriction orders that make conditions for release more stringent, under the Mental Health Act 1983 section 41. The vast majority of such orders are made without any limit on their duration. A restriction order is made when the court is satisfied that it is necessary to protect the public from serious harm. The effect of the imposition of a restriction order is to make release dependent on a judgement by either the Home Secretary or a Mental Health Review Tribunal that the offender's detention is no longer necessary to protect the public from serious harm.

    5.86      On average, offenders subject to a restriction order spend 9 years in hospital before discharge.[68] Offenders successfully pleading diminished responsibility who are sent to prison (about 20% of the total) receive sentences of up to 10 years.[69] The sentencing range is thus not that dissimilar to the range found in successful provocation cases.

    5.87      Our recommendation was supported by the majority of consultees, but produced a divergence of opinion. Several consultees thought that, whatever the effect of a provocation plea was on the verdict, diminished responsibility should reduce first degree murder to manslaughter (not just to second degree murder). We have already explained in Part 2 why we regard that solution as liable to produce injustice in many cases, through splitting juries as to the appropriate verdict. Pleas of provocation and diminished responsibility may be run in tandem, even though they are (in our view) rightly kept separate as defences. They involve a significant number of overlapping issues. For example, both pleas typically arise in the context of a rage, quarrel or fight.[70] As we indicated in Part 2, for this and other reasons it would not be acceptable for these pleas to have to end in different verdicts, forcing the jury to choose which one is successful.

    5.88      Further, if diminished responsibility reduced first degree murder to manslaughter it would have to be a partial defence to second degree murder as well. Otherwise, those who have killed with the fault element only for second degree murder would irrationally be labelled more harshly (because they could not plead diminished responsibility) than those who killed with the fault element for first degree murder but successfully pleaded diminished responsibility. As we have already indicated in Part 2, however, to extend the number of crimes to which evidence of diminished responsibility bears on verdict, rather than on sentence, would be wrong. Amongst other things, it would fly in the face of the way in which medical experts, whose evidence is so crucial, perceive their role in presenting their evidence. They regard their evidence as primarily of relevance to the sentence and not to the verdict, because it can be presented in a nuanced way when relevant to the sentence, in a way it cannot be when it must relate to the verdict.

    5.89      In their response to our CP, the Royal College of Psychiatrists made clear to us how important it was, from the point of view of medical experts in homicide cases, that psychiatric evidence was wherever possible made relevant to sentencing decisions, rather than to distinctions between offences and hence to verdicts.[71] Their reasoning was as follows:

    [W]here the law does not attempt to construct 'discrete' defined 'mental condition constructs', within an adversarial legal process, but allows for a 'graded' approach to justice within sentencing, there is far less mismatch between law and psychiatry. That is, abandonment of 'trials of mental responsibility', and substitution of judicial consideration of medical evidence expressed in its own terms, is likely not only to all but abolish the 'mismatch' but also to enhance justice, so far as it depends upon the application of medical evidence.

    5.90      To make psychiatric evidence of diminished responsibility relevant to the verdict in second degree murder cases, and not just to the sentence, would be to ensure that the problem of mismatch that concerns the Royal College is retained – indeed extended[72] – within a revised structure. We believe it is right to ensure that the mismatch is kept to a minimum. That means, putting aside insanity cases, ensuring that psychiatric evidence is made relevant to the verdict only when the mandatory sentence of life imprisonment is at issue, namely in first degree murder cases.

    5.91      A minority of consultees thought that the defence should be abolished. The Criminal Cases Review Commission (CCRC) supported this suggestion. Their argument was that they came across a large number of cases in which the basis for an applicant's appeal against a conviction for murder was fresh evidence that he or she was suffering from diminished responsibility. The claim is either that the abnormality of mind was not spotted or accurately assessed before trial, or that (possibly more rarely) a new classification of abnormality of mind has been identified that explains their conduct, that was not known about at the time of the trial. In the view of the CCRC:

    In both types of cases, it is hard to assess where the interests of justice lie when considering whether, and if so how, the matter should be remedied at the appeal stage, when, usually, a jury is no longer able to assess the matter…In the CCRC's view, these faults do not lie in the old provision or in the Law Commission's proposed definition, but in the existence of the partial defence itself…If it is considered necessary to avoid mandatory life sentences for particular cases, it ought to be recognised that this is at the cost of considerable inconsistency, confusion, and possible resulting injustice in a number of other cases.

    5.92      The CCRC's view about evidence of mental disorder is to some extent consistent with submissions about such evidence that we received from, for example, Justice for Women. They thought that prison psychiatrists might not always be best placed to identify before trial symptoms of post-traumatic stress disorder or borderline personality disorder in women who have killed violent partners.[73]

    5.93      We believe that many of the problems identified by the CCRC and Justice for Women would be addressed by improving assessment procedures prior to trial. Some of these were discussed in our CP.[74] The Criminal Bar Association was one of those consultees who thought that proper consideration of the issue at the plea and case management hearing was essential. The Royal College of Psychiatrists stressed the importance of experts on both sides conducting a full and independent examination of D, if possible, rather than relying on consideration of the other side's report.

    5.94      Medical experts also commonly suggested that the partial defence should be abolished as a trial matter. Instead, they thought discretion should rest with the judge to determine the appropriate sentence post trial, if he or she found that D was suffering from diminished responsibility. This would (in very broad terms) accord with the way in which diminished responsibility is handled in France and Germany: as a sentencing matter only. It is a solution recently endorsed by the state of Victoria.[75]

    5.95      Whatever its merits may be, this solution could make significant inroads on the mandatory nature of the life sentence for murder. If they were entitled to do so, it seems likely that a significant proportion of Ds would indicate at a plea and case management hearing that, in the event that they were convicted of first degree murder by the jury, they would then like the judge to consider whether they had been suffering from diminished responsibility.[76] If a finding by the judge of diminished responsibility at that later stage freed him or her from the obligation to pass a life sentence, the mandatory nature of the life sentence for first degree murder would have become defeasible.

