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

The Law Commission


You are here: BAILII >> Databases >> The Law Commission >> Partial Defences to Murder (Report) [2004] EWLC 290(3) (06 August 2004)
URL: http://www.bailii.org/ew/other/EWLC/2004/290(3).html
Cite as: [2004] EWLC 290(3)

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



     
    PART 3
    PROVOCATION
    Structure of this part
    3.1     The subject of the defence of provocation is central to this project. In this part we set out our recommendations and the thought process which has led us to make them. To assist readers to follow that thought process, we begin by outlining the structure of this part.

    3.2    
    In paras 3.13 – 3.14 we refer to the relevant parts of Consultation Paper No 173 without recapitulating their contents.

    3.3    
    In paras 3.15 – 3.17 we refer to the provisional conclusions which we published in April 2004.

    3.4    
    In paras 3.18 – 3.19 we comment on the response of many consultees about the need for a wider review of the law of murder.

    3.5    
    In paras 3.20 – 3.31 we identify the major sources of consultees' dissatisfaction with the present law – the lack of a clear rationale and the unsatisfactoriness of its key components (the concept of provocation, the supposed requirement of a sudden and temporary loss of self-control and the supposed objective test). We explore each of those factors in the paragraphs which follow, but not in the same detail as in Consultation Paper No 173.

    3.6    
    In paras 3.32 – 3.46 we address the question whether the defence should be abolished or retained in some form.

    3.7    
    In paras 3.47 – 3.59 we consider the case for reshaping the defence along the lines of the EMED (Extreme Mental or Emotional Disturbance) defence under the American Law Institute Model Penal Code.

    3.8    
    In paras 3.60 and following we set out our approach to the reform of provocation. We deal with:

    3.9    
    In para 3.163 we refer to the question whether there should be a separate partial defence of excessive force in self-defence.

    3.10    
    In paras 3.164 – 3.166 we refer to the proposal for merging provocation and diminished responsibility.

    3.11    
    In para 3.167 we consider the question of the burden of proof.

    3.12    
    In para 3.168 we set out our recommendations.

    Consultation Paper[1]
    3.13     Part III of our Consultation Paper described the history of the development of the defence of provocation.[2] In Part IV we summarised the present law and considered its defects. In Part V we considered the law of provocation in other common law jurisdictions. In Parts IX and X we considered the topics of excessive use of force in self-defence and abused women who kill. In Part XI we referred to previous recommendations for the reform of provocation, and in Part XII we set out options for reform. The principal options discussed were abolition of the defence of provocation, its retention in a modified form or merger of provocation and diminished responsibility into a single partial defence.

    3.14     The first question which we posed was

    Do consultees agree:
    (1) that the law of provocation is unsatisfactory; and
    (2) that its defects are beyond cure by judicial development of the law?[3]
    An overwhelming majority of consultees agreed with both parts of the question, but there was a wide diversity of views about what should be done.
    Provisional Conclusions
    3.15     At the end of April 2004 we sent to all consultees and published on our website a document setting out our provisional conclusions.[4] In it we provisionally concluded on balance that, even if the mandatory sentence of life imprisonment for murder were abolished, the arguments for retaining some form of provocation defence outweighed the arguments for its abolition. We provisionally proposed[5] that the following principles should govern a reformed provocation defence:

    (1) Unlawful homicide that would otherwise be murder should instead be manslaughter if the defendant acted in response to
    (a) 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
    (b) fear of serious violence towards the defendant or another; or
    (c) a combination of (a) and (b); and
    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 a similar way.
    (2) In deciding whether a person of the defendant's age and of ordinary temperament in the circumstances of the defendant might have acted in the same or a similar way, the court should take into account all the circumstances of the defendant other than matters (apart from his or her age) which bear only on his or her general capacity for self-control.
    (3) The partial defence should not apply where
    the provocation was incited by the defendant for the purpose of providing an excuse to use violence, or
    the defendant acted in pre-meditated desire for revenge.
    (4) A person should not be treated as having acted in pre-meditated 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.
    3.16     We emphasised that this was not put forward as a statutory formula. Rather, we were seeking to identify the principles which should govern any legislative reform. If those principles were accepted, the drafting of legislation would be a matter for Parliamentary Counsel.

    3.17    
    We had 146 written responses to Consultation Paper No 173 and 30 responses to our provisional conclusions.

    The wider law of murder
    3.18    
    Many consultees have argued that there is a need for a broader review of the law of murder, and that the problems associated with the defence of provocation cannot be satisfactorily tackled in isolation from other parts of the law of murder. We agree.[6]

    3.19     Different systems of criminal law have different ways of grading homicides, but our law with its broad definition of murder, mandatory sentence and patchwork partial defences is a product of piecemeal development and reforms, rather than systematic thought. The expansion of provocation over the past half century can be seen as an attempt by courts and juries to avoid a conviction of murder with a mandatory life sentence in cases where the court has had some degree of sympathy for the defendant. A reformulation of provocation, without regard to the surrounding law of murder, would not be satisfactory in the long term and would leave the law still subject to the same pressures as have led to the past expansion of provocation.

    The major problems with provocation
    3.20    
    There was widespread dissatisfaction among consultees both with the theoretical underpinning of the defence of provocation and with its various component parts. It is not underpinned by any clear rationale. There is widespread agreement that the concept of provocation has become far too loose, so that a judge may be obliged to leave the issue to the jury when the conduct and/or the words in question are trivial. The concept of loss of self-control has proved to be very troublesome. The supposed requirement of a sudden and temporary loss of self-control has given rise to serious problems, especially in the "slow burn" type of case. There is much controversy about the supposed objective test (that the provocation was enough to make a reasonable person do as the defendant did), which has been interpreted by the majority of the House of Lords in Smith (Morgan)[7] in a way that may enable a defendant to rely on personal idiosyncrasies which make him or her more short tempered than other people.

    Rationale of the defence
    3.21     The rationale underlying the defence of provocation is elusive. As we said in Consultation Paper No 173,[8] a study of the cases and textbooks (including particularly Provocation and Responsibility (1992) by Dr Jeremy Horder)[9] suggests that the doctrine has never been truly coherent, logical or consistent. At the time of the Homicide Act 1957 there was theoretically an excusatory rationale of sorts, namely that the defendant had suddenly and temporarily lost his or her self-control as a result of provocation which might have caused a reasonable person to do the same. However, this rationale did not bear too close scrutiny, particularly in relation to the requirement of loss of self-control (to which we will return).

    3.22     In Consultation Paper No 173 we invited consultees to consider the moral basis of a defence of provocation.[10] We discussed possible justificatory and excusatory bases[11] and we asked the question:

    Do consultees consider that, morally speaking:
    (1) a killing with the intent required for murder should be classified as murder notwithstanding any amount of provocation or loss of self-control; or
    (2) there ought to be a partial defence, leading to a conviction for manslaughter, based:
    (a) on the narrower (justificatory) ground; or
    (b) on the broader (excusatory) ground? [12]
    3.23     The views of those consultees who responded to this question were fairly evenly split between those who considered that provocation should in no circumstances amount to a partial defence to murder, those who considered that there should be a partial defence on a narrower (justificatory) ground and those who considered that there should be a partial defence on a broader (excusatory) ground. However, that statement needs to be amplified and qualified for two reasons. First, a good number of respondents were unhappy with our use of the labels justificatory and excusatory. Secondly, the overwhelming majority of those who considered that no amount of provocation should ever provide a partial defence to murder also considered that the abolition of the defence should be conditional on the abolition of the mandatory sentence.

    3.24    
    In seeking the views of consultees about what should be the underlying basis of a defence of provocation, if any, we are concerned that we may not have helped by the way we formulated our reference to excusatory and justificatory bases. But, however imperfectly the question may have been phrased, it served to produce helpful arguments. We will return to this subject when we discuss whether there should be a partial defence of provocation and on what rationale.

    The provoking conduct
    3.25    
    In Consultation Paper No 173 we explained that because the word "provoked" in section 3 of the 1957 Act has come to be interpreted as meaning no more than "caused", conduct can qualify as provocation although it is of a minor character or even entirely lawful.[13] We did not put a specific question to consultees about this, but one highly experienced judge expressed the views held by many respondents when she wrote:

    The scope of provocation has been so enlarged that a judge is obliged to leave it when … the conduct and/the words in question are trivial. The issue should only arise where circumstances are sufficiently grave to justify it. Such tightening up would not remove the last straw in the slow burn of domestic violence, although provocation must always be distinguished from revenge.
    Sudden and temporary loss of self-control
    3.26     In Consultation Paper No 173 we discussed this ingredient of the defence and the problems to which it has given rise.[14] There is no satisfactory definition of loss of self-control. In Oneby[15] it was said that to reduce a crime from murder to manslaughter the provocation had to arouse in the defendant "such a passion as for the time deprives him of his reasoning faculties". Similarly in Duffy[16] Devlin J spoke of provocation "rendering the accused so subject to passion as to make him or her for the moment not master of his mind".[17]

    3.27     However, the equation of loss of self-control with deprivation of reasoning faculties has been only partial. The courts have rejected the argument that a reasonable person who has lost self-control cannot be fully responsible for their conduct. In Phillips v The Queen[18] Lord Diplock said:

    Before their Lordships, counsel for the appellant contended, not as a matter of construction but as one of logic, that once a reasonable man had lost his self-control his actions ceased to be those of a reasonable man and that accordingly he was no longer fully responsible in law for them whatever he did. This argument is based on the premise that loss of self-control is not a matter of degree but is absolute; there is no intermediate stage between icy detachment and going berserk. This premise, unless the argument is purely semantic, must be based upon human experience and is, in their Lordships' view, false. The average man reacts to provocation according to its degree with angry words, with a blow of the hand, possibly if the provocation is gross and there is a dangerous weapon to hand, with that weapon.[19]
    3.28     The term loss of self-control is itself ambiguous because it could denote either a failure to exercise self-control or an inability to exercise self-control. To ask whether a person could have exercised self-control is to pose an impossible moral question. It is not a question which a psychiatrist could address as a matter of medical science, although a noteworthy issue which emerged from our discussions with psychiatrists was that those who give vent to anger by "losing self-control" to the point of killing another person generally do so in circumstances in which they can afford to do so. An angry strong man can afford to lose his self-control with someone who provokes him, if that person is physically smaller and weaker. An angry person is much less likely to "lose self-control" and attack another person in circumstances in which he or she is likely to come off worse by doing so. For this reason many successful attacks by an abused woman on a physically stronger abuser take place at a moment when that person is off-guard.

    3.29    
    The courts have responded to the criticism that the law of provocation treats an angry strong person more favourably than a frightened weak person by extending the concept of loss of self-control to include "slow-burn" cases but in so doing they have made the concept of loss of self-control still more unclear.

    3.30    
    In summary, the requirement of loss of self-control was a judicially invented concept, lacking sharpness or a clear foundation in psychology. It was a valiant but flawed attempt to encapsulate a key limitation to the defence - that it should not be available to those who kill in considered revenge.

    The reasonable person test
    3.31    
    In Consultation Paper No 173 we discussed at length the reasonable person test[20] and the way in which the law has developed in a series of cases up to Smith (Morgan).[21] We asked whether consultees favoured the approach of (a) the majority in Smith (Morgan),[22] (b) the New South Wales Law Reform Commission[23] or (c) of the minority in Smith (Morgan),[24] or whether they had any alternative suggestion. Views were divided between the three options with few respondents offering any alternative formulation.