    5.96      It may be said that this suggestion would make no more of an inroad on the mandatory life sentence than is currently produced by the fact that the prosecution now accepts a very high proportion (70-80%) of pleas of guilty to manslaughter by reason of diminished responsibility, when there is sound medical evidence to back those pleas.[77] It is, however, one thing to avoid a trial by accepting a plea, when there is medical evidence about the offender's state of mind that is decisive in that regard. In such cases, the evidence bears directly on trial considerations. It is quite another matter not to consider that evidence until the trial ends in a guilty verdict or a plea of guilty to murder is entered. In such cases, the direct bearing the evidence has is on sentencing considerations. The latter procedure involves a direct challenge to the mandatory nature of the sentence for murder (and is hence not a solution within our terms of reference) in a way that the former does not. Moreover, such a procedure may seem to undermine or bypass the jury's verdict.

    5.97      In relation to any possible abolition of the defence of diminished responsibility, and its re-emergence as a sentencing matter post-trial, His Honour Judge Jeremy Roberts QC pointed out to us a way in which inroads on the mandatory nature of the sentence for murder could be kept to a minimum. This could be achieved if the only discretion the judge was given to depart from the mandatory sentence for murder was where the disordered offender was dangerous: where there was evidence of diminished responsibility of a kind justifying the passing of a hospital order coupled with a restriction order.

    5.98      This solution is worthy of further consideration. It does, however, have an odd consequence. This concerns offenders whose disorder would have been severe enough to warrant a reduction of the offence from first degree murder, if the partial defence of diminished responsibility still existed, but who do not constitute a threat to the public at the time of sentencing. These offenders include perhaps the most deserving of lenient treatment, such as a battered woman or a mercy killer who had the full consent of the victim. These would then become the very offenders guaranteed to receive the mandatory life sentence for first degree murder, even if they were suffering from diminished responsibility.

    5.99      More broadly, we believe that there is a case on 'labelling' grounds for reducing the offence from first degree murder, when there is evidence of diminished responsibility. The mitigating factor of an abnormality of mental functioning can be such that a killer acting under its influence does not deserve to be labelled a first degree murderer.

    5.100      To illustrate this, we can employ three examples used in our CP:

    (1) A mentally sub-normal boy is cajoled into playing a minor role in a killing by the elder brother he idolises.
    (2) A woman kills her husband having been physically and mentally abused by him over many years, to the point where she admits she has 'lost touch with reality'.
    (3) A severely depressed husband is taken past breaking point by his wife's persistent demands that he 'put her out of her misery', during a terminal illness that has progressively worsened as he has cared for her over many years. At her request, he suffocates her with a pillow.

    5.101      These cases may seem like cases that should fall into the lowest tier within a homicide structure but it must be recognised that, like provocation, diminished responsibility covers a wide range of cases. Many cases, although worthy of mitigation on the grounds of diminished responsibility, will not be so deserving as these cases. That is just one of the reasons, explored earlier (paragraphs 5.87 to 5.90), why we are not persuaded that a manslaughter verdict ought to be the outcome of a successful plea of diminished responsibility.

    5.102      An analogous, alternative approach has been suggested to us. Given that the most popular sentence following a successful plea of diminished responsibility is a hospital order, there might be a case for treating 'killing under diminished responsibility' as a discrete – treatment orientated – offence. Amongst a minority of our consultees, this reform was supported by, for example, the Law Society. There are few instances world-wide of this way of dealing with diminished responsibility, so its likely success is not easy to assess. We have already indicated in Part 2 that one reason for rejecting this approach is that it is right to label as a 'murderer', albeit as a second degree murderer, someone who kills with the fault element for first degree murder, even if, leaving aside complete defences such as insanity, the extenuating circumstances warrant reduction of the offence.

    5.103      There was very wide support for the view that, as a matter of principle, the effect of a successful plea of diminished responsibility should be to reduce first degree murder to the middle-tier offence, second degree murder. This view was supported by, for example, the Higher Courts' Homicide Working Party, the Association of Chief Police Officers, the Police Federation, Victims' Voice, the Public Affairs Council of the Church of England, several individual judges in the higher courts, several individual academics, the Crown Prosecution Service (and by the vast majority of individual prosecutors surveyed[78]), Justice, Pro-Life, and the Royal College of Psychiatrists.

    5.104      For those who disagree with our thinking on this point, there are other reasons for rejecting the alternative approach. One reason for rejecting this option has already been given, namely that the effect of a plea of provocation and a plea of diminished responsibility should be the same, because they may be pleaded together and involve important overlapping issues (paragraph 5.87 above). The jury should not be forced to choose between them as a matter of law.

    5.105      An equally significant difficulty with the alternative approach is that the prosecution would come under the obligation to show beyond reasonable doubt that the offender was suffering from diminished responsibility. As D is not obliged to submit to an examination by the prosecution's medical experts, a contested case would become impossible to prosecute. Even if D initially supplied the prosecution with a favourable medical report, leading the prosecution to a charge of 'killing under diminished responsibility' rather than first degree murder, that would not prevent a later change of heart as to defence tactics. D may decide, further into the case, to deny the offence altogether. That, again, could unfairly put the prosecution in difficulty because the medical report may, on its own, be insufficient to support proof of diminished responsibility beyond a reasonable doubt in a contested hearing. The court might permit the prosecution to amend the indictment to include a charge of first degree murder in such circumstances, but we see no reason why the law should be permitted to become this complex.

    5.106      Finally, in practice, this approach is not really necessary, even in the interests of the accused. The prosecution in England and Wales generally accepts a plea of guilty to manslaughter by reason of diminished responsibility, if it is backed by sound expert medical opinion, far more often than is usual in some other jurisdictions.[79] Therefore, it seems unlikely that more contested trials would be avoided by giving the prosecution the option of charging 'homicide by reason of diminished responsibility'.

    THE DEFINITION OF DIMINISHED RESPONSIBILITY

    5.107      We recommend that:

    The definition of 'diminished responsibility' should be modernised, so that it is both clearer and better able to accommodate developments in expert diagnostic practice.

    5.108      Diminished responsibility is currently defined as follows:

    [S]uch abnormality of mind (whether arising from a condition of arrested or retarded development of mind or any inherent causes or induced by disease or injury) as substantially impaired [the defendant's] mental responsibility for [his or her] acts or omissions in doing or being a party to the killing.[80]

    5.109      There has been consistent criticism of the way in which diminished responsibility was defined in the 1957 Homicide Act. Lord Justice Buxton has gone so far as to describe the wording of the Act as "disastrous" and "beyond redemption".[81] There are two principal problems.

    5.110      First, the definition says nothing about what is involved in a "substantial impairment [of] mental responsibility". The implication is that the effects of an abnormality of mind must significantly reduce the offender's culpability. The Act neither makes this clear, nor says in what way the effects of an abnormality of mind can reduce culpability for an intentional killing, such that a manslaughter verdict is the right result.

    5.111      Secondly, the definition has not been drafted with the needs and practices of medical experts in mind, even though their evidence is crucial to the legal viability of any claim of diminished responsibility. 'Abnormality of mind' is not a psychiatric term, so its meaning has had to be developed by the courts from case to case. Further, diagnostic practice in diminished responsibility cases has long since developed beyond identification of the narrow range of permissible 'causes' of an abnormality of mind stipulated in the bracketed part of the definition. In any event, the stipulated permissible causes have never had an agreed psychiatric meaning. The outmoded stipulation of permissible causes has become as much a hindrance as a help. As Dr Madelyn Hicks put it to us:

    [A]ttempting to specify the cause of mental disorders…is irrelevant [and] misleading, and in fact there are almost always multiple causes stemming from the interaction between genetic vulnerability and life events.[82]

    5.112      To address these problems, in our CP[83] we provisionally proposed a new definition of diminished responsibility, developed from a definition adopted in the state of New South Wales in 1997. That definition received very broad support when we first proposed a version of it in 2004. Once again, it has had the support of a majority of consultees, but we have also had some helpful suggestions for improvement. In response to comments and analysis that we have received, we recommend adoption of the following definition:

    (a) a person who would otherwise be guilty of first degree murder is guilty of second degree murder if, at the time he or she played his or her part in the killing, his or her capacity to:
    (i) understand the nature of his or her conduct;[84] or
    (ii) form a rational judgement;[85] or
    (iii) control him or herself,
    was substantially impaired by an abnormality of mental functioning arising from a recognised medical condition, developmental immaturity in a defendant under the age of eighteen,[86] or a combination of both; and
    (b) the abnormality, the developmental immaturity, or the combination of both provides an explanation for the defendant's conduct in carrying out or taking part in the killing.

    5.113      It is envisaged that this definition would improve the present law in the following ways.

    5.114      First, the law will no longer be constrained by a fixed and out-of-date set of causes from which an abnormality of mental functioning ('mental functioning' is a term preferred by psychiatrists to 'mind') must stem. The issue will be whether the abnormality was brought about by a 'recognised medical condition'. The Royal College of Psychiatrists supported this change, saying:

    The presence of such a restriction is, we believe, consistent with the general nature and purpose of 'diminished responsibility' as a defence and would ensure that any such defence was grounded in valid medical diagnosis. It would also encourage reference within expert evidence to diagnosis in terms of one or two of the accepted internationally classificatory systems of mental conditions (WHO ICD10 and AMA DSM) without explicitly writing those systems into the legislation…Such an approach would also avoid individual doctors offering idiosyncratic 'diagnoses' as the basis for a plea of diminished responsibility. Overall the effect would be to encourage better standards of expert evidence and improved understanding between the courts and experts.

    5.115      In that regard, at the Royal College's suggestion, we must correct an impression that we gave about the nature of conditions capable of falling within the new definition. In the CP we made reference to 'serious' mental conditions, but that is potentially misleading. A condition could be serious in the sense that it is untreatable and liable to worsen, or serious in the sense that great stigma is attached to its sufferers, without meeting the conditions for reducing the offence from first degree murder under the new provisions.

    5.116      In law it is sufficient that a condition is 'more than trivial'. What matters is that it has the effect of substantially impairing D's capacities.[87] It follows that, alongside the familiar psychotic disorders that fall within the scope of the provisions, neurotic disorders, for example, may also do so. An example of the latter would be a post traumatic stress disorder suffered by a woman due to violent abuse suffered over many years.

    5.117      Secondly, we believe the provisions make clearer the relationship between the role of the expert and the role of the jury. It is for the experts to offer an opinion on:

    (1) whether D was suffering from an abnormality of mental functioning stemming from a recognised medical condition; and
    (2) whether and in what way the abnormality had an impact on D's capacities, as these are explained in the new provisions.

    5.118      It is then for the jury to say whether, in the light of that (and all the other relevant) evidence they regard the relevant capacities of D to have been 'substantially impaired'. On this point, the Royal College of Psychiatrists commented:

    Although it is common for the courts to accept, or even encourage a psychiatric expert to comment upon whether the defendant should be seen as 'substantially impaired', the College believes that this should be resisted…Our belief is that this restriction should apply irrespective of the 'side' which is calling an expert. We believe that this can be achieved without in any way leaving the jury 'floundering' with materials and issues with which they are not in a position to cope…Hence, for example, an expert can inform the jury in a murder trial of the likely effects of the defendant's disorder in terms of the various mental capacities, yet still fall short of saying whether these 'amounted to' substantial impairment of mental responsibility.