    Abolition or reformulation of provocation?
    3.32     As we have said, there was overwhelming agreement among consultees that the law of provocation is unsatisfactory and that its defects are beyond cure by judicial development of the law. We asked consultees whether they favoured:

    (1) abolition of the defence of provocation, whether or not the mandatory sentence is abolished;
    (2) abolition of the defence of provocation, conditional upon abolition of the mandatory sentence, or
    (3) retention of the defence of provocation, whether or not the mandatory sentence is abolished.[25]
    3.33     Consultees were nearly unanimous that the defence should not be abolished while the mandatory sentence remained. This view accords with the opinion which we expressed in Consultation Paper No 173.[26] We know of no common law system where provocation has been abolished as a defence to murder but a mandatory sentence of life imprisonment retained, nor of any law reform body which has made such a recommendation.

    3.34     The major differences were between those who favoured abolition of the defence (conditional on abolition of the mandatory sentence) and those who favoured its retention in a reformed version. The arguments involved questions of principle about whether provocation should be capable of providing a partial defence to murder and questions as to whether the defence is capable of being reformed satisfactorily. Among those who favour retention of some form of provocation defence, there were diverse views about the directions which reform should take.

    3.35    
    As long as it is the Government's position that it does not intend to consider abolition of the mandatory sentence, it could be argued that it is pointless for us to consider the arguments for and against abolition of the defence if we were no longer to have the present mandatory sentence. However, this is an issue on which consultees have devoted a good deal of time and effort in setting out their views. It is right that we should summarise them and state our conclusion. Moreover, the debate regarding the best model of reform within the constraints of the mandatory sentence has been enhanced by the depth of principled discussion in relation to the position of provocation if the mandatory sentence were to be abolished.

    3.36    
    Powerful arguments can be advanced for and against the abolition of provocation as a defence. Abolitionists argue that a person who is sane and who kills another person unlawfully, with the intent required for murder, ought to be guilty of murder however great the provocation may have been. Provocation may be a mitigating circumstance which should be taken into account in passing sentence, but not in defining the offence. Assessing sentence requires a balanced appraisal of all the circumstances of the case (aggravating as well as mitigating), and this is a judicial rather than a jury function. Not only is it inappropriate that provocation should be singled out among other possible mitigating circumstances as providing a special partial defence, but there are great difficulties in trying to define what may amount to provocation and how serious it has to be in order to amount to a partial defence.

    3.37    
    Those who argue for the retention of some form of provocation defence, whether or not the mandatory sentence is retained, say that there are moral and practical reasons for doing so. Where the defendant's conduct was precipitated by really serious provocation, it is morally right that this should be reflected in the way that society labels and sentences the defendant; and it is desirable that the factual and evaluative question whether the defendant was provoked in that sense should be taken by the jury. A short sentence (or even in some circumstances a non-custodial sentence) for a provoked killing will be more understandable by, and acceptable to, the public if it results from a conviction by a jury of an offence not carrying the title of murder, than a decision by a judge after a conviction for murder. The existence of such a partial defence is justifiable in the law of murder, although there is no similar partial defence to non-fatal offences of violence, not only because the sentence for murder is fixed by law but also because of the unique gravity and stigma attached to murder. The real problem with provocation is not the underlying concept, but the way it has developed. It needs to be reshaped.

    3.38    
    The debate has generated interesting discussion about the moral qualities of the emotions of anger and fear. One school of thought holds that anger cannot ethically afford any ground for mitigating the gravity of deliberately violent action, or at any rate violent action which threatens life. The counter argument is that anger can be an ethically appropriate emotion and that in some circumstances it may be a sign of moral weakness or human coldness not to feel strong anger. That does not legitimise a violent response; one of the functions of the legal system is to channel legitimate anger at wrongdoing in ways that are considered just and proportionate. Nevertheless, a killing in anger produced by serious wrongdoing is ethically less wicked, and therefore deserving of a lesser punishment, than, say, a killing out of greed, lust, jealousy or for political reasons.

    3.39     The paradigm case of provocation involves inter-action between two people, the provoker and the provoked. The emotion aroused in the provoked contains a cognitive component, viz. the belief that the provoked has been wronged by the provoker. If that belief is justified, it does not justify the provoked person in giving vent to his or her emotions by resorting to unlawful violence, however great the provocation. Two wrongs do not make a right. However, the argument states that there is a distinction in moral blameworthiness between over reaction to grave provocation and unprovoked use of violence.

    3.40    
    A just system of law should reflect this. If it be right that a killing under grave provocation is made less wicked by that factor, the question whether a distinction should be drawn in the classification of the offence or at the sentencing stage is not a matter of simple ethics.

    3.41    
    Some have said that it is unfortunate that one is forced to refer to provocation as a 'defence', when the critical question is whether a killing under provocation should be categorised as a less grave offence than murder. There is general agreement that provocation (subject to what is meant by that word) should be capable of making a significant difference in the sentence passed on the defendant. That result could be achieved by a variety of routes: by labelling a killing under provocation as a separate offence; by labelling it as murder, but with different statutory sentencing provisions for different categories of murder; or by labelling it as murder with the same sentencing provisions for all cases of murder and leaving it to the judge to set the appropriate sentence having regard to the provocation.

    3.42    
    As some consultees have pointed out, the problems about what should or should not be regarded as provocation would not disappear by abolishing the defence of provocation, but would be faced at a separate stage of the proceedings. Some, including Victim Support, consider that there would be real advantages in terms of justice and transparency if issues of provocation were considered as part of the sentencing process, if necessary through a Newton hearing. The family would have an opportunity to counter allegations raised by the defendant against the deceased and to set the matter in context in a way which is sometimes not presently possible. The court would then pass sentence based on full information about any relevant background circumstances, and this process would be more understandable than the present. We recognise fully the importance that the public generally and all affected by the criminal justice process should be able to understand how it works.

    3.43    
    We are also not surprised, as was made clear to us by representatives of both Victim Support and Support After Murder and Manslaughter (SAMM), that victims' families are often confused about how it may come about that a person charged with murder may be convicted of manslaughter, either on a plea accepted by the prosecution or by a verdict of the jury. The law of murder and manslaughter is complex because of the way in which it has developed. That is part of the reason why we believe that it is high time for a wider review of the law. We are also highly sympathetic to the argument that if an issue of provocation is raised by a defendant, the rules of evidence and procedure should enable the prosecution to put before the court any additional relevant material. The problem is not so much one of admissibility but of ensuring, through rules of procedure and pre-trial case management, that the nature of the defence is sufficiently disclosed in advance.[27]

    3.44     We have considerable respect for the arguments of those who advocate the abolition of the defence of provocation and the mandatory sentence, although (as we have noted) that would not remove all problems relating to provocation but would move them to a different stage of the proceedings in which the jury would not be involved. Abolition of the defence was recommended by the Model Criminal Code Officers Committee of the Standing Committee of Attorneys General of Australia in 1998[28] and by the New Zealand Law Commission in 2001.[29] However, through the course of the debates which we have had both internally and with consultees, we have come to favour the reform rather than the abolition of the defence. Most civilised systems of law have gradations of homicide which allow for the existence of extenuating circumstances and, although the wider structure of the law of homicide is outside our terms of reference, we see a case for retaining grave provocation within it as a form of extenuating circumstances.

    3.45     Some whose first option would be to abolish the defence (conditional on abolition of the mandatory sentence) would, as their second choice, prefer to leave the defence as it is than to attempt statutory reform of it. The main argument for this approach is that, although at a conceptual level the problems associated with the defence are very difficult, at a practical level it works satisfactorily in the majority of cases. We should wait to see what difference Smith (Morgan)[30] makes in practice before deciding whether and, if so, how the law in this area should be reformed. We are not persuaded by the "leave it alone" arguments. The great majority of consultees agreed with the views expressed in Consultation Paper No 173 that the law of provocation is unsatisfactory and beyond cure by judicial reform. We believe that the defence is capable of significant improvement, but we add an important caution. Bluntly, we think that provocation has got out of hand because of pressures on the defence which result from the present scope of the offence of murder and the mandatory sentence. We have a real and serious concern that reforming the law of provocation without a wider review of the law of murder may in the long run fail to achieve its objective, because the same pressures are liable to lead in practice to a stretching either of the reformed provocation defence or possibly of diminished responsibility in cases where the judge and jury have a degree of sympathy for the defendant.

    3.46     Before we set out our approach, we consider the possibility of reforming the law on the lines of the EMED provisions of the American Law Institute Model Penal Code, 1985.[31]

    Extreme mental or emotional disturbance (EMED)[32]
    3.47     Clause 210.3(1)(b) of the MPC provides:

    [A] homicide which would otherwise be murder [is manslaughter when it] is committed under the influence of extreme mental or emotional disturbance for which there is reasonable explanation or excuse. The reasonableness of such explanation or excuse shall be determined from the viewpoint of a person in the actor's situation under the circumstances as he believes them to be.
    3.48    
    In Consultation Paper No 173[33] we asked:

    Should the concept of "loss of self-control" be retained or should it be replaced by a test of acting "under extreme emotional disturbance" or some similar phrase?[34]
    3.49     A majority of judges and academics were opposed to a test of extreme emotional disturbance, principally on the ground that it was too vague. However, a significant number of respondents, including particularly representative bodies of the legal profession, women's groups and JUSTICE, thought that a test involving extreme emotional disturbance would be preferable to a test based on loss of self-control.

    3.50    
    We have looked at the experience of EMED in the USA, and we have been greatly helped in this regard by a paper written for us by Professor Sanford Kadish (Alexander F. and May T. Morrison Professor of Law Emeritus, University of California).[35] The EMED defence has been a source of controversy in those states (known as "reform states") which have adopted the MPC or a version of it.[36] In particular, controversy has arisen over fundamental aspects of the defence. One issue concerns the focusing of the requirement for a "reasonable explanation or excuse" on the extreme emotional disturbance but not the defendant's conduct. It is not difficult to imagine circumstances in which there may be a reasonable explanation for a person being in a very disturbed emotional state, but no reasonable explanation or excuse for killing another person.

    3.51     The defence is rooted in the excuse based category of defences, founded on the defendant's state of mind (whatever may have caused it), although some states have adopted the MPC in an amended form narrowing the causes of the defendant's disturbance. (New Hampshire requires that the extreme mental or emotional disturbance be "caused by extreme provocation".)