    5.119      The Royal College here affirms the view that an expert should not express an opinion on what lawyers call the 'ultimate issue', the question whether D did or did not meet all the criteria to warrant reducing first degree murder to second degree murder, on the grounds of diminished responsibility. In the Royal College's opinion, the recommended definition of diminished responsibility will not require medical experts to express a view on the ultimate issue and hence trespass on the jury's territory.

    5.120      In that regard, Professor Nigel Eastman, a member of the Royal College and a medical expert with many years' experience of giving evidence in murder cases, pointed out that there is an additional reason to avoid a situation in which the medical expert gives an opinion on the ultimate issue. This is that such an opinion will inevitably be based, in part, on assumptions about the factual underpinning for the diminished responsibility claim when that background may itself be a contested matter between prosecution and defence. As he put it:

    The reasons I believe that it is wrong for a doctor to give an opinion on the ultimate issue of 'diminished' is not merely that it is 'legally wrong' within the 'law of evidence'. It is that, to do so, usually means taking a view about matters of fact. And that seems an entirely wrong thing for an expert to do. It also opens up enormously the lines of cross examination, since then the doctor will be asked questions which arise directly out of his/her 'assumption' of a particular version of past and offence-related events, which the 'other side' may dispute.

    5.121      Thirdly, the new provisions seek to make clear what impact on capacity the effects of an abnormality of mental functioning must have, if the abnormality is to be the basis for a successful plea of diminished responsibility. It might be helpful to illustrate this by example, although it is not our case that on the actual facts given below the defence should necessarily succeed.

    (1) Substantially impaired capacity to 'understand the nature of his or her conduct':
    (a) a boy aged 10 who has been left to play very violent video games for hours on end for much of his life, loses his temper and kills another child when the child attempts to take a game from him. When interviewed, he shows no real understanding that, when a person is killed they cannot simply be later revived, as happens in the games he has been continually playing.
    (2) Substantially impaired capacity to 'form a rational judgement':
    (a) a woman suffering from post traumatic stress disorder, consequent upon violent abuse suffered at her husband's hands, comes to believe that only burning her husband to death will rid the world of his sins;
    (b) a mentally sub-normal boy believes that he must follow his older brother's instructions, even when they involve taking take part in a killing. He says, 'I wouldn't dream of disobeying my brother and he would never tell me to do something if it was really wrong';
    (c) a depressed man who has been caring for many years for a terminally ill spouse, kills her, at her request. He says that he had found it progressively more difficult to stop her repeated requests dominating his thoughts to the exclusion of all else, so that 'I felt I would never think straight again until I had given her what she wanted.'
    (3) Substantially impaired capacity to 'control him or herself':
    (a) a man says that sometimes the devil takes control of him and implants in him a desire to kill, a desire that must be acted on before the devil will go away.

    5.122      Fourthly, it has never been entirely clear whether, under the existing law, the abnormality of mind must, in some sense, 'cause' D to kill. The law simply states that the abnormality of mind must substantially impair D's mental responsibility for his acts in doing or being a party to the killing.

    5.123      Leading experts on this area of law, such as Professor Mackay,[88] advised against the introduction of a strict causation requirement. The Royal College of Psychiatrists did not object to the requirement, as such, but cautioned against creating a situation in which experts might be called on to 'demonstrate' causation on a scientific basis, rather than indicating, from an assessment of the nature of the abnormality, what its likely impact would be on thinking, emotion, volition, and so forth.

    5.124      The final choice of particular words is a matter for those drafting the legislation. However, we have framed the issue in these terms: the abnormality of mind, or developmental immaturity, or both, must be shown to be 'an explanation' for D's conduct. This ensures that there is an appropriate connection (that is, one that grounds a case for mitigation of the offence) between the abnormality of mental functioning or developmental immaturity and the killing. It leaves open the possibility, however, that other causes or explanations (like provocation) may be admitted to have been at work, without prejudicing the case for mitigation.

    DEVELOPMENTAL IMMATURITY

    5.125      We recommend that:

    It should be possible to bring in a verdict of diminished responsibility on the grounds of the developmental immaturity of an offender who was under 18 at the time he or she played his or her part in the killing.

    5.126      This part of our recommendation for reform of the defence is so important that it warrants discussion under a separate heading.

    5.127      The recommendation was supported by a large majority of our consultees. It had the support of, for example, the Royal College of Psychiatrists, the Judges of the Central Criminal Court, the Criminal sub-committee of the Council of Her Majesty's Circuit Judges, the Criminal Bar Association, the NSPCC, the Youth Justice Board, Justice, and the Law Society.

    5.128      A very important aspect of this recommendation is that evidence of developmental immaturity can be combined with evidence of an abnormality of mental functioning, to make a case for a verdict of second degree murder. As we explained in the CP, experts may find it impossible to distinguish between the impact on D's mental functioning of developmental immaturity, and the impact on that functioning of a mental abnormality. To force experts – as the law currently does – to assess the impact of the latter, whilst disregarding the effect of the former, is wholly unrealistic and unfair.[89]

    5.129      We recognise that our recommendation may prove to be a controversial one, in spite of the level of support expressed for it. Some consultees considered that it was too generous to those who had killed with the fault element for first degree murder.[90] By way of contrast, some consultees expressed doubts about the proposal because it did not go far enough to shield child Ds from conviction for serious offences of homicide.

    5.130      It is important to recognise the nature and limits of what we are suggesting. In England and Wales, criminal liability for murder can be imposed on an offender if he or she was at least 10 years of age at the time of the offence. We are not suggesting that imposing liability for murder on a child of this age is always unfair or inappropriate. Some 10-year-old killers may be sufficiently advanced in their judgement and understanding that such a conviction would be fair.[91]

    5.131      What we are suggesting is that it is unrealistic and unfair to assume that all children aged 10 or over who kill must have had the kind of developed sense of judgement, control and understanding that makes a first degree murder conviction the right result (provided the fault element was satisfied). Instead, our recommendation is that it should be for the jury to decide in the individual case whether D had such a sense of judgement, control, or understanding. Moreover, it will be for the D to prove that his or her capacity for judgement, control and understanding was substantially impaired by developmental immaturity.

    5.132      D may wish to prove substantial impairment by developmental immaturity through appeal either to biological factors, or to social and environmental influences, or to a combination of both. For example, D may wish to give evidence that his or her power of control over his or her actions was substantially impaired by a biological factor such as poor frontal lobe development. This is because, as the Royal College of Psychiatrists have put it:

    Biological factors such as the functioning of the frontal lobes of the brain play an important role in the development of self-control and of other abilities. The frontal lobes are involved in an individual's ability to manage the large amount of information entering consciousness from many sources, in changing behaviour, in using acquired information, in planning actions and in controlling impulsivity. Generally the frontal lobes are felt to mature at approximately 14 years of age.[92]

    5.133      As this final sentence implies, however, in an individual case involving a child under 14 years of age, it would be open to the prosecution to seek to rebut evidence of poor frontal lobe development by arguing that this particular D had matured to a sufficient degree to be fairly convicted of first degree murder. The jury should be trusted to reject implausible claims, as they are with other defences based on expert evidence.

    5.134      So far as social and environmental influences on young killers are concerned, two studies have shown how these influences may, in extremely rare circumstances, lead children or young persons to kill.[93] One study has shown that seriously disturbed children who murder have home backgrounds in which there was commonly paternal psychopathy, alcohol abuse, absence from the home, and a history of violent behaviour by fathers. Another study has shown depressive illnesses in the mothers of juvenile murderers, and histories of serious sexual and physical abuse of juvenile murderers.

    5.135      Such evidence might be employed by D to suggest that he or she was developmentally immature to the extent of having substantially impaired judgement, control, or understanding. It would be open to the prosecution to seek to rebut such evidence, on the facts of an individual case. Again, the jury should be trusted to reject implausible claims. In that regard, we are very grateful to Professor Sue Bailey,[94] who has given us a template providing a guide to the ways in which the nature and degree of developmental immaturity can be assessed by experts in the field (Appendix F). There should be no shortage of experts sufficiently well-versed in the subject to provide expert opinions of the kind that would be needed by both sides, were diminished responsibility to be reformed as we recommend.

    5.136      Offenders under the age of 18 commit a very small proportion of the 850 or so homicides that occur each year (around 4%). We find it hard to imagine that more than a low proportion of the even smaller number from this group finding themselves charged with first degree murder will be in a strong position to claim that they were suffering diminished responsibility, in virtue of developmental immaturity alone. Even those who do succeed in persuading the jury of the merits of their claim – and they are likely to be the younger age group, who commit the fewest homicides – will still be convicted of second degree murder, and can be sentenced to anything up to and including life imprisonment.

    5.137      Our recommendation is, therefore, likely to affect only a very few cases, and only by reducing the crime from top-tier homicide to middle-tier homicide. However, for the few cases that do meet the criteria, we believe our recommendation meets requirements of justice recognised as fundamental in civilised legal systems across the world.

    DIMINISHED RESPONSIBILITY AND INSANITY

    5.138      Our recommendations do not affect the availability of the defence of insanity as a complete defence to murder. Professor Mackay quite properly asked us how we saw the recommended change to the definition of diminished responsibility affecting the relationship between diminished responsibility and insanity.[95] It could be argued that our recommendation is a more restrictive version of the existing law, meaning that to prove diminished responsibility D would have to show that he or she was 'borderline insane'. That would mean not only a narrower range of Ds who could use evidence of mental disorder as a defence to murder in some form, but also a very fine line between the conditions for the partial defence (diminished responsibility), and the conditions for a complete acquittal (insanity).

    5.139      It is possible that our recommendations will actually broaden the scope of the diminished responsibility plea. For the first time, the definition makes provision in general terms for 'recognised medical conditions' to form the basis for a plea. The nature and scope of such conditions will evolve with changing diagnostic practice, and this may enlarge somewhat the range of mentally disordered Ds who might make a successful plea. It must be kept in mind, however, that just as some new medical conditions giving rise to abnormalities of mental functioning may come to be recognised (potentially broadening the reach of defence), so other medical conditions hitherto accepted may be shown to have been unsoundly based (narrowing the reach of the defence).

    5.140      It is also important to recognise how legal change may affect changes in prosecution policy, generating an impact in practice on the scope of the defence, whatever its limits in theory. As we have already indicated (paragraph 5.106 above), there is currently a high rate of acceptance by the prosecution of a plea of guilty to manslaughter by reason of diminished responsibility, when the plea is based on sound medical evidence. If our recommended definition is accepted, it may subsequently be judged right for the prosecution to look more sceptically at some diminished responsibility claims than others, and be more willing to test some rather than others at trial.

    5.141      It is possible to imagine, for example, a sceptical view being more commonly taken of claims by jealous husbands, who have killed their wives or children, that their capacity to form a rational judgement was substantially impaired by clinically recognised depression. Prosecutors may well think it right to show a much greater readiness to test claims of this kind before the jury[96] than they might show in a case where, for example, a child assessed before trial as severely subnormal has become involved in a killing at the urging of his domineering father. We take no stance on this issue ourselves.