    3.52    
    Victoria Nourse,[37] addressed this controversy in a study of jurisdictions in the USA, some of which have kept a 'traditional' law of provocation while others have moved to a more 'liberal' scheme based on the MPC. Professor Nourse presents a strongly reasoned case why wrongdoing by the victim is an important part of the proper moral foundation of the defence of provocation, providing a justifiable reason for the defendant's sense of outrage, and therefore a partial excuse (but not a justification) for the defendant's over-reactive response. She has written:

    The most persuasive scholarly defences of provocation have all invoked examples … in which the defendant's emotion reflects the outrage of one responding to a grave wrong that the law otherwise punishes. Commentators frequently use examples of men killing their wives' rapists or children who kill abusive parents as clear cases of provoked murder. When, for example, the MPC drafters sought to justify their expansion of the defence, they relied on a case involving forcible sodomy.
    The problem comes when we focus on cases in which the emotion is based on less compelling 'reasons' – when women kill their departing husbands or men kill their complaining wives. Under conventional liberal theory, if extreme emotion is shown, these cases should be handled no differently from cases where victims kill their rapists and stalkers and batterers. The quantity or intensity of the emotion provides the excuse, not the reasons for the emotion. This focus on emotion, to the exclusion of reason, reflects a very important assumption made by liberal theories of the defence, that emotion obscures reason. When we distinguish the rapist killer from the departing wife killer, we acknowledge a very different view of emotion, one in which emotion is imbued with meaning. Both the departing wife killer and the rapist killer may be upset, but the meanings embodied in their claims for emotional understanding are quite different. In distinguishing these cases based on the reasons for the claimed emotion, we acknowledge a view of emotion in which emotion is not the enemy of reason but, instead, its embodiment… .
    In the past two decades, the idea of emotion as the natural enemy of reason has been seriously questioned by brain scientists and psychologists, by rhetoricians and philosophers, by classicists and even by legal scholars. That both brain scientists and philosophers may now agree that emotion reflects or assists our reasoning processes tells us something that law, and life, already reflect. When we see that someone is angry we do not call … [a] psychiatric expert for a diagnosis, we simply ask "why?" We expect reasons, and they are typically attributions of wrongdoing and blame.[38]
    3.53     She goes on to argue:

    Conventional understandings of criminal law place defences in two mutually exclusive categories: as excuse or justification. In the excuse category are defences, such as insanity, that focus on state of mind; these defences do not embody judgements that what the defendant did was 'right' or 'justified', but that the defendant was less blameworthy. In the 'justification' category are defences, such as self-defence or necessity, which assume that what the defendant has done, overall, was 'right' or 'warranted'. Traditionally, 'excuse' and 'justification' have been viewed as mutually exclusive categories: a defendant cannot both be excused and justified because an excused action presupposes that the action was wrong and therefore unjustified. This assumes, however, a crucial feature of the inquiry – that we are evaluating acts and acts alone. To say that an act cannot be both justified and excused is to say something about acts, not emotions. It is perfectly consistent to say that one's emotions are justified or warranted even when one's acts are not. Indeed, as I have noted above, we may easily say that passionate killings are not justified even if we believe that the emotions causing some killings are, in some sense, the 'right' emotion.[39]
    3.54     Provocation, she argues, is on the cusp because it applies (or should apply) in a case where the defendant's sense of outrage is warranted, but not the manner or scale of reaction. She terms this a 'warranted excuse'.

    3.55    
    Moving on to the practical application of the EMED test, Victoria Nourse observes that:

    Jurors are told to put themselves in the defendant's position, to adopt his or her perspective and, yet, at the same time, to be 'reasonable'. They are asked to exercise independent 'moral judgement', and, at the same time, adopt the defendant's vantage point. In practice, this has done little to resolve the problem and much to confound judges and jurors. After days of deliberation in a case in which a defendant killed a man who had parked in his parking place, one jury summed up its conclusion about the EED defence by sending a note to the judge, asking, Whose norms apply, his or ours?[40]
    3.56     Victoria Nourse also cites many other cases of EMED defences being left to the jury where there was no reasonable ground to regard the defendant as having been seriously wronged by the victim. She says:

    Reform has permitted juries to return a manslaughter verdict in cases where the defendant claims passion because the victim left, moved the furniture out, planned a divorce, or sought a protective order. Even infidelity has been transformed under reform's gaze into something quite different from the sexual betrayal we might expect – it is the infidelity of a financee who danced with another, of a girlfriend who decided to date someone else, and of the divorcee found pursuing a new relationship months after the final decree. In the end, reform has transformed passion from the classical adultery to the modern dating and moving and leaving. And because of that transformation, these killings, at least in reform states, may no longer carry the law's name of murder.[41]
    3.57     Professor Kadish has pointed out that Victoria Nourse's study was of cases in which manslaughter was left to the jury as a possibility, and that in the cases instanced by her the defence failed. However, we would not favour developing a test which opened the possibility of a defence in such a broad range of circumstances.

    3.58    
    Professor Kadish has also pointed out that many "reform states" have not adopted the second part of the EMED provision in the MPC ("the reasonableness of such explanation or excuse shall be determined from the viewpoint of a person in the actor's situation under the circumstances as he believes them to be"), but the mental and emotional character of the defendant remains central to the question whether he or she was acting under the influence of extreme mental or emotional disturbance for which there was reasonable explanation or excuse.

    3.59    
    In the USA there is no general equivalent of the defence of diminished responsibility. In that context the development of the EMED defence with its concentration on psychiatric evidence is supported by some scholars (including Professor Kadish). We would not recommend importing a defence based on EMED. We think that it is too vague and indiscriminate. We favour as the moral basis for retaining a defence of provocation that the defendant had legitimate ground to feel seriously wronged by the person at whom his or her conduct was aimed, and that this lessened the moral culpability of the defendant reacting to that outrage in the way that he or she did. It is the justification of the sense of outrage which provides a partial excuse for their responsive conduct.

    Our approach to reform of provocation
    3.60    
    A number of approaches could be taken in seeking to define what may found a defence of provocation. It could be limited to, or exclude, precisely defined categories of conduct (providing the maximum certainty of outcome and minimum evaluative role for the judge and jury) or broader principles could be adopted.

    3.61    
    In our provisional conclusions we set out[42] principles on which we thought a reformed defence of provocation should be based. The comments which we have received have been mainly supportive, with certain qualifications, but some consultees have been critical of the principles. Having considered these responses, for which we are grateful, we have not changed our conclusions in substance. From the comments which we have received, we recognise that there are matters which require further explanation. There have also been helpful suggestions how some of our proposed principles might be clarified or improved. We also identify and address the principal criticisms made of the approach put forward in our provisional conclusions.

    3.62     A major criticism by some was that our provisional conclusions failed adequately to explain the rationale underlying our proposed approach. We will address that issue first and then explain our proposed principles. We emphasise, as we did in our provisional conclusions, that what we are putting forward are principles, rather than a statutory formula. If our approach is accepted, the drafting of legislation will be a matter for Parliamentary Counsel.

    Rationale of provocation
    3.63    
    Putting it in broad and simple terms, we think that the moral blameworthiness of homicide may be significantly lessened where the defendant acts in response to gross provocation in the sense of words or conduct (or a combination) giving the defendant a justified sense of being severely wronged. We do not think that the same moral extenuation exists if the defendant's response was considered, unless it was brought about by a continuing state of fear. (There are also strong policy reasons for the law not to treat vendettas as partial excuses.) We do not suggest that these are the only circumstances which could significantly extenuate moral responsibility for homicide, but we do think that they fall into a distinct category. Another distinct category is where the defendant suffers from some mental condition by reason of which he or she ought not to be regarded as fully responsible for his or her actions. Because we see these as essentially different, we do not favour amalgamating them in a single defence, but we will refer to this in further detail later. There are also other circumstances which may significantly extenuate moral responsibility for homicide. An example is the genuine case of mercy killing, but that falls outside the terms of our present review. This project is confined by our terms of reference to the present partial defences and the possibility of a partial defence of excessive force in self-defence.

    3.64    
    We would not favour extending provocation to cover cases where there was no gross provocation in the sense suggested above, but where the defendant acted under extreme or extraordinary stress.[43] We recognise that, tragically, there are cases in which a person who is not ordinarily violent may do a sudden act of serious violence under stress, not intending to kill but in fact resulting in death. Such a person will be guilty of murder if there was an intention to cause serious harm; but that is a consequence of the broad definition of murder in England and Wales, which, as we have already said, we consider to be in need of review.

    Provocation: how to reshape it
    3.65     We think that the defence as it presently operates is in some respects too broad and in other respects too narrow. We think that it is too broad in that it can apply to conduct by the victim which is blameless or trivial. It is too narrow in that it provides no defence to a person who is subjected to serious actual or threatened violence, who acts in genuine fear for his or her safety (but not under sudden and immediate loss of self-control) and who is not entitled to the full defence of self-defence (either because the danger is insufficiently imminent or their response is judged to have been excessive). We are satisfied from consultees' responses that this is a real and not merely an academic problem, particularly in cases of defendants who have been victims of long-term abuse.

    3.66    
    In principle, we consider that the first pre-requisite of a defence of provocation should be that the defendant acted in response to (1) gross provocation or (2) fear of serious violence towards himself or herself or another, or (3) a combination of (1) and (2) (the trigger).

    3.67    
    The second should be that a person of ordinary temperament, i.e. ordinary tolerance and self-restraint, in the circumstances of the defendant might have reacted in the same or a similar way (the objective test).

    The trigger: gross provocation
    3.68    
    We consider that the essence of gross provocation is that it is words and/or conduct which caused the defendant to have a justifiable sense of being seriously wronged. The preferred moral basis for recognising a partial defence of provocation is that the defendant had legitimate ground to feel strongly aggrieved at the conduct of the person at whom his/her response was aimed, to the extent that it would be harsh to regard their moral culpability for reacting as they did in the same way as if it had been an unprovoked killing. It is important to note that there are two aspects to this.

    "Caused" the defendant to have a … sense of being seriously wronged
    3.69    
    The first is that the words or conduct should have caused the defendant to have a sense of being seriously wronged and therefore to react as he or she did. (This follows from the requirements that the defendant acted in response to gross provocation, and that to be gross provocation there must have been words or conduct, or both, which caused the defendant to have a sense of being seriously wronged.)

    "Justifiable" sense of being seriously wronged
    3.70    
    The second aspect is that the defendant's sense of being seriously wronged should have been justified. In deciding whether there was gross provocation in the sense of words and/or conduct which caused the defendant to have a justifiable sense of being seriously wronged, we do not intend the test to be purely subjective, i.e. what the defendant thought. It is for the jury to decide whether there was gross provocation in the relevant sense. In making that judgement the jury must of course consider the situation in which the defendant found him or herself and take into account all the characteristics of the defendant which they consider to be relevant. Taking into account the circumstances and characteristics of the defendant does not mean that if the defendant considered it to be gross provocation, the jury must therefore accept that it was gross provocation. The jury may conclude that the defendant had no sufficient reason to regard it as gross provocation, or indeed that the defendant's attitude in regarding the conduct as provocation demonstrated an outlook (e.g. religious or racial bigotry) offensive to the standards of a civilised society. Dr Horder gives an example:

    Reconsider the imaginary case of Terreblanche, whose deep-rooted beliefs include the belief that it is the gravest of insults for a coloured person to speak to a white man unless spoken to first. If he became enraged and killed a coloured person for speaking to him in this way, is the jury to be directed to take his beliefs into account – qua characteristics – in judging the gravity of the provocation (assuming there is held to be a 'real connection' between the provocation and those beliefs, or that the victim deliberately spoke to him as a challenge to those beliefs)? Would not such a direction be an outrageous compromise of society's commitment to racial tolerance?[44]
    3.71     Our answer to the question posed in the last sentence is yes. No fair-minded jury, properly directed, could conclude that it was gross provocation for a person of one colour to speak to a person of a different colour. In such a case the proper course would therefore be for the judge to withdraw provocation from the jury. (We discuss the role of judge and jury below.) [45]

    Does the provocation need to emanate from the deceased?
    3.72     We have not stipulated that the provocation should come from the deceased. Our only reason for not doing so is that there may be cases of accident or mistake where we consider that the defence should be capable of applying. We discuss this topic below.[46] Except in such cases, we are satisfied that where D kills V in response to provocation by a third party the objective test would preclude this defence. No person of ordinary tolerance and self-restraint would deliberately respond to provocation from one person by using violence to another.