    5.142      Finally, there is the legal relationship between diminished responsibility and insanity. In theory, the definition of insanity means that whether a defect of reason (stemming from a disease of the mind) amounts to insanity in law is an 'all or nothing' matter. Either D shows that the defect of reason led him or her not to know the nature or quality of his or her act, or that the act was wrong, or the defect of reason did not have that effect.[97] By way of contrast, the definition of diminished responsibility has always meant that whether an abnormality of mind (stemming from one of the required causes) amounts to diminished responsibility in law is a matter of judgement and degree. 'Substantial' impairment of mental responsibility is what is required, and that by its nature is a matter of degree. So, under our recommendations, there will remain an important theoretical distinction between the insanity plea (involving an all-or-nothing question) and the plea of diminished responsibility (involving a judgement of degree as to whether there was a substantial impairment by an abnormality of mental functioning).[98]

Ý
Ü   Þ

Note 1    Detailed supporting argument for those recommendations can be found in the report itself, Partial Defences to Murder (2004) Law Com 290.    [Back]

Note 2    Much fuller detail can be found in Sentencing Advisory Panel: Manslaughter by Reason of Provocation (2005), 18 to 19.    [Back]

Note 3    See Appendix C.    [Back]

Note 4    Or, depending on the age of the offender, a sentence of custody for life, or detention at her Majesty’s pleasure.    [Back]

Note 5    See Scott (unreported, 20 October 2000), cited by the Sentencing Advisory Panel: Manslaughter by Reason of Provocation (2005), 11; Atland [2003] EWCA Crim 474.    [Back]

Note 6    [1978] AC 705.    [Back]

Note 7    Germany is an example of such a jurisdiction.    [Back]

Note 8    Partial Defences to Murder (2004) Law Com No 290.    [Back]

Note 9    See our earlier report, Partial Defences to Murder (2004) Law Com No 290.    [Back]

Note 10    Indeed, the Court of Appeal has said that when the offence has been reduced to manslaughter by the jury on the grounds of provocation, the judge should not treat the case as one involving mere jealous rage, for example, but as one involving nothing less than loss of self-control: A-Gs Reference (Nos 74 to 76 of 2002) (Suratan & Others) [2002] EWCA Crim 2982, [2003] 1 Cr App R (S) 42.    [Back]

Note 11    See, for example, the contrast between the views of Devlin J in Duffy [1949] 1 All ER 932n, and of Lawton LJ in Ibrams (1982) 74 Cr App R 154 (both taking a narrow view of the requirement) and the views of Lord Lane CJ in Ahluwalia [1992] 4 All ER 859 (taking a broader view of the requirement).    [Back]

Note 12    For a helpful discussion, see Simester and Sullivan, Criminal Law: Theory and Doctrine (2nd edition 2003) pp 344 to 348.    [Back]

Note 13    Oliver Quick and Celia Wells, in “Getting Tough with Defences” [2006] Criminal Law Review 514, 522, claim in relation to our proposals that, ”The CP retains the notion of loss of self-control, and its association with male temper killings…”. We do not believe there is a basis for this claim in anything that we have proposed or recommended. We respectfully refer them to our criticisms of the loss of self-control requirement in Partial Defences to Murder (2004) Law Com No 290, paras 3.28 to 3.30, and 3.135 to 3.137, criticisms that we continue to believe are valid.    [Back]

Note 14    See Appendix C.    [Back]

Note 15    Above, paras C.7 to C.8    [Back]

Note 16    Partial Defences to Murder (2004) Law Com No 290, paras 3.47 to 3.59.    [Back]

Note 17    Above, Appendix F.    [Back]

Note 18    Above, paras 3.28, and 3.104 to 3.107.     [Back]

Note 19    See eg, Thornton [1992] 1 All ER 306; Ahluwalia [1992] 4 All ER 889.    [Back]

Note 20    See eg, Salma Choudry, “Pakistani Women’s Experience of Domestic Violence in Great Britain”, Home Office Research Findings No 43 (1996).    [Back]

Note 21    [1995] Criminal Law Review 739.    [Back]

Note 22    Appendix C.    [Back]

Note 23    Thomson v HMA (1986) SLT 281. It would not, however, be desirable to follow the rule in Scotland that the cumulative effect of provocation other than that immediately preceding the final provocation cannot be considered.    [Back]

Note 24    As the older English cases required.    [Back]

Note 25    [1995] Criminal Law Review 739.    [Back]

Note 26    [1995] UKPC 23, [2005] 2 AC 580.    [Back]

Note 27    Smith (Morgan) [2001] AC 146.    [Back]

Note 28    [2005] 2 AC 580, 594 to 595.    [Back]

Note 29    Such a provocation is, of course, very minor and could not in itself be regarded as sufficient to reduce the offence to second degree murder. Under our recommendations, we would expect such a provocation (unsupported by other provocation of a substantial kind) to fail at the first hurdle, namely at the stage when the judge decides whether or not to put it to the jury.    [Back]

Note 30    See Camplin [1978] AC 705; Morhall [1996] AC 90.    [Back]

Note 31    It is not the function of the reasonable person requirement to lay down, as a matter of law, what kinds of provocation can and cannot be treated as “gross” or as giving rise to a “justifiable sense of having been seriously wronged”, to use our definition of provocation from 1(a) in para 5.11 above. The jury should be trusted to evaluate the relative grossness of provocation, in whatever form it comes, according to their own sense of justice in an individual case.    [Back]

Note 32    See Partial Defences to Murder (2004) Law Com No 290, paras 3.109 to 3.134    [Back]

Note 33    Save only in the rare case when they are relevant to the gravity of the provocation because D has been taunted about them. See A-G for Jersey v Holley [2005] UKPC 23, [2005] 2 AC 580.    [Back]

Note 34    Weller [2004] 1 Cr App R 1. The decision itself does not sit well with our recommendations.    [Back]

Note 35    That may be easier for them to do, under our recommendations, than it would be under the existing law because we are replacing the requirement that an abnormality of mind (mental functioning) stem from specified causes with a requirement instead that it stem from a ’recognised medical condition’: see para 5.112 below.    [Back]

Note 36    John Gardner, “The Gist of Excuses” (1998) 1 Buffalo Criminal Law Review 575, 579 to 587 (Gardner’s emphasis).    [Back]