    Does the defendant need to be the sole or immediate sufferer from the provocation?
    3.73     We do not intend that the defence of provocation should be restricted to cases where the defendant is the sole or most immediate sufferer from the provocation. It would be an absurdly narrow approach to suggest, for example, that it could not apply to a parent whose child was raped but only to the child. The Western Australian Criminal Code[47] limits provocation to a wrongful act or insult done to the defendant or "to another person who is under his immediate care, or to whom he stands in a conjugal, parental, filial or fraternal relation, or in the relation of master and servant." We see problems in trying to define a precise list, and most jurisdictions have not done so. We think that a jury would be able to recognise that where there is a close personal connection between the defendant and the person directly wronged, the defendant may well have a feeling of suffering jointly from the wrongdoing so as to fall within the principle of the formula which we have proposed.

    3.74     We have considered whether provocation should be confined to cases of serious and unlawful actual or threatened violence to the person; or criminal conduct; or provocation by conduct only, as distinct from words. In common with many respondents, we see real difficulties with inflexible provisions of that kind for reasons which we develop in the following paragraphs. We begin with the general observation that it is very difficult to foresee and catalogue in specific terms every form of conduct which might deservedly be thought to qualify for the defence, but that the governing principles ought to be capable of articulation.

    The impact on domestic violence
    3.75    
    Our terms of reference ask us to have particular regard to the impact of the partial defences in the context of domestic violence, and we have done so. The responses which we have received for the most part have not been polarised on male – female lines. There is not, for example, a single 'feminist' position. We have had very helpful responses from a number of women's groups, and these expressed a range of views. One theme which emerged strongly is that a person may feel imprisoned in an abusive and humiliating relationship without being the victim necessarily of serious physical violence. One group proposed abolishing the defence of provocation and replacing it with a partial defence of self-preservation on the grounds of domestic violence, but they recommended the definition of domestic violence used in the New Zealand Domestic Violence Act 1995, which defines it as physical, sexual or psychological abuse, including but not limited to intimidation and harassment. This group was not alone in recommending that any reforms should include a comprehensive legal definition of domestic violence based on the definition in the New Zealand Domestic Violence Act 1995.

    3.76    
    Another group argued for an approach to domestic violence which may include 'physical, sexual, emotional or financial abuse'. They stressed that concentration on the purely physical can lead to a failure to understand the position in which vulnerable people may find themselves.

    3.77    
    In considering the impact of the law of provocation in the context of domestic violence, our approach remains as stated in Consultation Paper No 173:[48]

    Domestic violence is an extremely worrying problem. The law must deal with it in a way which is fair and shows proper respect for human life. At the same time it would be wrong to introduce special rules relating to domestic killings unless there is medical or other evidence which demonstrates a need and a proper basis on which to do so.
    3.78     As a matter of principle, the criminal law should be gender neutral unless it is absolutely necessary to depart from that principle. Our proposals do not depart from that principle. Our provisional conclusions have received a considerable degree of support from those women's groups who have commented on them.[49]

    "Gross Provocation" need not involve a risk of physical violence
    3.79     Turning to cases other than domestic violence, we would not want to narrow the law of provocation as it currently affects householders. There is undoubtedly a very strongly held view among many members of the public that the law is wrongly balanced as between householders and burglars.[50] We think that much of the public anxiety is possibly based on a misunderstanding of the present law and of the highly unusual facts of the Tony Martin case. However, we do accept that many members of the public are genuinely worried about what may happen to them if they use force against intruders and are subsequently judged to have gone too far. Sometimes burglaries involve the most vile acts of desecration of a person's home and of belongings which may be cherished for highly personal reasons. If, for example, a householder confronted a burglar responsible for such behaviour and immediately attacked him, causing fatal injury, we think that it would be a harsh law which precluded the jury from considering provocation and we doubt whether such a law would command public support or respect.

    3.80     In our Consultation Paper[51] we gave other examples of cases in which the defendant would not have been at risk of serious physical violence, but where the deceased's conduct was nevertheless such that we can see a moral case for allowing a partial defence of provocation. Some respondents suggested other examples. One consultee said:

    We do not agree that words alone should never constitute provocation since there may be some circumstances where a person is subjected to a campaign of abuse which when combined with a complete lack of options of escape can amount to provocation. A good example of this is someone subject to racism in the workplace … The racist conduct may fall short of violent conduct, threats of violence or threats to kill, but may nevertheless count as seriously inhumane and degrading treatment and amount to provocation.
    3.81     Some consultees expressed concern that "gross provocation" and "words and/or conduct which caused the defendant to have a justifiable sense of being seriously wronged" were unacceptably vague.[52] Others supported the proposals as well balanced.

    3.82     Professor Andrew Ashworth wrote:

    I think the proposals represent a fine attempt to tread the difficult lines, and enable some flexibility while maintaining a much more structured approach than would have resulted from combining provocation with DR or introducing the EMED doctrine.
    3.83    
    Lord Justice Kennedy wrote:

    I agree with paragraph 18 of the Provisional Conclusions[53] as to the "trigger". The defendant must have legitimate ground to feel strongly aggrieved at what has been said or done to him, and I would not like to see what is capable of amounting to provocation being at all closely confined. Juries can deal with that.
    3.84     We will return to this issue after setting out our proposal about the roles of the judge and jury.[54]

    The trigger: response to fear
    3.85     In paragraph 3.66 above, we referred to a defendant who acts in response to fear of serious violence to him or herself or another. Lord Hoffmann in Smith (Morgan)[55] suggested that provocation under the present law is not confined to anger, but may include fear; and in many cases there will be an overlap between conduct which is gross provocation and conduct which causes the defendant to fear serious violence. However, the proposed partial defence would go beyond the traditional limits of provocation.

    3.86     It is important to emphasise that before the defence arises the jury will, where it has been raised, first have had to exclude self-defence, which operates as a complete defence.

    3.87    
    We recognise the need for the scheme of defences and partial defences to murder to be coherent and also to do justice.

    3.88    
    Consider the case of D, who has suffered regular violent assaults by V. D eventually responds to an assault by fatally stabbing V. D has a complete defence if D's conduct was proportionate to the immediate risk. If it was disproportionate, D has a partial defence only if D acted in sudden and immediate loss of self-control. Consultees with experience of such cases have told us of the "catch 22" dilemma which faces D (and D's lawyers) in such a case. If the defence want to obtain an acquittal, it is unhelpful to present D as somebody who lost self-control. Running a defence of provocation based on loss of self-control resulting from years of abuse makes it correspondingly difficult to present D's acts as proportionate to the immediate risk. We are told that sometimes in such cases a defendant may plead guilty to manslaughter, although arguably D's conduct may have been in lawful self-defence, for fear of the risk of conviction of murder with the mandatory sentence.

    3.89    
    The dilemma facing D in such a case is not confined to those of domestic violence. One consultee[56] drew our attention to a recent newspaper report[57] about the case of Osborn, who stabbed an intruder to his friend's home late one bank holiday evening. The defendant was watching television with four friends. The intruder was under drug induced illusions and was covered in blood as a result of smashing windows in the street with his head and fists. Hearing the commotion, one of Osborn's female companions had gone outside and was assaulted by the intruder, who pursued her into the house. He was ejected once but forced his way back inside. Osborn stabbed him five times with a knife. One of the stab wounds punctured his lung with fatal results. Osborn was charged with murder, but the prosecution offered to accept a plea of guilty to manslaughter on the ground of provocation. He took that option and was sentenced to five years imprisonment.[58] According to the newspaper article, Osborn was advised by his counsel that it was very unlikely that a jury would reject his plea of self-defence, but they could not exclude a small chance that the jury would decide that his use of the knife was disproportionate, in which case he would be convicted of murder and sentenced to life imprisonment. Osborn decided that he could not face the risk and therefore pleaded guilty to manslaughter.

    3.90     Under our proposal, the jury would first consider whether D acted in lawful self-defence. If the jury is satisfied that the killing was unlawful, they would then consider whether D was entitled to a partial defence of provocation under either limb. We believe that this represents a coherent and just approach. D would not be in the dilemma identified in the previous paragraphs and therefore under pressure to plead guilty to manslaughter in circumstances where the jury might well have concluded that D's conduct was lawful.

    3.91    
    Consider also the case where D is the victim of long-term abuse by V. After a grave attack or threat of attack, D waits until V is asleep and then kills V. D acts in fear and despair, thinking that this is the only way of being able to live a life free from a risk of grave violence from V. Under present law D is guilty of murder (unless entitled to a partial defence of diminished responsibility) because D's acts were not done at a time of imminent danger.

    3.92    
    Some take the view that D ought properly to be regarded as guilty of murder and are worried about the "floodgates" risk if a partial defence were available to D in such circumstances. We think that such a result would be overly harsh and we believe that the floodgates risk can be satisfactorily met by the objective test which we discuss below.

    3.93    
    In informal discussions our proposal has been subject to criticism from two sources: academics and the senior judiciary. At the heart of each there is a common concern: whether conduct in response to provocation and conduct in response to fear should be joined in a single defence, or whether there should be a separate partial defence for those who kill in fear, but who do not attract the full defence that they acted in lawful self-defence.

    3.94    
    A number of academics favour separate defences, because they say that the situations are morally different. Their concern is that we are inappropriately linking two conceptually different partial defences: provocation and excessive force in self-defence.

    3.95    
    The partial defence of provocation arises in the case of conduct by the defendant which, it is said, the actor acknowledges is unlawful but for which there is a sufficient excuse or justification from an external source so as to permit a different label to be attached to the defendant's conduct.

    3.96    
    By contrast, a partial defence of excessive force in self-defence would apply to a person who thinks that he or she is acting lawfully but who has miscalculated the extent of the action open to him or her and so has fallen into unlawfulness. This is sufficient to prevent the label of murder to be attached to the defendant's conduct.

    3.97    
    It is, therefore, said to be conceptually confused to join them together as we suggest.

    3.98    
    While we respect this view, we question how far it corresponds with the likely thought processes of a defendant. It will be seldom that a defendant does other than merely react to the external stimulus. We also think that there is medical, practical and moral justification for the proposed combination.

    3.99    
    First, the Royal College of Psychiatrists said in their response to Consultation Paper No 173:

    [W]e would point out that the approach adopted within the document to the relationship between provocation and self-defence, with the suggestion of a new partial defence of 'excessive self-defence', is based, at least partly, upon a legal misrepresentation of psychology and physiology. Hence, one way of reading the proposal to abolish the provocation defence 'in favour' of the new partial defence of self-defence is that it rests upon the assumption that 'anger' cannot be a justification for 'responsive violence', but 'fear' can be. However, this assumes that the two emotions of anger and fear are distinct. In medical reality they are not. Physiologically anger and fear are virtually identical, whilst many mental states that accompany killing also incorporate psychologically both anger and fear. Hence, the abused woman who kills in response even to an immediate severe threat will also be driven at least partly by anger at the years of abuse meted out to her, and perhaps her children. Again, the woman who waits until the man is 'helpless' to kill him, is likely not merely to be angry but also fearful that eventually he will kill her, and/or her children, and that there is no way of preventing it other than by the death of the man (partly because her cognitions have been so distorted by the years of abuse that she does not perceive the options for escape, for example legal options, at all in the same way as an ordinary person would do). Any legal solution to the current perceived problems with partial defences to murder which rested upon the assumption that fear and anger can (even usually) be reliably distinguished must, from a medical perspective, therefore fail.
    3.100    
    Secondly, from a practical perspective, it is desirable to try to keep jury directions as broad and simple as possible.

    3.101    
    Thirdly, from a moral viewpoint, there is a common element namely a response to unjust conduct (whether in anger, fear or a combination of the two).

    3.102    
    A different criticism made by one consultee is that the expression "serious violence" needs further clarification and may be too narrow. This consultee proposed that the wording should be "fear of serious violence or significant harm towards the defendant or another". Whatever form of words is used, there may be borderline cases but we think that the concept of serious violence is one which a jury should not find it difficult to grasp or to apply in a sensible way. To introduce the alternative "or significant harm" would in our view tend to blur the test. Conduct may be harmful in a broad sense without involving any physical threat and without being deliberate. We think that this would be too wide. Those criticisms could be met by adapting the phrase to "significant intentional physical harm" but that is not far removed from "serious violence". On consideration, we prefer a straightforward test of "serious violence". This would obviously include sexual as well as other physical violence, and what was judged to be serious would have to be considered in the context of the relationship between the defendant and the victim.

    3.103    
    The question has been raised whether any statement of a reformed provocation defence should make express reference to emotions other than fear. We recognise that provocation may give rise to a range of emotions, e.g. anger, fear, disgust and despair. It may be appropriate for a judge to refer to these matters in his summing up, when inviting the jury to think about how provocation might cause a person to respond with violence, but we do not think it necessary to incorporate a list of all such emotions in a definition of provocation.

    The trigger: combination of gross provocation and fear
    3.104    
    HHJ Stewart QC has brought to our attention the case of Annette and Charlene Maw[59] in which he appeared as junior counsel for the defendants. The case aroused considerable public interest at the time. Annette was 21 and Charlene was 18. They lived with their father, mother and younger brother in Bradford. The father was a drunkard and a bully and habitually treated their mother with violence. On the night of the offence he had a blood alcohol level equivalent to at least ten pints of cider. He was in an ugly mood and assaulted both defendants and their mother. There was a fight in a bedroom which ended only when the mother smashed a mirror over the father's head. The women came downstairs, leaving him in the bedroom. The girls then decided that they could stand no more of his treatment and would finish him off by stabbing him. Before they did anything, the father came downstairs, made a grab at Annette and hit her. He then hit the mother. Annette called for Charlene to fetch a knife. Charlene did so and handed it to Annette, who stabbed the father below the neck, severing the jugular vein. They called the police and initially told a false story that the father had been trying to knife Annette.

    3.105     The sisters were charged with murder but pleaded guilty to manslaughter. Their plea was accepted with the approval of the trial judge. The basis of the plea was not specified in open court. There was good reason for this. As a matter of law they were guilty of murder. They were not acting under a sudden loss of self-control and there were other ways in which they could have protected themselves against the immediate risk than by killing the father. As Lord Lane CJ said in the Court of Appeal "unhappily, the case was not merely a simple one of self-defence nor was it simply one of the agony of the moment."

    3.106    
    The trial judge sentenced both defendants to three years' imprisonment. There was a public outcry that the sentences were too severe. On appeal, Charlene's sentence was reduced to six months imprisonment and Annette's sentence was upheld.

    3.107    
    It is unsatisfactory that the case could only be treated as manslaughter by prosecutorial discretion which involved turning a blind eye to the law with the connivance of the judge. Under our proposals, the conviction for manslaughter would have been available on a principled and transparent basis.

    3.108    
    HHJ Stewart QC has commented on our provisional conclusions:

    I agree that the trigger should be gross provocation by words or conduct or fear of serious violence to self or another which caused the defendant to have a justifiable sense of being seriously wronged in the situation in which he found himself. This would cover the Maw situation to which I have referred in earlier correspondence.
    The objective test
    3.109    
    Even if there was gross provocation, it by no means follows that an ordinary person would have reacted in the way that the defendant did. Most people from time to time suffer gross provocation without resorting to lethal violence. The defence should only be available if a person of ordinary temperament, i.e. ordinary tolerance and self-restraint, in the circumstances of the defendant might have reacted in the same or a similar way.

    3.110    
    In deciding whether a person of ordinary temperament in the circumstances of the defendant might have reacted in the same or similar way, the court should take into account 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" (to adopt a succinct expression used by Professor Glanville Williams[60] in his analysis of the decision in Camplin).[61] The only qualification which we would make is that the court should have regard to the defendant's age, because capacity for self-control is an aspect of maturity, and it would be unjust to expect the same level of a twelve-year-old and an adult (as was recognised in Camplin). In other words, we prefer the minority position in Smith (Morgan),[62] which also accords broadly with the law in Australia, Canada and New Zealand, and with the recent provisional recommendations of the Irish Law Reform Commission.

    3.111     This is not to deny a defence to an abused person whose temperament may have been changed as a result of the provocation, for that would be a matter which bore not simply on the defendant's general temperament independent of the provocation but on the effect of the provocation itself. As Lord Millett said in Smith (Morgan)[63] about such a case:

    The question for the jury is whether a woman with normal powers of self-control, subjected to the treatment which the accused received, would or might finally react as she did… . It does not involve an inquiry whether the accused was capable of displaying the powers of self-control of an ordinary person, but whether a person with the power of self-control of an ordinary person would or might have reacted in the same way to the cumulative effect of the treatment which she endured.[64]
    3.112     We think that the objective test should apply in the case of a person responding to fear of serious violence as in the case of a person responding to provocation. It might be argued that self-restraint is a relevant factor when considering provocation but not when considering the position of a person acting in fear, but we would disagree. Ordinarily it would not be even partially excusable for a person in fear, but not in imminent danger, to take the law into his or her own hands. We would not, for example, want a partial defence to be available to criminal gangs who choose to deal with threats of violence from rival gangs by striking first. Our proposals regarding the role of the judge and jury[65] would properly preclude such a defence from being left to the jury in those circumstances (on the basis that no properly directed jury could reasonably conclude that a gangster who chose to act in such a way could satisfy the objective test).

    3.113     Some concerns have been expressed about our formulation of the objective test on the basis that persons of ordinary temperament do not kill in face of provocation. The "reasonable person" test in the law of provocation has always involved this problem. In Campbell[66] Lord Bingham CJ observed that "it is not altogether easy to imagine circumstances in which a reasonable man would strike a fatal blow with the necessary mental intention, whatever the provocation".[67] Nevertheless juries have recognised that there may be circumstances in which an ordinary person may be driven to use fatal violence in response to provocation.

    3.114     It seems to us that there must be some objective assessment of the response and its causes. As Lord Hoffman said in Smith (Morgan):[68]

    A person who flies into a murderous rage when he is crossed, thwarted or disappointed in the vicissitudes of life should not be able to rely upon his anti-social propensity as even a partial excuse for killing.[69]
    3.115     Moreover our proposals involve abandoning the loss of self-control test, which has proved very unsatisfactory, and this makes the need for an objective test still greater.

    3.116    
    It is clear from our consultation process that different judges and practitioners have had different experiences of how the Smith (Morgan) test has worked in practice. Some have said that juries in their experience appear to have no difficulty with the present test. Others have reported otherwise and cases have reached the Court of Appeal in which juries and the court do appear to have had difficulties.

    3.117    
    In Lowe[70] the judge, after consultation with counsel, directed the jury that the prosecution had to make them sure that the defendant "was not acting under provocation" and gave them an explanation of provocation in which she said:

    The law expects everyone to exercise control over their emotions and it is for you to decide whether the circumstances may have been such as to make [the defendant's] loss of control sufficiently excusable to reduce the gravity of the offence from murder to manslaughter. That is because you represent the community and decide what may count as a sufficient excuse. Apply what you consider to be appropriate standards of behaviour, on the one hand making allowance for human nature and the power of emotions but, on the other hand, not allowing someone to rely on his own particular inability to control himself.
    3.118     The Court of Appeal commented:

    It is apparent that the direction owes more than a little to the speech of Lord Hoffmann in R v Smith. Whatever may be said about its correctness there can be little doubt that it achieves the object of explaining the principles in simple terms. Even so, following retirement the jury came back with two questions. The first was: "can we have clarification of the term 'not acting under provocation'" and "does there have to be a certain degree of provocation and does it matter when the provocation took place?"
    3.119    
    In Weller[71] the judge's direction on the objective element of provocation read as follows:

    The second aspect is this. The fact that someone may have lost their self-control as a result of some provocative act cannot by itself be a defence to murder, because if it were it would mean anybody who found it difficult to control their emotions or their temper could kill and then say, "Well, I lost my self-control. I'm not guilty of murder." The law is not that stupid, members of the jury. The law expects people to control their emotions. It expects people to exercise reasonable restraint. Even if you are an unusually excitable person the law expects you to control yourself. So that is why you have the second aspect and that is why the section keeps referring to the role of the jury and what a reasonable person would do. So the law says, right, you the jury, you decide, representing the community as you do, you decide whether the circumstances were such or may have been such as to make the loss of self-control excusable so that you reduce the offence from murder to manslaughter. You apply the appropriate standards of behaviour and again you consider all the circumstances. You of course make allowances for human nature and the power of emotions but you have to consider and decide what society expects of a man like this defendant in his position. If you are sure his behaviour was not a reasonable reaction, if you are sure his behaviour was inexcusable, then the verdict will be one of guilty of murder. If it was or may have been excusable your verdict would be "not guilty of murder but guilty of manslaughter by reason of provocation".
    3.120     The Court of Appeal noted that following retirement it was clear that the jury had difficulty in following the direction on provocation, because they sent a note. The issue on appeal was whether the judge ought to have directed the jury to take into account the defendant's obsessive and jealous nature in determining whether or not the provocation was sufficient to provide such excuse as might reduce the offence from murder to manslaughter. The judge had declined to direct the jury that they should do so, but the Court of Appeal concluded that it was inconceivable that the jury would not have taken those matters into account, since they had not been told to disregard them. The court therefore did not find it necessary to decide whether the judge was right or wrong in declining to tell the jury that they should take those matters into account.

    3.121    
    In Rowland[72] the Court of Appeal referred to Weller, and to the commentary on that case by Professor Ashworth in the Criminal Law Review,[73] which expressed concern at the "evaluative free-for-all" resulting from the speeches of the majority in Smith (Morgan) as applied in Weller, and to the problems set out in Consultation Paper No 173 at paragraphs 1.34-1.53 and 4.16-4.173. The court continued:

    Whatever those concerns, however, until the outcome of the Law Commission's review is embodied in a change in the law, the practical problems remain of providing a good working guide for judges faced with summing up in cases where the defence of provocation is raised in circumstances where no fixed standard of self-control can be provided by the judge against which to measure the defendant's conduct.
    3.122     The court concluded that before speeches the trial judge should discuss with counsel the terms of the appropriate direction and added:

    In this connection, the judge should bear in mind … that, in addition to cases where particular factors clearly have a bearing on the issue, there may be difficult borderline cases, particularly as between mere bad temper or excitability on the one hand and identifiable mental conditions and personality traits on the other. In such cases, after prior discussion with counsel, the judge should be careful to include all potentially relevant factors at the appropriate point in his summing up to the jury.
    3.123    
    In serious or complex cases it is always wise for the judge to identify with counsel before summing up any issues or points of law which may be in doubt. However, the basic principles of provocation ought to be sufficiently clear so that there should be no need for counsel and the judge to embark on a difficult exercise discussing how the law should be put and how the jury should be directed, in particular, regarding the defendant's personality traits.