Note 37    For the Commission’s view about cases of mistaken provocation more generally, see Partial Defences to Murder (2004) Law Com No 290, paras 3.153 to 3.160. Where D has lost self-control and reacted more or less spontaneously, there is a case for asking the jury to treat a mistaken provocation as if it were a true provocation, even if the mistake seems in hindsight unreasonable. The contrast is with duress cases, where there will typically have been more time to reflect on whether there has been a mistake about the nature of the duress: hence a requirement of reasonableness in relation to that mistake (see Part 6).    [Back]

Note 38    [1990] Criminal Law Review 122.    [Back]

Note 39    The discussion of such cases by Lord Nicholls in A-G for Jersey v Holley [2005] UKPC 23; [2005] 2 AC 580, 594, accords with the way in which we envisage our proposals operating in practice.    [Back]

Note 40    See Judicial Studies Board, Model Directions (Provocation); Rowlands, The Times 12 January 2004. See also the CP, para 6.135.    [Back]

Note 41    Buckner (1641) Style 467; Hopkin Huggett (1666) Kel 59; Goffe (1672) 1 Vent 216; Mawgridge (1707) Kel 119, 135 to 137 (by Holt CJ).    [Back]

Note 42    Robertson v HMA (1994) SCCR 589; Gilllon v HMA [2006] HCJAC 61.    [Back]

Note 43    This ’ordinary person’ restriction does not apply to the relevant limb of the provocation defence in Scotland.    [Back]

Note 44    Criminal Law Act 1967, s 3.    [Back]

Note 45    See Clegg [1995] 1 AC 482.    [Back]

Note 46    By analogy with duress cases, the defence should not be available where the D had, without reasonable excuse, knowingly exposed him or herself in advance to the risk of being threatened with violence. Such cases should be excluded by the judge’s use of his or her new power – see (5) in para 5.11 – to withdraw the issue from the jury when no reasonable jury, properly directed, could find that the defence succeeds.    [Back]

Note 47    Subject to any control that may be exercised by the judge over the options put to the jury: see now Coutts [2006] UKHL 39; [2006] 1 WLR 2154.    [Back]

Note 48    This proposal has been discussed with the Police Federation, who indicated to us that they could see logic in both courses being available to a D such as D in Example 2, as we are suggesting.    [Back]

Note 49    Coutts [2006] 1 WLR 2154, 2159.    [Back]

Note 50    For a fuller discussion, see the CP, paras 6.128 to 6.140.    [Back]

Note 51    Although any additional, ’defensive’ motivation would count in favour of the provocation plea, insofar as it was generated by a fear of serious violence.    [Back]

Note 52    Susan S M Edwards, “Sex and Gender in the Legal Process” (1996), 394 to 395; Partial Defences to Murder (2004) Law Com No 290, para 3.28.    [Back]

Note 53    Oliver Quick and Celia Wells, “Getting Tough with Defences” [2006] Criminal Law Review 514, 524.    [Back]

Note 54    Partial Defences to Murder (2003) Law Com Consultation Paper No 173, para 6.15.    [Back]

Note 55    Assuming that they are ‘stand alone’ provocations and not simply the ’last straw’ in a long history of provocation that, seen as whole, could reasonably amount to gross provocation of a kind that satisfies our suggested test in para 5.11 above.    [Back]

Note 56    For example, a question arises under the Housing Act 1996 s 198(2A) of what it would be for someone to be exposed to ’violence’ if housed in a particular area. Does that include, say, harassment under the Protection from Harassment Act 1997, offences concerned with making threats, public order offences, or ’psychic’ assaults?    [Back]

Note 57    See Criminal Justice Act 1967, s 8.    [Back]

Note 58    Camplin [1978] AC 705.    [Back]

Note 59    Partial Defences to Murder (2004) Law Com No 290, Appendix C.    [Back]

Note 60    For analysis of such an example against an American law background, see Susan D Rozelle, ‘Controlling Passion: Adultery and the Provocation Defence’ (2005) 37 Rutgers Law Journal 197.    [Back]

Note 61    Partial Defences to Murder (2004) Law Com No 290, Appendix C, annex 3.    [Back]

Note 62    See the discussion in Smith and Hogan, Criminal Law (11th ed 2005) p 337.    [Back]

Note 63    See, on this point, the speech of Baroness Hale, in Z [2005] UKHL 22; [2005] 2 AC 467.    [Back]

Note 64    The number of successful pleas has dropped dramatically in recent years. In 1980 to 1986, there were 70 to 85 successful pleas per year: R D Mackay, Mental Condition Defences in the Criminal Law (1995) p 181.    [Back]

Note 65    See Partial Defences to Murder (2004) Law Com No 290, Appendix C (research conducted by Professor Barry Mitchell).    [Back]

Note 66    For the judicial approach to guidance on sentencing, see Chambers (1983) 5 Cr App R (S) 190. It is now possible to make a ’hospital and limitation direction’, under the Crime (Sentences) Act 1997, s 45, if the offender is suffering from a psychopathic disorder. This means that a court can send an offender to prison, whilst also directing that he or she be admitted to hospital for treatment. The direction is very rarely used: see Home Office Statistical Bulletin 13/2002 (recording three cases in 2000 and in 2001).    [Back]

Note 67    In a majority of cases where diminished responsibility has been successfully pleaded, the result will be one or other of a hospital order without limit on time, or a sentence of life imprisonment: see Partial Defences to Murder (2004) Law Com No 290, Appendix B (research conducted by Professor R Mackay). There will be some cases that will rightly be met with a determinate sentence of imprisonment. An example might be a case in which the effects of the abnormality of mind are much less severe, or non-existent, by the time the trial has been concluded.     [Back]

Note 68    See Andrew Ashworth, Sentencing and Criminal Justice (4th edition 2005) p 376.    [Back]

Note 69    See Andrew Ashworth, Principles of Criminal Law (4th edition 2003) pp 284 to 285.    [Back]