    3.124    
    We also think that it is not satisfactory that the jury should be asked to decide what are the legal standards which differentiate murder from manslaughter. That should be a matter of law on which the jury should be given a clear direction. Our proposal does so.

    3.125    
    Alternatives have been suggested along the lines proposed by the New South Wales Law Reform Commission,[74] which would require the jury to consider whether:

    Taking into account all the characteristics and circumstances of the accused, he or she should be excused for having so far lost self-control as to have formed an intention to kill or inflict grievous bodily harm … as to warrant the reduction of murder to manslaughter.[75]
    3.126     We invited consultees' views on this option. It had some distinguished supporters, but the majority were against it. It provides the jury with no yardstick by which to decide whether the offence was murder or manslaughter. As Professor Alan Norrie put it:

    Option B really does subjectivise the law, or at least it seriously understates the complementary objective requirement.
    3.127    
    The test under our proposal is not whether the defendant's conduct was reasonable, but whether it was conduct which a person of ordinary temperament might have been driven to commit (not a bigot or a person with an unusually short fuse). We believe that a jury would be able to grasp and apply this idea in a common-sense way. Because the test is not whether the defendant's conduct was reasonable, there is no illogicality in providing only a partial defence.

    3.128    
    It has also been suggested that the test should be whether a person of ordinary temperament in the circumstances of the defendant would (rather than might) have reacted in the same or similar way. The difficulty is that if a test including the word "would" were strictly applied, it would be near to impossible, because even under extreme provocation killing is not the probable reaction of a person of ordinary temperament.[76] We can see the counter argument that, read literally, a test including the word "might" could include almost every homicide, since the possibility always exists, however remote, that a person might completely take leave of his or her senses. We do not believe, however, that a jury would understand the test in that way.

    3.129     It has been argued by some that in applying the objective test the court should take into account not only the defendant's actual age but also their mental age.

    3.130    
    In the case of those whom the law regards as adults, although we recognise the logic of that argument, we do not support it for policy reasons. From our discussions with psychiatrists, we understand that mental age is a complex subject. People's cognitive and emotional development is infinitely variable. Many people who kill are emotionally immature. To introduce a test which would lead to psychiatric and psychological evidence about the particular intellectual and emotional development of a defendant, in order to provide some kind of bench mark by which to judge whether his or her conduct was that of a person of ordinary temperament for such level of development, would be complicated and would go a significant way to undermining the objective test. A person who is a psychopath or suffers from retarded development of mind may be eligible for a defence of diminished responsibility. We do not think that factors of that nature should be taken into account in adjusting the objective test for the purposes of the defence of provocation.

    3.131    
    Children who kill present special problems. Their position is special by reason of their youth, and the criminal law must take proper account of this both in its substance and in its procedures. Children mature at different rates and it is frequently the case that children who kill come from significantly disturbed backgrounds. In the case of children who commit homicide, a Consultation Paper on tariffs in murder cases issued by the Sentencing Advisory Panel in November 2001 recognised that:

    What does appear to be a common factor among these young offenders is that they tend to come from seriously dysfunctional families, many are the victims of abuse, and they are often themselves seriously disturbed.
    3.132    
    This is reinforced by comments which we have received from psychiatrists in response to Consultation Paper No 173. Dr. Eileen Vizard wrote:

    [T]here is a very robust evidence base which shows that children who kill are very much more disturbed than ordinary child delinquents. These children are co-morbid (have several psychiatric disorders at the same time) [have] many serious psychiatric disorders and abusive past experiences which impair their emotional and cognitive development and their moral judgement. Children who murder are, by any standards 'unreasonable', they are also seriously disturbed and developmentally immature.
    3.133    
    A recent report by the Royal College of Psychiatrists on Child Defendants[77] contains, and refers to, important material on this subject. It states

    Intelligence is a somewhat blurred concept, consisting of many different facets. In children intellectual development is a changing, dynamic process which is affected by other aspects of the child's development and which can be helped or hampered by environmental and other factors in the child's life … . Improved cognitive or thinking capacities are only one aspect of the maturational and learning processes which need to occur to turn the naturally impulsive, self-centred, short-term thinking toddler into a reasonably self-controlled, reflective young adult, able to take a long-term view… .
    The child with general learning disabilities (mental retardation) functions overall at a lower mental age. However, there are also difficulties from the psychological perspective in the casual use of the term "mental age", although this can be a useful legal concept. Whatever the cause of the child's disability, its effect is usually to give uneven, superimposed selective deficits. The result may be to leave the individual with misleading islands of ability that may encourage the interviewer to see the child as more competent than is the case….
    The acquisition of self-control will overlap with the development of other characteristics. For example, understanding the emotional consequences of their actions does much to shape the behaviour of the normally developing young child and adolescent. Another important aspect of emotional development is the capacity of children to monitor their own behaviour and to alter it accordingly. Gaining self-control, deferring impulses, recognising the impact of behaviour on others and monitoring one's own behavioural patterns are also part of the process of developing insight into how one's own mind works, how one's behaviour affects relationships with other people and how to make the right choices about behaviour in the future.
    These aspects of emotional development are highly relevant in the assessment of child defendants, many of whom are emotionally immature, impulsive and lacking in insight into the impact of their criminal behaviour on others … .
    Another important child–related aspect of social development is moral development and this is of particular relevance to children facing criminal charges, where certain assumptions are made about moral understanding. Moral development is also a crucial issue to be addressed within clinical psychological assessment and also within a psychiatric assessment of the child defendant. The moral development of children is a complex issue which has implications for both the understanding of the seriousness of the offence and the presence or absence of any subsequent empathy for the victim or remorse for the crime.
    3.134     Permitting the jury to take account of the age of the defendant in assessing provocation allows for the child of normal development (who very seldom kills) but not for the child with significant developmental problems. There is also a problem with the defence of diminished responsibility, which does not include developmental immaturity unless of such a degree as to amount to an "abnormality of mind". We have considered the possibility of proposing special versions of provocation and diminished responsibility for the child with particular difficulties, but there are real problems in the notion of judging such a defendant by the standards of an ordinary child suffering from the particular child's personal disabilities. We consider that there is an imperative need for a review of the law of homicide in relation to child and young person defendants, but this goes beyond our present terms of reference. One possibility would be to introduce in their case a single offence of culpable homicide, but we have not consulted about this.

    Exclusion of considered revenge
    3.135    
    The defence should not be available if the defendant acted in considered desire for revenge. In our provisional conclusions we used the words "premeditated" but it has been suggested to us that "considered" might be a better word and on reflection we are happy to accept that suggestion. A defendant who acts in considered desire for revenge is to be distinguished from a defendant who acts on impulse or in fear or both. There may be borderline cases on the facts, but we think that the distinction is one which a jury would be able to recognise and apply.

    3.136    
    We prefer this approach to the present requirement that the defendant should have acted in sudden and temporary loss of self-control (Duffy).[78] The Duffy test has been troublesome, and those who have commented on our provisional conclusions have largely welcomed its removal. From its earliest years, provocation was limited to a person who acted "on the sudden" (as it is put in some jurisdictions)[79] and not in considered revenge, but we think that Devlin J's test in Duffy[80] was with hindsight an unsatisfactory way of doing so. The rule has operated harshly in so-called "slowburn" cases, where the defendant is acting out of fear rather than a considered desire for revenge. However, attempts to redress the hardship by stretching the requirement for sudden and temporary loss of self-control to include slowburn cases have had undesirable side effects, as illustrated by Baille.[81] In that case the defendant armed himself with a sawn off shot gun and cut throat razor, drove to the home of the deceased and shot him, because he had been supplying drugs to the defendant's teenage sons. His conviction for murder was quashed by the Court of Appeal on the ground that provocation ought to have been left to the jury as a possible defence on the slowburn approach. In our view it was a clear case of considered revenge.

    3.137     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. Without such a principle the extension of the defence to those who kill in fear could be nugatory, because (as the Royal College of Psychiatrists has pointed out) "many mental states that accompany killing also incorporate psychologically both anger and fear"[82], in which case the killing may have an element of retaliation. In practice, we think that a jury would be able to differentiate between a defendant who killed out of fear, although angry at the same time, and a defendant who was not truly killing out of fear for their future safety (or that of the children) but for other reasons.

    Exclusion of self-induced provocation
    3.138     The defence should not be available if the gross provocation relied upon was incited by the defendant for the purpose of providing an excuse to use violence.

    3.139    
    It has been pointed out to us, correctly, that this only deals with self-induced provocation in a narrow sense. In Consultation Paper No 173 we said as follows[83]:

    Self-induced provocation might be taken to refer to two different situations. In its narrower sense it would refer to a situation in which a defendant has formed a premeditated intent to kill or cause grievous bodily harm to the victim and incites provocation by the victim so as to provide an opportunity for attacking him or her. In that situation the "provocation" by the deceased will not in truth have been the cause of the fatal attack, which the defendant already intended. In a broader sense, self-induced provocation could also include a situation in which the defendant exposes himself or herself to the likelihood of provocation and then retaliates by killing the provoker. The conduct which exposes the defendant to the provocation might in itself be morally laudable (for example standing up for a victim of racism in a racially hostile environment), morally neutral or morally evil (for example blackmail).
    We can see strong arguments for a rule of law precluding self-induced provocation in the narrower sense from affording a partial defence to murder, and we can see no good argument to the contrary.
    To exclude from a defence of provocation all forms of conduct which might fall within the broader sense of self-induced provocation would in our view go too far. While there is much to be said, for example, in denying the defence to criminals whose unlawful activities expose them to the risk of provocation by others, we see considerable problems in trying to devise a rule of law which would differentiate satisfactorily between forms of self-induced provocation in the broader sense which should, and which should not, preclude a defence of provocation. The circumstances are too potentially variable for a clear and simple rule.
    We are not putting any particular question to consultees on the topic of self-induced provocation, but we would be interested in any observations by consultees who disagree with our comments on this subject.
    3.140     We had no comments on those observations and we adhere to them. Where an issue arises about self-induced provocation in the broader sense, it would be for the jury to take a common-sense view whether the defendant's conduct met the requirements of the objective test.

    Role of judge and jury
    3.141    
    A judge should not be required to leave the defence of provocation to the jury unless there is evidence on which a reasonable jury, properly directed, could conclude that it might apply.

    3.142    
    If provocation is to be defined by general principles rather than specific categories, this proposal is important. The restoration of this power to the trial judge (which was removed by section 3 of the Homicide Act 1957), coupled with the supervision of the appellate courts, will enable the law to set boundaries in a reasoned, sensitive and nuanced way, whereas an inflexible statutory formula would have no room for development.

    3.143    
    Consider, for example, the decision of the High Court of Australia in the leading case of Stingel.[84] The defendant stalked a former girlfriend. She obtained a court order preventing him from approaching her, but he ignored it. After a party he found her (according to his account) in a car with another man having sex. He was sworn at and told where to go. He fetched a knife from his car and killed the man. The judge withdrew the defence of provocation from the jury and the High Court upheld his decision. In Smith (Morgan)[85] Lord Hoffmann, agreeing with the decision, said:

    Male possessiveness and jealousy should not today be an acceptable reason for loss of self-control leading to homicide, whether inflicted upon the woman herself or her new lover. In Australia the judge was able to give effect to this policy by withdrawing the issue from the jury. But section 3 prevents an English judge from doing so.[86]
    3.144     Under our approach provocation should not be left to the jury in such a case because we do not see how any reasonable jury, properly directed, could conclude there had been gross provocation or that a person of ordinary tolerance and self-restraint might have acted in the same way as the defendant.

    3.145    
    Unfortunately our empirical evidence about the success rate of provocation defences before juries is not as good as our empirical evidence about the success rate of diminished responsibility. The evidence which we have tends to suggest that juries are less prone than is sometimes thought to return verdicts of manslaughter on grounds of provocation where the provocation alleged is simple separation or infidelity, but in our view such cases ought not to be left to the jury. To leave such a case to the jury would imply that a properly directed jury could reasonably conclude that a person of ordinary tolerance and self-restraint might respond to such a situation by killing the other person. We do not believe this to be so. More than fifty years ago in Holmes[87] Lord Simon said that Othello would be guilty of murder, even if Iago's insinuations had been true, and we think that this should be so.

    3.146     There are also cases in which a defendant relies on additional taunts or insults. Any study of the history of the defence of provocation shows that it has changed as public values have changed, and that the change of social attitudes is a gradual process. Public opinion should not necessarily decide what the law should be, for public opinion may not be carefully thought out and the law may itself help to shape public opinion, but it should properly be taken into account. As part of our research we commissioned Professor Mitchell (Professor of Criminal Law and Criminal Justice at Coventry University) to interview a small group of individuals drawn from various parts of the country, who reflected a wide cross-section of backgrounds and personal circumstances, and a subgroup of next-of-kin of victims of unlawful homicide (contacted through the organisation Support After Murder and Manslaughter (SAMM)). Interviews were conducted with 62 respondents (including 15 SAMM respondents) and the interviews lasted on average for 1 hour 15 minutes.[88] The sample was small but nevertheless provided an interesting pointer towards public attitudes. The interviews were recorded and contemporaneous notes made. Interviewees were given scenarios of various homicides and questioned about their assessment of the seriousness of each scenario. One scenario involved a husband who killed his wife because she had been having an affair. A variant involved the husband being taunted by the deceased about his sexual inadequacy when he confronted her about the affair. Just over half the respondents thought that this lessened the gravity of the crime, giving as their reason that the husband reacted spontaneously to the taunt. Interestingly, there was no significant difference between the replies of male and female respondents.

    3.147     It is a sad commonplace that when relationships break up there are often arguments and mutual recriminations. We think that it would be seldom that words spoken in such a situation could legitimately make the other party feel severely wronged, to the extent that a person of ordinary tolerance and self-restraint in such a situation might have used lethal violence; but there may be cases where one party torments another with remarks of an exceptionally abusive kind or where one party's behaviour puts quite exceptional emotional pressure on the other. Unless the law is reduced to a formula which removes any evaluative function from the judge and jury (which we would not favour) there are bound to be borderline cases.

    3.148    
    Examples of other cases which under our approach ought not to be left to the jury are Baille,[89] Doughty[90] and Dryden.[91]

    3.149     We return to the question, discussed above, whether our approach to provocation is too vague. We think that to attempt to set the bounds of provocation within narrowly defined categories, for example, limiting it to cases of unlawful violence, would be too restrictive and that too inflexible an approach would rapidly prove to be unsatisfactory.

    3.150    
    Our approach has been to seek to set out broad principles, to rely on the judge to exercise a judgement whether a reasonable jury could regard the case as falling within those principles and then to rely on the jury to exercise its good sense and fairness in applying them.

    3.151    
    This will place significant responsibilities on the judge, but his or her decision whether or not to withdraw provocation from the jury will be given after argument from the prosecution and the defence and will require to be supported by reasons.

    3.152    
    Nobody pretends that this is an easy area of the law, but we believe that this scheme would be workable and it would be a real improvement on the present law.

    Accident or mistake
    3.153    
    In an earlier draft of our proposals, which we circulated for informal discussion, we referred to "grossly provocative words or conduct or what the defendant [reasonably] believed to be grossly provocative words or conduct". This provoked an interesting and difficult debate with a number of academics about whether the word "reasonably" should be included.

    3.154    
    However, we have come to the view that it is not necessary to include any such provision. We have been influenced in this by Professor Ashworth, who commented:

    I feel that the reasonableness requirement is out of place when we are thinking of people who are acting out of fear or anger and are therefore likely to be in a somewhat disturbed emotional state. Moreover, we have managed without a reasonableness requirement for mistake in provocation cases for almost a hundred years and probably longer.
    3.155    
    Our research confirms that this has indeed been one part of the law of provocation which appears not to have been problematical and we do not wish to create problems. At common law, where there has been something which amounts to provocation, but the defendant made a mistake as to who was responsible for it, the defendant has been given the benefit of the mistake. (This was the case in Brown (1776),[92] where the defendant mistakenly believed that the victim was part of a violent street mob.) We would expect the same to apply if, for example, through poor sight or poor hearing, the defendant believed mistakenly that he was being insulted or attacked. East in his Pleas of the Crown (1803) stated:[93]

    Yet still if the party killing had reasonable grounds for believing that the person slain had a felonious design against him, and under that supposition killed him; although it should afterwards appear that there was no such design, it will only be manslaughter, or even misadventure; according to the degree of caution used, and the probable grounds for such belief.
    3.156     The passage is not entirely clear, with its reference to "manslaughter or even misadventure according to the degree of caution used and the probable ground for such belief." East referred to the case of Brown, commenting that "the circumstances being such as might reasonably have induced him to believe that the deceased was one of the [attackers], it was still but the same degree of offence", although it does not appear from the report of Brown that the reasonableness of the defendant's belief was an issue in the case.

    3.157    
    In Letenock[94] the defendant, a soldier, was convicted of murder after drunkenly killing a corporal. His defence was that he acted under the mistaken belief that the corporal was about to attack him. The judge directed the jury that his drunkenness was irrelevant unless he was so drunk as to be incapable of knowing what he was doing. The Court of Criminal Appeal quashed his conviction for murder and substituted a verdict of manslaughter. Lord Reading CJ said that:

    The only element of doubt in the case is whether or not there was anything which might have caused the applicant, in his drunken condition, to believe that he was going to be struck.[95]
    This suggests that the question is whether there was any intelligible basis for the defendant's belief. If so, the defendant is entitled to be judged on the facts as he believed them to be, whether or not his belief was reasonable.[96]
    3.158     We would not expect or intend provocation to be available as a defence in a case where the defendant had no intelligible basis for believing in the supposed provocation. Professor Mackay's study on the diminished responsibility plea in operation has shown a number of cases where defendants have killed under a paranoid delusion about threatening or insulting conduct by the victim. In such cases we think that diminished responsibility, not provocation, is the appropriate defence.

    3.159    
    The common law has also adopted a merciful position towards a defendant who in response to provocation attempted to attack the provoker, but by mistake hit the wrong target. Such a defendant was treated in the same way as if he or she had hit the intended target.[97]

    3.160     We think that the way in which the courts have dealt with accident and mistake in relation to provocation has been sound. There is no need, therefore, to supplement the common law in these cases.

    Duress
    3.161    
    The Law Commission has in the past considered and made recommendations about duress as a defence to murder,[98] but these have not yet been accepted. We have not consulted again on duress as part of this project.

    3.162     We therefore exclude from our proposals a defendant who kills or takes part in the killing of another person under duress of threats by a third person. We wish to make it clear that this does not represent a policy judgement that such a person should not be entitled to a defence or partial defence to murder. On the contrary, we have in the past advocated that duress should be available as a defence to murder. If, for example, a terrorist hijacks a motorcar and forces the driver to drive at gunpoint to a place where the driver knows that the terrorist intends to carry out a murder, and the terrorist does so, under English law both the terrorist and the driver are guilty of murder.[99] There is a strong case for arguing that this is unjust and that the driver should either have a complete defence or be guilty of a lesser offence. We believe that the matter needs to be considered, but any further consideration of the subject would fall within a wider review of murder.

    Excessive force in self-defence
    3.163     Since our proposal for provocation is that it should be recast in a way which would include (subject to safeguards) excessive force in self-defence, we do not propose a separate partial defence of that kind. This subject is discussed more fully in Part 4.

    Merger of provocation and diminished responsibility into a single defence
    3.164    
    The proposal by Professors Mitchell and Mackay for a merger of these defences (discussed in Consultation Paper No 173 at paras. 12.77-12.81) has stimulated a lively debate in recent issues of the Criminal Law Review.[100]

    3.165     It attracted a small amount of support from consultees, but a far greater number were opposed to it. These included the Royal College of Psychiatrists who wrote that:

    [W]e … agree emphatically with your conclusion in your paragraph 61 [of your provisional conclusions paper], that the defences of provocation and diminished responsibility should not be merged into a single defence. Diminished responsibility and its underlying concepts of mental abnormality are complex enough already. We also think that provocation is probably a concept that the jury can understand relatively easily but we are not impressed that the complexities of mental abnormality are easily understood by jurors.
    3.166    
    A merger of the two defences would not be compatible with our present thinking about the way in which the defence of provocation should be reshaped, and we do not recommend it.

    Burden of proof
    3.167    
    In Consultation Paper No 173 we raised the question whether the prosecution should continue to bear the legal burden of disproving the defence of provocation. Of those respondents who addressed this issue, the majority thought that the burden should remain on the prosecution. We agree with this view, particularly in the light of two changes which we are recommending. One is that the judge should not leave the issue to the jury unless there is material on which a properly directed jury could reasonably conclude that the defence was available. The other is the extension of provocation to include cases where a defendant acts in fear of serious violence. If the burden is on the prosecution to disprove a full defence of self-defence, it seems fair in principle that the same should apply to the partial defence. Moreover, since the jury will in many cases have to consider both the full defence and the partial defence, we think that it would make for unnecessary complexity, as well as possible unfairness, if the jury were to be given different directions on the burden of proof in relation to the two defences.

    Recommendations
    3.168    
    We recommend that the defence of provocation should be reformed in accordance with the following principles:

    1) Unlawful homicide that would otherwise be murder should instead be manslaughter if
    the defendant acted in response to
    (a) 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
    (b) fear of serious violence towards the defendant or another; or
    (c) a combination of (a) and (b); and
    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 a similar way.
    2) In deciding whether a person of ordinary temperament in the circumstances of the defendant might have acted in the same or 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) The partial defence should not apply to a defendant who kills or takes part in the killing of another person under duress of threats by a third person.[101]
    6) 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.