Note 70    See Partial Defences to Murder (2004) Law Com No 290, Appendix B (research conducted by Professor R Mackay).    [Back]

Note 71    Indeed, their opposition to the mandatory life sentence for murder is founded on their judgement that, “its retention greatly enhances the difficulty in usage of expert psychiatric evidence in homicide trials”: (Response of the Royal College of Psychiatrists).    [Back]

Note 72    In so far as it would then bear on the question of whether killing through an intention to cause injury, or a fear or risk of injury, in the awareness of a serious risk of causing death should be second degree murder or manslaughter. To date, psychiatric evidence in such cases has been relevant only to the appropriate sentence, because these are manslaughter cases under the existing law.    [Back]

Note 73    For a case study, see the CP, Appendix F.    [Back]

Note 74    Paras 6.105 to 6.106.    [Back]

Note 75    Victorian Law Reform Commission, Defences to Homicide (2004), 241.    [Back]

Note 76    Defendants might also more often plead guilty to first degree murder, to reach this stage.    [Back]

Note 77    See Partial Defences to Murder (2004) Law Com No 290, Appendix B (research conducted by Professor R Mackay).    [Back]

Note 78    See the CP, Appendix B.    [Back]

Note 79    For example, Professor Mackay told us that, between 1990 and 1993, whereas some 70% of diminished responsibility cases went to trial and only 30% of Ds had pleas of diminished responsibility accepted in New South Wales, only 23% of such cases went to trial in England and Wales.    [Back]

Note 80    Homicide Act 1957, s 2.    [Back]

Note 81    See Partial Defences to Murder (2004) Law Com No 290, para 5.43 n 40. The late Professor Griew said, “the wording is altogether a disgrace”: “Reducing Murder to Manslaughter: Whose Job?” (1986) 12 Journal of Medical Ethics 18. See, the CP, paras 6.36 to 6.41.    [Back]

Note 82    The CP, para 6.40. Dr Hicks is a Consultant Psychiatrist and Honorary Lecturer, Institute of Psychiatry, King’s College London.    [Back]

Note 83    Para 6.2.    [Back]

Note 84    This wording replaces the wording of ’understand events’, at the suggestion of the Criminal Cases Review Commission, and (independently) of Professor Mackay. The aim is to ensure that the accused’s lack of understanding of, say, global political events, is not relevant to his or her plea.    [Back]

Note 85    This wording replaces “judge whether his or her actions were right or wrong”. The Royal College of Psychiatrists, whilst content for this phrase to appear, considered that “form a rational judgement” was apt to cover cases the original phrase was not. An example might be one in which a deluded D killed someone he believed to be the reincarnation of Napoleon. D might realise that it is morally and legally wrong to take the law into one’s own hands by killing, and yet be suffering from a substantially impaired capacity to form a rational judgement. Professor Mackay also cast doubt on the ’right/wrong’ formula.    [Back]

Note 86    This new part of the definition is discussed from para 5.125 onwards.    [Back]

Note 87    Lloyd [1967] 1QB 175    [Back]

Note 88    Professor of Criminal Policy and Mental Health, De Montfort University, Leicester.    [Back]

Note 89    It would, of course, be possible to meet this point simply by stipulating that evidence of developmental immaturity can only form part of a case for diminished responsibility if it augmented or accompanied the effects of an abnormality of mind.    [Back]

Note 90    This objection may to some extent by met by lowering the age at which D ceases to be eligible to plead developmental immaturity as part of his or her case that he or she was suffering from diminished responsibility. We recommend 18 years of age (as against the Royal College of Psychiatrists’ recommendation of 21 years of age), but it would clearly be possible to lower the upper age limit to, say, 16 years of age.    [Back]

Note 91    The Police Federation suggested that, ”there needs to be recognition that children are more advanced in present day life than at any time previously in history and to ignore this fact would be an injustice to the families of the deceased.” Theirs is, however, a controversial claim so far as moral development is concerned. The Royal College of Psychiatrists has suggested, by way of contrast, that, “Given the eagerness to please and suggestibility of learning-disabled children who may have learned legal or psychological phrases off by heart, the automatic recitation of such jargon should not be taken to indicate a good understanding of the process but, rather, an attempt to take some control over a new situation by attempting to sound competent or streetwise”, Child Defendants (Occasional Paper OP56, March 2006).    [Back]

Note 92    Royal College of Psychiatrists, Child Defendants (Occasional Paper OP56, March 2006), 38.    [Back]

Note 93    Royal College of Psychiatrists, ‘Child Defendants’ (Occasional Paper OP56, March 2006), 41, citing studies by S Bailey, “Adolescents who murder” (1996) 19 Journal of Adolescence, 19 to 39; G Boswell, “Young and Dangerous: The Background and Careers of Section 53 Offenders” (1996).    [Back]

Note 94    Lancashire School of Health and Postgraduate Medicine, University of Central Lancashire.    [Back]

Note 95    See M’Naghten’s Case (1843) 10 Cl & Fin 200, 210, ”the jurors ought to be told in all cases that…to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature or quality of the act he was doing, or if he did know it, that he did not know he was doing what was wrong.”    [Back]

Note 96    And/or to use information obtained, when possible, through the disclosure provisions brought in by the Criminal Justice Act 2003, s 35.    [Back]

Note 97    We say, ’at least in theory’, because it is not our case that, in practice, the defence of insanity is inflexibility understood as an all-or-nothing matter in all cases in which it is pleaded. For more detailed discussion, see R D Mackay, Mental Condition Defences in the Criminal Law (1995) pp 100 to 108.    [Back]

Note 98    This explanation broadly follows the way in which insanity and what English lawyers call diminished responsibility are distinguished in French law where, whilst insanity is a complete defence (as in England), diminished responsibility is a matter for sentencing.    [Back]

Ý
Ü   Þ


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/other/EWLC/2006/304(5).html