    Ý
    Ü   Þ

Note 1    Partial Defences to Murder Consultation Paper No 173 (2003).    [Back]

Note 2    See also Consultation Paper No 173 paras 1.24 - 1.32.    [Back]

Note 3    Consultation Paper No 173, para 12.5.    [Back]

Note 4    Partial Defences to Murder Provisional Conclusions on Consultation Paper No 173.    [Back]

Note 5    Ibid, para 58.    [Back]

Note 6    We discuss the need for a broader review in Part 2 and do not repeat the arguments here.    [Back]

Note 7    [2001] 1 AC 146.    [Back]

Note 8    Para 1.23.    [Back]

Note 9    Dr Horder is among those to whom we are particularly grateful for his help in the course of this project.    [Back]

Note 10    Paras 12.10 to 12.20.    [Back]

Note 11    Historically English law distinguished justifiable homicide from excusable homicide, but the practical significance of the distinction diminished in 1828 on the abolition of forfeiture of a defendant’s possessions to the Crown, which had previously applied in the case of excusable but not justifiable homicide. In modern scholarship a good deal has been written about the concepts of justifactory and excusatory defences. Essentially, justifactory defences are those which recognise that the conduct was legitimate in the circumstances e.g. self-defence. Excusatory defences involve recognition that although the conduct was not legitimate, the actor lacked personal culpability for some reason or another, e.g. because of a disability.     [Back]

Note 12    Consultation Paper No 173 para 12.19.    [Back]

Note 13    Paras 4.8 – 4.11.    [Back]

Note 14    Paras 4.15 – 4.28.    [Back]

Note 15    (1727) 2 Ld Raym 1485; 92 ER 465.    [Back]

Note 16    [1949] 1 All ER 932.    [Back]

Note 17    Ibid.    [Back]

Note 18    [1969] 2 AC 130.    [Back]

Note 19    Ibid, at pp 137-138.    [Back]

Note 20    Paras 4.29 – 4.150.    [Back]

Note 21    [2001] 1 AC 146.    [Back]

Note 22    According to this approach, it is for the jury to decide whether the circumstances were such as to make the defendant’s lost of self-control sufficiently excusable to reduce the gravity of the offence from murder to manslaughter, applying the appropriate standards and deciding what degree of self-control everyone is entitled to expect that his or her fellow citizens will exercise in society as it is today, but taking into account such characteristics of the defendant (but not defects of character) which affected the degree of control which society could reasonable have expected of him or her and which it would be unjust not to take into account.    [Back]

Note 23    New South Wales Law Reform Commission, Partial Defences to Murder: Provocation and Infanticide: Report 83(1997) at para 2.81. This report proposed a test whether “taking into account all the characteristics and circumstances of the accused, he or she should be excused for having so far lost self-control as to have formed an intention to kill or inflict grievous bodily harm … as to warrant the reduction from murder to manslaughter.”     [Back]

Note 24    Under this approach the test should be whether a person with ordinary powers of self-control might have acted as the defendant did. In assessing the gravity of the provocation, the jury is to take the defendant as he or she is. They must then consider whether that degree of provocation might have caused a person of the defendant’s age, and with the powers of self- control of an ordinary person, to have reacted as the defendant did.    [Back]

Note 25    Para 12.26.    [Back]

Note 26    Paras 12.23 –12.25.    [Back]

Note 27    The topic of defence disclosure has been addressed in section 33 of the Criminal Justice Act 2003.    [Back]

Note 28    Modern Criminal Code Officers Committee of the Standing Committee of Attorneys-General, Discussion Paper, Draft Model Criminal Code: Chapter 5, Fatal offences Against the Person (1998).    [Back]

Note 29    Some Criminal Defences with Particular Referenceto Battered Defendants: Report 73 (May 2001).    [Back]

Note 30    [2001] 1 AC 146.    [Back]

Note 31    Referred to as “the MPC”.    [Back]

Note 32    In places our discussion relates to extreme emotional disturbance (EED). The difference between these two expressions is that the former includes extreme “mental” disturbance. The terms are sometimes used interchangeably by commentators on the Model Penal Code.    [Back]

Note 33    Para 12.37.    [Back]

Note 34    In a recent article James Chalmers has suggested that under US case law EED has not, in substance, proved to be an alternative to loss of self-control but a restatement in different guise – “Merging Provocation and Diminished Responsibility: Some Reasons for Scepticism” [2004] Crim LR 198, at p 204.    [Back]

Note 35    This paper is included in Appendix F to this Report.    [Back]

Note 36    Of some 34 jurisdictions that revised their criminal codes in the post MPC era none adopted the MPC proposal as a whole, although 5 adopted it almost whole, omitting the term “mental”. Those 5 states were Arizona, Arkansas, Connecticut, Kentucky and New York. About a dozen other states adopted some of the Code’s features but only with significant alterations, either explicitly requiring a provocative act or rejecting the subjectivity of “the actor’s situation” standard, or both, and in some other ways.    [Back]

Note 37    “Passion’s Progress: Model Law Reform and the Provocation Defense” (1997) 106 Yale LJ 1331.    [Back]

Note 38    Ibid, at pp 1390-91.    [Back]

Note 39    Ibid, at p 1394.    [Back]

Note 40    Ibid, at p 1372.    [Back]

Note 41    Ibid, at pp 1332-33.    [Back]

Note 42    Paras 18–59.    [Back]

Note 43    We have discussed whether “extreme emotional disturbance” should be the test in paras 3.47 to 3.59. We discuss whether “abnormality of mind” for the purpose of the defence of diminished responsibility should include, for example, people who are physically and mentally exhausted by over work or lack of sleep, or distracted by shock or grief, in paras 5.72-5.75.    [Back]

Note 44    Provocation and Responsibility (1992) p 144.    [Back]

Note 45    Paras 3.141 – 3.152.    [Back]

Note 46    Paras 3.153 – 3.160.    [Back]

Note 47    Section 245.    [Back]

Note 48    Para 1.67.    [Back]

Note 49    Association of Women Barristers; Justice for Women (but who prefer the majority view inSmith (Morgan)); Refuge (but who would like to see inclusion of a partial defence comprising realistic fear of serious harm or death coupled with extreme emotional distress); Rights of Women (but who would like to have a definition of domestic violence enshrined in English law).    [Back]

Note 50    When a BBC radio programme recently invited listeners to send in their suggestions for a new law on any subject, the proposal which received the greatest number of votes was to legitimise any use of force by householders towards burglars. See further http://www.bbc.co.uk/radio4/today/.    [Back]

Note 51    Para 12.25(3).    [Back]

Note 52    One consultee suggested that “gross” might be misinterpreted by a jury as meaning crude rather than grave.    [Back]

Note 53    Reproduced at para 3.66 above.    [Back]

Note 54    Para 3.151.     [Back]

Note 55    [2001] 1 AC 146, 168.    [Back]

Note 56    Colonel David Whitaker.    [Back]

Note 57    Sunday Telegraph, 9 May 2004, at p 20.    [Back]

Note 58    Currently subject to appeal.    [Back]

Note 59    Court of Appeal (Criminal Division) 3 December 1980, No. 4795/R/80.    [Back]

Note 60    Textbook of Criminal Law (2nd ed 1983) p 540.    [Back]

Note 61    [1978] AC 705.    [Back]

Note 62    [2001] 1 AC 146.    [Back]

Note 63    Ibid.    [Back]

Note 64    Ibid, at p 213.    [Back]

Note 65    See paras 3.141 – 3.152.    [Back]

Note 66    [1997] 1 Cr App R 199.    [Back]

Note 67    Ibid, at p 207.    [Back]

Note 68    [2001] 1 AC 146.    [Back]

Note 69    Ibid, at p 169.    [Back]

Note 70    [2003] EWCA Crim 677.    [Back]

Note 71    [2004] 1 Cr App R 1.    [Back]

Note 72    [2003] EWCA Crim 3636.    [Back]

Note 73    [2003] Crim LR 724, at pp 725-727.    [Back]

Note 74    Partial Defences to Murder: Provocation and Infanticide: Report 83 (1997).    [Back]

Note 75    Ibid, at para 2.82    [Back]

Note 76    There is also the point made by Professor Glanville Williams that since the law’s reasonable person is an invention, there is no sense expressing the test in terms of statistical probability Textbook of Criminal Law (2nd ed 1983) p 537.    [Back]

Note 77    Royal College of Psychiatrists, Occasional Paper OP 56, July 2004. (As yet unpublished.)    [Back]

Note 78    [1949] 1 All ER 932.    [Back]

Note 79    See for example, Canadian Criminal Code, s.232(2); Northern Territory of Australia Criminal Code, s.34 (2)(c).    [Back]

Note 80    [1949] 1 All ER 932.    [Back]

Note 81    [1995] Crim LR 739.     [Back]

Note 82    See para 3.99.    [Back]

Note 83    Paras 12.59 - 12.62.    [Back]

Note 84    (1990) 171 CLR 312.    [Back]

Note 85    [2001] 1 AC 146.    [Back]

Note 86    Ibid, at p 169.    [Back]

Note 87    [1946] AC 588, 598.    [Back]

Note 88    Professor Mitchell’s report “A brief empirical survey of public opinion” set out in Appendix C.    [Back]

Note 89    [1995] Crim LR 739.     [Back]

Note 90    [1986] 83 Cr App R 319. The defendant killed a crying baby. There would be a strong case for arguing that the defendant should be guilty of manslaughter rather than murder if he acted without intent to kill (as would be the case under the law of Scotland). But under our approach a crying baby could not be regarded as “provocation”.    [Back]

Note 91    [1995] 4 All ER 987. The defendant killed a court official enforcing a court order. The court official acted in a perfectly proper manner and was only doing his duty. Under our approach his behaviour could not be regarded as provocation.    [Back]

Note 92    1 Leach 148, 168 ER 177.    [Back]

Note 93    Vol. 1 p 273.    [Back]

Note 94    (1917) 12 Cr App R 221    [Back]

Note 95    Ibid, at p 224.    [Back]

Note 96    The law relating to self-induced intoxication as applied to defences to crimes of specific intent is controversial. See O’Grady [1987] QB 995; O’Connor [1991] Crim LR 135; McAuley, “The Intoxication Defence in Criminal Law” (1997) 32 Irish Jurist 243; Simester and Sullivan, Criminal Law Theory and Doctrine (2000) p 561; Smith and Hogan, Criminal Law (10th ed 2002) pp 246-247. The Law Commission is reviewing the general principles of criminal law in relation to intoxication and its bearing on criminal liability as part of its work on a criminal code and will be publishing a consultation paper in 2005.    [Back]

Note 97    Gross (1913) 23 Cox CC 455; Porritt [1961] 1 WLR 1372.    [Back]

Note 98    Most recently in Consultation Paper No 218 (1993) Offences Against the Person and General Principles, pp 48-64 and 104 –107 (clauses 25 and 26 of the draft Bill).    [Back]

Note 99    This is the effect of R v Howe [1987] 1 AC 417.    [Back]

Note 100    Mackay and Mitchell “Provoking Diminished Responsibility: Two Pleas Merging into One?” [2003] Crim LR 745; Chalmers “Merging Provocation and Diminished Responsibility: Some Reasons for Scepticism” [2004] Crim LR 198; Gardner and Macklem “No Provocation Without Responsibility: A Reply to Mackay and Mitchell” [2004] Crim LR 213; Mackay and Mitchell “Replacing Provocation: More on a Combined Plea” [2004] Crim LR 219, and Susan Edwards “Abolishing Provocation and Reforming Self-Defence – the Law Commission’s Options for Reform” [2004] Crim LR 181.    [Back]

Note 101    Pending a wider review of the law of murder. See paras 3.161-3.162.    [Back]

Ý
Ü   Þ


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/other/EWLC/2004/290(3).html